**3. Magnetic nanostructures**

#### **3.1. Magnetic nanoparticles**

Magnetic NPs are usually made of a magnetic core bound within a shell that allows them to be functionalized with relevant ligands and gives them stability in solution [66]. Their main advantage is their ability to be manipulated using an external magnetic field, making them attractive for different biomedical applications. These include cell labeling and MRI contrast agents [67, 68], targeted drug delivery [69, 70], and cancer cell eradication [71, 72]. Iron oxide NPs such as magnetite (Fe<sup>3</sup> O4 ) and maghemite (γ-Fe2 O3 ) are the most widely used magnetic NPs [73]. Sufficiently small Fe oxide NPs exhibit superparamagnetism: NPs become magnetized under an external magnetic field, but loose do not possess any remanent magnetization once the field is removed [74]. Superparamagnetic Fe oxide NPs (SPIONs) can be manipulated and guided by a magnetic field without losing the stable colloidal suspension, when there is no field applied, a quality that is attractive for biomedical applications.

#### **3.2. Superparamagnetic iron oxide nanoparticles**

Initial cytotoxicity studies compared the cytotoxic effects of bare magnetite NPs with PEGcoated ones in primary human fibroblast hTERT-BJ1 cells, and found that, whereas cells treated with 40–50 nm PEG-coated NPs for 24 hours remained 100% viable for concentrations as high as 1 mg/mL, uncoated, 10–15 nm NPs reduced the viability to around 70% at a concentration of 250 μg/mL [75]. Similar results were observed with pullulan-coated and uncoated NPs within the same size range [76]. There, it was also shown that bare NPs significantly reduce cell attachment and disrupt the distribution of actin filaments and microtubules, while also being taken up at a higher rate compared to the coated ones. Uncoated NPs were also reported to have cytotoxic effects only at higher doses (100–250 μg/mL) in terms of cell viability and LDH leakage in rat liver BRL 3A cells [53]. In agreement with these findings, hydroxy-tetramethylammonium-coated SPIONs at higher concentrations (23 mM) did not induce a reduction inviability of kidney COS-7 cells, though the time of incubation tested was only of 4 hours [77].

the gene expression of matrix metalloproteinases (MMPs) MMP-3, MMP-11, and MMP-19, which play a key role in extracellular matrix degradation and can be activated through ROS were also reported [60]. Further, another parameter that has been observed is the effect of Ag NPs of different sizes on the differentiation of embryonic stem cells [61]. Differentiation into cardiomyocytes was inhibited in a dose-dependent manner, with Ag NPs of 20 nm having a

In recent studies, it has been proposed that the intracellular release of Ag ions from the NPs is one of the causes of their cytotoxicity. Singh et al. showed that, after being taken up through scavenger receptor-mediated phagocytosis in macrophages, intracellular dissolution of Ag NPs had a 50 times faster rate than in water, at around 5% of the total dose being dissolved [62]. It was suggested that Ag ions are a cytotoxic response initiator in human lung BEAS-2B cells [63]. Ag NW cytotoxicity, on the other hand, has not been extensively studied. In a comparative study of Ag NWs with a diameter of around 100 nm and lengths of 3, 5, 10, 14 and 28 μm, it was found that only the wires with a length of 28 μm could elicit a significant decrease in cell proliferation and membrane instability in THP-1 cells [64]. Using light microscopy and back-scattered electron imaging, it was also proven that NWs of 14 and 28 μm are not properly internalized, resulting in a frustrated phagocytosis, or an inability to engulf its target, which is in turn an initiator of the inflammatory response. In a different approach, red blood cells exposed to Ag NWs of 2 μm in length and a diameter of 40 nm were confirmed to suffer

structural changes, aggregation, and hemolysis in a dose-dependent manner [65].

Magnetic NPs are usually made of a magnetic core bound within a shell that allows them to be functionalized with relevant ligands and gives them stability in solution [66]. Their main advantage is their ability to be manipulated using an external magnetic field, making them attractive for different biomedical applications. These include cell labeling and MRI contrast agents [67, 68], targeted drug delivery [69, 70], and cancer cell eradication [71, 72]. Iron oxide

NPs [73]. Sufficiently small Fe oxide NPs exhibit superparamagnetism: NPs become magnetized under an external magnetic field, but loose do not possess any remanent magnetization once the field is removed [74]. Superparamagnetic Fe oxide NPs (SPIONs) can be manipulated and guided by a magnetic field without losing the stable colloidal suspension, when there is

Initial cytotoxicity studies compared the cytotoxic effects of bare magnetite NPs with PEGcoated ones in primary human fibroblast hTERT-BJ1 cells, and found that, whereas cells treated with 40–50 nm PEG-coated NPs for 24 hours remained 100% viable for concentrations

O3

) are the most widely used magnetic

) and maghemite (γ-Fe2

no field applied, a quality that is attractive for biomedical applications.

stronger effect compared to larger ones.

