Silver nanoparticles’ impact on the gene expression of the cytosolic adaptor MyD-88 and the interferon regulatory factor IRF in the gills and digestive gland of mytilus galloprovincialis

Abstract Silver nanoparticles (AgNPs) have been reported as stressors for the bivalves' immune system at different regulatory levels, impacting the detection step and receptors, and other mediators, as well as effector molecules. However, studies on how AgNPs impact the transmission of signals from receptors and whether they have an effect on mediators and transcription factors are still scarce. This study aims to investigate the effect of 12 hours of in vivo exposure to 100 µg/L of AgNPs on the gene expression of the cytosolic adaptor Myeloid, the differentiation protein 88 (MgMyD88-b), and the interferon regulatory factor (Me4-IRF) in the gills and digestive gland of Mytilus galloprovincialis, before and after blocking two major uptake pathways of nanoparticles (clathrin- and caveolae-mediated endocytosis). The results illustrate a tissue-specific gene expression of the MgMyD88-b and the Me4-IRF in the gills and digestive gland of M. galloprovincialis. In the gills, AgNPs did not significantly impact the expression of the two genes. However, blocking the caveolae-mediated endocytosis decreased the expression of Me4-IRF. However, inhibition of clathrin-mediated endocytosis in the digestive gland recorded a significant decrease in the expression of MgMyD88-b. Overall, the inhibition of the AgNPs’ uptake routes have highlighted their potential interference with the immune response through the studied mediators’ genes, which need to be studied further in future investigations.


Introduction
Bivalve' mollusks, are organisms of a biological, ecological, and economic importance since they could be ideal candidates for critical, basic and applied research investigations in a large number of clades across the whole spectrum of biology and beyond like neuroendocrinology, host-pathogen interactions and symbiosis (Balbi et al. 2020, Destoumieux-Garz on et al. 2020, innate immunity, as well as in environmental biomonitoring (Fern andez Robledo et al. 2019). This current study aims to investigate the impact of nanoparticles as environmental stressors on living organisms and their implications in environmental species like bivalves. Probing nanoparticles (NPs) across different environmental species has reported their potential impact affecting multiple molecular components and signaling pathways across living species, including invertebrates like the marine mussels M. galloprovincialis (Ale et al. 2019, Duroudier et al. 2019a. M. galloprovincialis are abundant in marine environments, and as filter-feeding organisms able to filter a large volumes of water, subsequently could retain a large amount of chemicals and particles, including nanoparticulate entities, have been proposed as target group in nano-ecotoxicology (Moore 2006, Canesi and Corsi 2016, Fern andez Robledo et al. 2019. Transcriptomic and proteomic studies have reported the influence of environmental contaminants on the molecular components of the immune system in species of environmental importance, including M. galloprovincialis and Paracentrotus lividus (Balbi et al. 2014, Pinsino et al. 2015, Châtel and Mouneyrac 2017, Granger et al. 2017, Duroudier et al. 2019a. However, environmental contaminants including polycyclic hydrocarbon, and nanoparticulate entities have been reported to induce pathogen-free chronic inflammation, also called "sterile" inflammation, and tissue damages (De Vico and Carelle 2012, Carella et al. 2015, Boraschi et al. 2017, Bouallegui et al. 2017b, 2018a, Alijagic et al. 2020. Such an inflammatory response was reported to be due to excessive production of reactive oxygen radicals and oxidative stress damages which have been illustrated through lysosomal membrane disruption, anti-oxidative stress enzymatic inhibition/induction, lipid peroxidation, and reduced glutathione depletion (Katsumiti et al. 2015, Châtel and Mouneyrac 2017, Bouallegui et al. 2017b, 2018a. They have also been shown to induce modifications and damages to the histology of tissue structures due to the excessive infiltration of hemocytes into the gills and digestive gland epithelia, as described in De Vico and Carelle (2012), Carella et al. (2015), and Bouallegui et al. (2017bBouallegui et al. ( , 2018a. In addition, modulation of the gene expression of immune response molecules like metalloproteins C1q containing domain proteins, and Toll-like receptors has been recorded (Katsumiti et al. 2015, Auguste et al. 2019, Jimeno-Romero et al. 2019. Toll like receptors are a kind of pattern recognition receptors (PRRs) that could be