Single-cell Transcriptomics Uncovers The Impacts Of Titanium Dioxide Nanoparticles On Human Bone Marrow Mesenchymal Cells

Nanomaterials have been utilized in various fields,such as biomedicine,electronic devices,environmental protection,as well as food industry.The application of nanomaterials has become more and more extensive.As a result,the safety issues and potential biological effects remains a major concern of nanotechnology.Due to unique physicochemical properties such as physicochemical stability,biochemical inertness,environmental friendliness,and low toxicity,titanium dioxide nanoparticles（TiO₂ NPs）have been widely used in nanotechnology and nanomedicine.TiO₂ NPs may directly enter the human body through occupational exposure and the use of products containing TiO₂NPs,and may also affect human health indirectly through TiO₂ NPs emitted into the environment during production.Therefore,the widespread application of TiO₂ NPs has raised concerns regarding the health risks of TiO₂ NPs exposure on humans.Controversy about the potential risks of TiO₂ NPs,especially genotoxicity,may be related to the complexity of physicochemical properties of TiO₂ NPs,the heterogeneity of physiological systems,and the limitations of existing nanotoxicity detection methods.Besides,the morphology,particle size,and surface area of TiO₂ NPs are critical for their internalization,subcellular localization,accumulation and elimination,and toxicity.Therefore,there is an urgent need to develop more sensitive methods for the safety assessment of TiO₂ NPs and gain insight into the mechanism of the potential toxicity.The safety assessment and toxicological mechanism research of nanomaterials benefit from the development of omics technologies.Different omics techniques,including transcriptomics,proteomics,epigenetics,and metabolomics,provide a global understanding of the complex responses of nanomaterials exposure.Some transcriptomic studies of TiO₂ NPs exposure have revealed the new potential mechanisms of toxicity and identified biomarkers of TiO₂ NPs exposure.However,due to the inherent heterogeneity of cells and the complexity of the physicochemical properties of TiO₂ NPs,the gene expression profiles provided by population-level RNA sequencing（RNA-seq）still cannot fully address the controversy over the toxicity of TiO₂ NPs.Single-cell RNA sequencing（sc RNA-seq）has the ability to detect cellular responses to nanoparticle exposure at the single-cell level and to identify heterogeneous responses that may be masked in population-level assays.Therefore,sc RNA-seq could further elucidate the transcriptional signatures of cellular responses after nanoparticles exposure and resolve the current controversy.Based on a microfluidic system,we capture bone marrow mesenchymal stem cells（BMSCs）after exposure to 5,14,54,135,and 228 nm TiO₂ NPs.Single cell c DNA was obtained by REPLI-g WTA Single Cell Kit,and c DNA library was prepared by the NEBNext Library Prep kit,and then sequenced on the Illumina Hi Seq 2000 platform.The transcriptional characteristics of BMSCs after exposure to TiO₂ NPs were preliminarily revealed at the single-cell level.On this basis,the pivotal pathways of cells in response to TiO₂ NPs exposure,the effects of particle size on cell response patterns,the heterogeneous gene expression profiles,and the generation and scavenging of reactive oxygen species（ROS）were discussed.The transcriptional landscape at single-cell level provides new insights into the cytotoxicity of TiO₂ NPs and he underlying mechanisms.（1）Central role of genotoxicity-related pathways in cellular responses to TiO₂ NPs exposureAccording to the results of cell viability,apoptosis,hemolysis rate,and BMSCs surface markers,a dose of 100μg/m L was chosen to conduct sequencing and verifying experiments.After sc RNA-seq,quality control,and filter,we performed principal component analysis（PCA）with all detected genes.In the PCA plot,TiO₂ NPs exposed cells and unexposed cells were separated.In the exposed cells,5 and 14 nm TiO₂ NPs exposed cells are clustered together and separated from the larger size of 54,135 and 228nm TiO₂ NPs exposed cells.Compared with the control group,2058,2458,3435,3985,and 3545 differentially expressed genes（DEGs）were identified in different size exposure groups,respectively.Functional GO and KEGG analyses of all these DEGs revealed that some pathways,such as DNA repair,response to virus,DNA replication,and cell cycle were altered in the exposure of TiO₂ NPs.GO and KEGG enrichment analysis of shared DEGs in all sizes TiO₂ NPs of TiO₂ NPs treated cells showed that genotoxicity-related DNA damage and cell cycle pathways play central role in responses to TiO₂ NPs exposure.The numbers of DNA repair and cell cycle-related genes and their total expression levels increased with the size increasing.Subsequently,some genes in DNA damage and cell cycle pathways and important genes in DNA damage response（DDR）were verified by q PCR,western blotting,and immunofluorescence.The up-regulation of CINP,RPA3,and PRKACA,and the down-regulation of TYMS and MCM2 was consistent with the sequencing results.In addition,TiO₂ NPs treatment could induce the nuclear translocation of CINP and CHD1L.However,assessments such asγH2AX foci formation,cell cycle distribution,and Ed U incorporation,which are commonly used for genotoxicity assessment of nanomaterials,did not show obvious alterations at the same concentration.Taken together,our results demonstrate that TiO₂ NPs exposure can induce alterations in the expression of numerous genotoxicity-related genes,some of which are more sensitive than conventional genotoxicity assays.