RESEARCH ARTICLE www.small-journal.com Lung Persistence, Biodegradation, and Elimination of Graphene-Based Materials are Predominantly Size-Dependent and Mediated by Alveolar Phagocytes Thomas Loret, Luis Augusto Visani de Luna, Matteo Andrea Lucherelli, Alexander Fordham, Neus Lozano, Alberto Bianco,* Kostas Kostarelos,* and Cyrill Bussy* Graphene-based materials (GBMs) have promising applications in various 1. Introduction sectors, including pulmonary nanomedicine. Nevertheless, the influence of The respiratory system is a major portal GBM physicochemical characteristics on their fate and impact in lung has not of entry for microorganisms and xenobi- been thoroughly addressed. To fill this gap, the biological response, otics. Exposure to airborne pollutants and microorganisms can cause chronic lung in- distribution, and bio-persistence of four different GBMs in mouse lungs up to flammation and diseases such as asthma, 28 days after single oropharyngeal aspiration are investigated. None of the fibrosis, or cancer and lead to impairment GBMs, varying in size (large versus small) and carbon to oxygen ratio as well of lung functions or even death.[1,2] Tar- as thickness (few-layers graphene (FLG) versus thin graphene oxide (GO)), geting the respiratory system by direct ad- induce a strong pulmonary immune response. However, recruited neutrophils ministration of pharmaceuticals into the lungs is therefore representing a highly internalize nanosheets better and degrade GBMs faster than macrophages, valuable approach to improve health out- revealing their crucial role in the elimination of small GBMs. In contrast, large comes. However, controlling the deposition GO sheets induce more damages due to a hindered degradation and of inhalable drugs in the respiratory tract long-term persistence in macrophages. Overall, small dimensions appear to is quite challenging. In this context, drug be a leading feature in the design of safe GBM pulmonary nanovectors due to delivery systems based on nanomaterials an enhanced degradation in phagocytes and a faster clearance from the lungs offer new opportunities for enhanced de- livery and efficacy.[3,4] Such improvement for small GBMs. Thickness also plays an important role, since decreased has been demonstrated recently with a li- material loading in alveolar phagocytes and faster elimination are found for posomal formulation of antibiotics that has FLGs compared to thinner GOs. These results are important for designing been commercialized to target pulmonary safer-by-design GBMs for biomedical application. macrophages infected by mycobacterium.[5] Although most drug delivery sys- tems are currently based on liposomal T. Loret, L. A. V. de Luna, A. Fordham, K. Kostarelos, C. Bussy T. Loret, L. A. V. de Luna, A. Fordham, C. Bussy Nanomedicine Lab 2.0 Lydia Becker Institute of Immunology and Inflammation School of Biological Sciences Faculty of Biology Faculty of Biology Medicine and Health Medicine and Health The University of Manchester The University of Manchester Manchester Academic Health Science Centre Manchester Academic Health Science Centre Manchester M13 9PT, UK Manchester M13 9PT, UK M. A. Lucherelli, A. Bianco E-mail: kostas.kostarelos@manchester.ac.uk; CNRS cyrill.bussy@manchester.ac.uk Immunology T. Loret, L. A. V. de Luna, A. Fordham, K. Kostarelos, C. Bussy Immunopathology and Therapeutic Chemistry National Graphene Institute UPR 3572 The University of Manchester University of Strasbourg Manchester M13 9PL, UK ISIS Strasbourg 67000, France The ORCID identification number(s) for the author(s) of this article E-mail: a.bianco@ibmc-cnrs.unistra.fr can be found under https://doi.org/10.1002/smll.202301201 N. Lozano, K. Kostarelos © 2023 The Authors. Small published by Wiley-VCH GmbH. This is an Catalan Institute of Nanoscience and Nanotechnology (ICN2) open access article under the terms of the Creative Commons Attribution CSIC and BIST License, which permits use, distribution and reproduction in any Campus UAB medium, provided the original work is properly cited. Bellaterra, Barcelona 08193, Spain DOI: 10.1002/smll.202301201 Small 2023, 19, 2301201 2301201 (1 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com formulation,[6] two-dimensional (2D) materials such as ter single intratracheal administration, Bengston et al. reported graphene-based materials (GBMs) have recently emerged as that both materials were genotoxic in BAL immune cells but new drug delivery candidates owing to their unique combination not in lung tissues.[24] They also reported that rGO could cause of physicochemical and biological properties. On the one hand, chronic lung inflammation whereas GO-induced inflammation their high surface area to volume ratio is a major feature allowing was only acute.[24] The importance of lateral dimensions was not only high binding capacity,[7] but also the grafting of vari- also highlighted in several studies that reported an increased ous moieties on their surface for multimodal/multifunctional inflammatory profile for larger materials (either GNPs or GO capability.[8] On the other hand, GBMs have the capacity to enter sheets).[14,25–28] Although tissue recovery was observed in most in various cell types and ability to biodegrade, two essential fea- of the aforementioned studies, large GO sheets were shown to tures for a successful intracellular delivery.[9,10] Among the most cause long lasting lung injuries and fibrosis, owing to a reduced common types of GBMs explored for biomedical application clearance from the alveolar region.[25,29] Equally, we have previ- is graphene oxide (GO). GO is an amphiphilic material that ously reported that the mechanisms involved in the adverse ef- possesses a surface rich in oxygen-containing groups conferring fects of GO are size-dependent, with the number of molecular stability in aqueous suspension.[7,11] Due to its amphiphilic na- pathways indicative of adverse outcomes decreasing with the lat- ture, binding of both hydrophilic and hydrophobic molecules to eral dimensions of the GO sheets.[27] Going further, we ascribed GO via covalent and non-covalent functionalization is possible. this size-dependent activation of adverse outcome pathways in In comparison, graphene, whether it is categorized according the lung to the slower tissue recovery caused by repeat exposure to the number of layers as graphene nanoplatelets (GNPs) or to micrometric GO compared to nanometric GO.[28,30] few-layer graphene sheets (FLGs), is hydrophobic, as it possesses As highlighted above, inhalation of GBMs has raised several only sp2 carbon regions. For this reason, it needs to be chem- health concerns. It follows that proposing GBMs as the next ically functionalized or exfoliated to become stable in aqueous platform for pulmonary drug delivery can be perceived as chal- suspension. Although chemical functionalization of either GO lenging. Nevertheless, GBMs could still be promising nanocarri- or FLG can improve specificity and bioavailability, it may com- ers for the respiratory tract if the balance between benefits and promise the biocompatibility of such drug delivery systems.[7] risks is accurately and systematically evaluated. Indeed, GBMs Prior to the translation of drug delivery systems based on GBMs offer a unique opportunity for enhanced targeted delivery to the into clinical practice, a comprehensive evaluation of their safety alveolar space, owing to their distinctive biological properties. profile would therefore be critical to address all health concerns Namely, their deep lung penetration, their natural internalization and fulfil regulatory requirements, especially when considering in macrophages, their accumulation but at the same time contin- pulmonary application.[12,13] uous elimination from the lungs, and the overall rapid lung re- In respect to inhalation, the pulmonary impact of GBMs have covery reported in several if not most of the existing studies could been assessed before,[14] albeit primarily for the purpose of repli- all be deemed advantageous features.[14,19] Noticeably, targeting cating unintended exposure to airborne GBMs at the workplace. alveolar cells could be of interest in the context of lung infec- In several in vivo studies, single pharyngeal aspiration of GNPs tions by pathogens, such as mycobacteria that undergo intracel- or FLGs was demonstrated to induce acute lung inflammation lular replication in alveolar macrophages, or for cancer therapy, followed by progressive tissue recovery by day 7 or 28.[15–17] Im- as cancer cells tend to grow in the alveolar cavity in many lung portantly, regardless of the GBMs used in these studies, there was carcinomas.[31] Once accumulated in macrophages, drugs carried no tissue remodeling, although most materials were shown to by GBMs would be expected to be slowly released to nearby lung persist in the respiratory tract. Conversely, Schinwald et al. high- cells upon degradation of the GBM nanovectors.[27,32] lighted that large and thick GNPs may be responsible for frus- To exploit this opportunity and address all concerns and re- trated phagocytosis in lung macrophages, a predictive indicator maining challenges, the in vivo fate and behavior of GBMs in often leading to tissue remodeling and fibrotic lesions, but found relation to their biological impact on lungs will need to be sys- no evidence of long term damages despite the material bioper- tematically evaluated. Indeed, regulators would only approve a sistence in a follow-up study.[15,18] In another study, Park et al., GBM-based nanovector for pulmonary delivery if it is demon- reported biological effects in lungs for up to 90 days after admin- strated that this vector is inert to the lungs and rapidly eliminated istration of GNPs, including an increase in the number of re- from the airways. In this respect, a thorough study of GBM fate cruited cells, elevated levels of pro-inflammatory mediators, and over time is missing due to the difficulty of tracking GBMs in an increase in apoptotic cells in broncho-alveolar lavage (BAL) vivo, as they are made of carbon, the main element of organic fluids.[19] In contrast, no significant inflammation was surpris- matter. In addition, although previous studies have reported that ingly reported after chronic inhalation of GNPs for either 5 or GBMs can degrade in vitro[33] and in tissues in vivo,[27,34] there 28 days, even at the highest dose and despite the persistence of is no definitive information explaining how GBMs are getting materials.[20,21] eliminated from the lungs. Both degradation by immune cells Similar to GNPs, pulmonary exposure to GO sheets was also and clearance mediated by immune cell migration or lung fluid reported to induce acute inflammation in rodent lungs.