triggered by pathogens aiming to activate the immune system (Auguste et al. 2019). However, not much information is available about the effects of NPs on mediator molecules that transmit signals from molecules at the external interfaces (i.e., PRRs) to cascading elements aimed to activate immune defenses (Lee and Kim 2007, Kawai and Akira 2010, Rauch et al. 2013, Toubiana et al. 2014, Châtel and Mouneyrac 2017, Ispanixtlahuatl-mer az et al. 2017. Previous studies have demonstrated that inflammatory lesions caused by silver nanoparticles AgNPs interfere with oxidative stress damages (Bhattacharya et al. 2017, Ispanixtlahuatl-mer az et al. 2017, Bouallegui et al. 2017b, 2018a. Aquatic environments commonly reported as final sinks of most chemicals waste made by humans. AgNPs have received special attention due to its exceptional broad spectrum bactericidal properties, relatively low cost of manufacturing AgNPs, unique properties and ability to form diverse nanostructures, and used in a diverse range of consumer products like food storage, coating materials, fabrics and clothing, toothbrushes, and antimicrobial coatings (Yu et al. 2013), aiming to predict the effect of their potential release through wastewater into aquatic environments and which end in living organisms. However, AgNPs toxicity is closely related to the release of Ag þ ions, which are considered very toxic to aquatic species. Although, it is difficult to distinguish the effect of AgNPs from that of Agþ (Ale et al. 2019). Ag þ ions have been commonly considered as the most toxic form of silver in water bodies prior to the focus on AgNPs (Lekamge et al. 2018). The reported estimated silver concentrations are of 0.03-0.1 ng/L in open seas, and of 0.14-1.29 ng/L in coastal regions, while calculated environmental concentrations of AgNPs are from 40 to 320 ng/L in surface waters, and of 2.3 lg/L in marine waters, as reported by Ale et al. (2019). Overall, the experimental design of this study is in line with previous experiments proposing new results that allow us to deepen further understanding and appraisal of AgNPs interactions with mussels' immune system (Bouallegui et al. 2017a(Bouallegui et al. ,b, 2018a. In so doing, our previous results of different exposure times to a single dose (100 mg/L as a commonly reported concentration) (Katsumiti et al. 2015) for different AgNPs sizes, before and after inhibition of endocytic uptake routes, showed a significant cytotoxic effect on immune cells with changes in the percentages of different sub-populations of hemocytes involved, and significant variations in the histopathological indices of the inflammatory response. AgNPs showed an impact on the inflammation morphology and intensity, while redox proteomics showed redox-based changes in the proteome of the gills and the digestive gland. Overall, it was concluded that AgNPs' impact was size-as well as exposure time-dependent with a greater effect caused by the smaller size. Endocytic routes were deeply involved in determining redox-based changes and immune response activation (Bouallegui et al. 2017a(Bouallegui et al. ,b, 2018a. It is worth noting that clathrin-and caveolae-mediated endocytosis are described as receptordependent routes of NP's entry into cells (Santos et al. 2011, Khan et al. 2015, Bouallegui et al. 2018a. In accordance with that, further studies demonstrated that the epithelium of tissues at the external interfaces play a role of oxygen exchange while the sub-epithelial tissues of the digestive gland that have a nutrient extraction role are among the most hemocyte-rich tissues and are enriched with patterns of the innate immune response (e.g., PRRs) Pales Espinosa 2016, Lau et al. 2017). Immune defense factors associated with the surfaces of the epithelial tissues and the freewandering hemocytes in bivalves' open-circulatory system, make these contact surfaces a first checkpoint for any invading microorganisms or any source of danger like waterborne pollutants (Fern andez Robledo et al. 2019). Such enriched epithelia are responsible for activating what is known as peripheral immunity (or also as local immunity), which thereafter has to activate and is involved in adjusting the systemic immune response Pales Espinosa 2016, Wang et al. 2016).
In the current study we aim to highlight how AgNPs could impact the gene expression of the cytosolic adaptor Myeloid differentiation protein 88 (MgMyD88-b) and the interferon regulatory factor (Me4-IRF) as molecular mediators that deliver a signal from interaction sites through the surfaces' receptors like TLRs to the downstream cascading reaction (Jeong andLee 2011, Granger et al. 2017) which may hamper a proper evaluation of the immune response.