216 Cytotoxicity

**3. Magnetic nanostructures**

**3.1. Magnetic nanoparticles**

NPs such as magnetite (Fe<sup>3</sup>

O4

**3.2. Superparamagnetic iron oxide nanoparticles**

Ma et al. studied the uptake of 30 nm aminosilane-coated NPs by human lung cancer SPC-A1 and human lung WI-38 cells and found that the intracellular Fe content was 15 times higher for the cancerous cells compared to normal counterparts [78]. As with other NPs, they are likely endocytosed through phagocytosis and found within endosomes and lysosomes. Human monocytes-macrophages were also found to endocytose SPIONs and retained them inside lysosomes, remaining highly viable with no apparent activation of pro-inflammatory cytokines for up to 14 days following the incubation with 0.4 mg/mL SPIONs [79].

Using bare 20–30 nm magnetite NPs, Karlsson et al. tested other parameters of cytotoxicity in the human alveolar A549 cells, such as DNA damage and intracellular ROS [80]. No DNA damage or intracellular ROS were found for doses up to 40 μg/cm2 , although a slight oxidative DNA lesion was found at this dose. In contrast, another study showed that uncoated SPIONs elicited a significant level of apoptosis on mouse fibroblasts (L929), whereas PVA-coated ones did not show a loss of cell viability, apoptosis, necrosis, or cell cycle arrest for up to 72 hours of incubation and concentrations up to 200 mM [81]. However, an increase in the concentration to 400 mM did induce apoptosis and cell cycle arrest, possibly due to DNA damage through oxidative stress. Naqvi et al. obtained similar results for Tween 80-coated NPs in macrophage J774 A1 cells: >95% cell viability for low concentrations (25–200 μg/mL) and low incubations times, with a decrease to 55–65% for higher concentrations (300–500 μg/mL) associated to an apoptotic death pathway through ROS generation [82].

In contrast to previous findings, both citric acid and dextran-coated NPs were found to produce a dose-dependent cytotoxicity in human umbilical vein endothelial cells (HUVECs) [83]. Concentrations as low as 0.1 nM decreased the cell viability for both NPs to around 80% after 24 hours and increasing the value to 20 nM would decrease the cell viability to less than 15%. Additionally, as shown by Soenen et al. [84], actin filaments and microtubules appeared disrupted, thinner, and less organized and vinculin adhesion points were diminished. Further, NPs also reduced the migration and vasculogenesis capabilities of HUVECs. Similar results regarding cell attachment and cytoskeleton morphology were also reported in a multiparametric study with NPs with different coatings on various cell lines [84].

With an aim to understand the differences in cytotoxicity between the charges provided by different coatings on SPIONs, a study showed that when different functional groups were added in order to provide either a positive or negative charge on SPIONs, cell viability and cell membrane integrity remained above 85% up to 24 hours for doses as high as 1000 ppm on L929 fibroblasts for all the coatings tested [85]. As observed for other types of NPs, the positively charged NPs were more readily taken up than negatively charged ones. Similarly, ROS generation was not significantly different. However, the positively charged and highest negatively charged NPs showed DNA damage starting from concentrations of 200 ppm. In agreement with this, another study using HCM (heart), BE-2-C (brain), and 293T (kidney) cell lines reported similar results [86]. There, bare, positively, and negatively charged NPs all showed a dose-dependent response for doses up to 36 mM, with positively charged NPs being more cytotoxic for the three cell lines, suggesting a cell-specific response. Gene expression analysis showed that genes that were mainly altered were those related to apoptosis, cell cycle, and cell proliferative responses, most probably due to ROS.

endosomal or lysosomal endosomes (**Figure 4**) [99]. It was concluded that NWs are degraded by cells and cut into shorter pieces, possibly by the decrease in pH occurring in lysosomal compartments [100]. Along with no significant decrease in cell viability, no ROS were found after 4 hours of incubation for doses up to 170 NWs per cell. In a recent study, NWs with iron core and iron oxide shell were compared to pure iron NWs and tested on HCT 116 cells. The experiments confirmed the high cell viability values found for iron NWs before and revealed even higher values for the core/shell NWs [101]. An additional advantage of the core/shell NWs is the possibility to tune their magnetic properties to the specific requirements of vari-

Review of *In vitro* Toxicity of Nanoparticles and Nanorods: Part 1

http://dx.doi.org/10.5772/intechopen.76365

219

Studies with Ni NWs first showed a similar distribution to that of Fe NWs: Ni NWs of 20 μm in length and 200 nm in diameter activated cell membrane receptors associated with metalloproteins, thereby being internalized, triggering lysosomal function in the process and localizing inside them around the cell nucleus [102]. Lamellipodium extensions, due to cell tethering and re-alignment, were also a consequence of Ni NW internalization, possibly due to a cell stiffening response. The same group then studied the biocompatibility of Ni NWs on human monocyte THP-1 cells using high content analysis [103]. Measuring cell viability

O3

membrane-bound compartments, identified as late endosomal or lysosomal endosomes, b) and d) are zoomed images from a) and b) respectively. Adapted with permission from Safi et al. [99]. Copyright 2011 American Chemical Society.

