Biomedicines, Vol. 10, Pages 3137: Polyelectrolyte Coating of Ferumoxytol Differentially Impacts the Labeling of Inflammatory and Steady-State Dendritic Cell Subtypes

1. IntroductionCellular immunotherapies have become an increasingly promising approach for the development of integrative and personalized therapies. Antigen-presenting dendritic cells (DCs) are considered particularly well suited for the development of such therapies, due to their unique capacity to initiate and orchestrate antigen-specific immune responses [1]. DCs are the first cells to be involved in antigen sensing and scavenging, followed by processing and subsequent antigen-specific T-cell priming. Additionally, DCs can activate further immune effector cells, including B cells, natural killer (NK) cells, and NKT cells [2]. Thus, DCs harmonize immune responses that eventually result in resistance to foreign pathogens and tolerance to self. Novel DC-based therapies, therefore, aim at establishing beneficial immune conditions in diseases such as cancer, chronic inflammation, autoimmunity, and transplant rejection. This is readily achieved using DC-targeted vaccines, mainly to induce anti-tumor immunity, while tolerogenic DCs are explored to silence autotoxic immune responses [3,4,5,6].In vivo monitoring of engraftment, position, and/or migration to the target site and function of transplanted cells could decisively contribute to the success of such therapies. Magnetic resonance imaging (MRI) of contrast agent-labeled cells, including DCs, has emerged as a well-suited imaging technique for tracking cells in vivo [7]. An outstanding feature of MRI is its capacity for long-term tracking of cells and their migration while achieving excellent high-resolution images of target tissue in a three-dimensional anatomical context. Stable labeling of DCs with iron-oxide-based magnetic nanoparticles (MNPs) as contrast agents has proven successful for MRI-based detection of cell deposits and their migration [8,9,10,11,12].Ferumoxytol is an FDA-approved iron-oxide-based MNP formulation currently used as a drug to treat iron-deficiency anemia in chronic kidney disease [13]. Ferumoxytol has also been used as a macrophage-imaging agent as well as a blood-pool agent with MRI [14]. However, ferumoxytol alone does not result in effective cell labeling [15,16]. Thus, in previous studies, we successfully established conditions that enabled us to manufacture colloidal stable ferumoxytol particles coated with polyelectrolytes (PE), resulting in enhanced cell labeling [16]. In this study, we mainly focused on the impact of PE-coated ferumoxytol MNPs on the labeling of different DC subsets, including inflammatory DCs, steady-state conventional DCs (cDCs), and plasmacytoid DCs (pDCs) [17]. DCs generated in vitro from monocytes or CD34+ hematopoietic stem/progenitor cells (HSCs) from blood or bone marrow using the granulocyte-macrophage colony-stimulating factor (GM-CSF) resemble a DC subset that only occurs in vivo under inflammatory conditions and is hence referred to as inflammatory DCs [17]. These patient-derived DCs represent the prevailing DC subtype used in autologous cell-based immunotherapies so far [6]. However, employing primary existing cDC and pDC subtypes under steady-state conditions for therapeutic approaches is considered a potentially better-suited alternative [3]. Here, we investigated the cellular uptake of PE-coated and unmodified ferumoxytol into inflammatory DCs and steady-state cDCs/pDCs generated by in vitro cultures from mouse bone marrow. We assessed the impact of the PE coatings on cell viability, labeling efficiency, and intracellular iron content of MNP-labeled cells. Furthermore, we investigated the immunophenotypic alterations of the various DC subtypes upon MNP incorporation. In summary, our results reveal a differential impact of PE-coated and uncoated MNPs upon uptake in inflammatory DCs and steady-state DC. 4. Discussion

Surface coating of MNPs using polyelectrolytes has great potential to tailor MNP properties for use in various biomedical applications, including cell labeling and tracking by MRI. In this study, we demonstrated that specific PE coatings cause different cellular responses in distinct DC subpopulations. We found a selective labeling capacity of PE-coated MNPs dependent on the DC subtype, together with a differential impact on the cytotoxic as well as immunomodulatory consequences of both uncoated and PE-coated ferumoxytol.

