Investigation of sex-based differences in the immunotoxicity of silver nanoparticles

Brandon Canup; Paul Rogers; Angel Paredes; Wimolnut Manheng; Beverly Lyn-Cook; Tariq Fahmi

doi:10.1080/17435390.2024.2323070

. Author manuscript; available in PMC: 2025 Sep 19.

Published in final edited form as: Nanotoxicology. 2024 Mar 5;18(2):134–159. doi: 10.1080/17435390.2024.2323070

Investigation of sex-based differences in the immunotoxicity of silver nanoparticles

Brandon Canup ^a, Paul Rogers ^b, Angel Paredes ^c, Wimolnut Manheng ^d, Beverly Lyn-Cook ^a, Tariq Fahmi ^a

PMCID: PMC12445140 NIHMSID: NIHMS2110169 PMID: 38444264

Abstract

The growing application of silver nanoparticles (AgNPs) in consumer, healthcare, and industrial products has raised concern over potential health implications due to increasing exposure. The evaluation of the immune response to nanomaterials is one of the key criteria to assess their biocompatibility. There are well-recognized sex-based differences in innate and adaptive immune responses. However, there is limited information available using human models. The aim was to investigate the potential sex-based differences in immune functions after exposure to AgNPs using human peripheral blood mononuclear cells (PBMCs) and plasma from healthy donors. These functions include inflammasome activation, cytokine expression, leukocyte proliferation, chemotaxis, plasma coagulation, and complement activation. AgNPs were characterized by dynamic light scattering and transmission electron microscopy. Inflammasome activation by AgNPs was measured after 6- and 24-hours incubations. AgNPs-induced inflammasome activation was significantly higher in the females, especially for the 6-hour exposure. No sex-based differences were observed for Ag ions controls. Younger donors exhibited significantly more inflammasome activation than older donors after 24-hours exposure. IL-10 was significantly suppressed in males and females after exposure. AgNPs suppressed leukocyte proliferation similarly in males and females. No chemoattractant effects, no alterations in plasma coagulation, or activation of the complement were observed after AgNPs exposure. In conclusion, the results highlight that there are distinct sex-based differences in inflammasome activation after exposure to AgNPs in human PBMCs. The results highlight the importance of considering sex-based differences in inflammasome activation induced by exposure to AgNPs in any future biocompatibility assessment for products containing AgNPs.

Keywords: Inflammasome, Immunotoxicity, silver nanoparticles, sex-based difference, peripheral blood mononuclear cells

Introduction

Silver nanoparticles (AgNPs) have been used in various consumer, healthcare, and industrial products due mostly to their anti-microbial properties (Royce et al., 2014, Ferdous and Nemmar, 2020). Among these products, AgNPs have been used in food containers, toothbrushes, bandages, and dental implants, which present a higher risk of translocation of AgNPs and Ag ions into the body (von Goetz et al., 2013, Mackevica et al., 2017, Ninan et al., 2020). With the growing application and exposure to AgNPs, there is a need to evaluate the potential safety hazards associated with short-term and long-term exposures, which have been reviewed and highlighted in various reports (Xu et al., 2020). For example, AgNPs have been reported to cause DNA damage, oxidative stress, and inflammation (Kim et al., 2009, Khan et al., 2019, Nallanthighal et al., 2017). The cytotoxicity and proinflammatory properties of AgNPs have been attributed to the dissociation into Ag ions, which has been reported to be directly correlated with the size of AgNPs, with smaller sizes dissociating more than larger sizes (Mishra et al., 2016, Zhang et al., 2011).

Sex-based differences in the innate and adaptive immunity have garnered increased importance due to the observed differences in genetic, inflammatory, autoimmune, and infectious diseases (Zychlinsky Scharff et al., 2019, Chamekh and Casimir, 2019, Pratap et al., 2020, Zhang et al., 2021a). The sex hormones estrogen and testosterone have been associated with some of the observed differences in immune response (Luczak and Leinwand, 2009, Kovats, 2015). Emerging evidence has also indicated sex-based differences in immune response to nanomaterials. After intraperitoneal injection, gold nanoparticles caused higher liver toxicities and upregulated red and white blood cell counts in male mice compared to female mice while female mice showed an increased spleen index and markers for kidney damage compared to male mice (Chen et al., 2013). After oropharyngeal aspiration, nickel nanoparticles induced more lung inflammation in male mice than in female mice (You et al., 2020). Another study showed that male and female mice presented different immune response patterns after short term and chronic exposure, including differences in cytokine production, neutrophil recruitment, and lung pathologies, when exposed to multiwalled carbon nanotubes and crystalline silica (Ray and Holian, 2019). These reports of sex-based differences in immune response to different nanomaterials emphasize the importance of further investigations into other commonly used nanomaterials, such as AgNPs.