[22–24] Im- flows (interstitial and mucocilliary) could be responsible for GBM portantly and comparing different GO materials, Li et al. demon- elimination from the airways and alveolar cavity. Moreover, due strated that higher degree of oxidation correlated with higher to the limited number of comparative studies between FLG and level of lung inflammation.[20] These results agree well with an- GO, it remains undetermined whether one or the other should other study in which GO sheets exhibited higher inflammation be favored in the design of future lung delivery systems based on levels than GNPs when the two materials were compared.[22] GBMs. In particular, how different physicochemical characteris- In addition, when comparing GO and reduced GO (rGO) af- tics such as dimension, thickness, or carbon to oxygen ratio (or Small 2023, 19, 2301201 2301201 (2 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com the combination of such features) may influence the degradation, fluids (Figure 2a,b for numerical results, Figure S1, Supporting elimination, and toxicological/biocompatibility profile of GBMs Information, for pictures). At day 1, we found a clear influx of in the lungs remain poorly understood. neutrophils for the four tested GBMs compared to the vehicle To address some of these knowledge gaps, we evaluated the control. The difference was statistically significant only for SGO pulmonary impact of four different GBMs over 28 days after sin- and SFLG. Some eosinophils were also identified, but the influx gle oropharyngeal aspiration in mice. Thin GO was compared to was statistically significant only for LFLG. After 7 days, the num- thicker FLG in order to understand the importance of the carbon ber of neutrophils was back to normal for all conditions, but not to oxygen ratio as well as material thickness in these processes. for the positive control, namely MWCNTs (MWCNT-7; Mitsui- Two lateral dimensions for each type of GBMs were also tested 7), a well-established nanomaterial causing inflammation, fibro- to evaluate the importance of size. A single exposure scenario sis and DNA damages.[39] The number of eosinophils was also was specifically chosen to investigate the potential of the differ- decreasing for all GBMs except LGO, which saw an increase at ent materials to degrade in cells over 28 days after exposure and day 7 in comparison to day 1 in a similar manner to the positive track the administered materials in the lungs. We measured the control. Both neutrophil and eosinophil levels were back to nor- cell loading and intracellular degradation of the selected GBMs mal at day 28 for all four GBMs, but not for the positive control using Raman microscopy, in all cells present in the alveolar space MWCNTs that still presented noticeable eosinophils in the alve- in vivo. Using this data set, we could also determine their kinet- olar space, as expected for this nanomaterial causing an allergy- ics of elimination from the alveolar space over the recovery period like response.[28] (from day 1 to 28). Finally, we examined the biological effects in Importantly, lymphocytes were not significantly present in respect to GBMs’ physicochemical characteristics and correlated BAL fluids following exposure to any GBMs. This is in clear con- the biological effects with material fate, in order to determine the trast with the MWCNTs that sustained a significant lymphocyte combination of features leading to the best biocompatibility pro- influx at each time point (Figure 2). Similarly, the number of file in the lungs. mononucleated macrophages did not change after exposure to any of the four GBMs (Figure 2b). However, we observed a signif- icant increase in the number of multinucleated macrophages for 2. Results and Discussion the LGO at day 28 compared to the vehicle control. Multinucle- 2.1. Graphene-Based Material Production and Characterization ated macrophages are the product of macrophage fusion that oc- curs when foreign bodies or materials cannot be internalized and We produced four different types of GBMs from graphite, fol- eliminated properly due to large dimension.[40] Moreover, pres- lowing two well-established methods previously reported by our ence of multinucleated macrophages in lungs has been previ- groups. GO sheets (1–2 nm thick due to hydration) with either ously associated with frustrated phagocytosis and clearance issue nanometric (SGO; average ≈60 nm) or micrometric (LGO; av- following the inhalation of high aspect ratio materials (HARN) erage ≈8 μm) lateral dimensions were synthetized by the Hum- such as long MWCNTs.[41,42] Herein, only the largest material mers’ method.[35] FLG sheets (3–7 layers and thickness >5 nm) of namely LGO (≈8 μm), but not LFLG (≈1 μm), led to the formation nanometric (SFLG; average ≈200 nm) and micrometric (LFLG; of multinucleated macrophages, suggesting that only this mate- average ≈1 μm) lateral dimensions were produced using ri- rial could not be cleared properly from the lungs, in agreement boflavin as an exfoliating agent.[36,37] with the HARN frustrated phagocytosis paradigm and its associ- The main physicochemical characteristics of the four materi- ated 5 μm threshold. [18,41,42] als are reported in Figure 1a–e. The results demonstrated that for each category of GBMs, either GO or FLG, we achieved two dis- tinct size distributions with minimal overlapping, allowing the 2.2.2. Biological Response in Lung Tissues comparison between different lateral dimensions. We also con- firmed that GOs and FLGs had distinct C:O ratio (with GO being We then determined how the lung tissue responded to GBM the more oxidized materials, irrespective of lateral dimensions) presence, using histopathological analysis of lung sections and distinct thickness (with FLG being the thicker materials with (Figure 3a,b; Figure S2, Supporting Information). First, the thickness going up to 15–20 nm in some part of the flakes as bronchial thickness was evaluated to assess the impact of GBMs seen in Figure 1a, regardless of lateral dimensions). In order to on the bronchial tract. No variation in bronchial thickness was perform a biological assessment without bias, we also confirmed found for any of the GBMs compared to the negative control (wa- that all four GBMs were endotoxin free using a method previ- ter for injection). Only the positive control (MWCNTs; MWCNT- ously published.[38] 7) induced a significant increase in bronchial thickness at day 7 and 28 (Figure 3b). Second, we assessed the presence of immune cell infiltrates in 2.2. Immune Infiltration and Tissue Changes in Response to the lung sections. Cell infiltrates were identified for all materi- Graphene-Based Materials Presence in Lungs als (Figure 3a; Figure S2, Supporting Information) at day 1, but significant only for LGO and positive control. This is in agree- 2.2.1. Immune Infiltration in the Alveolar Space ment with the BAL analysis showing immune cell recruitment (Figure 2a). However, we noted some variations between the dif- To assess the biological impact of the four different GBMs on ferent GBMs tested. Exposure to SGO or any of the two FLGs lungs, we first evaluated the influx of immune cells in the alve- led to a diffuse infiltration of granulocytes into the lungs. In con- olar space, using differential staining of cells contained in BAL trast, LGO (and the positive control MWCNTs) induced a more Small 2023, 19, 2301201 2301201 (3 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com Figure 1. Structural and morphological characterization of the GBMs. a) Atomic force microscopy (AFM) images of SGO, LGO, and LFLG. b) Scanning electron microscopy (SEM) images of GOs. c) Transmission electron microscopy (TEM) images of FLGs. d) Raman spectra of the GBMs. e) Summary of GBMs physicochemical characteristics; C/O ratio was evaluated using X-ray photoelectron spectroscopy (XPS), Lateral dimensions were evaluated using AFM (GOs and LFLG), SEM (GOs), TEM (FLGs) and optic microscopy (LGO). Thickness was evaluated using AFM (GOs and LFLG), Raman (FLGs) or TEM (FLGs). Small 2023, 19, 2301201 2301201 (4 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com Figure 2. Variation of immune cell populations in the alveolar space. Mice were exposed by oropharyngeal aspiration to 30 μg of GBMs or controls (positive control: MWCNTs; negative control: water for injection). Alveolar cells were collected at day 1, 7, and 28, cyto-spun on slides and then stained with colorimetric dyes for cell phenotyping. a) Percentage of each immune cell population identified. b) Total number of immune cells, eosinophils, neutrophils, mononucleated macrophages, multinucleated macrophages, and lymphocytes in BALF. Two-way ANOVA followed by Tukey’s multiple com- parisons test were used to evaluate statistical differences in GBM number compared to the negative control. Statistical analysis performed separately for the GBMs (p < 0.05:* , p < 0.01:** , and p < 0.001:*** ) and for the positive control (p < 0.05:# , p < 0.01:## , and p < 0.001:### ) (n = 3). localized and distinctive infiltration of granulocytes around 28, no significant immune cell infiltrates were found for any of bronchia and blood vessels. This was materialized by an in- the GBMs (Figure 3b). This rapid lung recovery highlighted that creased number and size of immune cell infiltrates at day 1 for the four GBMs had a moderate impact on lungs, especially when LGO in comparison to other GBMs (Figure 3b). Yet, the LGO- compared to the persisting inflammatory effects of MWCNTs at induced infiltrates were still in lower amount compared to the day 28. MWCNT positive control. The more diffuse immune responses Besides immune cell infiltrates, we also noted the formation found at day 1 for SGO and both types of FLGs could be attributed of granuloma-like structures, which were more apparent for to a better distribution of these materials across the lungs, includ- LGO compared to other GBMs (Figure 3a; Figure S2, Supporting ing in alveoli, as suggested by the Raman imaging analysis of Information). Interestingly, high amounts of materials were lung sections (Figure S3, Supporting Information). In compar- detected in these structures using Raman microscopy (Figure ison, LGO sheets formed large material agglomerates localized S3, Supporting Information). Moreover, some of these structures primarily next to bronchia (brown agglomerates in H&E images, were still present at day 28 for LGO, though fewer compared to Figure S2, Supporting Information), suggesting that their larger day 7, suggesting that localized accumulation and persistence of dimensions may hinder their transport from the upper airways to materials may be associated to the formation of these granuloma- the alveolar space and favor material agglomeration. At day 7 and like structures. The greater number of granuloma-like structures Small 2023, 19, 2301201 2301201 (5 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com Figure 3. Transient immune structures attributable to GBM presence in the lungs. Mice were exposed by oropharyngeal aspiration to 30 μg of GBMs or controls (Positive control: MWCNTs; Negative control: water for injection). Collected left lungs were inflated with formalin, embedded in paraffin and then processed. a) Lung sections (5 μm) were stained in Haematoxylin-Eosin for histopathological analysis. Immune infiltrates (blue arrow) and granulomatous-like structures (green arrow) were identified. b) Sizes of immune infiltrates and granulomatous-like structures and bronchial thickness were recorded and comparisons were performed between the different GBMs and controls. Two-way ANOVA followed by Tukey’s multiple comparisons test were used to evaluate statistical differences in GBM number compared to the negative control. Statistical analysis performed separately for the GBM (p < 0.05:* , p < 0.01:** , and p < 0.001:*** ) and for the positive control (p < 0.05:# , p < 0.01:## , and p < 0.001:### ) (n = 3). found in LGO exposed animals in comparison to the other GBMs of macrophages that fused and formed cell clusters in order may be explained by an impaired clearance of LGO materials to eliminate the LGO agglomerates more efficiently.