Engineered iron oxide MNPs are increasingly harnessed as advanced tools for medical applications, including cell labeling and tracking, targeted drug release, non-invasive monitoring of therapy, and vaccination [14,29]. Moreover, iron-oxide-based MNPs are considered to be most suitable for combining a number of these applications into a single multifunctional formulation, thus providing an accurate nano theranostics tool.Currently, ferumoxytol is the only FDA-approved iron-oxide-based MNP formulation, initially launched as an iron replacement therapy. It has recently been investigated extensively as a contrast agent in MRI since it shows fewer side effects, such as allergic or idiosyncratic reactions, than other contrast agents [14]. Moreover, ferumoxytol does not entail a risk for the development of nephrogenic systemic fibrosis and thus may substitute gadolinium-based contrast agents as a blood pool agent in a number of MRI applications. Accordingly, attempts have been made to use ferumoxytol as an MRI contrast agent for labeling and tracking cells in vivo, including mesenchymal stromal cells, neural stem cells, and immune cells such as T cells, monocytes, and DCs [10,24,27,30]. The uptake of ferumoxytol by macrophages in vivo is also being explored as a novel imaging approach for the assessment of lymph nodes, tumors, and vascular lesions. For example, in a preclinical model of autoimmune myocarditis, the iron oxide MNPs ingested by macrophages improved the distinction of areas of severe inflammation by MRI compared to conventional T2-weighted and gadolinium-enhanced MRI [31]. However, in a more recently published clinical study in patients with acute myocarditis, ferumoxytol-enhanced MRI was unable to identify myocarditis by the detection of macrophage activity [32]. As a possible explanation for this contradictory finding, the authors discuss the limited uptake capacity of ferumoxytol by macrophages compared to other iron oxide MNPs. This is in line with previous studies that found no efficient labeling of cells with ferumoxytol alone or in combination with protamine [10,15]. However, a combination of ferumoxytol with protamine and heparin was described to result in improved labeling of neural stem cells, bone marrow stromal cells, monocytes, and T cells, with an increase in T2 relaxivity compared to ferumoxytol alone [27].We frequently use layer-by-layer (LbL) assembly of polyelectrolytes for coating of MNPs to improve cellular responses, including uptake, intracellular localization, and processing of MNPs, and correspondingly, the MRI properties of labeled cells. For example, the coating of oleate-stabilized MNPs with PDADMAC resulted in a more dense agglomeration of MNPs within DC endosomal compartments, resulting in a larger magnetic susceptibility effect (T2*) when compared with loosely packed MNPs that were coated with polystyrene sulphonate or chitosan [18]. Notably, the observed differences in MRI contrast-agent properties of PE-coated MNPs were not correlated with the total amount of iron taken up by cells, as chitosan-coated MNPs yielded the highest iron concentration in DCs, but exhibited inferior performance in MRI [18]. Here, we used the positively charged polyelectrolytes PEI and PDADMAC for the coating of ferumoxytol. PEI and PDADMAC have proven to be excellently biocompatible as PE coating for MNPs and allow stable colloidal coating of ferumoxytol [16,18]. Remarkably, the PE coating of ferumoxytol significantly improves the labeling of steady-state DCs by up to fourfold. In contrast, uncoated ferumoxytol was already incorporated by ~50% of the inflammatory GM-DCs, and the PE coating did not result in substantially improved cell labeling. Patient-derived GM-DCs are, to date, the prevailing DC subtype used in autologous cell-based immunotherapy studies [6]. Another approach, however, is to harness the available steady-state cDCs and pDCs in vivo by directly targeting the specific subsets and activating their subset-specific properties depending on the type of disease [3]. Accordingly, attempts are made by, for example, using antibodies that recognize and bind to unique subsets as carriers for antigens, drugs, or immune regulatory factors [33]. Moreover, very similar approaches have emerged using nano-carrier-based delivery systems, including iron oxide-based MNPs [34]. A wide range of nanoscale materials have been developed that can serve as platforms for assembling various antigens, adjuvants, and other immunomodulatory reagents bound to the surface of and/or enveloped in such nanocarriers. Interestingly, some of these antigen and adjuvant factors represent polyelectrolytes that target specific immune pathways. One such pathway is activated by toll-like receptors (TLRs), and TLR agonists are intensively explored as molecular adjuvants for vaccination. This includes negatively charged nucleoside analogs that act as specific TLR agonists, such as double-stranded RNA or the synthetic analog poly(I:C) for TLR3, bacterial or viral single-stranded RNA for TLR7, or unmethylated DNA oligonucleotides (ODN) containing CpG motifs for TLR9. For example, simultaneously applied poly(γ-glutamic acid)-based NPs loaded with a tumor model antigen (OVA) or with poly(I:C) induced higher anti-tumor activity compared to the activity without NPs [35]. Even more elegantly, the group of Jewell et al. [36] used LbL-assembly of positively charged OVA peptide (SIINFEKL) and negatively charged poly(I:C) around calcium carbonate templates where the core was finally removed using a chelator to create hollow capsules. These immune polyelectrolyte multilayers (iPEMs) were able to activate steady-state DCs to a greater extent than dose-matched soluble factors alone [34].The TLR specificity of agonists can be further exploited to target specific DC subsets since DC populations express non-overlapping sets of TLRs [28]. For example, in humans, TLR9 is expressed by pDCs but not inflammatory DCs, and thus CpG ODN acts particularly on pDCs to induce type I interferon (IFN) production.