To date, the majority of investigations of sex-based differences in the biological response to AgNPs have been performed in mouse and rat models. AgNPs can adversely affect both male and female reproductive systems through alteration in spermatogenesis in male mice and endometrial receptivity in female mice (Fathi et al., 2019, Ajdary et al., 2021). Male and female rats presented slight liver toxicity after an oral treatment with 60 nm AgNPs, while more accumulation of the particles was observed in the female kidneys (Kim et al., 2008). Kidney and lung accumulation of 20 nm AgNPs was observed more in female mice after intravenous injection (Xue et al., 2012). Daily gavage of rats with 10, 75, and 110 nm AgNPs showed that smaller sized AgNPs absorbed more readily into tissues with the female rats showing more accumulation in the kidney, liver, jejunum, and colon than the male rats (Boudreau et al., 2016). Reactive oxygen species (ROS) were found to be higher in the liver and lung of female mice and glutathione was found to be higher in the lung and brain of male mice upon exposure to 10 nm AgNPs (Tariba Lovaković et al., 2021). Oral treatment with AgNPs highlighted that there was similar hepatotoxicity in male and female mice (Heydrnejad et al., 2015). Female rats had more upregulated tight junction mRNA expression and TNF-α than male rats after oral gavage with 18 – 30 nm AgNPs (Orr et al., 2019). Oral gavage with 10, 75, and 110 nm AgNPs in rats showed size dependent sex-based differences in antibacterial activity, bacterial population, and gene expression in the intestinal tract (Williams et al., 2015). An ex vivo human study using excised human ileum tissue exposed to AgNPs showed that males had more upregulation in genes related to epithelial permeability than females, but the trends observed for the chemokines and cytokines were not found to be significant (Gokulan et al., 2020).

The inflammasome is an important part of the innate immune system, which recognizes various pathogen-associated molecular patterns and damage-associated molecular patterns and also promotes the formation of the multi-protein inflammasome complex (Jo et al., 2016). The formation of the inflammasome complex ultimately activates caspase-1 and leads to the maturation and secretion of IL-1β. Inflammasome activity can either play a protective role to promote host defense response or promote various diseases, such as cancer and metabolic and inflammatory disease through dysregulation (Zhen and Zhang, 2019, Sharma and Kanneganti, 2021). Nanoparticles, such as silica and titanium oxide, have been shown to activate the inflammasome through the NOD-like receptor family pyrin domain containing 3 (NLRP3) pathway (Gómez et al., 2017, Kolling et al., 2020). The unintentional inflammasome activation by nanoparticles could potentially result in excessive IL-1β, leading to proinflammatory conditions (Silva et al., 2017). Experimental evaluation of nanoparticle-induced inflammasome activation is generally performed using a two-step process: priming, followed by activation (De Nardo and Latz, 2013). However, depending on the cell model utilized, the necessity of the priming step is still under debate (Gritsenko et al., 2020). Monocytes play an important role as one of the first immune responders to potentially harmful stimuli and AgNPs have been reported to activate the inflammasome in the monocytic cell line THP-1 and peripheral blood mononuclear cells (PBMCs) (Murphy et al., 2016); however, the sex-based differences in inflammasome activation induced by AgNPs has not been evaluated.

Investigating other innate immune functions is important when evaluating nanoparticles. Leukocyte proliferation occurs in response to antigens or mitogens exposure, which is important for host response to pathogens and infections (Datta and Sarvetnick, 2009). Dysregulation of the mitogen-induced leukocyte proliferation processes has been shown with gold and silver nanoparticles, indicating potential immunosuppressive properties (Devanabanda et al., 2016). Chemotaxis is the unidirectional migration of motile cells, such as immune cells, to chemoattractant chemicals or stimuli (Stock and Baker, 2009). Nanoparticles, such as gold nanoparticles, have the potential to influence the motility of immune cells either by presenting chemoattractant or chemorepellent properties (Zhang et al., 2021b). Coagulation, or clotting, occurs during vascular injury and prevents excess bleeding, and dysregulation of this process could have beneficial or adverse effects (Palta et al., 2014). Different formulations of the nanoparticles, such as liposomes, dendrimers, and metal, have been shown to either stimulate coagulation or deplete coagulation factors, thereby hindering coagulant activity (Ilinskaya and Dobrovolskaia, 2013). The complement cascade is activated to protect from infection and inflammation through immune cell recruitment and enhancing phagocytic activity of immune cells (Nesargikar et al., 2012). Similarly, exposure to nanomaterials has been linked to increasing hypersensitivity reactions through the activation of the complement (Szebeni, 2014). The potential of nanoparticles interference in these innate immune functions could potentially lead to dysregulation and potential health risks.

As previously noted, the availability of data from human models are very limited for sex-based differences in immune response to AgNPs. In this study, we aimed to provide ex vivo data to evaluate the effects of AgNPs on immune cell function and compare differences in responses between males and females. We investigated sex-based differences in inflammasome activation, cytokine expression, leukocyte proliferation, chemotaxis, plasma coagulation, and complement activation after exposure to 30 nm AgNPs using human PBMCs and plasmas from healthy donors. The inflammasome activity demonstrated distinct sex- and age-related differences, with the females presenting more inflammasome activation after AgNPs exposures. Ag ion controls did not present a difference in inflammasome activation suggesting that the sex-based differences observed in the inflammasome activation may be due to AgNPs interaction. Leukocyte proliferation was suppressed in females and males after phytohemagglutinin-M (PHA-M)-induced proliferation. AgNPs did not affect the chemotaxis, plasma coagulation, or complement activity. To our knowledge, this is the first report to show a sex-based difference in inflammasome activation by AgNPs.

