Size exclusion chromatography is a separation technique that sorts molecules based on their size. It employs a porous matrix within a column, where smaller molecules enter the pores and elute later, while larger molecules bypass the pores and elute earlier. This method is particularly advantageous for separating protein-nanoparticle complexes from unbound proteins due to its gentle nature, which maintains the integrity of the complexes. After the incubation of the nanoparticles with a biological fluid, the mixture is loaded onto a size exclusion column. The choice of column material and pore size is crucial, as it determines the separation efficiency. For protein corona isolation, columns with a pore size suitable for separating nanoparticle-protein complexes from free proteins are typically used. The sample is eluted through the column using an appropriate buffer. As the sample moves through the column, nanoparticle-protein complexes, being larger, elute first, followed by smaller unbound proteins; the eluted fractions are collected systematically (Cedervall et al., 2007; Böhmert et al., 2020; Kristensen et al., 2021).

This technique takes advantage of the unique properties of magnetic nanoparticles (Sakulkhu et al., 2015; Bonvin et al., 2017). By incorporating magnetic materials into nanoparticles, an external magnetic field can be applied to quickly and selectively separate the nanoparticles and their associated proteins from a complex biological mixture. Nanoparticles are engineered with a magnetic core, typically composed of iron oxide or other magnetic materials. These nanoparticles are then exposed to biological fluids, such as blood serum or plasma, to allow the formation of the protein corona. Once the protein corona has formed, an external magnetic field is applied to the mixture. The magnetic nanoparticles respond to this field, allowing them to be rapidly separated from the non-magnetic components in the solution. The separated nanoparticles, along with their protein corona, are collected for further analysis. This method has also been used for lipid nanoparticles (Francia et al., 2024): Cullis and co-authors prepared iron-oxide-loaded lipid nanoparticles that allow a fast and precise approach for isolating lipid nanoparticle–corona complexes, preserving the integrity of the particles.

Field flow fractionation (FFF) operates on the principle of separating particles based on differences in their size, shape, and mass. Unlike traditional chromatographic methods, FFF does not rely on a stationary phase; instead, it uses a field (such as a cross-flow, thermal, or centrifugal field) perpendicular to the flow of the sample to achieve separation (Tadjiki et al., 2024). The external field pushes the particles towards the channel bottom, while diffusion acts in the opposite direction. Equilibrium is reached when these opposing forces balance out. At this point, particles of different sizes form distinct clouds: smaller particles create more diffuse clouds and move faster through the channel, while larger particles form narrower clouds and move slowly.

Centrifugation is the most used method for protein corona isolation and it is the method we currently use in our laboratory (see Figure 1). In our laboratory we focused on studying the hard corona that coated the nanoparticles: after incubation with serum or plasma, amorphous silica NPs were diluted with ice-cold PBS in polycarbonate tubes (Beckman Coulter, cat. number 355603), then immediately recovered by ultracentrifugation (45 min, 100,000 g at 4°C, XL-70 Ultracentrifuge Beckman, fixed angle 50 Ti rotor). They were subsequently washed twice with 10.5 mL of ice-cold PBS at pH 7.4. In our work (Fedeli et al., 2015) we have also isolated corona proteins using another method, the centrifugation over a sucrose cushion (Tenzer et al., 2013); even with this procedure, the protein pattern of the nanoparticle corona was similar to that obtained with a simple centrifugation of samples.

For poly lipoic-nanoparticles and organic silica NPs, after incubation at 37°C, a centrifugation was sufficient to recover NPs: samples were therefore centrifuged for 30 min at 13,000 rpm at 4°C. This was followed by three consecutive washes with 1 mL of cold PBS, centrifuging the nanoparticles each time at 13,000 rpm for 30 min at 4°C, in order to obtain the hard corona. For poly lipoic-nanoparticles we observed a very low recovery of proteins, resulting in them being almost corona-free (Trzciński et al., 2021).

In each experiment, mock samples are always included, incubating serum or plasma in the absence of NPs, to estimate the nonspecific protein background after centrifugation and washings.