Silver nanoparticles' characteristics and preparation
Poly-vinyl-pyrrolidone (PVP)-coated AgNPs (<50 nm, 99.5% trace metal based, 99.1% purity) previously used in Bouallegui et al. (2017a) were used in the current study (See Supplemental material 1 for description of the synthesis method and representative characterization measurements). Briefly, the TEM analysis (TECNAI G20, Ultra-Twin, FSB, Bizerte, Tunisia) showed that AgNPs <50 nm (Ag50) are homogeneously spherical with an approximate primary size of 50 nm and a size distribution with a median size of 41.6 ± 18.82 nm. The XRD pattern recorded on a D8 Advance diffractometer (Bruker, Bizerte) showed the crystalline nature of the AgNPs where the diffraction peaks matched the face centered cubic (fcc) phase of silver. The UV-Vis spectrum (T60; PG-instruments, Leicestershire, UK) of the colloidal AgNPs stock solution were prepared using artificial sea water (ASW [salinity ¼ 35%, pH 8.0]) as previously described in Bouallegui et al. (2017a) and was performed prior to exposure. This clearly confirmed (kmax ¼ 400 nm) that the AgNPs have a homogenous dispersion in aqueous solutions. The AgNP stock solution was mixed several times by inversion and an aliquot removed as a working solution that was sonicated for 15 min in alternating cycles (2 Â 30 sec) in an ultrasonic bath (100 W; 40 KHz; VWR, Strasbourg, France).

Mussels and experimental design
Tissues used in the current study were aliquots previously collected, flash frozen and cryopreserved at À80 C until their use in this molecular analysis (Bouallegui et al. 2017a). The experimental exposures are described as follows: Mytilus galloprovincialis with an average shell length of 75 ± 5 mm were collected from an aquaculture farm located in the Bizerte Lagoon (in northeast Tunisia) and immediately transported to the laboratory and maintained in oxygenated artificial sea water (ASW) (salinity 35%, pH 8) in static tanks under standard conditions (aeration, photoperiod: 12/12 h; T ¼ 16 C) where they were allowed to acclimatize (48 h) changing the water every 12 h before exposure. As previously reported in Bouallegui et al. (2017a), mussels were sorted into groups of ten individuals each (n ¼ 10, exposure rate ¼ 1 mussel/0.5 L ASW/tank). Each group was exposed for 12 h to 100 mg/L of AgNPs < 50 nm (Bouallegui et al. 2018a,b). The groups designed to probe the effect of uptake routes were exposed to pharmaceutical inhibitors prior to the exposure to AgNPs (Katsumiti et al. 2015). For inhibitor-treated groups, effective concentration ranges used were confirmed based on a previous study by Khan et al. (Khan et al. 2015). Groups aimed at assessing clathrin-mediated endocytosis blocking were incubated for 3 h with 100 mM of inhibitor amantadine (Sigma, Steinheim, Germany). They were then placed in AgNP exposure solutions for 12 h (without amantadine and containing 100 mg/L of AgNPs). To block caveolae-mediated endocytosis, selected groups were exposed to 50 mM nystatin/0.05% (v/v) Dimethyl-sulfoxide (DMSO) (Sigma, Steinheim, Germany) for 1 h a priori, with AgNPs being added for another 12 h (Khan et al. 2015). Control groups (n ¼ 10 mussels each) constituted ASW, ASW with inhibitors (each inhibitor apart), DMSO (vehicle), and DMSO with AgNPs. Controls were maintained as comparative standards to normalize any undesirable effects. All exposures were done in triplicate. The mussels were dissected and then the gills and digestive gland were collected from the controls and the exposed groups, flash-frozen and kept at À80 C until they were processed in the molecular analysis.