Materials and methods

Cell culture and plasma

Human acute monocytic leukemia THP-1 cells were purchased from American Type Culture Collection (ATCC, Manassas, VA). Human acute monocytic leukemia cells with deficient NLRP3 activity (THP-1defNLRP3) and human acute monocytic leukemia cells with deficient Caspase-1 activity (THP-1defCaspase-1) cell lines were purchased from InVivoGen (InVivoGen, San Diego, CA). The selective antibiotic Hygromycin B Gold (InVivoGen) was added at 200 µg/mL to every other passage of the deficient THP-1 cells to maintain the selection pressure. PBMCs from 14 male and 15 female healthy donors were obtained from the healthy donor inventory of Precision for Medicine (Frederick, MD) from donors ages 20 – 54 years old and were stored in liquid nitrogen upon receipt. THP-1 and PBMCs were cultured in complete RPMI-1640 media (Sigma-Aldrich, St. Louis, MO) containing 10% v/v fetal bovine serum (FBS; ATCC) and 1% penicillin-streptomycin (ThermoFisher, Waltham, MA). After thawing, the PBMCs were rested for a minimum of 20 hours prior to experimentation. Cells were maintained and all exposures were conducted in a humidified incubator with 5% CO₂ at 37 °C. Fresh frozen pooled plasmas with citrate were obtained from Precision for Medicine, one pooled from 3 male and one pooled from 3 female healthy donors. Frozen plasmas were gently thawed on ice prior to experimentation.

Nanoparticle Characterization

Citrate coated, 30 nm AgNPs (nanoComposix, San Diego, CA) were sonicated for 10 minutes in a water bath sonicator (Branson Ultrasonics 5510, Danbury, CT) at 40 kHz prior to characterization or treatment. The average hydrodynamic size was measured using dynamic light scattering (DLS) and the zeta potential was measured using a Malvern Zetasizer (Malvern Instruments, Malvern, UK). The sonicated AgNPs were diluted by adding 10 µL of the 1 mg/mL AgNPs to 990 µL of distilled water. Diluted samples were placed in a 4-mL polystyrene cuvette (Malvern) for DLS measurement and a 1 mL clear zeta potential cuvette (Malvern) for zeta potential analysis. Data were collected and determined using the Malvern Zetasizer software 7.13. AgNPs were analyzed for their primary size using a JEM-2100F transmission electron microscopy (TEM, Peabody, MA) after loading and drying onto carbon-coated nickel grid (Polysciences, Warrington, PA). The primary sizes were determined using the open-source ImageJ software.

AgNP dose formulations were prepared in 2 mM sodium citrate and used after sonication. For silica nanoparticles experiments, 20 nm NanoXact Silica Nanospheres (nanoComposix) were prepared in molecular grade water and were used for cell treatments following a 10-minute bath sonication.

Transmission electron microscopy and energy-dispersive X-ray spectroscopy

To identify silver within the samples, the cells were pelleted in agarose and then stained and embedded in Durcapan resin (Electron Microscopy Sciences, Hatfield, PA) following the established serial block face embedding procedure (Deerinck et al., 2022) with acetonitrile replacing ethanol through the dehydration series in that protocol. Once embedded, the samples were sectioned between 50–70 nm thick onto 200 mesh copper EM grids using a Leica EM UC6 ultramicrotome (Leica Microsystems, Deerfield, IL). The sections/grids were then carbon-coated on both sides using an Edwards Auto 306 carbon evaporator (Sanborn, NY) and the grids individually examined using a JEOL JEM-2100 electron microscope with an EM-24511SIOD bright field STEM (JEOL, Japan) and Oxford 80 T-MaxN energy-dispersive X-ray (EDS) detector (Oxford Instruments, Santa Barbara, CA.). To quickly determine if silver was present in the sample, the electron beam of the microscope was expanded over the cells being imaged and an EDS spectrum collected over that region. Once the presence of silver was confirmed, the JEM-2100 microscope was then placed in STEM mode (scanning transmission electron microscopy) which persistently scans and rescans the area, digitizing the image each time. Using this mode and persistently scanning the area of interest for 24 hours, the Oxford EDS detector was able to record and map the characteristic X-ray spectra from the elements present at each pixel in the sample using the Oxford Aztec software. The software was then used to compare and overlap the initial electron STEM image with the locations of the identified silver signal from the cells in that image. Specifically, the characteristic kα X-ray signal for silver, located at 22.166 keV, was mapped onto the electron image of the cell being examined.

