4.2.4. Impact of SPIONs on Monocytes and Macrophages

Since macrophages play an important role in immunosurveillance supported by major phagocytic activity and given their significant presence in tumors, it is essential to assess the impact of SPIONs, as vectors of anti-cancer therapies, on these cells and their precursors (monocytes), whether from safety (cytotoxicity, inflammation) or functional perspectives (polarization, biological responses). In this sense, several recent studies have examined the impact of SPIONs based on in vitro models of monocytes and macrophage cells. All in vitro macrophage models described in this review that were used to study SPIONs are detailed in Figure 4. These macrophage models can also be found in the first two columns

*Pharmaceutics* **2022**, *14*, x FOR PEER REVIEW 21 of 39

4.2.4. Impact of SPIONs on Monocytes and Macrophages

phage polarization and biological responses.

resident macrophages [147].

of Table 3, a table that depicts the effects of SPIONs on monocyte/macrophage polarization and biological responses. cells. All in vitro macrophage models described in this review that were used to study SPIONs are detailed in Figure 4. These macrophage models can also be found in the first two columns of Table 3, a table that depicts the effects of SPIONs on monocyte/macro-

and pancreatic cancers showed that a significant part of TAMs also derived from tissue-

Since macrophages play an important role in immunosurveillance supported by major phagocytic activity and given their significant presence in tumors, it is essential to assess the impact of SPIONs, as vectors of anti-cancer therapies, on these cells and their precursors (monocytes), whether from safety (cytotoxicity, inflammation) or functional perspectives (polarization, biological responses). In this sense, several recent studies have examined the impact of SPIONs based on in vitro models of monocytes and macrophage

**Figure 4.** Some of the in vitro models based on murine or human macrophages. Created with Bio-Render.com. **Figure 4.** Some of the in vitro models based on murine or human macrophages. Created with BioRender.com.

One of the first parameters to take into account when evaluating the impact of SPI-

ONs on monocytes or macrophages is whether these nanoparticles can undergo rapid uptake. In general, monocytes or macrophages are able to uptake SPIONs relatively rapidly (few hours) [114,170]. Wu et al. demonstrated in primary human monocyte cells that SPI-ONs can be identified in phagosomes or in cytoplasm [171]. However, there are two noteworthy items regarding the cellular uptake of SPIONs: their size and their charge. Indeed, SPIONs with a size up to 150 nm show a high uptake (Table 3), whereas those with a size One of the first parameters to take into account when evaluating the impact of SPIONs on monocytes or macrophages is whether these nanoparticles can undergo rapid uptake. In general, monocytes or macrophages are able to uptake SPIONs relatively rapidly (few hours) [114,170]. Wu et al. demonstrated in primary human monocyte cells that SPIONs can be identified in phagosomes or in cytoplasm [171]. However, there are two noteworthy items regarding the cellular uptake of SPIONs: their size and their charge. Indeed, SPIONs with a size up to 150 nm show a high uptake (Table 3), whereas those with a size above 200 nm showed limited cellular uptake [172]. Zhang et al. [114] demonstrated that the surface charge of SPIONs influenced their uptake rate by murine macrophages. Thus, positively charged SPIONs (+) have a higher rate of uptake than negatively charged SPIONs (−), and negatively charged SPIONs (−) in turn displayed a higher uptake rate than neutral SPIONs (N). Sharkey et al. [173] have also demonstrated that positively charged SPIONs (DEAE-Dextran) provided the best uptake when compared to negatively (CM-Dextran) or neutral (Dextran) ones.

Another important parameter to consider is the impact of SPIONs on cell viability in order to take advantage of the beneficial effects provided by anti-cancer therapies while minimizing the harmful adverse effects potentially induced by SPION vectors, especially since SPIONs cytotoxicity remains unclear [171]. In fact, several variables considerably complicate the evaluation of SPION cytotoxicity. These variables include, for instance, the duration of cell exposure to SPIONs. In order to reduce SPION-related cytotoxicity, Sharkey et al. [173] reduced from 24 h or 48 h to 4 h the incubation time of SPIONs with bone marrow derived-macrophages and no significant decrease in cell viability or increase in cytotoxicity was observed.







assays (e.g., iron assays) were performed to determine whether SPIONs are internalized in cells, while cytotoxicity assays (e.g., ATP assays) attempted to evaluate the degree of SPION toxicity.