[43] Similar from the lungs, as evidenced by the longer lasting presence of observations have been made for pathogens such as mycobac- LGO in lung sections analyzed by Raman spectroscopy (Figure terium or other foreign bodies that find their way to the alveolar S3, Supporting Information) compared to the other three GBMs. space.[43] In addition, the presence of granuloma-like structures We theorized that these granulomatous structures are made attached to the alveolar or airway walls is in agreement with the Small 2023, 19, 2301201 2301201 (6 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com presence of multinucleated macrophages in BAL fluids pensions were carefully homogenized to achieve a good spatial (Figure 2a,b; Figure S1, Supporting Information), further distribution of the isolated cells on the glass slide during cyto- suggesting that LGO sheets suffer from a clearance issue, spinning. This step was important to ensure that the image of likely due to their larger dimensions. Importantly, despite their these cells would be representative of the whole alveolar cell pop- ostensible presence in LGO exposed lungs, the number of ulation present in the airways at the considered time point. Af- granulomatous structures was still measurably lower than after ter fixation, cyto-spun BAL cells were first imaged by Raman to exposure to the positive control MWCNTs. evaluate the presence of GBMs. GBM presence was determined Taken together, these results demonstrated that GBMs caused using the specific Raman signatures of the four fully character- an acute immune response in the lungs, as shown previously in ized materials tested here (i.e., D and G bands for GO; G and 2D other studies, including ours.[16–18,22,24,44] Noticeably, the recov- bands for FLG; Figure 1d). Following Raman imaging, the same ery of the lungs from this acute inflammation was rapid, even glass slides were subsequently stained using Diff-Quick stains after LGO exposure. This is in sharp contrast to the response and imaged by bright-field optical microscopy in order to phe- found for the positive control, namely long and rigid MWCNTs, notype the different immune cells present in BAL fluids. Finally, which have been repeatedly reported to induce chronic inflam- Raman maps (G band only) and bright-field images were over- mation, fibrosis, tissue remodeling, and even cancer.[41,42] How- lapped (Figure 4a) to correlate the cell identity with GBM pres- ever, the presence of multinucleated macrophages, granuloma- ence. Using this approach, we identified that only macrophages like structures, and persistent material agglomerates in LGO- (blue arrow) and neutrophils (purple arrow) had internalized treated mouse lungs warrants further investigations, in partic- materials, since we recorded positive Raman signatures only in ular to assess the long-term fate and possible consequences of these two cell types. Importantly, we did not identify GBMs in these biopersistent materials in the lungs. any of the other cells present in BAL fluids (i.e., eosinophils, lymphocyte, and detached bronchial cells). These results are in agreement with the expected phagocytic ability of neutrophils 2.3. Graphene-Based Material Alveolar Accumulation and macrophages. For each GBM, we then determined the num- ber of macrophages or neutrophils that contained a positive Ra- Most of the biological outcomes reported above for LGO could be man signature (i.e., how many cell in each category contained associated to a greater persistence in the alveolar space of these GBMs) and evaluated the average material loading per positive materials compared to the other three GBMs. To assess alveo- cell (i.e., how much GBMs each loaded cell contained). Subse- lar persistence, we used Raman imaging to measure GBM pres- quently, we calculated the total material loading per cell type, in- ence at each time point and evaluate the evolution of materials’ cluding both positive and negative cells (i.e., how much GBMs presence over time as indicator of persistence, clearance, and/or is present in either macrophages or neutrophils when the whole degradation (Figure S3, Supporting Information), as previously BAL cell count is taken into consideration; Figure 4b,c), as previ- reported.[27,45,46] However, this method does not provide a quanti- ously reported.[28] Using the three parameters all expressed as tative measurement of the amount of GBMs present in the whole percentage (i.e., number of positive cells, average loading per lungs. Indeed, only few lung sections were scanned, and within a positive cell, total material loading per cell type; see further de- given section, only specific regions of interest (ROI) were imaged. tails about how these different values were calculated in the ex- Such analysis is therefore only indicative of the relative amount perimental section), we finally compared the material loading in of GBMs and its variation over time (i.e., a decrease of Raman sig- macrophages or neutrophils over time and between the different nature with time suggesting a reduction of GBM lung burden). types of GBM used. Moreover, Raman analysis of lung sections is not directly pro- viding cell-specific measurement. To overcome this limitation, we decided to image by Raman spectroscopy the alveolar phago- 2.3.2. GBM Accumulation in Alveolar Macrophages cytes isolated from BAL fluids, as previously reported.[28] Alveo- lar phagocytes (tissue resident and recruited upon inflammation) At day 1, 7, and 28 after exposure, there was a higher number of are known to internalize various nanomaterials after pulmonary Raman positive pixels per surface area of a considered cell for exposure. Furthermore, once nanomaterials have reached the GOs compared to FLGs (Figure 4b). This result suggests that alveolar space, they are expected to remain in this compartment there was a greater material loading per BAL macrophage for for some time, especially if not cleared rapidly.[18,19] Therefore, GOs than for FLGs. At day 1, this trend was found for the two we reasoned that evaluating GBM loading in BAL phagocytes at GOs in comparison to the two FLGs, but appeared statistically each time point and over time in a semi-quantitative way would significant only when comparing SGO to SFLG. At day 7, a sig- represent a valuable approach to assessing the fate of GBMs (i.e., nificant increase was measured for both GOs in comparison to persistence, degradation, and clearance) and would be comple- both FLGs. After 28 days, the difference in cell loading was sig- mentary to the qualitative assessment done at organ level on tis- nificant only for LGO, but not for SGO, in comparison to the sue sections. two FLGs. All time points considered, these results demonstrated the accumulation of all types of GBMs in macrophages, but with 2.3.1. Graphene-Based Material Accumulation in Alveolar relatively more cell loading for GOs compared to FLGs. The dif- Phagocytes ferences in material thickness between GOs and FLGs could ex- plain the variation in cell loading. Here, both SGO and LGO were First, BAL fluids were collected from GBM exposed lungs at 1, made mostly of monolayers, while both types of FLG were mostly 7, and 28 days after oropharyngeal aspiration. The BAL cell sus- few layered (Figure 1a,e). Thus, for the same mass of materials Small 2023, 19, 2301201 2301201 (7 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com Figure 4. Evaluation of GBM presence in alveolar phagocytes. BAL fluids were collected and alveolar cells were cyto-spun on slides. Fixed cells were first analyzed by Raman scanning microscopy and second stained with colorimetric dyes. Then, materials’ signal (G bands for GO and FLG) and cell phenotyping images were overlapped to evaluate GBM internalization in alveolar phagocytes. a) Raman signal in alveolar cells at day 1, 7, and 28 after exposure. b) Internalization of GBMs in macrophages at day 1, 7, and 28. c) Internalization of GBMs in neutrophils at day 1. d) Proportions of internalized Small 2023, 19, 2301201 2301201 (8 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com administered (assuming that the delivered dose to the lungs is over, these parameters are material specific, hence would vary identical between GOs and FLGs), the number of particles per with different GO production. In an in vivo scenario, other pa- lung surface area, hence the total lung surface area potentially rameters would also need to be factored in, including a different covered by FLGs, was expected to be less in comparison to GOs. type of biomolecule coronation, as well as a more dynamic envi- At the cell level, this also implies that for the same mass of mate- ronment. These parameters specific to the in vivo environment rials delivered to macrophages, GOs would occupy a larger space may explain why macrophage uptake was not size-dependent at (surface area/volume) within the cell than would FLGs. These re- the earliest time point in the present study. sults are important because higher macrophage loading has often been associated with higher biological responses.[47] 2.3.3. GBM Accumulation in Neutrophils When comparing the material loading between days 1 and 7, we measured a clear reduction in FLG signal in BAL Neutrophils are typically recruited from the blood to the respi- macrophages, while the decrease in GO signal was less pro- ratory tract upon induction of an innate immune response fol- nounced (Figure 4b). After 28 days, the four GBMs were still lowing lung exposure to airborne particles, in order to help re- present in BAL macrophages (Figure 4b), but we noted a clear de- ducing the lung burden.