Cells are the basic functional unit of living organisms. Unlike plant cells (prokaryotic), animal cells (eukaryotic) do not have enclosing cell walls, and they are surrounded only by cell membranes ^[1][80] (Figure 1a). The cellular membrane is a thin (5–10 nm thickness) and permeable lipid bilayer that controls the flow and movements of ions and molecules between the interior of cells (cytosol) and the extracellular environment ^[2][76]. To retain the structural integrity of cells, a specialized cellular structure is required. This cellular structure, the cytoskeleton, determines the mechanobiological properties of cells (such as stiffness) and influences the shape, division, and functions of cells ^[3][4][81,82]. In addition to cytoskeletal proteins, other cellular components such as the membrane, the nucleus, and the cytoplasm could impact the mechanic of cells to some extent ^[5][21].

A schematic drawing of a eukaryotic cell is illustrated in Figure 1a. The nucleus, the mitochondria, endoplasmic reticulum, Golgi apparatus, and cytoskeleton are the main components of the internal part of a typical eukaryotic cell ^[6][83]. The nucleus of the cell is the largest organelle among subcellular components ^[7][84] and is located within the central region of cells and includes two regions: the internal region containing DNA and proteins and the outer boundary of the nucleus or karyotheca, which is a lipid bilayer similar to the membrane of cells. Regulating the gene expression is the main role of the nucleus, and to some extent, contributes to the cell mechanics ^[8][85]. The cytoplasm of eukaryotic cells includes all the material within the cell and outside the nucleus, such as proteins, protein complexes, and organelles ^[9][86]. Cells are dynamic living systems, and their mechanobiological properties allow them to sense microenvironmental changes and convert stimuli and changes into biological signals ^[5][10][21,87].

The cytoskeleton is made with a complex network of protein fibers and biopolymers embedded in the cytoplasm. In addition to maintaining the integrity of cells, the cytoskeleton provides pathways for molecular motor proteins to shuttle cargo between different regions of cells and generate and transmit cellular forces ^[3][81]. In response to the mechanical changes in their microenvironments, cells can either reinforce their cytoskeleton by polymerizing their structural proteins or fluidize their cytoskeleton to reduce their stiffness. Microtubule (MT), intermediate filament, and actin filament (F-actin) are three major fiber parts of the cytoskeleton ^[11][88] (see Figure 1b–d). MTs (diameter ≈25 nm), composed of two subunits (α and β tubulins), are stiff and hollow structures of the cytoskeleton, radiating outward from the central organelle. Intermediate filaments provide the strength, integrity, and organization of both the cell and nucleus. The intermediate filaments (diameter ≈10 nm) are composed of various proteins known as protofilaments (protein lamin, vimentin, keratin). These proteins are bundled around each other in a rope-like structure to form the final intermediate filaments. Intermediate filaments have a Young’s modulus between 1 and 5 GPa, and their length is between 1 and 3 μm. Intermediate filaments within the cytoplasm act as “stress absorbers” and organizes the position of organelles in cells ^[12][89].

Actin filaments are the main structural component of cytoskeleton, and with the help of non-muscle myosin II proteins, they provide the required forces for the movement and contraction of cells ^[13][14][90,91]. The actin filaments are composed of two different actin chains: F-actin and G-actin, which are twisted around each other. G-actin monomers are polymerized to form stiff F-actin with a modulus elasticity between 1 and 2 GPa ^[2][15][76,92]. The dimeter of F-actin varies from 5 to 9 nm and has a length in the order of ten micrometers. F-actin filaments are linked to each other during cell migration to form branches at a 70-degree angle from the original filament, enabling the cell membrane to protrude outward ^[2][76]. With the aid of non-muscle myosin II, two or more F-actin filaments are bundled in parallel to provide stress fibers. Myosin II is a molecular motor protein that makes F-actin filaments slide past each other to generate forces within cells ^[13][90]. Myosins directly impact cell mechanics, elasticity, cells adhesion, and mechanosensing ^[16][93]. The force generated by myosin is transmitted through focal adhesions, aggregates of cytoplasmic proteins at the inner surface of the membrane, to the interface of the integrin and extracellular matrix, and these forces are considered as traction forces to help cells move forward during cell migration ^[2][12][76,89].

Among these three different components of the cytoskeleton, actin filament plays the most important role in the structural integrity and deformability of the cell. Intermediate filaments are also able to tolerate some reasonable extent of deformations by engaging in shear stress. MTs play an important role in the cytoskeleton stability but contribute less to the mechanical integrity than the two other filaments ^[3][17][18][81,94,95].