（2）Distinct response patterns in different sizes TiO₂ NPs and heterogeneous responses within the same nanoparticles exposureUnsupervised hierarchical clustering analysis showed that the smaller sizes TiO₂NPs（5 nm and 14 nm）exposed cells were clustered together,and that the bigger sizes TiO₂ NPs（54 nm,135 nm,and 228 nm）exposed cells were grouped in another cluster,indicating significant differences in cellular responses between smaller and bigger TiO₂NPs.Functional GO and Reactome Pathway analysis revealed that response to virus and the corresponding cytokine and interferon signaling pathway were the key to this difference.The expression of representative DEGs were validated by q PCR,western blotting,and immunofluorescence.Smaller TiO₂ NPs exposure dramatically increased the expression of antiviral activity related DEGs,such as IFITM1,MX1.Although IL-5levels were elevated in the 135 nm group,and IL-17 and TNFαlevels were elevated in the 228 nm group compared with the control group;no significant differences were observed between the smaller and bigger sizes TiO₂ NPs groups.Therefore,exposure to5 nm and 14 nm TiO₂ NPs may lead to changes in cellular gene expression similar to antiviral defenses,and IFITM1,MX1 may serve as biomarkers of cellular responses in<15 nm TiO₂ NPs exposure.Heat map and unsupervised hierarchical clustering analysis showed that some cells（S28＿5 nm,S39＿14 nm,S59＿54 nm,S81＿228 nm and S89＿228nm）were divergent cells due to the different responses from other cells.GO analysis revealed that some specific pathways were also involved in the response to TiO₂ NPs exposure,such as cytoskeleton organization,response to estradiol stimulus,and axon guidance.Some cells showed extensive down-regulation of metabolic processes and transcription regulation;on the contrary,some cells showed marked up-regulation of metabolism-related pathways.Under the same treatment conditions,apoptosis-related signaling pathways were up-regulated in some cells and down-regulated in others.In summary,particle size is one of the key factors affecting cellular responses,but exposure of TiO₂ NPs with the same particle size may also induce heterogeneous cellular responses.（3）TiO₂ NPs exposure resulted in increased cellular ROS and unaffected osteogenic differentiation potential in BMSCsAlthough only 10 DEGs were enriched in oxidative stress-related pathways,and the numbers and their total expression levels of oxidative stress-related genes were low;we confirmed that TiO₂ NPs exposure can lead to ROS over-production in BMSCs by staining with Cell ROX Green,DHE and Mito SOX Red.Corresponding elevated levels of Superoxide dismutases（SOD）and decreased ratio of GSH/GSSG（glutathione/oxidized glutathione）were found in all TiO₂ NPs exposed group.In addition,TiO₂ NPs exposure also resulted in increased phosphorylation levels of NRF2 and increased m RNA and protein expression of its transcriptional target genes HO-1,NQO1,GGT1 and GCLC.Compared with untreated BMSCs,TiO₂ NPs exposure did not lead to obvious inhibition in osteogenic differentiation,nor affect the protein levels of main regulators of osteogenic differentiation,such as BMP4,OGN,RUNX2,ERK1/2,AKT,FAK,and phosphorylation levels of ERK1/2,AKT,and FAK.Taken together,the above results suggest that TiO₂ NPs treatment under these sizes and concentrations could induce a controllable ROS over production,but did not cause oxidative stress and reduction of osteogenic potential in BMSCs.（4）The accumulation of TiO₂ NPs in vivo may elevate the levels of ROSBalb/c mice were treated with TiO₂ NPs by intravenous injection at the concentration of 100 mg TiO₂ NPs/kg body weight.The hematoxylin-eosin（H&E）staining in tissue sections showed that TiO₂ NPs mainly accumulated in the livers and spleens.The accumulations in the bone marrow,lung,hearts,kidney,and brain were also observed.The 14 nm,54 nm and 135 nm TiO₂ NPs injection led to a higher TiO₂ NPs accumulation in the livers,spleens,and bone marrows,compared with the 5 nm and 228 nm groups.The accumulation of TiO₂ NPs in these organs did not lead to obvious damage and inflammation,but induced a significant up-regulation of SOD levels and a significant du-regulation in the GSH/GSSG ratio in the liver compared with controls.In addition,we also found a slight increase in the number of white blood cells due to the injection of TiO₂NPs,suggesting that TiO₂ NPs injection may trigger an immune response in vivo.However,TiO₂ NPs exposure did not affect thrombocytes and red blood cell indices,such as thrombocyte counts,red blood cell counts,hematocrit,hemoglobin,mean corpuscular hemoglobin,mean corpuscular hemoglobin concentration,and mean corpuscular volume.Taken together,these results indicate that TiO₂ NPs exposure in vivo may lead to inflammation and oxidative stress risks but no changes in hematopoietic homeostasis.