In this particular case, Sharkey et al. aimed at labeling macrophages with SPIONs before injecting them in mice in order to visualize SPIONs-labelled macrophages by MRI. Therefore, reducing incubation time for ex vivo labeling is possible (for imaging purposes) [173]. However, for studies that aim at evaluating treatments with SPIONs (therapy purposes), systemic administration of SPIONs does not allow the control of the incubation time. In order to reduce cytotoxicity, experiments have shown that SPIONs coated with biocompatible polymers such as dextran, polyethylene glycol, or starch were less cytotoxic [178,179]. Most of the SPIONs listed in Table 3 were coated with these molecules. Another means of decreasing cytotoxicity is to choose biocompatible iron oxides cores such as magnetite (Fe3O4), maghemite (γ-Fe2O3), or hematite (α-Fe2O3) [179] and adapt their concentration below 100 mg/mL [170].

There is no doubt that SPIONs exert an important modulation of macrophages' biological responses. Kodali et al. showed in a bone-marrow-derived macrophages model that 1052 genes were differently expressed between macrophages treated with SPIONs and controls [180]. The challenge resides in the understanding of which cell signaling pathways are involved. Several studies clearly demonstrated that SPIONs activate the MAPK signaling pathway through the phosphorylation of the downstream mediator ERK1/2 [171,174,176] (Figure 5, 1). One of the most important signaling pathways implied in cell proliferation is the MAP kinase pathway. This signaling pathway can also be activated in case of stress such as DNA damage or heat shock. In this case, the effects of this pathway will be more oriented toward differentiation or apoptosis rather than cell proliferation [181]. Three studies clearly demonstrated that SPION treatment activates the MAPK signaling pathway [171,174,176]. Indeed, downstream mediator ERK1/2 was phosphorylated in those studies. Interestingly, the activation of other MAPK downstream mediators (other than ERK1/2) has been shown to be highly dependent on the type of SPION coating or cellular model used in these studies. As such, PEI-coated SPIONs activated p38 and JNK downstream mediators in RAW 264.7 macrophages [174] as well as dextran-coated SPIONs in primary peripheral blood monocytes [171]. However, DMSA-, APS-, and AD-coated SPIONs induced no phosphorylation of p38 nor JNK in THP-1 cells (monocyte cell line) [176] (Figure 5, 1). Studies have demonstrated that the action of SPIONs on macrophages is, at least in part, mediated by the family of toll-like receptors (TLRs). TLRs are receptors present on cells of the innate immune system, mainly monocytes and macrophages. There are so far ten TLRs that have been discovered in humans [182]. The ligands recognized by these receptors are very variable, either by their structure (LPS, LTA, peptidoglycans, flagellin, RNA, DNA) or by their origin (derived from bacteria, viruses, parasites, or *fungi*) [183]. It has been demonstrated that there is a crosstalk between MAPK and TLR signaling pathways in THP-1 cells, especially with TLR4 [184], the receptor that binds LPS [185] (Figure 5, 2). Moreover, since PEI was linked to the activation of TLR4 [186], Mulens-Arias et al. have demonstrated that the TLR4 signaling pathway is also activated by PEI-coated SPIONs, at least partly, since an inhibitor of TLR4 (CLI-095, also known as TAK-242) reduced IL-1β and VEGFA mRNA induction upon PEI-coated SPION treatment [174]. Jin et al. likewise demonstrated that TLR4 was involved following a SPION treatment [175]. Finally, another signaling pathway that has been described as being activated by SPIONs is the AKT signaling pathway, which could be activated by metabolic stress, such as ROS production [187], which will be discussed below. Indeed, Rojas et al. showed that DMSA-, APS- and AD-coated SPIONs activated the AKT signaling pathway in murine bone marrow-derived macrophages [176] (Figure 5, 1). One point that remains to be elucidated is whether SPIONs activate these various signaling pathways through their interaction with cell membrane receptors (e.g., TLR4) or classical internalization (e.g., phagocytosis), or both. Other signaling pathways should also be studied in detail, such as G protein-coupled receptors, knowing that there is a link between ROS production and AMPK phosphorylation [187], or cytokines and JAK (janus kinases) protein activation, since their triggering has already been shown in an in vitro model of human endothelial cells following a nanoparticle treatment [188].

**Figure 5.** Impact of SPION treatment on polarization markers of macrophages. Created with Bio-Render.com. **Figure 5.** Impact of SPION treatment on polarization markers of macrophages. Created with BioRender.com.