Various techniques can be implemented to measure the mechanobiological properties of single cells, such as viscosity and elasticity. The elastic modulus and viscosity modulus are typically used to express the mechanical properties of cells ^[19][11][25,88]. In the elastic modulus, the applied forces are related to cell deformation, while in the viscosity, time-dependent stress relaxation is measured in response to a step displacement ^[5][20][21,96]. Sufficient and controlled forces need to be applied to the cells to measure their mechanical properties. Based on the types of forces, different microrheological tools have been developed to measure mechanical properties. The most used methods for experimental measurements are shown in Figure 2. Classical methods such as Atomic Force Microscopy (AFM) ^[21][97], micropipette aspiration (MA) ^[22][98], optical tweezer (OP) ^[23][99], and magnetic twisting cytometry (MTC) ^[24][100] are preferred because of their high-resolution measurements. However, they are tedious, and the measurements take a long time. With MEMS (micro-electromechanical systems) ^[25][101] and microfluidic devices ^[26][102], mechanobiological properties can be measured at a higher speed, but their resolution is not as high as that of classical methods, and most of them are able only to measure deformability-related parameters, not elastic and viscosity modulus ^[27][103]. To enhance the accuracy of these methods, in parallel to experimental measurements, computational analyses need to be carried out; however, they may impose a level of complexity ^[2][76]. Table 1 shows the limitation and advantages of different techniques. The technique can be chosen based on the type of cells and depending on the specific desired information.

Any changes in the cytoskeletal structure of cells could lead to the alterations of the mechanobiological properties of cells. In order to understand the effects of NPs on the mechanobiological properties of cells, we need to study the physiochemical interactions between NPs and the three main filamentous proteins: intermediate filament, actin filament, and MT. In some studies, the disruption of subcellular structures has been reported due to the NPs uptake; however, the consequences of those changes to fundamental biological processes have been less investigated. The cytoskeleton is responsible for the basic functions of cells: (a) to preserve the morphology of cells, (b) to anchor organelles, (c) to physically connect cells to the microenvironment, (d) to produce internal forces for cells movement, (e) to help cells for division, and (f) endocytosis ^[3][81]. Therefore, any changes in the cytoskeleton organization could induce cellular dysfunction (see Table 2).

Most NPs are thought to penetrate cells through forming vesicles, and these membrane-bound vesicles transport NPs along MT to intracellular compartments. During this process, the NPs might have indirect interactions with cytoskeletal proteins and change their organizations. It is not clear how they interact with those proteins while they are encapsulated inside lysosomes and endosomes ^[66][17]. However, there are some evidence showing that NPs could directly interact with the cytoskeletal proteins. It has been found that carbon nanomaterials enter cells by adhesive interaction, enabling them to freely swim in the cytoplasm and directly interact with the subcellular structures of cells. For example, Lundqvist et al. ^[67][59] found the presence of MT in the protein corona formed around the SiO2 NPs, suggesting the direct NPs–proteins interactions. Direct or indirect interaction with NPs may negatively affect the biological functions ^[68][69][70][137,138,139]. Tian et al. ^[71][140] showed that single-wall carbon nanotubes could enter cells and alter cell morphology by disturbing the actin networks. They observed that these NPs cause an irregular actin network in comparison to untreated cells. Various NPs-related parameters such as the shape, size, surface chemistry, concentration, and incubation time are important in assessing the toxicity of nanomaterials in cytoskeleton. The shape of the NPs can induce different effects on the cytoskeletal structure of cells. It has been shown that unlike silica NPs with small aspect ratios, silica nanorods with large aspect ratios can largely change the organization of the actin filament, particularly in the vicinity of the cell membrane, resulting in serious damages to the cytoskeletal structures ^[72][73][74][141,142,143]. Ibrahim et al. ^[75][144] used different techniques such as SEM, TEM, and immunofluorescence analysis to study the cytoskeletal changes in osteoblast-like cells underexposure of titanium-based orthopedic and dental implants NPs (nano-Tio₂). Smaller particles were found to be more disruptive to the actin and microtubule cytoskeletal network in comparison to larger particles. In another work, Holt et al. ^[76][145] used fluorescence lifetime microscopy to study the interactions of single-wall carbon nanotubes with HeLa cells. They showed that nanotubes preferentially interact with F-actin compared to G-actin and dramatically change their distribution. NPs even could disrupt the MT and actin network at non-toxic concentrations. Liu et al. ^[77][146] showed that bare gold NPs with the size of 20 nm alter the microfilament arrangement of endothelial cells more than NPs with the size of 5 nm. In this study, five types of gold NPs with different sizes and surface coatings were used to determine the viability and cytoskeletal change of endothelial cells. They found that gold NPs do not affect the viability of cells; however, the force balance between intracellular tension and paracellular forces is broken in 20 nm bare gold NPs-treated cells. In another study, the sub-lethal concentration of silver NPs was used to investigate cytoskeletal changes in neural cells ^[78][147]. They found that the percentage of AgNP-treated cells containing inclusions is doubled compared to control cells, indicating a significant disruption of actin filaments.

