Due to the flow-through nature of QCM-D experiments, changing experimental variables such as solution pH, ionic strength, and valence is facile and many of the studies have primarily focused on studying the effect of solution chemistry and other environmental variables on the aggregation and deposition of NPs on SLB coated surfaces (Chen and Bothun, 2013). Yi and Chen (2013) investigated the effect of solution chemistry on the deposition kinetics of carboxylated MWCNTs on zwitterionic 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) SLBs. They showed that MWCNTs do not deposit on DOPC in presence of NaCl due to repulsive hydration forces between the carboxyl groups of the nanotubes and the SLB; however, in presence of CaCl₂, favorable deposition was achieved. The authors related this favorable deposition to the bridging effect of Ca²⁺ ions. They also investigated the interaction of MWCNTs with lipid vesicles by forming SVLs on gold coated QCM-D sensors. By means of cryogenic transmission electron microscopy (cryo-TEM) and SVL studies, they concluded that MWCNT attachment did not compromise the integrity of the lipid bilayers.

In a similar work, Liu and Chen (2015) also investigated the effect of solution chemistry on the interaction of GO, another carbon nanomaterial, with DOPC SLBs and SVLs. They showed that GO deposits on DOPC SLBs in presence of NaCl and CaCl₂ and similar to MWCNTs, presence of Ca²⁺ ions resulted in a larger mass and faster deposition kinetics due to their bridging effect. Similar to MWCNTs, no disruption of SVLs was detected; however, dye leakage assay proved pore formation in the presence of GO nanosheets. Thus, QCM-D is not sensitive enough to detect small liquid leakage due to reversible pore formation and complementary techniques such as fluorescence microscopy and dye leakage assays are necessary to confirm the integrity of the lipid bilayers.

In an early study, Zhang and Yang (2011) investigated the effect of Ca²⁺ ion and net electric charge of lipids on the interaction of anionic and cationic CdSe quantum dots (QDs). Using QCM-D, they concluded that the interaction of these QDs with SLBs are purely electrostatic in nature and can be controlled by solution chemistry, such as pH and presence of Ca²⁺ ions. One particular observation was that the QDs favorably deposit on SLBs within a pH window where they are electrostatically attractive to each other. They also showed that the range of the favorable pH window can be controlled by the presence of Ca²⁺ ions.

A clearer picture on the effect of various cations and anions on the interaction of NPs with SLBs is given by Jing et al. (2014). They probed the interaction of semi-hydrophobic carboxylated polystyrene latex NPs with zwitterionic L-α-phosphatidylcholine (α-PC) SLBs using QCM-D. In another study by the same researchers, these NPs had been shown to induce pores in α-PC SLBs (Jing and Zhu, 2011). The degree of hole formation was investigated using fluorescence microscopy (Figure 1D). It was concluded that the degree of SLB-NP interactions, as demonstrated by the amount of hole formation by NPs, follows the Hofmeister series (the order by which ions cause protein precipitation from solution) for various anions such as CH₃COO⁻, Cl⁻, NO3-, and SCN⁻, while for cations such as Cs⁺, Rb⁺, Na⁺, N(CH3)4+ such order is not observed (Jing et al., 2014).

Once they interact with their surrounding environments, most NPs are transformed by processes such as dissolution, adsorption, oxidation, and reduction, occasionally leading to drastic changes of the properties of the “pristine” original NPs. Natural organic matter (NOM)—natural macromolecules that are the by-products of decay—often covers the surface of NPs once they are introduced into aquatic or soil environments (Lowry et al., 2012). In the same fashion, once they enter the body, NPs are covered by proteins, lipids and other biomolecules that are readily available in the bloodstream or extra-cellular matrix to form a corona on the NP surface (Pearson et al., 2014). The environmental and biological transformations of NPs can completely change the NP identity, and hence, its interaction with cell membranes. QCM-D is a powerful tool to probe such interactions and in this section, recent studies on this topic are reviewed.

Bailey et al. (2015) investigated the effect of NOM on the interaction of gold NPs with zwitterionic α-PC SLBs. Polymethylmethacrylate (PMMA) was used as a model NOM. They showed that while bare NPs barely interacted with the bilayers, those coated with PMMA readily deposited on the SLBs, hence they likely may be more cytotoxic in presence of NOM. They also concluded that small PMMA-coated NPs could translocate in the space between the lipid leaflets; however, the QCM-D frequency and dissipation shift data are often not sufficient to express the spatial distribution of NPs and other complementary techniques such as fluorescence microscopy or AFM is required for a confident determination of the location of NPs (Chen and Bothun, 2013).