SPIONs have also been described as directly impacting macrophage iron uptake as well as the expression level of iron-related proteins [173,176,177]. SPIONs are incorporated and degraded inside macrophages. Since the core of SPIONs is composed of iron, their degradation results in an increase in intracellular iron concentration. This iron accumulation in macrophages is thought to promote an M1-like phenotype [189]. M1-like macrophages display an iron storage phenotype. Consequently, these cells express higher levels of proteins involved in iron retention such as ferritin (a multimeric protein that is the main iron storage complex in cells [190]) or transferrin receptor 1 also known as CD71 (a transmembrane protein involved in iron uptake thanks to its binding to iron-loaded transferrin [191]). Conversely, M2-like macrophages present an iron export phenotype with an increase in ferroportin (a transmembrane protein involved in iron release [190]). In this context, Laskar et al. have demonstrated that SPIONs increase the expression of ferritin on THP-1 and human monocyte-derived macrophages [177]. Moreover, SPION treatment caused a decrease in transferrin receptors in M2 bone marrow-derived macrophages as well as in THP-1 monocyte-derived macrophages [176]. Moreover, ferroportin-1 expression was also decreased after 48h following AD- and APS-coated SPION treatment in THP-1 monocyte derived-macrophages [176]. Taken together, these results demonstrate that SPIONs will tend to cause iron accumulation in macrophages, a feature mainly observed in M1-like macrophages. SPIONs degradation by macrophages may result in free iron atoms in the cytoplasm SPIONs have also been described as directly impacting macrophage iron uptake as well as the expression level of iron-related proteins [173,176,177]. SPIONs are incorporated and degraded inside macrophages. Since the core of SPIONs is composed of iron, their degradation results in an increase in intracellular iron concentration. This iron accumulation in macrophages is thought to promote an M1-like phenotype [189]. M1-like macrophages display an iron storage phenotype. Consequently, these cells express higher levels of proteins involved in iron retention such as ferritin (a multimeric protein that is the main iron storage complex in cells [190]) or transferrin receptor 1 also known as CD71 (a transmembrane protein involved in iron uptake thanks to its binding to iron-loaded transferrin [191]). Conversely, M2-like macrophages present an iron export phenotype with an increase in ferroportin (a transmembrane protein involved in iron release [190]). In this context, Laskar et al. have demonstrated that SPIONs increase the expression of ferritin on THP-1 and human monocyte-derived macrophages [177]. Moreover, SPION treatment caused a decrease in transferrin receptors in M2 bone marrow-derived macrophages as well as in THP-1 monocyte-derived macrophages [176]. Moreover, ferroportin-1 expression was also decreased after 48h following AD- and APS-coated SPION treatment in THP-1 monocyte derived-macrophages [176]. Taken together, these results demonstrate that SPI-ONs will tend to cause iron accumulation in macrophages, a feature mainly observed in M1-like macrophages.

[192]. These atoms can in turn promote reactive oxygen species (ROS) production in a non-enzymatical way (Fenton chemistry, Figure 5, 3) [193]. In macrophages, ROS are associated with a pro-inflammatory M1-like phenotype since their production is used to SPIONs degradation by macrophages may result in free iron atoms in the cytoplasm [192]. These atoms can in turn promote reactive oxygen species (ROS) production in a nonenzymatical way (Fenton chemistry, Figure 5, 3) [193]. In macrophages, ROS are associated with a pro-inflammatory M1-like phenotype since their production is used to destroy pathogens by a mechanism known as respiratory or oxidative burst [194] triggering inflammation [195] via the activation of the NF-κB (nuclear factor-κ B) signaling pathway. PEI-, DMSA-, APS-, and AD-coated SPIONs have been described as inducing ROS production in murine (RAW 264.7 macrophages and bone marrow-derived macrophages) or human (THP-1 monocyte-derived macrophages) macrophages [174,176]. In addition, depending on the type of coating, ROS production levels may vary. AD-coated SPION treatment resulted in more ROS production than SPIONs coated with DMSA in murine macrophages derived from bone marrow [176]. ROS overproduction forms an integral part of oxidative stress that can be deleterious to cells, especially macrophages. The impact of SPIONs on ROS production must be carefully assessed in order to avoid cytotoxic effects linked to oxidative stress and maximize their safety [192].

Other macrophage biological responses have been demonstrated following SPION treatment. DEAE-dextran-coated SPIONs have no impact on macrophage phagocytic activities [173]. This is particularly interesting since one of the main roles of macrophages, phagocytosis, allows them to monitor their microenvironment against possible pathogens and ensure clearance of cellular debris leading to tissue homeostasis [196]. A treatment that would dampen this key feature could therefore prove to be deleterious to the organism. Mulens-Arias et al. demonstrated that PEI-coated SPIONs induced podosome formation in RAW 264.7 macrophages [174], and Gonnissen et al. showed that starch-coated SPIONs led to the disruption of the cytoskeleton in human monocytes [172]. These results all point in the same direction and may suggest that even though SPIONs do not impact the phagocytic activity of macrophages, they could somehow stimulate it indirectly since there is a link between phagocytosis and the formation of podosomes and their transient disruption in human macrophages [197].