In vitro alternations in MT and F-actin concentrations and cytoskeletal destabilization have also been observed in cells, particularly under high concentrations of NPs. For example, Ogneva et al. ^[79][148] showed reduced F-actin content in silicon-treated mesenchymal stem cells compared to control cells (Figure 5a). Pisanic et al. ^[80][149] studied the effects of NPs concentrations on neuron cells. They found that by increasing the concentration of metal oxide NPs, the density of actin filaments is reduced, preventing them from getting mature under the stimulation of nerve growth factors. Mironava et al. ^[81][37] revealed that the cellular uptake of gold NPs disrupts actin fibers of human dermal fibroblast cells, and in contrast to the extended actin in control cells, in treated cells, actin filaments are broken and appeared as dotes (Figure 5b). However, no significant changes were found in actin or beta-tubulin protein levels. Choudhury et al. ^[82][150] studied the binding of nanosphere gold NPs to MT in the cell-free systems as well as in human lung carcinoma cells (A549) using Raman measurement, Fourier transform infrared (FTIR), and other imaging techniques. Their findings showed that gold NPs depending on their size and concentration might inhibit the polarization of MT. They also observed that MT networks are damaged and shrunken upon interaction with gold NPs compared to control cells.

NPs might have different affinities to different subcellular structures depending on their physicochemical properties ^[84][85][152,153]. Wen et al. ^[86][154] found that silver NPs tend to bind actin rather than tubules under electrostatic interactions. They used imaging techniques to visualize the organization of actin and tubulin proteins after treating with silver NPs (size 30 nm). They observed that the secondary structures of actins and tubules are changed due to the interaction with NPs, and alpha-helices of both proteins are decreased while their beta-sheets are increased. NPs-induced cytoskeletal changes could also cause significant morphological changes ^[87][88][155,156]. Rasel et al. ^[89][157] observed morphological changes of osteoblast cells after treating with boron nitride NPs, while they do not have adverse effects on the viability and the metabolism of cells. Ali et al. ^[90][158] showed that gold nanorods could change the cytoskeletal structure of oral squamous cell carcinoma. They observed morphological changes in cytoskeleton protrusions (filopodia and lamellipodia) when incubating cells with integrin-targeted gold nanorods. Qin et al. ^[83][151] found that the NPs-treated breast cancer cells have reduced the number and length of filopodia compared to control cells, causing them to lose their adhesion to the extracellular matrix (Figure 5c). Patra et al. ^[91][159] observed that gold NPs damage the cytoskeletal structure and induce profound morphological changes in human carcinoma cells (A549). Subbiah et al. ^[92][160] studied the morphological changes of A549, NIH3T3, and HS-5, and they found that silver NPs may induce changes in the topography of cells lines and treated cells appeared more rounded than untreated cells. Morphological changes could be influenced by concentration or incubation time. Wu et al. ^[93][161] proved that the density of filamentous proteins is reduced by increasing the concentration and exposure time of gold NPs in human aortic endothelial cells, causing topographic changes in the cell surfaces. In another work, Pernodet et al. ^[94][162] found that citrate-gold NPs profoundly affect the cell morphology of human dermal fibroblasts when the concentrations and exposure time are increased. They observed that the density of actin filament decreases in the presence of NPs by extending the exposure time, showing that the actin fibers are depolarized due to the cellular uptake of NPs.

In summary, in order to study the toxicity of nanomaterial in cells, the interaction of nanomaterial with subcellular structures, particularly cytoskeleton, needs to be taken into account. Nanomaterials, even under low concentration due to direct and indirect interactions with filamentous networks of cells, could change the main cellular structure and lead to mechanobiological changes in cells.