[54] Besides the interaction with BAL crease in cell loading for SGO, SFLG, and LFLG in comparison macrophages, the four GBMs could hence interact with or ac- to day 1 or 7. For LGO, strong Raman signal indicative of high cumulate in recruited neutrophils at day 1 (Figure 4a,c). Since material loading was still detectable in BAL macrophages after neutrophils were no longer present in BAL fluids beyond day 1, 28 days. Interestingly, when comparing the two FLGs (small ver- we could only estimate GBM accumulation in neutrophils at this sus large), there was no difference in the cell loading at any time time point (Figure 2). Interestingly, BAL neutrophils had accu- point. In contrast, there was a clear difference between SGO and mulated more SGO than any other tested materials (Figure 4c), LGO in the evolution of the material contained in macrophages with ≈70% of the recruited neutrophils demonstrating a positive over the 28-day recovery period. The total material loading of the Raman signal for SGO. In addition, all three measured param- macrophage population was the parameter showing the most eters were statistically significant for SGO compared to LGO. striking difference between the two GOs, with both the num- There was also a significant increase in the average material ber of positive macrophages and the average loading per positive loading and total neutrophil loading for SGO compared to the macrophage decreasing over time for SGO but not for LGO. This two FLGs. Comparing the two FLGs, the difference in lateral di- last result clearly demonstrated that the larger material (≈8 μm) mension had no significant impact on the cell loading, despite a had greater persistence in lungs compared to the other GBMs, slightly higher accumulation for SFLG compared to LFLG. These suggesting the existence of a size-threshold for GBMs’ persis- results highlighted again that lateral dimension is the leading fea- tence in the lungs, as reported previously for high aspect ratio ture in comparison to thickness or C:O ratio for GBM accumula- nanomaterials, such as fibers.[48] tion, here in neutrophils. Overall, these set of results underline the importance of ma- The presence of nanomaterials in neutrophils, including terial thickness and lateral dimension in the internalization and GBMs, has been reported before in several studies in vitro[55] and retention of GBMs in macrophages. Internalization of GBMs by in blood samples.[54,56] In these studies, the interaction of neu- macrophages has been explored before in several in vitro or in trophils with nanomaterials was shown to induce various down- vivo studies.[34,49–52] Noticeably, some of us highlighted the ben- stream effects, from their activation to the induction of apopto- efit of using Raman microscopy to estimate in vitro the accumu- sis, through inhibition of apoptosis. Moreover, it has also been lation of GO with varying lateral dimensions in macrophage cell reported that neutrophils could reverse migrate to lymph nodes, lines.[49] In this study, there were more macrophages positive for blood vessels, and even the bone marrow.[57] Since we found here the smaller GO (90%) than positive for the larger GO (60%) after a greater affinity of the recruited neutrophils for SGO, both the 24 h of exposure, suggesting that smaller materials have greater further biological implications of this accumulation/interaction internalization prospect than larger materials. These results (ob- and the long-term fate of SGO relocated to secondary organs via tained with a GO of different origin than the one used here and phagocyte reverse migration should be the focus of future inves- that possessed a smaller lateral dimension than the large GO with tigations. In particular, it would be important to assess in detail an average of 1.32 μm) contrast with our present data showing how much GBMs relocate from the lungs to secondary organs that at day 1 there was little to no difference between large and and the mechanisms (i.e., fluid drainage or cell-mediated migra- small materials in terms of cell loading, irrespective of the GBM tion) leading to this translocation. type considered (Figure 4b). Nevertheless, comparing in vitro to in vivo results is difficult as materials behave very differently in 2.3.4. Comparing GBM Accumulation in Macrophages and cell culture medium and in the complex environment of the air- Neutrophils ways. Sedimentation and agglomeration are often neglected vari- ables in vitro, although they are critical parameters that could Neutrophils and macrophages are both phagocytic cells. How- affect material deposition and internalization by cells.[53] More- ever, their internalization capacity is different, as confirmed by GBMs in macrophages and neutrophils at day 1. Two-way ANOVA (for macrophages) or one way ANOVA (for neutrophils) followed by Tukey’s multiple comparisons test were used to evaluate, at each time-point, statistical differences between GBMs respective large and small sizes (USGO vs LGO; SFLG vs LFLG: p < 0.05:* , p < 0.01:** , and p < 0.001:*** ) or chemical characteristics (SGO vs SFLG; LGO vs LFLG: p < 0.05:# , p < 0.01:## , and p < 0.001:### ) (n = 3). Small 2023, 19, 2301201 2301201 (9 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com our direct comparison between the two cell types for their respec- lyzed the spectra of the Raman maps (see Experimental Section tive average material loading per cell and total loading per cell for further details) (Figure 5a). type at day 1 (Figure 4b,c). When comparing the proportion of materials internalized at this time point in these two cell popula- tions, we found that macrophages had a greater cell loading than 2.4.1. Alveolar Clearance of GBMs neutrophils, regardless of the material considered (Figure 4d). This difference can be ascribed not only to the greater number of To estimate the overall alveolar clearance of GBMs over time (i.e., macrophages in the alveolar space but also to their larger size and how much material disappears from the alveolar space due to uptake capacity in comparison to neutrophils (Figure 2a). Consid- engulfment in alveolar phagocytes), we used the values of total ering that neutrophils were not present in the BAL fluid beyond material loading in macrophages and neutrophils calculated at day 1, we inferred that macrophages were across all time points day 1, 7, and 28 (Figure 4b,c; see Experimental Section for fur- the main cell type involved in GBM elimination from the alveolar ther details). At day 1, we were able to use the combination of space. total material loading values obtained for macrophages and neu- Unexpectedly, alongside the specific patterns of internalization trophils. However, at day 7 and 28, only the macrophage value described for LGO at day 1 in either macrophages or neutrophils, was available since neutrophils were no longer present in the we also found a substantial number of free LGO sheets (green alveolar space (Figure 2a). Day 1 constituted the starting point arrow) that were not present in any phagocyte (Figure 4a). How- for these assessments (i.e., maximum material loading value for ever, this was not the case for SGO or the two FLG materials, each GBM equals to 100%, clearance equals to 0%). Days 7 and suggesting that the combined alveolar phagocytes (macrophages 28 were the only two time points at which clearance from the and neutrophils) were able to engulf all materials present in the alveolar space could be determined in comparison to day 1 (with alveolar space in the case of SGO, SFLG, and LFLG, but not LGO. alveolar clearance expressed in percentage as = 100% – % of re- The larger lateral dimension of LGO sheets leading to a lower maining materials) (Figure 5a). On days 7 and 28, compared to LGO cell loading in both neutrophils and macrophages in com- day 1, we found clear and progressive clearance for three of the parison to the other GBMs (Figure 4d) could explain the pres- GBMs, namely SGO, SFLG, and LFLG (Figure 5a). The alveolar ence of free LGO sheets at day 1 (Figure 4a). Indeed, we rea- clearance appeared to be slightly faster for the two FLGs (SFLG: soned that macrophages present at day 1 in the alveolar space ≈30% remaining at day 7 and only ≈14% at day 28; LFLG: ≈44% may have reached their maximum loading capacity, and while remaining at day 7 and only ≈17% at day 28) compared to SGO the recruited neutrophils were able to engulf or interact with the (≈57% remaining at day 7 and only ≈21% at day 28). In contrast, remaining materials in the case of SGO, SFLG, and LFLG, they LGO clearance from the alveolar space was significantly slower, could not support the macrophages and engulf all remaining ma- with no apparent elimination found between days 1 and 7 (≈98% terials in the case of LGO. On the other hand, the presence of free remaining, clearance of 2%), and still ≈80% of the materials re- LGO materials at day 1 may explain why LGO were identified in maining after 28 days, resulting in only a 20% alveolar clearance macrophages for a longer period of time, barely decreasing from compared to day 1 (Figure 5a). These results not only confirm the day 1 to 28 (Figure 4b), if we theorized that new macrophages persistence of LGO in macrophages and hence in the lungs but were recruited to engulf the remaining LGO sheets. Future stud- also highlight the role of lateral dimensions, material thickness, ies should aim at interrogating the turnover and fate of LGO and carbon to oxygen ratio in the speed of material clearance from laden macrophages to reveal whether the laden macrophages the alveolar space. present at day 1 are the same as those present at day 28 and reveal the fate of the free LGO sheets in the lungs and beyond. Taken together, these results highlighted that lateral dimen- 2.4.2. GBM Degradation in Alveolar Phagocytes sion is a dominant physicochemical feature with respect to the accumulation of GBMs in alveolar phagocytes. Since we demon- To complement the alveolar clearance, we then evaluated the level strated that most of the free LGO sheets were internalized in of intracellular elimination (partial or complete degradation) of macrophages by day 7 (Figure 4a,b), it was unclear why a high GBMs in BAL cells (Figure 5b–e; Figure S3a,b, Supporting Infor- quantity of materials were still present in macrophages at day 28. mation). For this purpose, we used the decreases in Raman peak We hypothesized that this persistence may be due to impaired (D and G bands) intensities and changes in Raman peak inten- degradation of the LGO materials in macrophages and therefore sity ratios as indicators of such material alteration in cells. A clear decided to further assess the level of GBM degradation in alveolar change in intracellular GBM signature was found between days 1 phagocytes. and 28, as demonstrated by a decrease in the overall Raman inten- sity for all four GBMs (Figure 5b,d). In addition, both an increase in the ID /IG ratio and an alteration of the G peak were recorded 2.4. Alveolar Clearance and Degradation of Graphene-Based for the two GO materials (Figure 5b). These variations in peak Materials forms and intensities are known indicators of degradation or bio- transformation in biological systems for carbon-based materials, Considering the natural ability of alveolar phagocytes to remove, as previously reported by us and others.