We usually use a protein assay to quantify the proteins bound to the NPs surface. This measurement is essential to determine if the amount of protein bound to the nanoparticle changes depending on factors such as the coating and to ensure that a sufficient quantity of protein has been obtained for subsequent proteomic analyses (and thus, potentially set the experiment with a final reaction volume that is double or triple). It is also crucial in this case to prepare mock samples, using only the nanoparticles without incubation with serum and plasma proteins. It is also important to quantify the proteins in the supernatants obtained from the first centrifugation of the nanoparticles, after incubation with serum or plasma: this allows to estimate how much protein does not bind to the nanoparticles and how much forms the hard corona after the washes. After the first centrifugation step, supernatant must be collected; after the washing steps, NP pellets are resuspended in 25 μL of PBS 1x. We then quantified the total amount of protein by means of Bradford assay: 10 μL of resuspended pellets and collected supernatants were added in duplicate to a 96-well plate together with standard curve samples (prepared with known concentration of Bovine Serum Albumin). A volume of 200 μL of Bradford reagent (Merck) is subsequently added to each well. The absorbance of the samples and standard curve is analyzed using the spectrophotometer at wavelengths around 595 nm. In other studies, the amount of proteins was measured by micro-BCA (Lebreton et al., 2023) or BCA assay (Mohammad-Beigi et al., 2020; Meng et al., 2022). By adsorbing varying amounts of proteins onto their surface to form a protein corona, NPs can undergo steric stabilization to different degrees, leading to aggregation and resulting in a spectrum of color changes. These changes can be quantitatively characterized using UV-vis absorption spectroscopy. The linear working range of this method was found to be wider than that of the commercial Bradford Assay and comparable to the Micro BCA assay (Ho et al., 2015). Bradford is good to quantify a massive protein corona while for low-binding-NPs it is better BCA.

The gold standard technique to specifically identify the composition of the protein corona is mass spectrometry (Blume et al., 2020).

LC-MS/MS is employed for characterizing the protein corona’s composition, which involves identifying the various proteins adsorbed onto NPs using proteomic techniques. This investigation necessitates enzymatic digestion of proteins, performed in solution directly to the NPs-PC. Subsequently, NPs-PC undergoes digestion and LC-MS analysis. Protein identification is facilitated through common proteomics protocols, often involving protein sequencing. While MS can offer quantitative protein data, conventional LC-SI-MS/MS workflow typically yield semi-quantitative results using label-free methods, estimating protein abundances from peptide counts and ion intensities in a single LC-MS/MS analysis. Achieving absolute protein quantities within the corona necessitated specific standardization, a challenge particularly pronounced when analyzing numerous proteins, as in protein corona studies (Fuentes-Cervantes et al., 2023). After incubation in the physiological buffer, using the selected nanoparticle concentration and incubation time, we treat samples as visualized in Figure 6, as described also elsewhere (Ashkarran et al., 2022).

To load peptides we then use pico-frit columns (New Objective, NJ, United States of America), packed with C18 material (Phenomenex, CA, United States of America) followed by separation using an acetonitrile gradient containing 0.1% formic acid. The gradient ranges from 3% to 40% over a period of 45 min, at a flow rate of 250 nL/minute The ion source capillary is maintained at a temperature of 200°C, with a spray voltage between 1.2 and 1.3 kV. By analyzing the mass-to-charge ratio (m/z), individual peptides in the sample can be identified and matched to their respective proteins. For data analysis we use MaxQuant software and peptide identifications are cross-referenced with the UniProt Human and UniProt Mouse databases, depending on the serum source used in the experiment.

Below (Figure 7) is a representative figure of data obtained with this type of analysis (Tavano et al., 2018). Obviously, the shot gun analysis allowed us to more accurately calculate the relative quantities of different proteins in the protein corona of our nanoparticles (see Figure 4, right panel). Most of the proteins present in the protein corona studied via in-gel were comparable to those found through shotgun analysis. However, the shotgun approach enabled us to highlight the presence of complement proteins, various immunoglobulin isotypes, and a broad panel of apolipoproteins, which did not emerge in the in-gel analysis.