2.3. Expression of immune-related mRNA in mussels 30 mg of tissue samples (gills and digestive gland) from each experimental group (4 mussels from each group) were used to prepare tissue homogenate using a lysis buffer available from the Qiagen RNeasy mini-kit (Germany). The RNA was then isolated following the manufacturer's instruction protocol (Qiagen RNeasy mini-kit, Germany). The concentration and quality of the RNA was determined by measuring the absorbance at 260/280 nm using a UV-Vis spectrophotometer (T60; PG-instruments, Leicestershire, UK). cDNA was transcribed from the RNA (2 mg) within the reverse transcription step which was followed by the PCR reaction, both carried out in the same tube according to the manufacturer's protocol: QIAGEN One-step RT-PCR kit (Qiagen, Germany), containing a one-step RT-PCR kit enzyme mix (2 ml), a dNTP mix (10 mM), a 5X reaction buffer (containing 12.5 mM of MgCl 2 ), and the specific primer sets and conditions for each immune related mRNA (0.6 mM of each forward and reverse primer/ reaction/tube) for a total volume of 50 ml of PCR reactions (each/PCR tube). The elongation factor-a (EF-1a) specific primer was also amplified and confirmed as an internal control that admitted minimal changes between different exposure samples. The hot start of the RT-PCR program used for the immune related mRNA was initialized with a reverse transcription step (same tube step) for the first strand cDNA. Synthesis was performed at 50 C for 30 min followed by an initial PCR activation step at 95 C for 15 min, followed by 30 amplification cycles at 94 C for 30 s. The annealing temperature for each mRNA was as shown in Table 1, with an extension at 72 C for 1 min followed by a final extension step at 72 C for 10 min in the Applied Biosystems 2720 Thermal Cycler (Thermo-Fisher, USA).
The RT-PCR product was analyzed with 1.5% agarose gel electrophoresis, stained with 3% ethidium bromide, and visualized under ultraviolet light and documented using the Doc Print II system (VILBER LOURMAT, USA). Gene expression results were presented and semi-quantitatively determined from the ratio of band intensity to the internal control (EF-1a) of 4 biological replicates, using the ImageJ analysis program (from the NIH website by Scion Corporation, Frederick, MD). Each assay was carried out in triplicate for each reaction.

Statistical analysis
All assays were performed in triplicate. The gene expression rates were determined from three replicates of mRNA isolated from an average of four animals/treatments (extracts for each tissue). Results are presented as means ± SD of mRNA expression (relevant primer)/expression of EF-1a mRNA expression. The normal distribution and homogeneity of variance were tested using the Shapiro-Wilk and Bartlett tests prior to the statistical analysis. All samples showed a normal distribution. The statistical analysis was performed using a one-way analysis of variance (ANOVA) with a Tukey's HSD post-hoc test and significance determined at Ã p < 0.05 and ÃÃ p < 0.01.

Ethical statements
Permits are not required for the field collection of Mytilus galloprovincialis nor for their use in laboratory testing. All experimental trials have been conducted following the principles of the Declaration of Helsinki.
The authors confirm that all mandatory laboratory health and safety procedures have been complied with during the course of this experimental work.

Results
The exposure to AgNPs has proven to insignificantly modulate the gene expression of MgMyD-88-b and Me4-IRF in the gills. However, the inhibition of clathrin-mediated endocytosis (AgNPs þ Amantadine exposure) did not affect the modulation much. Although the inhibition of the caveolae-mediated endocytosis demonstrated an interference with MgMyD-88-b and significantly decreased the expression of Me4-IRF, while the vehicle (DMSO) did not significantly affect the expression of both genes compared to the control (Figure 1). Moreover, exposure to AgNPs when blocking the caveolae-mediated endocytosis recorded a significant decrease in the expression of Me4-IRF, but not for MgMyD-88-b (Figure 1).
In the digestive gland, exposures to AgNPs significantly increased the gene expression of Me4-IRF. However, the inhibition of the clathrin-mediated endocytosis (AgNPs þ Amantadine) decreased such expression (i.e., Me4-IRF) (Figure 2), and significantly increased the expression of the MgMyD-88-b expression (which did not show significant changes when exposed to AgNPs (i.e., without inhibition)) ( Figure 2).
In the digestive gland of mussels exposed to AgNPs, the gene expression of MgMyD-88-b when caveolae-mediated endocytosis was inhibited (AgNPs þ Nystatin) did not show any significant effect in all exposures. However, the expression of Me4-IRF showed a significant decrease (compared to DMSO) (Figure 2).

Discussion
Silver nanoparticles can exert toxicity through its particle form and through its Ag þ ions released during particles dissolution (Ale et al. 2019). The mechanisms of silver nanoparticles toxicity in mussels have been discussed in studies such as Katsumiti et al. (2015), and Ale et al. (2019). Although dissolution was not measured in this study, another study (Sikder et al. 2018) reports a dissolution rate constant (k) of 0.008 h-1 for a 100 mg Ag/L suspension of PVP-AgNPs in 20 ppt ASW measured over 120 hours (or 2.5 nm which is 13.51% of initial particle diameter; calculated according the graphs presenting data in Sikder et al. (2018)), therefore, the concentration and exposure time was chosen based on the estimated lower dissolution and it allows for comparison with other ecotoxicological studies. The focus on this study was on the NP entities' effects on signaling molecular pathways. However, the effects of the release of ions which is one of the main toxic mechanisms that explain NPs' toxicity, are beyond the objective of this study.