[27,33,34,45,60,61] Remark- process, and degrade foreign materials including carbon-based ably, we noted a faster alteration of the Raman spectra that we materials,[58,59] we evaluated the level of GBM alveolar clearance attributed to degradation for the smaller materials in comparison and degradation in both macrophages and neutrophils. To this to their larger counterparts. This was clearly illustrated with the end, we used Raman-based values, as calculated above, and ana- GO materials, with a more pronounced and faster increase in the Small 2023, 19, 2301201 2301201 (10 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License © 2023 The Authors. Small published by Wiley-VCH GmbH www.small-journal.com 2301201 (11 of 17) www.advancedsciencenews.com Small 2023, 19, 2301201 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com ID /IG ratio between days 1 and 7 for SGO (from 1.52 to 1.71) com- ment with previous studies suggesting that macrophages may pared to LGO (from 1.54 to 1.60) (Figure 5b; Figure S3, Support- be less potent than neutrophils at degrading GBMs in vivo due ing Information). Moreover, there was no increase in the ID /IG to their reduced enzyme-based oxidative degradation capacity.[62] ratio between days 7 and 28 for LGO, whereas a slight increase Overall, our results confirmed that neutrophils are better at de- was found for SGO. Interestingly, when comparing phagocytic grading materials, whereas macrophages are better at clearing cell types at day 1, we found a more obvious decrease in Raman materials, even if both cell types have degradation and clearance signal intensity in neutrophils than in macrophages for all the abilities. GBMs tested (Figure 5c,e). This cell type difference suggests a In respect to degradation mechanisms, a series of studies have faster degradation in neutrophils than in macrophages and un- described the different processes leading to the biodegradation derlines the important role of neutrophils in the elimination and of carbon nanomaterials, including GBMs.[63] Oxidative degrada- degradation of GBMs from the lungs during the early phase re- tion via the enzymatic activity of peroxidases, such as myeloper- sponse. Further evaluation would be necessary to identify poten- oxidase (MPO), is one of them and was demonstrated both in tial biodegradation by-products in macrophages and neutrophils. vitro and in vivo.[63,64] Using an acellular human MPO model from neutrophils, Kurapati et al. have reported that GO degrada- 2.4.3. Impact of GBM Physicochemical Characteristics on their tion was dependent on the level of material agglomeration, with Degradation and Elimination from the Lungs well-dispersed materials degrading better.[61] Degradation was also achieved when single- and few-layer graphene were exposed The findings of cell-mediated in vivo degradation of GBMs (via to human MPO, and was evidenced by a loss of crystalline and ge- macrophages and neutrophils) are in agreement with previous ometrical structure and an increase in ID /IG ratio with time.[33] In studies on the ability of various GBMs to be biodegraded. For in- the same study, the two materials were exposed to activated hu- stance, Girish et al. demonstrated for the first time in vivo the man primary neutrophils and shown to degrade within 5 days. important role of tissue-resident macrophages in the long-term Moreover, both a decrease in G band intensity and an increase degradation of carboxyl-functionalized graphene materials af- in ID /IG ratio were evidences that an MPO-mediated oxidative ter intravenous injection.[34] Using Raman confocal microscopy, degradation caused by the neutrophil degranulation (and pro- they evaluated the material degradation for up to 3 months af- duction of neutrophil extracellular traps (NETs)) had occurred.[33] ter IV injection in four different tissues of material accumulation Using the same concept, the degradability of large and small and for up to 7 days in macrophage cell lines to confirm the role of GOs by activated primary neutrophils was demonstrated by a fast macrophages in the degradation. The highest level of degradation decrease in Raman intensity, with D and G bands significantly was found in the spleen, the smallest in the lungs, with degra- reducing after 3 and 6 h.[60] Interestingly, MPO is produced in dation always starting from the edges of the material agglom- large quantity by activated neutrophils and to a lesser extent by erates. They confirmed that macrophage cell lines were able to macrophages. This difference in MPO expression capacity be- degrade materials to the same extent as the degradation found in tween neutrophils and macrophages would therefore explain the vivo. Similar to the present study, they reported an increase in the results reported here, with neutrophils degrading GBMs faster ID /IG ratio and a decrease in Raman intensity with time. In previ- than macrophages. This capacity difference also implies that neu- ous in vivo studies, we have also demonstrated that GO sheets ac- trophils may be more important than macrophages in the early cumulated in the spleen after IV injection were getting degraded phase of the response to nanomaterial exposure toward eliminat- over a 270-day period in the marginal zone macrophages, or GO ing as much materials as possible to reduce the lung burden.[65] It sheets present in the brain after intranasal administration were follows that materials that are not taken care of efficiently by the degrading over 28 days in the resident microglia and perivascular recruited neutrophils will be left to the attention of macrophages, macrophages.[45,46] which have a lower degradation capacity but greater phagocytic Taken together, these different studies agree with our findings ability. and confirm that recruited or resident macrophages, wherever As a consequence, the lung persistence of LGO sheets found they are located in the lungs, blood, or other organs, are essential here could be explained by a combination of events. First, there is cells in the long-term entrapment and/or elimination of GBMs a reduced accumulation of LGO sheets in both neutrophils and owing to their phagocytic and degradation capacities. Moreover, macrophages and an incomplete internalization in macrophages our present study also stresses the key role of neutrophils in this at day 1 due to their larger lateral dimensions (LGO sheets being cell-mediated degradation/elimination process, especially during on average ≈8 μm, while the other materials, including LFLG, the acute phase response, when neutrophils are recruited to the are below 1 μm). This suggests that there may be a size thresh- site of inflammation. In fact, neutrophils were shown here to be old for efficient internalization and elimination of GBMs by lung more efficient at degrading GBMs compared to macrophages, de- phagocytes, with GBM sheets above 5 to 10 μm inducing frus- spite their lower phagocytic ability. These results are in agree- trated phagocytosis in macrophages and precluding/reducing Figure 5. GBM clearance and intracellular degradation over time in lung phagocytes. Mice were exposed to 30 μg of GBMs and BALF were collected at day 1, 7, and 28. Alveolar cells were scanned by Raman and stained for cell phenotyping to evaluate the clearance and degradation of GBM. a) GBM clearance over time. b) Intracellular degradation of GO over time in alveolar phagocytes and c) at day 1 in neutrophils compared to macrophages. d) Intracellular degradation of FLG over time in alveolar phagocytes and e) at day 1 in neutrophils compared to macrophages. Raman signals were refined by removing the biological background. To evaluate GBM degradation, D/G ratio was calculated for each animal and variation of positive Raman signatures were evaluated. For each material and time-point, the corresponding average intensities and D/G ratio of 3 animals are presented in the graphs (30–3000 positive spectra for GBM signature were recorded for each animal, depending on the amount of materials detected). Small 2023, 19, 2301201 2301201 (12 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com internalization by recruited neutrophils, as previously reported terials and in particular SGO (the smallest and thinnest materials for HARNs such as nanotubes. [48] Second, LGO sheets en- of all) compared to other GBMs. Yet, more GBMs in total amount gulfed in neutrophils and macrophages are undergoing slower were found in macrophages compared to neutrophils, underlin- intracellular degradation compared to the other GBMs. In turn, ing the dominant role of macrophages in the isolation and subse- this slower degradation explains their slower alveolar clearance quent elimination or persistence of GBMs over time, especially (Figure 5a) and the persistence of a strong Raman signal in BAL during the inflammation resolution phase. macrophages at day 28 (Figure 4a,b). Future work should there- In respect to degradation and removal from the airways, SGO, fore aim at elucidating which molecular pathways in phagocytes SFLG, and LFLG were eliminated faster than LGO, the largest may be affected by LGO to explain the apparent reduced degrada- materials used in this study, as evidenced by a slower alveolar tion capacity of phagocytes that have internalized these materials. cell mediated clearance over time for this material. Importantly, Taken together, all the present findings point toward a link be- our results highlighted the essential role of neutrophils in the tween the extent of GBM degradation (elimination/persistence) early phase of material elimination and demonstrated that degra- and the degree of adverse effects, with LGO materials caus- dation of GBMs by neutrophils was size-dependent. At day1, ing the greatest impact among the four GBMs due to their re- we showed a stronger intracellular degradation in neutrophils duced degradation and greater persistence. Similar inferences than in macrophages, which combined with the greater affinity have been reached in previous works on carbon nanotubes, in of neutrophils for small sheets emphasized their crucial role in which inhibition of nanomaterial degradation was associated the elimination of small materials from the lungs at early time with more adverse outcomes than when the natural degrada- point. At later time points, the efficient clearance of SGO, SFLG, tion capacity was able to mitigate/reduce the effects of material and LFLG from the lungs was associated to a clear intracellular exposure.