From the wide spectrum of mass spectrometry techniques, TMT-MS shows outstanding characteristics (Zhang and Elias, 2017). Tandem mass tags (TMT) are reliable and efficient systems for multiplexed, relative protein quantitation of up to 18 samples from cells, tissues, or biological fluids. Each mass tagging reagent in a label set has the same nominal mass (i.e., they are isobaric) and chemical structure, consisting of a reactive group, a spacer arm (mass normalizer), and a mass reporter. The isobaric nature of TMT labeling reagents enables the simultaneous identification and quantification of proteins in various samples using tandem mass spectrometry. To have a trustful comparison of PCs between different particles, TMT-MS is the gold standard method (Liessi et al., 2021).

In our laboratory, we found out that TMT is better for comparative approaches but that can detect less proteins (Morbidelli, Papini, Tavano, unpublished data). In the case of this approach, all samples are prepared in duplicate, at the same time and under the same experimental conditions to allow a comparison using equal volumes of each preparation. Samples are reduced, alkylated and trypsin digested as specified for the label-free approach, dried under vacuum and suspended in equal volumes of 100 mM tri-ethyl-ammonium bicarbonate (TEAB). Peptides are then labeled with TMT tags (Thermo Fisher Scientific). Before pooling, the labeling efficiency is evaluated for every sample by performing a LC-MS/MS analysis (see Figure 8). The analysis or raw data is carried out against the relative section of the Uniprot database (as specified above). Trypsin is set as an enzyme with up to 2 missed cleavages allowed. Peptide and fragment tolerance is set at 10 ppm and 0.6 Da, respectively. Carbamidomethyl Cys and TMT labeling (N-term and K) are set as fixed modifications, while Met oxidation is set as variable modifications. The intensities of TMT reporter ions are used by the software to calculate the relative abundance of proteins across the different conditions.

The knowledge of the role of single proteins adsorbed to NPs would allow the design of nanoconstructs that, for example, bind proteins protective against cellular toxicity. To examine the specific roles of key human plasma proteins in forming the corona on amorphous silica nanoparticles, we combined purified fractions of human high-density lipoproteins (HDL), high-density lipoproteins (LDL), and Very Low Density Lipoprotein (VLDL) with purified human plasma IgG fractions (comprising IgG1, IgG2, IgG3, and IgG4 classes), Histidine-rich glycoprotein (HRG), Kininogen-1, Fibrinogen, and serum albumin (Fedeli et al., 2015). These proteins were mixed at concentrations reflecting their average levels in 10% human plasma to create a simplified protein medium (HRG 15 μg/mL; IgG 700 μg/mL; HSA 5 mg/mL; HDL 150 μg/mL; LDL 78 μg/mL; VLDL 12 μg/mL; Kin-1 8 μg/mL; Fibr 300 μg/mL). HRG was purified from human plasma, as explained in Fedeli et al. IgG, serum albumin, HDL, LDL, VLDL, kininogen-1, and fibrinogen were purchased by different companies. We observed (see Figure 9) that the protein corona formed on nanoparticles incubated with purified proteins displayed a polypeptide profile closely resembling that seen with human plasma. Notably, the absence of HRG in the corona mix led to an increased association of Kininogen with the nanoparticles. Conversely, the omission of other proteins did not markedly alter the corona composition, which remained predominantly characterized by a high abundance of HRG. This absence likely allows other proteins with lower affinity for the silica surface to bind to the NPs. Using this methodology allowed us to focus on studying the binding dynamics of various protein to the nanoparticle surface, depending on the presence or absence of these proteins in the biological medium.