Probing the effects of different types of engineered nanoparticles across a large spectrum of living organisms has developed much over the past decade (Pinsino et al. 2020). Such investigations have largely elucidated the effects of nanoparticles including AgNPs on living organisms at the molecular, cellular and whole organism levels utilizing different biomarkers like oxidative stress induction, immune response and embryo toxicity (Auguste et al. 2018). However, the first cell response after exposure to xenobiotics occurs at the gene transcription level (Duroudier et al. 2019a).
In general, bivalves are among the most studied invertebrate groups as an important target group for NP toxicity (Moore 2006, Canesi and Cors). Many investigations have focused on understanding the effects of AgNPs using conventional biomarkers (Jimeno-Romero et al. 2017, Duroudier et al. 2019a, and other methods like transcriptomics and proteomics (Rocha et al. 2016, Bouallegui et al. 2018b). However, complement information given by such tools still not enough. A major process of NPs to impact on immune response of different living organisms is their ability to mediate the modulation of signaling pathways at different checkpoints (Pinsino et al. 2015, Boraschi et al. 2017, Alijagic et al. 2020. Transcriptomic and proteomic studies on the digestive gland of mussels exposed for 1 and 21 days to 10 lg/L of AgNPs of 5 nm in different seasons, clearly indicating that exposure to AgNPs provoked significant alterations in the transcription levels of genes involved in the cytoskeleton reorganization (e.g., downregulations of the centrosome-associated 350-like isoform X2, calponin-1, and tropomyosin, alterations of paramyosin and actin proteins (Katsumiti et al. 2015, Duroudier et al. 2019a, Duroudier et al. 2019b. The same study documented alterations in cytoskeleton dynamics, which is a key process regulating vesicle and organelle transportation along the intracellular trafficking pathways. In addition, genes like Ras-and Rab-related, which are involved in endosomal trafficking were also found to be significantly Table 1. Specific primers and conditions used for the determination of immune-related mRNA expression.

Primers
Primer sequences (5' to 3') Annealing temp Amplicon size ( altered (Duroudier et al. 2019a). It's worth noting that in bivalves the endo-lysosomal trafficking system is also used as a nutrient digestion mechanism assured by the digestive gland. Subsequently, the alteration of such mechanisms would forcefully impact the hemostasis of exposed organisms. AgNPs have been documented to induce upregulations of the leucine-rich, repeat-containing, and low affinity immunoglobulin epsilon Fc receptor, and the serine threonine-kinase mitochondrial, simultaneously with downregulations of the calcium calmodulin-dependent kinase type IV. They have all been suggested to potentially experience interactions with bivalves' immune system through Toll like receptors (TLR), which may directly interact with immunoglobulin E (IgE) synthesis, and together with the signal input, promote the immune response (Duroudier et al. 2019a).