Inflammasome activation

Inflammasome activation by AgNPs was investigated using THP-1, THP-1defCaspase-1, and THP-1defNLRP3 cells following established protocols (McNeil, 2011). THP-1 cells were plated at 40,000 cells/well with 50 nM phorbol-12-myristate-13-acetate (PMA, Sigma-Aldrich) for 24 hours in a 96-well plate (Corning, Corning, NY). Cells were washed with media and treated with the 2 mM sodium citrate or 1, 10, or 30 µg/mL AgNPs for 24 hours after the PMA stimulation. Supernatants were collected for IL-1β ELISA analysis using an IL-1 beta/IL-1F2 DuoSet^® ELISA kits (RND systems, Minneapolis, MN). Experiments were performed in triplicate for each treatment for each cell line.

For the evaluation of the inflammasome activation by AgNPs in PBMCs, cells were plated at 100,000 cells/well in a 96-well plate (Corning). Cells were primed using 200 ng/mL lipopolysaccharides (LPS, Sigma-Aldrich) for 4 hours. Cells without LPS priming were also prepared to serve as non-primed controls. The cells were then exposed to 2 mM sodium citrate or 1, 10, or 30 µg/mL AgNPs for 6 and 24 hours with and without the LPS priming.

Silica is a potent activator of inflammasomes in PBMCs (Gómez et al., 2017). Thus, positive controls were performed using 30 µg/mL NanoXact Silica Nanospheres to verify the inflammasome activity of the PBMCs. Ag ions were tested by using silver nitrate (AgNO₃, Sigma-Aldrich). A stock of AgNO₃ was prepared in molecular grade water (Corning). PBMCs were incubated with water or 1, 10, or 30 µg/mL Ag ions for 6 and 24 hours. After the exposures, the supernatants were collected for IL-1β analysis by using an IL-1 beta/IL-1F2 DuoSet^® ELISA kits (RND systems). The AgNPs experiments were carried out using 6 donors per sex with n = 3 per treatment. The Ag ion experiments were performed using 3 donors per sex with n = 3 per treatment. Silica nano-sphere positive controls were collected from wells from 5 donors per sex.

The concentration of IL-1β in supernatant was measured following the manufacturer’s procedure. ELISA plates were prepared a day before supernatant collection using Nunc MaxiSorp^™ flat-bottom 96-well plates (ThermoFisher) by incubating with 4 µg/mL capture antibody overnight at room temperature. Plate washing was conducted using filtered phosphate buffered saline (PBS) with 0.05% Tween^® 20 (Sigma-Aldrich). Plates were then blocked with reagent diluent (RND systems) containing 1% bovine serum albumin (BSA) in PBS for 1 hour. Plates were washed, and samples were loaded to plates for incubation at 4 °C overnight. After the overnight incubation, the plates were washed and 200 ng/mL detection antibody was added for 2 hours at room temperature. Plates were washed again and then a working solution of streptavidin-horseradish peroxidase (HRP) was added to each well for 20 minutes at room temperature protected from light. After a final wash of the plates, 100 µL of a 1-Step^™ Ultra TMB-ELISA solution (ThermoFisher) was added to each well for 20 minutes protected from light. The reaction was stopped by the addition of 50 µl of a 2 N sulfuric acid (H₂SO₄). Plates were read on a Synergy 4 microplate reader (BioTek, Winooski, VT) using the Gen5 software 3.11 at 450 nm and reference at 540 nm. Data were fit to a four-parameter logistic (4PL) standard curve to determine the secreted IL-1β levels.

Cell viability

Cell viability was measured using a CellTiter-Glo^® Luminescent Cell Viability Assay (Promega, Madison, WI), following the manufacturer’s instructions. Cell density, media volume, and exposures were performed following the inflammasome activation experiment. THP-1 cells were exposed to the AgNPs for 24 hours, while the PBMCs were exposed to AgNPs for 6 and 24 hours. After the end of the incubation, plates were removed from the cell incubator and allowed to cool to room temperature. The plates were placed on a standard microplate vortex mixer (Fisher Scientific, Waltham, MA) for 2 minutes at 250 x g after the addition of 100 µL of the CellTiter-Glo^® reagent. The plates were allowed to incubate for 10 minutes at room temperature. The plates were then centrifuged in a Centrifuge 5430 R (Eppendorf, Hamburg, Germany) for 20 minutes at 2000 x g to remove possible interference of larger nanoparticles aggregates prior to reading on a Synergy 4 microplate reader. Cell viability was calculated using the equation $% c e l l V i a b i l i t y = (\frac{L u m i n e s c e n s e_{T r e a t e d} - L u m i n e s c e n s e_{B l a n k}}{L u m i n e s c e n s e_{U n t r e a t e d} - L u m i n e s c e n s e_{B l a n k}}) \times 100$ and data was represented as cell viability (% of control). THP-1 data were collected from n = 3 per treatment. PBMC data were collected from 3 donors per sex with n = 3 per treatment.