The process of selectively eliminating specific factors in order to better understand their roles in biological fluids is more compelling than studying the isolated factors. The selective elimination can be achieved through immune depletion using specific antibodies. In our studies about the protein corona of amorphous silica nanoparticles, human plasma was depleted of HRG, the most abundant protein found in the corona protein. The depletion was achieved by passing human plasma through a phosphocellulose resin (P11, Whatman), followed by collection of the flow-through. The collected plasma was then dialyzed against PBS (for 18 h at 4°C) and quantified for protein concentration. In the corona experiments performed with HRG-depleted plasma, similarly treated and dialyzed normal human plasma was used as control, with adjustments made for differences in protein concentration. To prevent coagulation, both plasmas were supplemented with sodium citrate (0.38% v/v) and were immediately used for corona experiments. We found (Fedeli et al., 2015) that in the absence of HRG, the protein corona of amorphous silica nanoparticles was enriched in Fibrinogen (4 times), albumin (2.5 times) and Apo A-I (2 times).

Complement activation induced by nanoparticles is a significant area of research in nanomedicine. The complement system consists of proteins that help to identify and eliminate pathogens. When nanoparticles are introduced into the body, they can trigger this system, leading to various immunological responses. nanoparticles can activate the complement system through multiple mechanisms, primarily dependent on their surface properties (La-Beck et al., 2021). Positively charged nanoparticles are more likely to attracts and bind complement proteins, leading to activation; functional groups on the nanoparticle surface, such as hydroxyl, amine, or carboxyl groups, can bind to complement proteins, triggering activation (Behzadi et al., 2017). The complement system can be activated via three main pathways: classical pathway, lectin pathway and alternative pathway (Moghimi et al., 2011). It is well-known that the activation of the classical and lectin pathways of the complement system is Ca²⁺-dependent, whereas Mg²⁺ is essential for the operation of the alternative pathway (Sahu and Lambris, 2001; Merle et al., 2015). In our study Tavano et al. (2018) we selectively chelated calcium by treating the nanoparticles during incubation with human serum with EGTA (10 mM) and Mg²⁺ (2 mM), inhibiting in this way the classical and lectin pathways; on the other hand, we chelated both calcium and magnesium using EDTA (10 mM), to inhibit the alternative pathway. To study changes in the composition of the corona protein due to complement activation by interaction with the nanoparticle surface, we performed shotgun proteomic experiments by incubating the nanoparticles with control serum or serum in the presence of the specific chelators. The results we obtained with organic silica nanoparticles are depicted in Figure 10: it is evident that the surface of the nanoparticle induces the deposition of C3, which is inhibited by both chelants; simultaneously, the deposition of IgM on the surface increases. The attachment of apolipoproteins to the nanoparticle surface is inhibited when complement activation is blocked, particularly for ApoE, ApoC3, and ApoA1.

Monocytes and macrophages, which are phagocytic cells within the reticuloendothelial system (RES) and mononuclear phagocyte system (MPS) are particularly exposed to nanoparticles following intravenous administration due to their specialized ability to capture particulate matter. The presence of a protein corona on the surface of nanoparticles signifies an intricate interplay between biomacromolecules and NPs, dictating their behavior in terms of pharmacokinetics and pharmacodynamics. As proteins interact with NPs, they induce modifications in their surface characteristics, rendering them susceptible to clearance by Mononuclear phagocyte system (MPS) (Khan et al., 2022). Studying how nanoparticles interact with these cells and how the protein corona influences these interactions is crucial for understanding nanoparticle behavior in biological systems (Yan et al., 2013; Cai et al., 2024; Liu et al., 2024). Nanoparticles that are effectively taken up by these phagocytic cells are likely to be swiftly cleared from circulation and accumulate in key filtering organs such as the liver and spleen (Gustafson et al., 2015). Conversely, nanoparticles that evade phagocytic uptake may have extended circulation times and potentially reach various other tissues.