[62,66,67] While we noticed that LGO had a lower impact degradation in macrophages. Conversely, the poor clearance of than MWCNTs, the induction of multinucleated macrophages, LGO was associated to an impaired degradation in macrophages granuloma-like structures, and persistent material agglomerates over time, revealing that efficient degradation in macrophages still present at day 28 is calling for further investigations about is crucial for a controlled elimination from the lungs. Taken to- the long-term impact of such persistence (i.e., beyond 3 or 6 gether, the degradation results suggested that small materials months after exposure). below 1 μm were degraded and eliminated faster and more ef- ficiently than large materials above 5 μm, hence explaining the 3. Conclusion lung bio-persistence of LGO compared to the other GBMs. In summary, the present findings stressed the importance of This comparative study between FLG and GO aimed to iden- each physicochemical feature such as lateral dimension, thick- tify the physicochemical characteristics suitable for the design ness, or carbon to oxygen ratio in the design of safe GBMs in of safe pulmonary nanovectors based on GBMs. To this end, we respect to lungs. They demonstrate that features that do not fa- investigated the lung tolerability, fate, and elimination profiles vor degradation and elimination lead to material bio-persistence, of four different GBMs (two graphene oxides and two few-layer which in turn triggers biological responses often associated with graphene; small and large) after single pulmonary administra- long-term adverse effects. Among the different GBMs tested tion in mice. here, LGO sheets showed the least biocompatible profile whereas Compared to a reference material such as MWCNT-7, none SGO, SFLG, and LFLG may represent suitable options for pul- of the GBMs induced a strong immune response in the lungs, monary drug delivery systems. In particular, SGO demonstrated despite a transient influx of neutrophils and eosinophils, but a limited innate response and a very good elimination profile due with no recruitment of lymphocytes. However, exposure to LGO to a greater ability to interact with alveolar immune phagocytes. caused a focal immune response, with formation of granuloma- like structures and presence of multinucleated macrophages, both of which were indicators of frustrated phagocytosis, often associated to impaired elimination and bio-persistence of mate- 4. Experimental Section rials. Carbon-Based Materials Preparation and Characterization: Large (LGO) To assess the extent of material elimination from the lungs and Small GO (SGO) sheets were produced from graphite powder using over time, we therefore evaluated by Raman microscopy the fate the modified Hummers’ method[35] and then suspended in sterile water of GBMs both across the lungs and in the alveolar space over 28 (Gibco, Thermo Fisher Scientific) before characterization (Figure 1). Full days after exposure. While all the materials tested (SGO, LGO, characterization of the GOs was performed as previously reported.[28] Large (LFLG) and Small FLG (SFLG) sheets were synthesized by ex- SFLG, and LFLG) were present in the lungs at day 1, only LGO foliation of graphite (Sigma Aldrich, LOT #BCBS5850V) using riboflavin- showed extended persistence at day 28. In the alveolar space, all 5′-phosphate as previously described.[36] Briefly, graphite was dispersed tested GBMs were found in both macrophages and neutrophils, in a 1 mg.mL−1 solution of riboflavin 5-phosphate sodium salt (Rib) in but not in other cells present in BAL fluids. Interestingly, GO cell milli-Q RNA-free water (1 g in 400 mL) and sonicated via cup horn son- loading appeared higher than FLG cell loading after 1 day, un- ication (Ultrasonic VC505) for 45 min (20% amplitude, 1 s sonication, derlining that for the same dose more alveolar phagocytes could and 1 s pause). The obtained dispersion was then centrifuged at 5000 be targeted using GO. We attributed this difference to a differ- RCF, collecting separately the supernatant and the precipitate. The super- natant was filtered on Millipore filters PTFE (pore size <100 μm), washed ence in thickness between the two materials. FLGs being thicker with water (200 mL) and ethanol (10 mL), and redispersed in water by occupied a smaller cell volume than GOs for the same mass of short bath sonication to produce the Small FLG dispersion. The precipi- material administered. Furthermore, we revealed that recruited tate was dispersed again in 1 mg.mL−1 solution of Rib, sonicated for 10 neutrophils at day 1 demonstrated a greater affinity for small ma- min by cup horn sonication and centrifuged at 2000 RCF. In this case, after Small 2023, 19, 2301201 2301201 (13 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com purification, the supernatant was discarded because of the small lateral each cell type for 500 cells counted per animal and condition (Figure 2a), size of the flakes. The precipitate was submitted to a new cycle of sonica- or multiplied by the total number of cells present in the BAL fluid (prior to tion and centrifugation at 500 RCF, to produce after purification the Large cyto-spinning) to report the total number of each cell type (Figure 2b). FLG dispersion. The dispersions were carefully stored in sealed glass vials Histopathology Analysis: Inflated lungs were fixed in 10% formalin in milli-Q RNA-free water until use. (4% formaldehyde; Merck–Sigma) for 24 h and then transferred in 75% The thickness of the GO and FLG flakes was experimentally determined ethanol (Thermo Fisher Scientific) prior embedding in paraffin. Embed- using AFM (Bruker MultiMode 8) equipped with the NanoScope v1.9 soft- ded lungs were cut using a microtome (RM2255, Leica Biosystems), and ware for analysis (Figure 1a). Size distribution of GOs was evaluated us- 5.0 μm thick sections were put on glass slides (Superfrost plus slides, ing SEM (Figure 1b) as reported previously.[28] The lateral size of FLGs Thermo Fisher Scientific). Lung sections were stained with Hematoxylin was evaluated by a statistical analysis on at least 200 TEM images, using a and Eosin using an automatic stainer (XL autostainer, Leica Biosystems). Hitachi 7500 transmission electron microscope (Hitachi High Technolo- For the histopathological analysis of the lungs, bright field images were gies Corporation, Tokyo, Japan) equipped with an AMT Hamamatsu digital acquired with a slide scanner (Pannoramic 250 Flash, 3DHistech Ltd). camera (Hamamatsu Photonics, Hamamatsu City, Japan) (Figure 1c). Ra- Immune cell infiltrates and granuloma-like structures were identified, and man spectra of GO and FLG sheets drop casted on quartz slides (Electron the bronchial thickness was measured. All the histopathological analyses Microscopy Sciences, Inc.) were acquired using a Raman confocal system were performed using CaseViewer (software version 2.4.0.11902, 3DHis- (XploRA Plus, HORIBA), equipped with a 638 nm LASER (slit at 100 and tech Ltd). hole at 100 for GO; slit at 300 and hole at 500 for FLG) (Figure 1d). For FLG, Evaluation of GBM Presence and Degradation Using Raman Spectroscopy: the shape of the 2D band was analyzed using a Lorentzian distribution to To evaluate the distribution of GBMs in lung sections and BAL cells, Ra- estimate the numbers of layers.[68] Briefly, monolayer graphene was made man spectroscopy-based confocal microscopy (XploRA Plus, HORIBA) of a single Lorentzian fit, with a half peak width of ≈20–30 cm−1 . The 2D was performed using GBM specific Raman signatures. Prior to scan- peak of multilayer graphene was made of two to more components, caus- ning, lung sections embedded in paraffin on glass slides (Superfrost plus, ing a clear increase in peak width (Figure S5, Supporting Information). In Thermo Fisher Scientific) were dewaxed, whereas ice-cold methanol fixed addition, the 2D/G band ratio was determined for FLGs to confirm the BAL cells cyto-spun on glass slides were prepared as mentioned above. thickness of the flakes; a 2D/G ratio > 1.5 indicates monolayer while a ra- In both cases, scanning was performed by Raman microscopy at a wave- tio < 1.2 indicates few-layer. The main physicochemical properties of GOs length of 638 nm and a grating of 600, with 3 μm of distance between each and FLGs are summarized in the Figure 1e. scanning point. Multi-walled carbon nanotubes (positive control; Mitsui, Japan, type The parameters used to identify materials were adapted to GBM Ra- Mitsui-7; kind gift from Prof. Ulla B. Vogel, National Research Centre for man spectral characteristics (D and G bands for GO; G and 2D bands for the Working Environment, Denmark) were heated at 180–200 °C for de- FLG) to optimize the detection. A slit and hole of 100 μm were used for pyrogenization overnight, dispersed in 0.5% bovine serum albumin (BSA) GO, whereas a slit of 300 μm and a hole of 500 μm were used for FLG. in water for injection (Gibco, Thermo Fisher Scientific), and then sonicated GBMs were identified based on their Raman signatures: D (≈1335 cm−1 ) in a water bath for 5–7 min at 80 W (VWR essential, UK). and G (≈1590 cm−1 ) bands for GO, as well as 2D (≈2650 cm−1 ) band for Endotoxin levels were measured for all materials following the method FLG. However, only the intensity of the G band was used for overlapping described in Mukherjee et al.