) NWs during 24 hours. The NWs were found inside

ous applications.

**Figure 4.** NIH/3T3 fibroblasts incubated with maghemite (γ-Fe2

More recent studies have focused on looking at other parameters to better understand the cytotoxic response. A size-dependent response was observed for uncoated NPs of 5 and 30 nm, with only the latter inducing a significant increase in ROS generation, whereas dextran-coated and PEG-coated did not have an effect on ROS levels [87]. Khan et al., on the other hand, studied the effects of SPIONs on autophagy, a homeostasis mechanism used to degrade proteins and organelles for multiple functions [88]. They proved that ROS induces autophagy through the mTOR pathway only on human alveolar cancer A549 cells, and not on normal human lung fibroblast IMR-90 cells, while the authors attributed the autophagy to be involved with cell death. Lastly, Singh et al. observed a different cytotoxic response and uptake related to the Fe redox state (magnetite vs. maghemite) in human lymphoblastoid MCL-5 cells [89]. While no significant difference was found between these two states in terms of uptake, a decrease in the serum concentration drastically increased the uptake for dextran-coated maghemite NPs and this specific kind of NPs was the only one reported to elicit a genotoxic response.

#### **3.3. Magnetic nanowires**

Magnetic NWs possess tunable lengths and diameters and can be functionalized to provide specific targeting and biocompatibility. Depending on the fabrication method and its parameters, as well as precursor materials, the magnetic properties of NWs can be finely modulated [66, 90]. Magnetic NWs have anisotropic structures with high aspect ratios, which allow them to exert torques when under a magnetic field [91]. Additionally, they possess higher magnetization values per unit of volume when compared to NPs, allowing them to exert larger forces [92]. These qualities have made magnetic NWs prime candidates for different biomedical applications, including cell separation and guidance [91–93], targeted drug delivery [94, 95], and cell eradication [15, 16, 96, 97].

The cytotoxicity of Fe NWs was first characterized by Song et al. on HeLa cells [98]. Using uncoated NWs of <10 μm in length and 50 nm in diameter, they determined that Fe NWs have no significant effect on the cell viability and proliferation for concentrations up to 10,000 NWs per cell and for incubation times up to 72 hours. Fe NWs were internalized either as single NW, bundles or as aggregates, mainly localizing in the cytoplasm and inside vesicles, but not inside cell nuclei. Later, Safi et al. proved the same intracellular distribution in fibroblast NIH/3T3 cells using maghemite NWs of <15 μm and identified such vesicles as late endosomal or lysosomal endosomes (**Figure 4**) [99]. It was concluded that NWs are degraded by cells and cut into shorter pieces, possibly by the decrease in pH occurring in lysosomal compartments [100]. Along with no significant decrease in cell viability, no ROS were found after 4 hours of incubation for doses up to 170 NWs per cell. In a recent study, NWs with iron core and iron oxide shell were compared to pure iron NWs and tested on HCT 116 cells. The experiments confirmed the high cell viability values found for iron NWs before and revealed even higher values for the core/shell NWs [101]. An additional advantage of the core/shell NWs is the possibility to tune their magnetic properties to the specific requirements of various applications.

on L929 fibroblasts for all the coatings tested [85]. As observed for other types of NPs, the positively charged NPs were more readily taken up than negatively charged ones. Similarly, ROS generation was not significantly different. However, the positively charged and highest negatively charged NPs showed DNA damage starting from concentrations of 200 ppm. In agreement with this, another study using HCM (heart), BE-2-C (brain), and 293T (kidney) cell lines reported similar results [86]. There, bare, positively, and negatively charged NPs all showed a dose-dependent response for doses up to 36 mM, with positively charged NPs being more cytotoxic for the three cell lines, suggesting a cell-specific response. Gene expression analysis showed that genes that were mainly altered were those related to apoptosis, cell