Cytokine secretion

Evaluation of the secretion of IL-6, TNF-α, and IL-10 from PBMCs after AgNPs exposure was determined following the same procedure for the inflammasome activation using AgNPs with PBMCs. PBMCs were exposed to the AgNPs for 6 hours. Supernatants were collected and stored in a −80 °C freezer. The ELISA evaluation was carried out using the DuoSet^® Human IL-6, DuoSet^® Human TNF-α, and DuoSet^® Human IL-10 (RND systems) following the manufacturer’s protocol similarly to the IL-1 beta/IL-1F2 DuoSet^® ELISA kits used in the inflammasome activation experiments. Data were fit to a 4PL standard curve to determine the secreted IL-6, TNF-α, and IL-10 levels. Data were collected from 6 donors per sex with n = 3 for each treatment.

Leukocyte proliferation

Leukocyte proliferation was measured following the NCL Method ITA-6.3 (Potter et al., 2020). PBMCs were plated at 100,000 cells/well in a 96-well plate. The cells were treated with 2 mM sodium citrate, PBS (negative control), dexamethasone (positive control), or 1, 10, or 30 µg/mL AgNPs for 24 hours. After the 24-hour exposure, the cells were then supplemented with 10 µg/mL phytohemagglutinin-M (PHA-M) for 72 hours. Proliferation was assessed using a CellTiter-Glo^® Luminescent Cell Viability Assay (Promega), following a similar procedure stated in the cell viability experimental section. Plates were centrifuged for 20 minutes at 2000 x g to remove possible interference of larger nanoparticles aggregates prior to reading on a Synergy 4 microplate reader. Data were analyzed using the equation $% P r o l i f e r a t i o n = \frac{(M e a n O D_{T e s t s a m p l e} - M e a n O D_{U n t r e a t e d c e l l s})}{M e a n O D_{U n t r e a t e d c e l l s}}$ Data were normalized to the average VC % Proliferation to determine the fold proliferation induction. Data were collected from 5 donors per sex with n = 3 for each treatment.

Chemotaxis

The chemoattractant property of the AgNPs was investigated following the ASTM E3238–20 protocol with the modification for using the CellTiter-GLO^® Luminescent Cell Viability Assay (ASTM, E3238–20 2020). PBMCs were cultured in starvation media (SM, RPMI-1640 supplemented with 0.2% BSA, 100 U/mL penicillin, and 100 μg/mL streptomycin sulfate) for 18 hours. After measuring the cell viability using a Cellometer^® K2 (Nexcelom, Lawrence, MA), a stock of the cells was prepared at 1 × 10⁶ viable cells/mL by centrifugation and resuspended in SM. The chemotaxis experiment was performed using MultiScreen^®-MIC plate (Millipore, Burlington, MA) consisting of a top filter plate and bottom feeding tray. The feeding tray was loaded with SM with a positive control of 20% FBS, a negative control using PBS, or 1, 10, or 30 µg/mL AgNPs. The filter tray was prepared by adding 50,000 cells in 50 µL into each well. An equivalent volume of the SM was added to the filter tray for the cell-free controls. The apparatus was assembled and placed into a humidified incubator with 5% CO₂ at 37 °C for 4 hours. The filter tray was removed and 100 µL of the CellTiter-GLO^® solution was added to each well. Samples were shaken at 250 x g for 2 minutes followed by a 10-minute incubation at room temperature. The solution from each well was transferred to a white, clear bottom 96-well plate. Plates were centrifuged for 20 minutes at 2000 x g and read using a Synergy 4 microplate reader. Sample fold chemotaxis induction was calculated by $f o l d c h e m o t a x i s i n d u c t i o n = \frac{S a m p l e c h e m o t a x i s}{B a c k g r o u n d c h e m o t a x i s}$ and data were presented as fold chemotaxis induction.

Plasma coagulation

Plasma coagulation, prothrombin time (PT) and activated partial thromboplastin time (aPTT) were measured using the healthy donor pooled plasmas following the NCL Method ITA-12 (Cedrone et al., 2020). In brief, plasmas were incubated with 2 mM sodium citrate or 1, 10, or 30 µg/mL AgNPs for 1 hour at 37 °C. Coagulation times were measured using a BFT-II Analyzer (Siemens, Erlangen, Germany). PT was measured by adding Dade^® Innovin^® reagent (Siemens) to the plasma and recording the coagulation time. aPTT was measured by incubating with Dade^® Actin^® FS Activated PTT reagent (Siemens) and adding calcium chloride solution (Siemens) to initiate the coagulation. The PT and aPTT of normal and abnormal control plasmas (Siemens) were measured before and after measuring the samples to validate the instrument accuracy.