[38] based on primary macrophages secre- Raman maps and bright-field images (Figure 4a; Figure S3a,b, Supporting tion of TNF-𝛼 and showing absence of significant contamination (data not Information). Background removal was performed to remove the noise shown). attributed to the biological signature of BAL cells or lung tissues. Raman In Vivo Exposure by Oropharyngeal Aspiration: C57BL/6J mice signatures obtained before and after background removal are presented (6–8 week old, female; Envigo, UK) were housed in ventilated cages in Figure S4 (Supporting Information) and Figure 5, respectively. with access to food and water ad libitum in an environment controlled After Raman scanning, the same BAL cell slides were stained with Diff- for humidity, temperature, and light. All experiments were performed Quick stain (Kwik-Diff, Shandon, Thermo Fisher Scientific) in order to in compliance with UK Home Office Animals (Scientific Procedures) overlap the Raman maps (G band only) with the optical microscopy im- Act 1986 (ASPA 1986) and under the project license no. P089E2E0A, ages of the stained immune cells and semi-quantify the level of internal- reviewed by the University of Manchester Animal Welfare and Ethical ization by either neutrophils or macrophages. The number of neutrophils Review Body. Animals were exposed by single oropharyngeal aspiration and macrophages present on each slide was determined by counting 150 to vehicle (water for injection, Gibco, ThermoFisherScientific) or 30 μg cells per animal for each condition. The number of Raman positive and of materials suspended in 30 μL of vehicle. Prior to administration, mice Raman negative cells for GOs and FLGs was evaluated for both cell types. were anaesthetized by inhalation of 4% isoflurane and then held on a For each positive cell, the relative quantity of material internalized per cell slanted board. The animals (3–4 animals per group) were kept for 1, 7, (i.e., average cell loading) was based on the number of pixels positive for and 28 days after the single exposure. GBM Raman signature and expressed as % of the total number of pixels Bronchoalveolar Lavage Fluids and Lungs Collection: At day 1, 7, and that represents the total surface of the considered cell (i.e., 100%). To esti- 28 after exposure, mice were euthanized by IP injection of pentobarbi- mate the overall “total material loading” per cell type (either macrophages tone (100 μL/animal). Left lungs were kept unwashed by ligaturing the or neutrophils) and condition (SGO, LGO, SFLG, and LFLG), the percent- respective trachea, and right lungs were washed with ice cold PBS (3 times age of positive cells was multiplied by the average loading in each positive 800 μL) to collect broncho-alveolar lavages (BAL) fluids. Left lungs were cell in order to take into consideration the negative cells in the analysis. collected and inflated with pure formalin (40% formaldehyde) for further To measure the proportion of macrophages and neutrophils involved in histopathological analysis. the elimination of GBMs, the total loading values for either macrophages Phenotyping of BAL cells: BAL cells were centrifuged at 1500 rpm for or neutrophils (expressed as % of cell surface) were multiplied by the re- 5 min at room temperature (Hettich GmbH) and supernatants were stored spective number of cells measured in BALF at day 1 (Figure 4a). To nor- at −80 °C. Cells were then suspended in PBS (Merck-Sigma), counted us- malize the results per surface area, values obtained for neutrophils were ing a haemocytometer (Marienfeld GmbH), and an equal number per an- divided by two in order to take into consideration a two-time smaller total imal and condition was cyto-spun on a glass slide (Superfrost plus slides cell surface compared to macrophages. Finally, total values (macrophages Thermo Fisher Scientific) at 600 rpm for 5 min (Hettich GmbH). Cells on and neutrophils combined) were calculated to determine the percentage slides were fixed in 100% ice-cold methanol for 10 min, dried and stored of internalization in either macrophages or neutrophils in respect to the at −20 °C. Differential cell staining was performed using Diff-Quick stain total amount of material internalized by BALF cells (i.e., the two cell types (Epredia Shandon Kwik-Diff Stains) following the provider’s instructions. combined). Neutrophils, eosinophils, mono- and multi-nucleated macrophages, and Alveolar clearance over time (i.e., material elimination at day 7 and 28 in lymphocytes were identified and counted using bright field optical mi- comparison to day 1, used as starting point) was evaluated using the calcu- croscopy (AxioObserver, Zeiss). Values are expressed as percentage of lated total material loading in BALF cells at day 1, 7, and 28. For day 1, the Small 2023, 19, 2301201 2301201 (14 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com total material loading in cells combined the results of both macrophages would like to acknowledge the following groups at ICN2 (Catalan Institute (at day 1) and neutrophils (at day 1). To set the total loading at 1 day at of Nanoscience and Nanotechnology, Barcelona, Spain): Advanced Elec- 100%, the total material loading in macrophages and the total material tronic Materials and Devices led by Prof. J. Garrido for the AFM and Raman loading in neutrophils (normalized by the proportion of neutrophil in BAL instrumentation. The ICN2 is funded by the CERCA programme, General- and by the surface of neutrophils/macrophages ( = 0.5)) for each animal itat de Catalunya and has been supported by the Severo Ochoa Centres were summed. To have a standard deviation, the value obtained for each of Excellence program [SEV-2017-0706] and is currently supported by the animal was divided by the average total loading in macrophages and neu- Severo Ochoa Centres of Excellence program, grant CEX2021-001214-S, trophils of the 3 animals. For day 7 and 28, total material loading was both funded by MCIN/AEI/10.13039.501100011033. based on loading of materials in macrophages only (as neutrophils were no longer detected in BALF at day 7 and 28). Values of remaining materi- als (%) in BALF cells at day 7 and 28 were normalized to the total quantity Conflict of Interest of materials calculated at day 1 in the combined macrophages and neu- trophils (i.e., values at day 1 were set at 100%). The authors declare no conflict of interest. Raman microscopy was also used to evaluate GBM intracellular degra- dation in BAL cells (macrophages and others cells) over time (from 1 to 28 days after exposure), based on the evolution of the D and G band in- tensities for all cells present in the ROI of the Raman scan. All the positive Author Contributions spectra identified by Raman spectroscopy were considered for the analy- T.L. And L.A.V.de L. contributed equally to this work. T.L. performed con- sis. To evaluate material degradation, variations in calculated ID /IG ratio ceptualization (lead), investigation (lead for biological analyses), formal and in Raman peak intensities and changes in spectral structure were con- analysis (lead), methodology (lead), visualization (lead), writing – origi- sidered. nal draft (lead), and writing – review and editing (lead). L.A.V.de L. per- Statistical Analysis: Data were expressed as mean ± standard devia- formed investigation (lead for biological analyses), formal analysis (lead), tion. Statistical analysis was performed using Graphpad Prism 8.0 (Graph- methodology (lead), visualization (lead), and review and editing (equal). Pad Software Inc., San Diego, CA). For BALF analysis and histopathology, M.A.L. performed investigation (lead of FLG materials) and characteriza- two-way ANOVA followed by Dunnett’s post-hoc test was used to evalu- tion (lead for FLG materials). A.F. performed characterization (lead for ate differences compared to the negative control (n = 3–4). As MWCNT GO materials) and investigation (technical assistance). N.L. performed exposures were very positive for most endpoints tested, an extra analysis investigation (lead for materials), resources (lead for GO materials), for- was performed separately to avoid false negative results for other mate- mal analysis (lead for GO materials), visualization (lead for materials), rials. For Raman analysis, one-way ANOVA (for neutrophils) or two-way supervision (supporting), and writing – review and editing (equal). A.B. ANOVA (for macrophages) followed by Tukey’s multiple comparisons test performed conceptualization (lead), resources (lead for FLG materials), were used to evaluate at each time-point statistical differences of GBM in- funding acquisition (lead), supervision (lead), and writing – review and ternalization between respective large and small sizes or between GO and editing (equal). K.K. performed conceptualization (lead), resources (lead FLG (n = 3). for GO materials), funding acquisition (lead), supervision (lead), and writ- ing – review and editing (equal). C.B. performed conceptualization (lead), investigation (supporting), resources (lead for MWCNT materials), fund- Supporting Information ing acquisition (lead), supervision (lead), visualization (supporting), and writing – review and editing (equal). Supporting Information is available from the Wiley Online Library or from the author. Data Availability Statement Acknowledgements The data that support the findings of this study are available from the cor- responding author upon reasonable request. This project has received funding from the European Union’s Hori- zon 2020 research and innovation program under grant agreement nos. 785219 (H2020-FET- GrapheneCore2) and 881603 (H2020-FET- GrapheneCore3). A.F. would like to acknowledge studentship co-funded Keywords by the Lloyds Register Foundation funded International Consortium on Nanotechnology (LRF-ICON) and the UK Research and Innovation Engi- alveolar phagocytes, carbon materials, clearance, degradation, in vivo, tox- neering and Physical Sciences Research Council (UKRI-EPSRC) Centre for icity Doctoral Training Programme Graphene NOWNANO (EP/L01548X/1). The authors also thank Dr Adrian Esteban Arranz for contributing to the Received: February 10, 2023 synthesis of the GO materials used in the present study. At the Faculty Revised: April 3, 2023 of Biology, Medicine and Health, the University of Manchester, UK, au- Published online: June 1, 2023 thors would like to thank P. Walker and G. Balko (Histology Facility), Dr P. March (Bioimaging Facility), Dr. N. Hodson (BioAFM Facility), and all staff in the Biological Services Facility. The microscopes of the Bioimag- ing and BioAFM Facilities used in the present study were purchased [1] B. Duchemann, I. Annesi-Maesano, C. J. De Naurois, S. Sanyal, P. Y. with grants from the UKRI Biotechnology and Biological Sciences Re- search Council (UKRI-BBSRC), the Wellcome Trust, and the University Brillet, M. Brauner, M. Kambouchner, S. Huynh, J. M. Naccache, R. of Manchester Strategic Fund. This work was also partly supported by Borie, J. Piquet, A. Mekinian, J. Virally, Y. Uzunhan, J. Cadranel, B. the Agence Nationale de la Recherche (ANR) through the LabEx project Crestani, O. Fain, F. Lhote, R. Dhote, N. Saidenberg-Kermanac’h, P. Chemistry of Complex Systems (ANR-10-LABX-0026 CSC). The authors A. Rosental, D. Valeyre, H. Nunes, Eur. Respir. J. 2017, 50, 1602419. wish to acknowledge the Centre National de la Recherche Scientifique [2] E. M. Conway, L. A. Pikor, S. H. Y. Kung, M. J. Hamilton, S. Lam, W. (CNRS) and the International Center for Frontier Research in Chemistry L. Lam, K. L. Bennewith, Am. J. Respir. Crit. Care Med. 2016, 193, 116. (icFRC). The authors wish to thank C. Royer and V. Demais for TEM anal- [3] S. Azarmi, W. H. Roa, R. Löbenberg, Adv. Drug Delivery Rev. 2008, 60, yses at the “Plateforme Imagerie In Vitro de l’ITI Neurostra”, CNRS UAR 863. 