More recent studies have focused on looking at other parameters to better understand the cytotoxic response. A size-dependent response was observed for uncoated NPs of 5 and 30 nm, with only the latter inducing a significant increase in ROS generation, whereas dextran-coated and PEG-coated did not have an effect on ROS levels [87]. Khan et al., on the other hand, studied the effects of SPIONs on autophagy, a homeostasis mechanism used to degrade proteins and organelles for multiple functions [88]. They proved that ROS induces autophagy through the mTOR pathway only on human alveolar cancer A549 cells, and not on normal human lung fibroblast IMR-90 cells, while the authors attributed the autophagy to be involved with cell death. Lastly, Singh et al. observed a different cytotoxic response and uptake related to the Fe redox state (magnetite vs. maghemite) in human lymphoblastoid MCL-5 cells [89]. While no significant difference was found between these two states in terms of uptake, a decrease in the serum concentration drastically increased the uptake for dextran-coated maghemite NPs and this specific kind of NPs was the only one reported to

Magnetic NWs possess tunable lengths and diameters and can be functionalized to provide specific targeting and biocompatibility. Depending on the fabrication method and its parameters, as well as precursor materials, the magnetic properties of NWs can be finely modulated [66, 90]. Magnetic NWs have anisotropic structures with high aspect ratios, which allow them to exert torques when under a magnetic field [91]. Additionally, they possess higher magnetization values per unit of volume when compared to NPs, allowing them to exert larger forces [92]. These qualities have made magnetic NWs prime candidates for different biomedical applications, including cell separation and guidance [91–93], targeted drug delivery [94, 95],

The cytotoxicity of Fe NWs was first characterized by Song et al. on HeLa cells [98]. Using uncoated NWs of <10 μm in length and 50 nm in diameter, they determined that Fe NWs have no significant effect on the cell viability and proliferation for concentrations up to 10,000 NWs per cell and for incubation times up to 72 hours. Fe NWs were internalized either as single NW, bundles or as aggregates, mainly localizing in the cytoplasm and inside vesicles, but not inside cell nuclei. Later, Safi et al. proved the same intracellular distribution in fibroblast NIH/3T3 cells using maghemite NWs of <15 μm and identified such vesicles as late

cycle, and cell proliferative responses, most probably due to ROS.

elicit a genotoxic response.

218 Cytotoxicity

**3.3. Magnetic nanowires**

and cell eradication [15, 16, 96, 97].

Studies with Ni NWs first showed a similar distribution to that of Fe NWs: Ni NWs of 20 μm in length and 200 nm in diameter activated cell membrane receptors associated with metalloproteins, thereby being internalized, triggering lysosomal function in the process and localizing inside them around the cell nucleus [102]. Lamellipodium extensions, due to cell tethering and re-alignment, were also a consequence of Ni NW internalization, possibly due to a cell stiffening response. The same group then studied the biocompatibility of Ni NWs on human monocyte THP-1 cells using high content analysis [103]. Measuring cell viability

**Figure 4.** NIH/3T3 fibroblasts incubated with maghemite (γ-Fe2 O3 ) NWs during 24 hours. The NWs were found inside membrane-bound compartments, identified as late endosomal or lysosomal endosomes, b) and d) are zoomed images from a) and b) respectively. Adapted with permission from Safi et al. [99]. Copyright 2011 American Chemical Society.

and membrane permeability, they found Ni NWs to be nontoxic for low incubation times (<10 hours) and concentrations (<100 NWs per cell). Hossain and Kleve then investigated the effects of Ni NWs on human pancreatic adenocarcinoma Panc-1 cells [104]. Using NWs of around 6.5 μm in length and 215 nm in diameter, a dose-dependent cytotoxic response was shown, including ROS generation and the induction of apoptosis and cell cycle arrest in the G0 /G<sup>1</sup> phase. A similar response was then reported for HeLa cells, along with mitochondrial membrane depolarization [105].

We reported the cytotoxicity of Ni NWs in human fibroblast WI-38 cells [106] and human colorectal carcinoma HCT 116 cells [107]. Whereas WI-38 cells showed no significant decrease of cell viability up to doses of 120 μg/mL for 24 hours of incubation and the viability of HCT 116 cells decreased significantly at the same incubation time for doses as low as 5 μg/mL. For both cell lines, NWs were internalized and appeared in the cytosol inside membrane-bound compartments, possibly lysosomes, as shown previously [102], with the internalization in HCT 116 cells taking place through the phagocytosis pathway. Apoptosis was also confirmed to be the cell death pathway, which would later progress into secondary necrosis and induce cell membrane instability and LDH leakage. Lastly, it was also confirmed that Ni2+ is released intracellularly following NW uptake, due to the acidic pH inside the lysosomes. Although the percentage of this intracellular dissolution is low compared to the total dose, it is plausible that the leeched Ni2+ contributes to the cytotoxic effects observed. It should be mentioned that Au-coated Ni NWs showed an improved biocompatibility, possibly due to the Au reducing the degree of dissolution, while also providing a functionalization layer [108, 109]. Similarly, we have reported the stabilization of Fe NWs by coating with a poly(MPC) homopolymer in order to increase dispersion and biocompatibility [110].