Total complement activation

Complement activity was measured following the NCL Method ITA-5.1 (Neun et al., 2020). Plasma samples (10 µL) were incubated with 10 µL of veronal buffer (Boston Bioproducts, Milford, MA) and 10 µL of test samples consisting of PBS (negative control), cobra venom factor (CVF, Quidel, San Diego, CA), or 1, 10, or 30 µg/mL AgNPs for 30 minutes at 37 °C. Samples were resolved by running on a 4–15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE, BioRad, Hercules, CA). Proteins were transferred from the gels onto 0.2 µm polyvinylidene difluoride (PVDF) membranes contained in the Trans-Blot Turbo Transfer Pack (BioRad) using a Trans-Blot^® Turbo^™ Transfer System (BioRad). Membranes were blocked with 5% nonfat milk in PBS with 0.1% Tween^®−20 (BioRad) for 1 hour. Intermittent washing between steps was performed using PBS with 0.1% Tween^®20. Immunoblotting was conducted using Complement C3 Polyclonal Antibody (Invitrogen, Waltham, MA) at 1:1000 in the blocking buffer for 1.5 hours to visualize the C3 cleavage products (~43 kDa). A secondary antibody incubation was carried out at 1:30,000 in blocking buffer using donkey anti-goat IgG (H + L) Secondary Antibody, HRP (Invitrogen) for 1.5 hours. Membranes were then incubated with Clarity^™ Western ECL Substrate (BioRad) for 5 minutes. Membranes were visualized using a ChemiDoc^™ Touch Imaging System (BioRad). Densitometry analysis of the blots was carried out using ImageJ by normalizing to the vehicle control.

Statistics

The statistical significance of the independent male and female experiments were determined using one-way ANOVA. The relationship between inflammasome activation fold values with independent variables Age, Sex, and Exposure levels across the 6- and 24-hour timepoints was investigated with a random effects model. An advantage of linear mixed effects models is that the mean response is modeled as a combination of subject-specific effects unique to an individual and fixed-effects shared by all individuals. In general, our regression model takes the following form:

I n f l a m m a s o m e fold values = A g e + Sex + T i m e + E x p o s u r e + T i m e * E x p o s u e + E x p o s u r e * T i m e + S e x * T i m e

The analysis of sex-based differences in the model was set up with male sex as the reference group. Age was set as a dichotomous variable with subjects less than 40 years old coded as the reference group. There were four levels of AgNPs exposure to include LPS, 1, 10, and 30 µg/mL AgNPs. These exposure levels were designed with LPS as the reference level. The model checked for interactions between exposure level, sex, and time. The regression coefficient, or estimate, refers to the estimated difference from the designated reference group. Statistical significance was defined at an alpha level of five percent (α ≤ 0.05).

Results

AgNPs Characterization and internalization

The characterization of the AgNPs was performed to determine their hydrodynamic size, zeta potential, and primary size. The hydrodynamic size was 32.80 ± 0.46 nm with a poly dispersion index of 0.134 (Figure 1A). The zeta potential of AgNPs was −25.5 mV (Figure 1B). The AgNPs presented a spherical size by TEM with an average primary size of 28.4 ± 4.9 nm (Figure 1C, D). The size distributions and charge were within the range reported by the supplier. Endotoxin levels were certified by the supplier to be below 2.5 EU/mL. In-house endotoxin measurements were in agreement.

Figure 1. — Physical characteristics and internalization of AgNPs. Hydrodynamic size and zeta potential was measured using a Malvern Zetasizer. (A) The hydrodynamic size was determined using DLS for the intensity and volume size distributions. Data were normalized to the peak size measured. Polydispersity index (PDI) is represented in the graph. (B) Zeta potential measurement was determined to be −25 mV. (C) Representative TEM image obtained using an EM-2100F transmission electron microscopy. (D) Size distribution analysis was performed using ImageJ. (E-H) Representative EDS data collected from male PBMCs and (I-L) representative data collected for female PBMCs exposed to 30 µg/ml AgNPs for 6 hours. (E, I) TEM images (F, J) EDS images of Ag represented in green. (G, K) Merged TEM and Ag EDS images. (H, L) EDS spectra.

Energy-dispersive X-ray spectroscopy (EDS) and electron microscopy is a common technique used to detect and identify the locations of elements in samples observed by electron microscopy. By placing the instrument in STEM mode, images are recorded by scanning the specimen with a small electron probe rather than by a single wide beam. The scanning produces digitized images and with an X-ray detector on the microscope, the locations of elements can be easily mapped onto the recorded images. This is because using EDS in STEM mode records a complete spectrum of all the identifiable elements at each pixel in the image.

Using a single wide flood beam (regular imaging mode) in the microscope condensed over the area of interest of both male and female PBMC samples, the presence of trace amounts of silver was detected using EDS. The EDS spectra from both sets of samples, both male and female, indicated the abundant presence of osmium, uranium, lead, and arsenic which are components of the stains used to prepare the sample for microscopy. Iron is also a component of the staining protocol, but its presence was ignored because its presence could also be attributed by secondary signals generated by the microscope column. Silver was detected but in a trace amount. Because silver is not a component of the microscope, the signal generated from the sample can confidently be attributed to the AgNPs. Regular imaging mode was used to quickly assess the presence of silver and after confirming, STEM/EDS mapping for the elements present in the sample was performed.