In our laboratory, we investigated how the protein corona can modulate the association and subsequent uptake of both amorphous silica nanoparticles and organic silica nanoparticles. We conducted these experiments with both human phagocytic cells (such as monocytes and macrophages) and human non-phagocytic cells (like lymphocytes and epithelial cells). We believe that working with primary cells is the best choice for studying in vitro mechanisms that closely resemble what occurs in vivo. Therefore, uptake experiments are conducted with phagocytic cells purified from fresh buffy coats from healthy donors, which are provided to us by the transfusion center of the Hospital of Padova. The buffy coats are processed on the same day they are prepared: through two successive centrifugation gradients (the first on Ficoll and the second on Percoll), we obtain monocytes and lymphocytes (see Figure 11), as described also elsewhere (Repnik et al., 2003; Meital et al., 2019; Domínguez-Andrés et al., 2021; Angioni et al., 2023). Neutrophils, on the other hand, are purified through preliminary dextran sedimentation, followed by centrifugation on a Ficoll-Paque gradient and then lysing of red blood cells through a hypotonic shock.

To study the interaction of amorphous silica nanoparticles with phagocytic cells, the nanoparticles were incubated with differentiated macrophages, cells were then washed, and the fluorescence associated with the cells was analyzed using flow cytometry (see Figure 12). To investigate the importance of the protein corona in cell-nanoparticle interactions, incubations were conducted in the presence of serum or plasma and also with only the HRG protein (Fedeli et al., 2014), the major component of the protein corona of amorphous silica nanoparticles (as shown in Figure 5). We found that NP uptake by macrophages after 3 h was significantly inhibited by the presence of plasma proteins and HRG alone, highlighting again that the protein corona was fundamental for the interaction of amorphous silica NPs with these cells.

Poly (ethylene glycol) (PEG) is a polymer known for its hydrophilic nature and flexibility, which effectively minimizes protein and biomacromolecule adsorption; it has been widely utilized to prolong the circulation duration of protein therapeutics, demonstrating significant success. Additionally, PEG coatings are commonly employed to mitigate nonspecific cellular uptake of nanoparticle drug carriers and other undesired interactions within biological contexts (Yang and Lai, 2017).

Incorporating PEG and PEG-containing copolymers onto nanoparticle surfaces significantly prolongs their blood circulation lifespan by several orders of magnitude. This technique establishes a hydrophilic protective barrier around the nanoparticles, effectively repelling opsonins protein absorption through steric repulsion forces, thereby obstructing and delaying the initial stage of opsonization (Owensiii and Peppas, 2006).

This stealth coating PEG and other polymers prevents adsorption of blood serum proteins, increases circulation time, and enhances the probability of particle permeation into tumor tissue (Mout et al., 2012).

Regardless of the polymer used for the coating, the overall charge of the surface plays a central role in the interaction between nanoparticles and the protein corona. Common terminal groups for polymers are positively charged amines, negatively charged carboxyl groups, a combination of both (zwitterionic), or non-charged methoxy groups. In particular, synthetic zwitterionic materials have been studied as PEG alternatives. In fact, they have a strong hydration rate that confers them stealth properties and low immunogenicity. In particular, NPs coated with amines exhibited quicker and more extensive protein absorption, but lower stability in serum. Carboxyl-PEG and methoxy PEG NPs demonstrated reduced protein corona assembly. Intriguingly, zwitterionic-PEG NPs exhibited the lowest protein absorption. The heightened interactions with methoxy-PEG NPs can be explained by the ability of proteins to interact with non-charged groups through hydrophobic interactions. Additionally, zwitterionic particles demonstrated the highest stability in serum (Rampado et al., 2020).

In the recent decades, an alternative for PEG has been searched to compensate for critical aspects of this molecule and to allow more chemical versatility, useful to add functionalities and modulate the surface features. Several alternatives have been proposed such as polyvinylpyrrolidone (PVP) (Escamilla-Rivera et al., 2019), zwitterions, peptides (Leal et al., 2020) and carbohydrate moieties.