3156, University of Strasbourg (Strasbourg, France). In addition, authors [4] J. Trousil, S. Aryal, G. Fuhrmann, Nanomedicine 2021, 16, 685. Small 2023, 19, 2301201 2301201 (15 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com [5] D. E. Griffith, G. Eagle, R. Thomson, T. R. Aksamit, N. Hasegawa, [25] X. Wang, M. C. Duch, N. Mansukhani, Z. Ji, Y. P. Liao, M. Wang, H. K. Morimoto, D. J. Addrizzo-Harris, A. E. O’Donnell, T. K. Marras, P. Zhang, B. Sun, C. H. Chang, R. Li, S. Lin, H. Meng, T. Xia, M. C. A. Flume, M. R. Loebinger, L. Morgan, L. R. Codecasa, A. T. Hill, S. Hersam, A. E. Nel, ACS Nano 2015, 9, 3032. J. Ruoss, J. J. Yim, F. C. Ringshausen, S. K. Field, J. V. Philley, R. J. [26] S. Vranic, A. F. Rodrigues, M. Buggio, L. Newman, M. R. H. White, D. Wallace, J. Van Ingen, C. Coulter, J. Nezamis, K. L. Winthrop, Am. J. G. Spiller, C. Bussy, K. Kostarelos, ACS Nano 2018, 12, 1373. Respir. Crit. Care Med. 2018, 198, 1559. [27] A. F. Rodrigues, L. Newman, D. Jasim, S. P. Mukherjee, J. Wang, I. [6] B. Pelaz, C. Alexiou, R. A. Alvarez-Puebla, F. Alves, A. M. Andrews, S. A. Vacchi, C. Ménard-Moyon, A. Bianco, B. Fadeel, K. Kostarelos, C. Ashraf, L. P. Balogh, L. Ballerini, A. Bestetti, C. Brendel, S. Bosi, M. Bussy, Adv. Sci. 2020, 7, 1903200. Carril, W. C. W. Chan, C. Chen, X. Chen, X. Chen, Z. Cheng, D. Cui, J. [28] T. Loret, L. A. V. de Luna, A. Fordham, A. Arshad, K. Barr, N. Lozano, Du, C. Dullin, A. Escudero, N. Feliu, M. Gao, M. George, Y. Gogotsi, K. Kostarelos, C. Bussy, Adv. Sci. 2022, 9, 2104559. A. Grünweller, Z. Gu, N. J. Halas, N. Hampp, R. K. Hartmann, et al., [29] B. Li, J. Yang, Q. Huang, Y. Zhang, C. Peng, Y. Zhang, Y. He, J. Shi, W. ACS Nano 2017, 11, 2313. Li, J. Hu, C. Fan, NPG Asia Mater 2013, 5,. [7] G. Reina, J. M. González-Domínguez, A. Criado, E. Vázquez, A. [30] L. A. V. de Luna, T. Loret, A. Fordham, A. Arshad, M. Drummond, A. Bianco, M. Prato, Chem. Soc. Rev. 2017, 46, 4400. Dodd, N. Lozano, K. Kostarelos, C. Bussy, Part. Fibre Toxicol. 2022, [8] M. A. Lucherelli, Y. Yu, G. Reina, G. Abellán, E. Miyako, A. Bianco, 19, 62. Angew. Chemie – Int. Ed. 2020, 59, 14034. [31] X. Liu, X. Xie, J. Jiang, M. Lin, E. Zheng, W. Qiu, I. Yeung, M. Zhu, Q. [9] K. Bhattacharya, S. P. Mukherjee, A. Gallud, S. C. Burkert, S. Bistarelli, Li, T. Xia, H. Meng, Adv. Funct. Mater. 2021, 31, 2104487 S. Bellucci, M. Bottini, A. Star, B. Fadeel, Nanomed. Nanotechnol., [32] C. Martín, K. Kostarelos, M. Prato, A. Bianco, Chem. Commun. 2019, Biol. Med. 2016, 12, 333. 55, 5540. [10] Y. Chen, J. Rivers-Auty, L. E. Cricǎ, K. Barr, V. Rosano, A. E. Arranz, [33] R. Kurapati, S. P. Mukherjee, C. Martín, G. Bepete, E. Vázquez, A. T. Loret, D. Spiller, C. Bussy, K. Kostarelos, S. Vranic, Nanoscale Adv. Pénicaud, B. Fadeel, A. Bianco, Angew. Chem., Int. Ed. 2018, 57, 2021, 3, 4166. 11722. [11] A. Servant, A. Bianco, M. Prato, K. Kostarelos, Bioorg. Med. Chem. [34] C. M. Girish, A. Sasidharan, G. S. Gowd, S. Nair, M. Koyakutty, Adv. Lett. 2014, 24, 1638. Healthcare Mater. 2013, 2, 1489. [12] S. F. Hansen, A. Lennquist, Nat. Nanotechnol. 2020, 15, 3. [35] A. F. Rodrigues, L. Newman, N. Lozano, S. P. Mukherjee, B. Fadeel, [13] D. A. Heller, P. V. Jena, M. Pasquali, K. Kostarelos, L. G. Delogu, R. C. Bussy, K. Kostarelos, 2D Mater. 2018, 5, 035020. E. Meidl, S. V. Rotkin, D. A. Scheinberg, R. E. Schwartz, M. Terrones, [36] H. Lin, D. K. Ji, M. A. Lucherelli, G. Reina, S. Ippolito, P. Samorì, A. Y. H. Wang, A. Bianco, A. A. Boghossian, S. Cambré, L. Cognet, S. R. Bianco, Small 2020, 16, 2002194. Corrie, P. Demokritou, S. Giordani, T. Hertel, T. Ignatova, M. F. Islam, [37] A. Ruiz, M. A. Lucherelli, D. Murera, D. Lamon, C. Ménard-Moyon, N. M. Iverson, A. Jagota, D. Janas, J. Kono, S. Kruss, M. P. Landry, Y. A. Bianco, Carbon N Y 2020, 170, 347. Li, R. Martel, S. Maruyama, et al., Nat. Nanotechnol. 2020, 15, 164. [38] S. P. Mukherjee, N. Lozano, M. Kucki, A. E. Del Rio-castillo, L. [14] B. Fadeel, C. Bussy, S. Merino, E. Vázquez, E. Flahaut, F. Mouchet, L. Newman, E. Vázquez, K. Kostarelos, P. Wick, B. Fadeel, PLoS One Evariste, L. Gauthier, A. J. Koivisto, U. Vogel, C. Martín, L. G. Delogu, 2016, 11, e0166816. T. Buerki-Thurnherr, P. Wick, D. Beloin-Saint-Pierre, R. Hischier, M. [39] K. Donaldson, F. A. Murphy, R. Duffin, C. A. Poland, Part. Fibre Toxicol. Pelin, F. Candotto Carniel, M. Tretiach, F. Cesca, F. Benfenati, D. 2010, 7, 5. Scaini, L. Ballerini, K. Kostarelos, M. Prato, A. Bianco, ACS Nano [40] A. K. McNally, J. M. Anderson, Adv. Exp. Med. Biol. 2011, 713, 97. 2018, 12, 10582. [41] K. Donaldson, C. A. Poland, F. A. Murphy, M. MacFarlane, [15] A. Schinwald, F. A. Murphy, A. Jones, W. MacNee, K. Donaldson, ACS T. Chernova, A. Schinwald, Adv. Drug Delivery Rev. 2013, 65, Nano 2012, 6, 736. 2078. [16] L. Mao, M. Hu, B. Pan, Y. Xie, E. J. Petersen, Part. Fibre Toxicol. 2015, [42] F. Murphy, N. R. Jacobsen, E. Di Ianni, H. Johnston, H. Braakhuis, 13, 7. W. Peijnenburg, A. Oomen, T. Fernandes, V. Stone, Part. Fibre Toxicol. [17] E. J. Park, G. H. Lee, B. S. Han, B. S. Lee, S. Lee, M. H. Cho, J. H. Kim, 2022, 19, 50. D. W. Kim, Arch. Toxicol. 2015, 89, 1557. [43] A. J. Pagán, L. Ramakrishnan, Ann. Rev. Immunol. 2018, 36, 639. [18] A. Schinwald, F. A. Murphy, A. Askounis, V. Koutsos, K. Sefiane, K. [44] S. Chortarea, O. C. Kuru, W. Netkueakul, M. Pelin, S. Keshavan, Z. Donaldson, C. J. Campbell, Nanotoxicology 2014, 8, 824. Song, B. Ma, J. Gómes, E. V. Abalos, L. A. V. de Luna, T. Loret, [19] E.-J. J. Park, S. J. Lee, K. Lee, Y. C. Choi, B.-S. S. Lee, G.-H. H. Lee, A. Fordham, M. Drummond, N. Kontis, G. Anagnostopoulos, G. D.-W. W. Kim, J. Appl. Toxicol. 2017, 37, 296. Paterakis, P. Cataldi, A. Tubaro, C. Galiotis, I. Kinloch, B. Fadeel, C. [20] J. H. Shin, S. G. Han, J. K. Kim, B. W. Kim, J. H. Hwang, J. S. Lee, J. H. Bussy, K. Kostarelos, T. Buerki-Thurnherr, M. Prato, A. Bianco, P. Lee, J. E. Baek, T. G. Kim, K. S. Kim, H. S. Lee, N. W. Song, K. Ahn, I. Wick, J. Hazard. Mater. 2022, 435, 129053. J. Yu, Nanotoxicology 2015, 9, 1023. [45] L. Newman, D. A. Jasim, E. Prestat, N. Lozano, I. De Lazaro, Y. Nam, [21] J. K. Kim, J. H. Shin, J. S. Lee, J. H. Hwang, J. H. Lee, J. E. Baek, T. G. B. M. Assas, J. Pennock, S. J. Haigh, C. Bussy, K. Kostarelos, ACS Kim, B. W. Kim, J. S. Kim, G. H. Lee, K. Ahn, S. G. Han, D. Bello, I. J. Nano 2020, 14, 10168. Yu, Nanotoxicology 2016, 10, 891. [46] L. Newman, A. F. Rodrigues, D. A. Jasim, I. A. Vacchi, C. Ménard- [22] M. C. Duch, G. R. S. Budinger, Y. T. Liang, S. Soberanes, D. Urich, S. Moyon, A. Bianco, C. Bussy, K. Kostarelos, Cell Rep 2020, 1, 100176. E. Chiarella, L. A. Campochiaro, A. Gonzalez, N. S. Chandel, M. C. [47] H. H. Gustafson, D. Holt-Casper, D. W. Grainger, H. Ghandehari, Hersam, G. M. Mutlu, Nano Lett. 2011, 11, 5201. Nano Today 2015, 10, 487. [23] R. Li, L. M. Guiney, C. H. Chang, N. D. Mansukhani, Z. Ji, X. Wang, [48] A. Schinwald, K. Donaldson, Part. Fibre Toxicol. 2012, 9, 34. Y.-P. P. Liao, W. Jiang, B. Sun, M. C. Hersam, A. E. Nel, T. Xia, ACS [49] J. Russier, E. Treossi, A. Scarsi, F. Perrozzi, H. Dumortier, L. Nano 2018, 12, 1390. Ottaviano, M. Meneghetti, V. Palermo, A. Bianco, Nanoscale 2013, [24] S. Bengtson, K. B. Knudsen, Z. O. Kyjovska, T. Berthing, V. Skaug, 5, 11234. M. Levin, I. K. Koponen, A. Shivayogimath, T. J. Booth, B. Alonso, A. [50] H. Lin, Z. Song, A. Bianco, J. Environ. Sci. Heal. – Part B 2021, 56, 333. Pesquera, A. Zurutuza, B. L. Thomsen, J. T. Troelsen, N. R. Jacobsen, [51] J. Ma, R. Liu, X. Wang, Q. Liu, Y. Chen, R. P. Valle, Y. Y. Zuo, T. Xia, S. U. Vogel, PLoS One 2017, 12, e0178355. Liu, ACS Nano 2015, 9, 10498. Small 2023, 19, 2301201 2301201 (16 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH 16136829, 2023, 39, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202301201 by Readcube (Labtiva Inc.), Wiley Online Library on [27/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com [52] R. G. Mendes, A. Mandarino, B. Koch, A. K. Meyer, A. Bachmatiuk, [61] R. Kurapati, J. Russier, M. A. Squillaci, E. Treossi, C. Ménard-Moyon, C. Hirsch, T. Gemming, O. G. Schmidt, Z. Liu, M. H. Rümmeli, Nano A. E. Del Rio-Castillo, E. Vazquez, P. Samorì, V. Palermo, A. Bianco, Res. 2017, 10, 1980. Small 2015, 11, 3985. [53] P. M. Hinderliter, K. R. Minard, G. Orr, W. B. Chrisler, B. D. Thrall, J. [62] V. E. Kagan, N. V. Konduru, W. Feng, B. L. Allen, J. Conroy, Y. Volkov, G. Pounds, J. G. Teeguarden, Part. Fibre Toxicol. 2010, 7, 36. I. I. Vlasova, N. A. Belikova, N. Yanamala, A. Kapralov, Y. Y. Tyurina, [54] S. Keshavan, P. Calligari, L. Stella, L. Fusco, L. G. Delogu, B. Fadeel, J. Shi, E. R. Kisin, A. R. Murray, J. Franks, D. Stolz, P. Gou, J. Klein- Cell Death Dis. 2019, 10, 569. Seetharaman, B. Fadeel, A. Star, A. A. Shvedova, Nat. Nanotechnol. [55] Y. J. Lu, Y. H. Wang, R. S. Sahu, J. P. Chen, B. S. Dash, P. J. Chung, H. 2010, 5, 354. W. Yang, E. Y. Chuang, T. L. Hwang, ACS Appl. Mater. Interfaces 2020, [63] G. P. Kotchey, S. A. Hasan, A. A. Kapralov, S. H. Ha, K. Kim, A. A. 12, 40141. Shvedova, V. E. Kagan, A. Star, Acc. Chem. Res. 2012, 45, 1770. [56] M. Cao, R. Cai, L. Zhao, M. Guo, L. Wang, Y. Wang, L. Zhang, X. Wang, [64] G. P. Kotchey, Y. Zhao, V. E. Kagan, A. Star, Adv. Drug Delivery Rev. H. Yao, C. Xie, Y. Cong, Y. Guan, X. Tao, Y. Wang, S. Xu, Y. Liu, Y. Zhao, 2013, 65, 1921. C. Chen, Nat. Nanotechnol. 2021, 16, 708. [65] S. P. Mukherjee, B. Lazzaretto, K. Hultenby, L. Newman, A. F. [57] S. De Oliveira, E. E. Rosowski, A. Huttenlocher, Nat. Rev. Immunol. Rodrigues, N. Lozano, K. Kostarelos, P. Malmberg, B. Fadeel, Chem. 2016, 16, 378. 2018, 4, 334. [58] D. Elgrabli, W. Dachraoui, C. Ménard-Moyon, X. J. Liu, D. Bégin, S. [66] G. Peng, M. F. Montenegro, C. N. M. Ntola, S. Vranic, K. Kostarelos, Bégin-Colin, A. Bianco, F. Gazeau, D. Alloyeau, ACS Nano 2015, 9, C. Vogt, M. S. Toprak, T. Duan, K. Leifer, L. Bräutigam, J. O. Lundberg, 10113. B. Fadeel, Nanoscale 2020, 12, 16730. [59] C. Bussy, C. Hadad, M. Prato, A. Bianco, K. Kostarelos, Nanoscale [67] A. A. Shvedova, A. A. Kapralov, W. H. Feng, E. R. Kisin, A. R. Murray, 2016, 8, 590. R. R. Mercer, C. M. St Croix, M. A. Lang, S. C. Watkins, N. V. Konduru, [60] S. P. Mukherjee, A. R. Gliga, B. Lazzaretto, B. Brandner, M. Fielden, B. L. Allen, J. Conroy, G. P. Kotchey, B. M. Mohamed, A. D. Meade, Y. C. Vogt, L. Newman, A. F. Rodrigues, W. Shao, P. M. Fournier, M. S. Volkov, A. Star, B. Fadeel, V. E. Kagan, PLoS One 2012, 7, e30923. Toprak, A. Star, K. Kostarelos, K. Bhattacharya, B. Fadeel, Nanoscale [68] G. S. Papanai, I. Sharma, B. K. Gupta, Mater. Today Commun. 2020, 2018, 10, 1180. 22, 100795. Small 2023, 19, 2301201 2301201 (17 of 17) © 2023 The Authors. Small published by Wiley-VCH GmbH