Because silver was present only in small amounts, detecting silver in the cells was not definitive because the signal generated from 50–70 nm thick sections was not significant enough to produce a strong indication in the maps for the 1–4 hour EDS scans. These short scans were enough to show the locations of the EM stains on or in the cells but produced only noise for silver over the same areas of the cells. The only indication that silver might be present in these short scans came from a short silver peak that was identified by the software at 2.94 keV but that was too close to the uranium peak at 3.165 keV to be conclusive (left-most red arrow, Figure 1H and L). To boost the signal from these scans, scanning of the sections was performed for 24 hours. Scanning the same area over many iterations for 24 hours produced enough signal to show the locations of silver in both male and female embedded PBMCs. Though very small amounts were detected in the nuclei, most of the silver was located in the cytoplasm (Figure 1. Male: E–G, Female: I–K). Interestingly, the silver appeared to stain the surface of the cell only in the male sample (Figure 1F and G). The green arrow shown in the spectra of both male and female PBMC samples show the location and strength of the characteristic X-ray signal (22.166 keV) for silver that was used to map silver in the electron image. The silver maps show the same outline and features of the cells they were mapped from, indicating a good signal (Figure 1F and G). To rule out the possibility that these mapped features might be coming from secondary sources as accumulated noise, a brief scan of a small region of the sample where silver indicated a strong presence was compared to two background regions where mapping showed very little silver. The spectra of the silver region compared to the two background regions showed a strong silver signal in only the area identified as containing silver in the 24-hour scan (Supplemental Figure 1).

Inflammasome activity

THP-1, THP1-defCaspase 1, and THP1-defNLRP3 are typical models used for evaluating the ability of nanoparticles to induce inflammasome activation through the NLRP3 pathway (McNeil, 2011). The 50 nM PMA-stimulated control THP-1 cells exhibited a concentration-dependent upregulation of IL-1β after 1, 10, and 30 µg/mL AgNPs treatments (Supplemental Figure 2A). THP-1 cells deficient for Caspase 1 or NLRP3 did not exhibit discernable AgNPs-induced inflammasome activation. Significant cell death was observed for the THP-1 control and THP-1defNLRP3 cells with increasing AgNPs concentrations (Supplemental Figure 2B). Notably, the deficient cells had reduced cytotoxicity from the AgNPs treatments when compared to the nondeficient THP-1 cells, with the THP1-defCapsase1 cells exhibiting the least overall cell death with the AgNPs exposures as compared to the other cell lines tested.

Silica particles were used as a nano-based positive control for inflammasome activation to validate the ex vivo inflammasome activation model with the optimized priming conditions and to test the activity of the commercially purchased PBMCs (Supplemental Figure 3). Silica nanoparticles (SiNPs) have been shown to activate the inflammasome significantly after priming in PBMCs (Gómez et al., 2017). Both the male and female PBMCs presented a significant up-regulation of IL-1β secretion after exposure to 30 µg/mL of 20 nm silica nanospheres after 6 and 24 hours (Supplemental Figure 3A and B).

The AgNPs-induced inflammasome activation was evaluated in male and female PBMCs after 6- and 24-hour exposure periods. The ELISA data are depicted as a box and whisker plot to show the distributions of the average IL-1β expression (pg/mL) and the average fold increase of IL-1β for male and female donors (Figure 2). Individual ELISA experiments can be found in Supplemental Figure 4. The one-way ANOVA statistical analysis was performed on each sex independently, comparing to the average LPS priming for the average and fold change of IL-1β to determine if males or females had significant inflammasome activity after the AgNPs treatment. The broad distribution of the average IL-1β levels indicated that there was a high degree of variability between the donors (Figure 2A and C). Due to this variability, a fold change was performed to isolate the contribution of the AgNPs in the inflammasome activation (Figure 2B and D). Of note, the vehicle control and AgNP treatments without priming did not induce any inflammasome activation as indicated by not detectable IL-1β levels (data not shown for the AgNPs treatments without LPS priming). LPS priming controls upregulated IL-1β for all tested PBMCs to different degrees (Figure 2A and 2C). Male PBMCs only showed significant inflammasome activation for the 30 µg/mL AgNPs 6-hour exposure with an average 2.4-fold increase in IL-1β over the priming control (Figure 2B). Inflammasome activity from AgNPs treatments in males was not significant for the 24-hour exposure in the male PBMCs. Female PBMCs had a non-significant average 1.7- and 1.8-fold increase in IL-1β for the 1 and 10 µg/mL AgNPs treatments for the 6-hour exposure, respectively (Figure 2B). There was a significant average 7.3-fold increase in IL-1β for the 30 µg/mL AgNPs treatment for the 6-hour exposure (Figure 2B). Similar to the male PBMCs after 24-hours, AgNPs did not have a significant effect on inflammasome activation in females, with only a 1.4-, 1.1-, and 1.4-fold increase for the 1, 10, and 30 µg/mL AgNPs treatments, respectively (Figure 2D).