Joh et al. worked on the modification of linear PEG chains in shorter hyperbranched polymers that could decrease the antigenicity of PEG guarantying stealth properties, presenting PEOGMA as an alternative to linear PEG (Joh et al., 2019). Another alternative is poly-glycerols (PGs) polymers. PG-based drugs present correct pharmacokinetics properties and low immunogenicity. However, it was proven that they tend to accumulate in the liver and kidneys. PGs have been shown to be less susceptible to oxidative or thermal stress than PEG (Hoang Thi et al., 2020).

Poly (acrylamide) and poly (methacrylamide) are well known materials already used for biomedical applications. They are biocompatible and make nanosystem stealth, decreasing the clearance. Although these polymers are biocompatible, their monomers are highly toxic. For this reason, they are not suitable as PEG alternatives (D’souza and Shegokar, 2016).

The poly-N-2-Hydroxypropyl-methacrylamide (PHPMA) has demonstrated an excellent preclinical efficacy as a carrier for chemotherapeutic drugs and has recently entered clinical trials. However, studies into the immune response of PHPMA remain elusive (Hoang Thi et al., 2020).

Among these alternatives, polyalkyloxazolines (PAOXA) and polyalkyloxazines (PAOZI) have been proposed by many studies and attract a wide interest (Adams and Schubert, 2007; Hoogenboom, 2009; Viegas et al., 2011). Oxazolines are hydrophilic, tertiary polyamides with the amide group in the side chain, as well as being structural isomers of polypeptides. Three isomers of oxazolines exist depending on the location of the double bond, 2-, 3-, and 4-oxazoline, where 2-oxazoline is by far the most studied. They can be further substituted on the 2, 4, or 5 position of the 2-oxazoline ring, however substitution on the 4th or 5th position is more difficult to polymerize due to steric crowding. POZs have structural similarities to PEG and polypeptides due to having (C-C-N) backbone repeat units and amide side chains. POZ has higher chemical stability due its N-vicinal C–H bond causing a lower polarization compared to PEG which has an O-vicinal C–H bond. Although PEGylated polymers are highly water soluble, amphipathic, very flexible and hydrated in water, non-toxic, and may be produced with low polydispersity, POZ can be further tuned in size and structure using a living cationic ring opening polymerization (CROP) as well as facile surface functionalization and fluorescent labeling via click chemistry thus overcoming PEGs low drug loading capacity.

In order to compare the biocompatibility of different coatings on the surface of organic silica nanoparticles, NPs were synthesized with no coating, a PEG coating, and a poly (2-methyl-2-oxazoline) (PMOXA) coating (Tavano et al., 2018).

The comparison of coatings was conducted by evaluating several parameters: the composition of the protein corona of the three types of nanoparticles (obtained through shotgun analysis), the assessment of complement activation (via ELISA assays), and the study of the uptake of different types of nanoparticles by macrophages.

Shotgun analysis was performed as described in chapter 5.3 (see Figure 13A). We found that the PMOXA coating induces the formation of a protein corona enriched in complement proteins (such as C3, C5, C6, C9, and complement factor B), with respect to uncoated or PEG-coated nanoparticles.

The activation of the complement system in human serum (see Figure 13B) was accessed by measuring the nanoparticle-induced release of complement activation products C5a (a protein fragment released from complement component C5), Bb (the fragment of complement factor B that results from activation of the alternative pathway), and sC5b-9 (the terminal complement complex) by specific ELISA kits, provided by Quidel. In this way we found that the PMOXA coating is responsible for the activation of the complement system and the subsequent release of C5a, Bb, and s-C5b9; instead, the PEG coating did not induce complement activation.

For the uptake experiments (Figure 13C), we preferred to pre-incubate the nanoparticles in 75% serum for 30 min at 37°C, then dilute 1 to 10 in cell medium (RPMI without serum), and finally incubate the nanoparticles with cells, to closely mimic what happens in vivo. We revealed that PMOXA coating results in an uptake of organic silica nanoparticles that is 5 times greater than that of the same uncoated nanoparticles and 2.5 times greater than that of PEG-coated nanoparticles.