Last, but certainly not least, it is also important to highlight that SPIONs exert an impact on macrophage polarization. Different polarization markers (M1 or M2) mainly from three molecule families (enzymes, membrane receptors, and cytokines) vary following SPION treatment (Figure 5, 4).

Firstly, in murine macrophages (RAW 264.7 or bone marrow-derived macrophages), SPIONs increased the expression of iNOS (M1 marker) and decreased the expression of Arg1 (M2 marker) [114,173].

Secondly, the expression of M1-like membrane receptors such as CD40 or CD80 in RAW 264.7 macrophages was increased following SPION treatment [174]. The expression of CD86, another widely used M1 marker, was increased as well when PEI-, DEAE-, Carboxydextrane-, DMSA- and APS-coated SPIONs were used regardless of whether this was assessed on murine or human macrophages [173,174,176,177]. However, no increase in CD86 expression was observed with dextran- or AD-coated SPIONs in human macrophages (THP-1) [170,176]. In addition, a decrease in the expression of M2-like membrane receptor CD206 expression was observed following DEAE-dextran-coated SPION treatment in murine macrophages from bone marrow [173]. An emphasis is needed in one experiment led by Zhang et al. [114]. Murine macrophages (RAW 264.7) were treated with SPIONs and then co-injected with HT1080 cells (fibrosarcoma cell line) in mice. Then, tumors were harvested and analyzed by immunohistochemistry. Compared to the control group (tumor cells co-injected with untreated macrophages), the treated group (tumor cells coinjected with macrophages pre-treated with SPIONs (+)) display an increase in CD80 (M1 marker) and a decrease in CD206 (M2 marker) in vivo. Moreover, SPION (+) pre-treated macrophages were shown to have an important tumor inhibition ability since tumor growth was reduced by at least threefold vs. the control group. These results showed that SPIONs (+) could repolarize macrophages and inhibit tumor growth.

Thirdly, cytokines (interleukins, chemokines, TNF-α, TGF-β, VEGF) expression was evaluated after SPION treatment. In general, SPION treatment had induced an increase in the expression of pro-inflammatory cytokines such as IL-1β, IL-2, IL-6, IL-12, IL-23α, CCL2, and TNF-α in murine or human macrophages [114,170,171,173–177]. Once again, the effects observed may be different for the type of coating and the cell model of use. Thus, in bone marrow-derived macrophages, Carboxydextrane-coated SPIONs induced an increase in IL-12 while DMSA-, APS-, and AD-coated SPIONs did not [175,176]. There is

one aspect, however, that must be underlined. Human primary peripheral blood monocytes were treated with dextran-coated SPIONs. This treatment induced the production of proinflammatory cytokines such as IL-1β and TNF-α at similar levels to those induced by LPS treatment [171]. This point needs to be further investigated, above all with a view to intravenous treatment with SPIONs. Moreover, the expression of anti-inflammatory cytokines, IL-10 and TGF-β, was also altered by SPION treatment in murine or human macrophages. However, no clear trend was observed. In murine macrophages (RAW 264.7), a decrease and an increase in IL-10 levels has been noted depending on the type of SPIONs being used [114,175]. In human macrophage models, an increase [176], a decrease [170], and no variation [172] in IL-10 levels have been described following SPION treatment. The expression of the other anti-inflammatory cytokine, TGF-β, was increased [176] in THP-1 monocyte-derived macrophages when treated with DMSA- and APS-coated SPIONs. It would be interesting to investigate the variation in this cytokine with different SPIONs in other cell models. Finally, in murine macrophages (RAW 264.7), VEGF expression has been found to decrease following SPION treatment. Since VEGF is considered a marker of M2 polarization involved in angiogenesis [198], it would be worth checking its variation in macrophages from human origin.

To summarize, SPIONs appear to globally induce a trend towards an M1-like proinflammatory phenotype (Figure 5) in macrophages (increase in M1- and decrease in M2 associated markers). Despite the fact that cytotoxicity and inflammation related to SPIONs remain issues to be improved, having nanoparticles in the context of cancer biology that would repolarize M2-like macrophages into M1-like macrophages may appear appealing.

In conclusion, given the great heterogeneity of SPIONs (size, surface charge, shape, coating, core composition), it is essential to evaluate the impact of newly synthesized SPIONs on the monocyte-macrophage axis, preferably on primary cell lines as they are closer to physiological conditions and human pathologies [199].