Figure 2. — Analysis of inflammasome induction by AgNPs using PBMCs from male and female donors. il-1β ELISA data obtained from the 6 male and 6 female experiments were consolidated to interpret sex-based differences in AgNPs on inflammasome activation in regard to exposure time and sex. PBMCs were primed with 200 ng/mL LPS for 4 hours prior to 6- and 24-hour AgNPs exposure. Supernatants were collected to measure the excreted IL-1β by ELISA. The average IL1-β levels of each individual ELISA are depicted in box-and-whisker plots for comparison. 2 mM sodium citrate is indicated by VC. (A) 6-hour exposure IL1-β (C) 24-hour exposure IL1-β. Data were normalized to the LPS priming controls to determine the effect of the AgNPs on inflammasome activation. (B) 6-hour exposure fold change (D) 24-hour exposure fold change. Data are depicted as the distribution of means for the 6 donor experiments per sex (n = 4 for each treatment). The ‘+’ represents the mean. *p < 0.05, **p < 0.01, and ****p < 0.0001 versus LPS priming controls by one-way ANOVA.

Further analysis of the data was performed by comparing two different age groups within the tested individuals: young (<40 years old) and old (≥40 years old) (Figure 3). Both male age groups exhibited inflammasome activity for the 6-hour 30 µg/mL AgNPs treatment, where there was a non-significant average 2.0- and 2.7-fold increase in inflammasome activity in the young and old males, respectively. The female PBMCs showed an average 1.9-, 2.2-, and 7.2-fold increase in the younger age group and 1.6-, 1.5-, and 7.5-fold increase in the older group in IL-1β for the 1, 10, and 30 µg/mL AgNPs treatments after the 6-hour exposure, respectively (Figure 3A). The young and old female age groups exhibited significant inflammasome activation for the 6-hour exposure to 30 µg/mL AgNPs. For the 24-hour exposure, younger and older male PBMCs did not show any significant inflammasome activation after exposure to AgNPs. The younger female age group continued to have sustained but not significant inflammasome activation with an average 1.5-, 1.4-, and 2.0-fold increase in IL-1β for the 1, 10, and 30 µg/mL AgNPs treatments, respectively, while the older female age group showed a non-significant 1.4-, 0.7-, and 0.9-fold change, respectively (Figure 3B).

Figure 3. — Sex and age analysis of AgNPs induced inflammasome activation in PBMCs from male and female donors. IL-1β data from the 6 male and 6 female donors was divided into distinct age ranges, n = 3 independent experiments for each age group. Data was normalized to the LPS priming controls to determine the effect of the AgNPs on inflammasome activation. (A) 6-hour AgNPs exposure. (B) 24-hours AgNPs exposure. Data are presented as means ± S.D. *p < 0.05 and ***p < 0.0005 versus LPS priming controls by one-way ANOVA.

The baseline IL-1β resulting from LPS priming was higher in males compared to females, especially in the 6-hour experiments (Figure 2A and C). To evaluate if there were significant sex-based differences in inflammasome activation by AgNPs, the data were further analyzed through applying a mixed effects model as described in equation 1 (EQ1). We accounted for between-subject correlation and initially intended to account for within-subject correlation across the repeated measures at the 6- and 24-hour marks but discovered a significant interaction between exposure levels and time as reported by the model’s Type 3 tests for fixed effects (F_3,42 = 17.34, p < 0.0001). Due to this significant interaction, we stratified the model on time, eliminating this variable from the model, creating two separate models one for each timepoint. Therefore, we no longer have a single longitudinal model that accounts for within-subject correlation, but two separate regression models for the fold values. The statistically significant result for each model is reported in Table 1. The 6-hour time point results show that female PBMCs are estimated to have 1.8-fold more inflammasome activation than male PBMCs when controlling for Age and Exposure. There is also a significant difference between 30 µg/mL AgNPs exposure compared to the LPS control baseline at the 6-hour measurement time. Overall, inflammasome activation was reduced after the 24-hour exposure. Female PBMCs were estimated to have 0.4-fold more inflammasome activation than male PBMCs when controlling for Age and Exposure. It was also found that the PBMCs from donors ages <40 have statistically more inflammasome activation than the PBMCs from donors ages ≥40 for the 24-hour exposure.

Table 1.

Statistically modeled results from the fold-change analysis obtained from male and female inflammasome activation by AgNPs.

Independent Factors	Levels	Estimate	p-value
6-Hour Statistical Results
Age Groups^a	≥40	−0.5254	0.9047
Sex^b	Female	1.7985	0.0002
Silver Exposure^c	1 µg/mL AgNP	0.1726	0.7809
	10 µg/mL AgNP	0.2668	0.6650
	30 µg/mL AgNP	3.8516	<.0001
24-Hour Statistical Results
Age Groups^a	≥40	−0.2724	0.0441
Sex^b	Female	0.3658	0.0080
Silver Exposure^c	1 µg/mL AgNP	−0.08921	0.6368
	10 µg/mL AgNP	−0.1841	0.3210
	30 µg/mL AgNP	0.1710	0.3561

Open in a new tab

The <40 years of age category is the baseline comparison group.

Male is the baseline comparison group.

LPS is the baseline comparison exposure.

Statistically significant results are indicated with bold text in the p value column.