To better understand the dependence of PMOXA-coated nanoparticle macrophage uptake on complement activation induced by the nanoparticle surface we used four approaches (Tavano et al., 2018): 1) the role of activation of the classical, lectin, and alternative pathways in NP uptake was studied as described above (chapter 6.3), using EGTA/Mg²⁺ or EDTA during the pre-incubation of nanoparticles with serum and during the incubation with macrophages; 2) the role of individual components of the complement activation cascade in the uptake of PMOXA-coated NPs was studied using commercially available sera depleted of C3, C1q, C4, and B factors, provided by Comptech (see Figure 14B); these depleted sera were used for pre-incubation with the nanoparticles (including in the experiment also a “super control”, where the depleted serum was reconstituted with the specific purified protein), diluted in RPMI (as described in Figure 13C) and then incubate with macrophages; 3) the role of the Ig deposition on the NP surface in the NP uptake was studied using sera depleted of IgG or IgM: IgM depletion was obtained using an anti-human IgM agarose matrix (Sigma); prechilled and PBS-equilibrated agarose beads were mixed one to one with cold human serum at 4°C for 60 min; the agarose beads were eliminated by centrifugation and the IgM depleted serum was recovered and immediately frozen in liquid nitrogen. To recover the bound IgM, beads were treated with 100 mM glycine (pH2.8) for 5 min and then centrifuged; after collecting the supernatants, 1M Tris-HCl was added to restore the physiological pH; then IgM solution was dialyzed overnight at 4°C against PBS. For IgG antibody depletion, human serum was incubated with an equal amount of protein A sepharose beads (Amersham) for 60 min at 4°C; sepharose beads were removed by centrifugation and IgG were recovered as described for IgM; IgG- or IgM depleted sera were used during the pre-incubation of the nanoparticles, then diluted and used for the incubation with macrophages; also in this case we prepare a “super control”, re-pleting serwith IgM or IgG; 4) the role of lectin pathway in the recognition of nanoparticles by macrophages has been studied using two sugars, N-acetyl glucosamine and mannose, which compete with mannose-binding lectin and ficolins, proteins involved in the lectin pathway, for binding to their substrates. The sugars, at a concentration of 25 nM, were used during the pre-incubation of the nanoparticles with the serum.

In this way, we were able to dissect the nanoparticle uptake process. As shown in Figure 14A, PMOXA-coated-NP uptake was completely inhibited in the presence of EGTA/Mg²⁺ and EDTA; the absence of C3 induces a 95% inhibition of PMOXA-coated NP uptake, whereas the absence of C1q, C4, and factor B results in a lesser inhibition (70%–80% depending on the complement factor, but still significant); neither Ig depletion nor lectin pathway blockade are significant in the uptake of PMOXA-coated NPs.

A crucial point we want to emphasize is the importance of using serum from different individual donors, rather than pooled sera, and to perform separate experiments with each donor’s serum. This approach takes in account of the interindividual variations in complement activity (Gaya Da Costa et al., 2018; Lipsa et al., 2023), which can significantly impact the efficiency of nanoparticle recognition and engulfment by phagocytes. By analyzing sera from multiple donors, it is possible to capture a broader spectrum of opsonic activity and better understand how these differences affect nanoparticle-phagocyte interactions. This allows to gain insights into the variability in immune response among different individuals and provides a more robust understanding of the role of complement proteins in nanoparticle uptake (La-Beck et al., 2021). For instance, PEG-coated nanoparticles were differently uptaken by macrophages, depending on the donor’s serum (Tavano et al., 2018). Another critical point is the choice of the source of the serum: in nanomedicine evaluations, murine models and materials derived from mice are commonly used. However, there are significant species-specific differences between humans and these animals, particularly in the innate immune system’s response to particulate matter (Li et al., 2021). Mice and human immune systems have differences that must be taken into account (Mestas and Hughes, 2004). In our experience, for example, PMOXA-coated nanoparticles, which are strongly phagocytosed by human macrophages, are not taken up at all by murine macrophages (Tavano et al., 2018). This highlights the importance of carefully testing the compatibility of nanomaterials with human cells and sera.