The synthetic polyelectrolytes PEI and PDADMAC used in our study elicited a slight adjuvant effect in steady-state DCs, comparable to ferumoxytol alone, but to a much lesser extent than the TLR4 agonist LPS. In contrast, all NP formulations were immunologically inert in inflammatory DCs, suggesting that GM-DCs, unlike FL-DCs, lack the respective recognition receptors.

A previously unrecognized immunomodulatory activity of ferumoxytol has been described in a recent study on its mechanism of tumor growth inhibition [37]. While ferumoxytol showed no direct cytotoxic effects on tumor cells, it was shown to expedite the recruitment of macrophages toward tumor cells and induce a phenotypic shift toward pro-inflammatory M1 polarization [37]. Tumor-associated macrophages generally develop toward an M2-like phenotype at a later stage of tumor progression. At the same time, M1 polarization requires activation by canonical IRF/STAT signaling pathways activated by IFNs and TLR signaling [38]. Ferumoxytol then elicits the production of reactive oxygen species (ROS) via the Fenton reaction in M1 macrophages, thereby inducing the apoptosis of cancer cells [37]. It is tempting to speculate that ferumoxytol is recognized and acts via TLRs on macrophages, but the precise mechanism has not yet been addressed. According to our data, however, it appears unlikely that ferumoxytol is recognized by TLR4, which is highly expressed in macrophages as well as GM-DC. In our study, both PE-coated and uncoated ferumoxytol did not result in activation or apoptosis of inflammatory DCs, in contrast to stimulation with the TLR4 ligand LPS. On the other hand, uncoated ferumoxytol increased apoptosis in steady-state DCs, probably by the generation of ROS through the Fenton reaction. Thus, the differential impact of ferumoxytol on the cytotoxicity of GM-DCs and FL-DCs further points toward DC-subset-specific pathways activated in responses to the MNPs. 5. Conclusions

LbL assembly of polyelectrolytes around MNPs represents a versatile means to tailor MNP surface properties. Here, we found that high MW PEI and low MW PDADMAC are well suited for the coating of ferumoxytol to improve the cell-labeling efficacy of various DC subtypes. PE-coated ferumoxytol was essentially non-toxic to labeled cells. Moreover, PE coating diminished the adverse cytotoxic effects of uncoated ferumoxytol in labeled steady-state DCs.

Polyelectrolyte coating of MNPs can be conceived to endow particle surfaces with desired properties. This can be employed, e.g., (i) for enhanced or even targeted uptake into cells, thereby augmenting their performance as contrast agents for MRI, and (ii) to provide additional functionalities such as immunoregulatory activities. For clinical use, immunomodulatory activities provided by the MNP formulation may have beneficial effects, e.g., as an adjuvant function in vaccination applications. However, the careful assessment of such immunomodulatory activities is necessary to avoid deleterious side effects of MNPs beyond cytotoxicity. The specific antigen uptake and presentation, as well as the immune regulatory function of distinct DC subpopulations, make them particularly important for the study of such MNP properties. Depending on the molecular pattern of the MNP shell, specific DC subsets—as shown here—can be expected to recognize MNPs as potentially foreign antigens, therefore providing the capability to biologically sense MNP shell chemistry. This can lay the foundation for further functionalization of MNPs that will combine efficient labeling of DCs for cell tracking with additional activities that impact DC function. In this respect, the combination of ferumoxytol coated with clinically approved polyelectrolytes should expedite a faster translation of this approach to clinical use.

留言 (0)

沒有登入
gif