Proteomic analyses showed that a putative C1q domain containing protein was also significantly altered after exposure of mussels to AgNPs, suggesting that the immune response of mussels was affected by the dietary exposure to AgNPs (Duroudier et al. 2019b). Previous studies (Jones et al. 2020), documented an important role allocated to such a category of molecules (C1q domain proteins) being members of mucosal surfaces covering epithelia of digestive gland and gill tissues, and involved a double-role as a nutrient selection process assured by bivalves as filter-feeder organisms to select suspended particles of nutrients (Jones et al. 2020). Overall, Duroudier et al. (2019b), taking into account the role of the C1q domain containing proteins in the immune response, suggested that the under-expression of the C1q domain proteins after exposure to AgNPs is due to a disrupted cell-mediated immunity (Duroudier et al. 2019b). Further studies documented differential regulations in the transcription levels of lysozyme, Mytilin B, Myticin B, the C1qdomain-containing protein (MgC1q), and the Toll-like receptor (TLR-I) as selected genes related to immune response, and were evaluated by RT-q-PCR in the digestive gland of CeO2NPs-exposed mussels, showing only significant variations in Mytilin B, while non-significant changes were recorded for lysozyme, Myticin B, and MgC1q, and almost no changes were seen in the TLR-I (Auguste et al. 2019). Similarly, Titanium dioxide nanoparticles (TiO 2 NPs) were reported to affect the protein expression of p38MAPK, extracellular receptor kinases (ERK), Toll-like receptor 4-like (TLR-4 like), HSP 70, and interleukin-6 (IL-6), in the urchin immune cells (Pinsino et al. 2015, Alijagic et al. 2020. Interestingly, receptor-ligand proteins like TLRs, act through adaptor molecules and activate various kinases and transcription factors mediating inflammatory response and apoptosis like caspase 8, Nuclear factor-ƦB (Nf-ƦB), JUN, and MAPK14 (Kawai and Akira 2010, Perkins et al. 2018). The current study shows the ability of AgNPs to modulate the expression of MgMyD88-b (significant downregulation in the digestive gland, and slight upregulation in the gills), which highlights the potential impact of AgNPs on mediators' components linking the sensing phase to the action phase. Such findings are in line with previous studies that have demonstrated the ability of metal based pollutants to modulate the expression of TLRs through the MyD88 adaptor (Balbi et al. 2014, Granger et al. 2017. Other studies reported upregulations of the 96-kDa endoplasmic reticulum (ER)-resident glycoprotein (GP96) in sea urchin hemocytes exposed to TiO 2 NPs (Pinsino et al. 2015). It is worth noting that GP96 could interact with TLRs to activate macrophages and is known to be induced by cytokines like interferon-Ç and interleukin-2 (Kawai andAkira 2010, Perkins et al. 2018). However, taking into account this possible activation of the interferon signaling pathway, the modulation of Me-4IRF by AgNPs recorded in our current results could be explained as such. Moreover, Granger et al. (2017) demonstrated the effect of exposure to cadmium on the expression of MyD88, TLR4-like, TLR2-like and TLR3 in the digestive gland from mussels relative to EF-1a in different experimental exposures. Data are presented as Means ± SD. Values significantly different from relevant compared groups at Ã p < 0.05 and ÃÃ p < 0.01, n ¼ 4.
the hemocytes of Mytilus edulis. Our results, have recorded the impact of AgNPs on the gene expression of MgMyD88-b and Me-4IRF, which might be linked to a putative activation of several TLR receptors (Jeong and Lee 2011, Rauch et al. 2013, Turabekova et al. 2014. Further, computational studies suggested as a mechanism of NP interferences with cell components the ability of nanoparticulate entities to associate with intracellular domains of receptors leading to block their dimerization, and/or mediating an unexpected hetero-dimerization and/or homo-dimerization (Turabekova et al. 2014). Blocking appropriate signals from being delivered to cascading mediators could hamper signal transmission (e.g., NPs ability to bind directly to extracellular domains of TLR4 and inhibit its activation) (Turabekova et al. 2014, Ispanixtlahuatl-mer az et al. 2017, Perkins et al. 2018. However, previous studies have demonstrated the ability of AgNPs to modulate the redox equilibrium of the proteome in Mytilus galloprovincialis tissues impacting its immune response (Bouallegui et al. 2018b,c). In this context, Jeong and Lee (2011) demonstrated the ability of sulforaphane, as a species resulting from a redox-modified-equilibrium, to bind directly to cysteine residues of the TLR4 extracellular domain and inhibit its interaction with its relevant ligand, which might generate chronic inflammation (Jeong and Lee 2011).
Throughout this study we have shown that the inhibition of clathrin-and caveolae-mediated endocytosis, considered as NPs uptake pathways, has significantly decreased the expression of MgMyD88-b and Me-4IRF in the gills and digestive gland of the Mediterranean mussel Mytilus galloprovincialis (in most cases). Interestingly, a potential involvement of the MyD88-independent pathway in recruiting specific receptors, such as for example the TLR3 which could be activated within intracellular vesicles formed by the clathrin-and caveolae-dependent endocytosis, both during the internalization of AgNPs, could interfere with the already activated immune response (Hromada-judycka 2014, Zhao et al. 2015, Perkins et al. 2018.

Conclusion
The current study has highlighted the possible interferences of AgNPs with the gene expression of mediator molecules which may alter the transmission signal from receptors at the cell surfaces (described as cellular sensors) aiming to activate the immune response. However, the results of the current study are preliminary and need to be further developed and integrated in other signaling pathways for a better understanding of the effect of NPs on the immune system of living organisms.