Recent studies have focused more on the biological transformation of NPs and their effect on their interaction with SLBs. Liu and Chen (2016) investigated the effect of phosphate ions, a prevalent species in biological fluids, on the interaction of silica, ceria and alumina NPs with DOPC SLBs using QCM-D. Although phosphate ions resulted in the favorable attachment of nano-sized silica to SLBs, favorable attachment could not be observed for the other NPs due to phosphate binding to the NP surface.

The effect of protein corona has also been investigated using carboxylated polystyrene latex and unmodified silica NPs (Lesniak et al., 2013). While polystyrene NPs were shown to deposit on 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) SLBs, coating them with a fetal bovine serum corona resulted in the absence of any interaction with the same SLB. Protein corona is believed to have significantly reduced the surface energy of the polystyrene NPs. Human serum albumin coated silver NPs were also shown to have much lower interaction with DOPC SLBs in contrast to their bare counterparts (Wang et al., 2016). The protein corona was shown to further stabilize the NPs by imparting more negative surface charge and also providing more steric hindrance due to the existence of large protein molecules.

The versatility of the QCM-D sensor surface provides a considerable advantage as it makes it possible to successfully form complex lipid structures. Incorporation of lipopolysaccharides (LPS) and phase segregated lipid rafts are among these complex structures. Jacobson et al. (2015) investigated the effect of LPS content on the interaction of anionic and cationic gold NPs with POPC SLBs. The outer membrane of Gram negative bacteria contains LPS; hence, incorporating it with POPC will increase the resemblance of the model membrane to that of Gram-negative bacteria. They found that cationic gold NP deposition increased by increasing the LPS content of the bilayer, while this was not the case for anionic gold NPs.

In an effort to better replicate the complex structure of mammalian cells, Melby et al. (2016) formed a lipid raft system incorporating highly ordered domains of sphingomyelin (SM) and cholesterol (Chol.) in a DOPC SLB using QCM-D (Figure 1C). Since the molecular packing and order of these small domains are different from that of the SLB matrix, they often have different responses toward biomolecules and NPs, hence, their role as gateways in cellular membranes. Although, these domains possess the same electrostatic charge as the matrix SLBs, cationic gold NPs were shown by means of QCM-D and AFM to preferentially deposit on the phase segregated domains due to their structural differences from the SLB matrix. The preferential attachment of NPs to the phase segregated domains could be used as a method for rational assembly of NPs into ordered structures, e.g., for targeted delivery applications. The assembly of NPs into a layered structure using QCM-D was also demonstrated for a GO and cationic SLB system for potential sensor applications (Frost et al., 2012).

Controlling and characterizing the interfacial interaction of SLBs with their underlying substrates is a new QCM-D based technique developed in our laboratory (Yousefi et al., 2016). This method is based on the electrostatic interaction between a charged SLB and substrate, whereby the surface charge of the substrate can be tuned by changing medium pH, resulting in controlled electrostatic interaction between the substrate and SLB. The degree of interfacial interaction can be monitored by changes in the QCM-D dissipation signal; a bilayer that is more weakly coupled to the substrate tends to dissipate more oscillatory energy, which is manifested by larger QCM-D dissipation shifts. We also developed a method for probing the integrity of bilayers upon their exposure to cationic and anionic polystyrene latex NPs; while anionic NPs did not impart any damage to the negatively charged SLB, cationic NPs disrupted the bilayers in a kinetic manner (Figure 1E). We showed that the gradual loss of response to pH modulation (that results in reversible changes in the degree of interfacial interactions) could be used as a powerful tool to probe the integrity of SLBs upon their exposure to NPs. This technique paves the way for development of sensors that can monitor bilayer integrity in situ. Using this technique, it was also shown that the degree of interfacial interaction between the SLB and its underlying substrate influences the bilayer response toward NPs, where weaker interfacial interactions results in faster disruption of the bilayers (Figure 1F).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer CD and handling Editor declared their shared affiliation, and the handling Editor states that the process nevertheless met the standards of a fair and objective review.