MALDI-MS Imaging Reveals Asymmetric Spatial Distribution of Lipid

Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST

Article

MALDI-MS imaging reveals asymmetric spatial distribution of lipid metabolites from bisphenol S-induced nephrotoxicity Chao Zhao, Peisi Xie, Ting Yong, Hailin Wang, Arthur Chung, and Zongwei Cai Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b04540 • Publication Date (Web): 12 Feb 2018 Downloaded from http://pubs.acs.org on February 14, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

MALDI-MS imaging reveals asymmetric spatial distribution of lipid metabolites from bisphenol S-induced nephrotoxicity

Chao Zhao a,b, Peisi Xie a, Ting Yong a, Hailin Wang b, Arthur Chi Kong Chung a, Zongwei Cai a,*

a

State Key Laboratory of Environmental and Biological Analysis, Department of

Chemistry, Hong Kong Baptist University, Hong Kong SAR, China. b

State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for

Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, China

* Address correspondence to Prof. Zongwei Cai, State Key Laboratory of Environmental and Biological Analysis, Department of Chemistry, Hong Kong Baptist University, Hong Kong SAR, China. Tel. +852-34117070; Fax. 34117348. E-mail: [email protected] 1

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract With the continuous exposure of environmental pollutants in organisms, determination of abundance variation and spatial distribution of lipids might expand our understanding of toxicological mechanisms occurring in kidney. Herein, an integrated method involving mass spectrometry (MS)-based lipidomics and matrix-assisted laser desorption/ionization-MS imaging (MALDI-MSI) was developed for the study of nephrotoxicity in mice exposed to 10 and 100 µg bisphenol S (BPS)/kg body weight/day. The BPS exposure remarkable perturbed abundances of 91 potential markers that mainly involved in five metabolic pathways. We elucidated the lipids spatial heterogeneity by using morphological analysis, probabilistic latent semantic analysis and co-registered multimodal three-dimensional (3D)-MSI. In morphological analysis, both 10 and 100 µg BPS induced significant nephrotoxicity to mice, including glomerular necrosis in renal cortex, cloudy swelling in renal medulla, interstitial collapsing in renal pelvis. Significant differential signaling lipids such as sphingomyelin (SM) (d22:0/20:4), ceramide (Cer) (d18:2/24:1) and sphingosine (d18:0) related to inflammation were found to be up-regulated and co-localized in renal cortex, medulla and pelvis, respectively. Also, seven significant differential lipids, which are considered involved in membrane homeostasis and cellular function, were found to be co-localized in renal cortex. The observed significant variations of morphology, lipid accumulation and metabolism in renal cortex implicated that lipids in renal cortex were more sensitive to BPS exposure than renal medulla and pelvis. Moreover, we reconstructed a 3D-MSI model of kidney and identified two heterogeneous-related sub-structures in renal cortex and pelvis upon the 100 µg BPS exposure. It might be used in novel specificity evaluation and early diagnosis for 2


Page 2 of 33

Page 3 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


environmental pollutants-induced kidney diseases. Keywords: Lipidomics; LC-MS/MS; MALDI- MSI; Bisphenol S; Nephrotoxicity.

3



Bisphenol S (BPS), a major bisphenol A (BPA) alternative, has been applied in industrialized production and daily applications, including electro planting solvent 1, phenolic resin1 and as a constituent of thermal paper.2 Due to its widespread use, BPS has been detected in various matrices.3-6 BPA is one of the most studied endocrine disrupting chemicals (EDCs). Exposure to BPA has been intimately linked with the incidence of chronic diseases on non-reproductive organs, including kidney diseases, diabetes and cardiovascular diseases.7,8 Human exposure to BPA has resulted in many adverse effects on renal function, contributing to progressive accumulate of renal damage.9 Previous investigations have demonstrated that BPA was well absorbed by oral route. Völkel et al. found urinary recovery in human volunteers of 84% of the dose in females and 97% in males by oral administration, demonstrating the extensive absorption of BPA through oral administration.10,11 Absorption of BPA through oral route caused extensive damage to kidney and liver in mice model.12 BPA also acts as a critical marker for renal diseases and displays nephrotoxicity. In kidney tissue, BPA up-regulates cytochrome P450 aromatase activity by using of steroidogenesis and increases estrogen metabolism, resulting in nephrotoxicity.13 BPS has estrogenic, androgenic, antiestrogenic and antiandrogenic activity similar to BPA. BPS also displayed other biological effects in vitro, including DNA damage and cellular dysfunction.14 However, toxicity and the molecular mechanisms underlying the nephrotoxicity of BPS in vivo remain elusive. Many studies showed that abnormal metabolism is one of the critical characteristics for the study of renal toxicology response to environmental pollutants, including the variation of metabolite abundance, metabolic pathway and metabolic phenotype. Following the exposure 4


Page 4 of 33

Page 5 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of environmental stimuli, the state of normal or lesion is a consequence of multi-interaction between small biomolecules (metabolites and lipids) and biomacromolecule (proteins and genes).15-18 Metabolites that are the end productes of metablic system provide precious information for the variation of orgainsm functions. Metabolic profiling is widely used in the assessment of nephrotoxicity induced by environmental xenobiotics.19 It has been proved useful for discovery and identification of renal toxicity-related markers both in vitro and in vivo.20,21 As a branch of metabolites, lipids are involved in structural components, energy storage and biochemical processing in living systems.22,23 Lipids deserve particular attention as they play regulatory and signaling roles on molecular recognition processes beyond genes and proteins, such as membrane architecture, functional lipid biosynthesis and membrane permeability.24-28 In addition to changes in the lipidome, revealing metabolites spatially in situ also contributes to understanding tissue microenvironments and toxicology of environmental pollutants in pathology and physiology. Matrix-assisted laser desorption/ ionization-mass spectrometry (MALDI)- mass spectrometry imaging (MSI) has been used to identify various metabolites and profile their spatial distribution in tissue sections by a label-free method.29 Moreno-Gordaliza et al.30 detected the spatial distribution of lipids on kidney sections from healthy control rats and the rats treated with antitumor drug cisplatin by using MSI. They found the distribution pattern of phospholipid species due to cisplatin-induced nephrotoxicity. Distribution variation of lipids showed structural and signaling processes in damaged kidney. Phospholipids have the potential to serve as therapeutic targets or markers of nephrotoxicity assessment. Liu et al.31 elucidated the mechanisms of renal fibrosis that is the critical 5



pathophysiological pathway in chronic kidney disease by MALDI-MSI in unilateral ureteral obstruction mouse model. Metabolic disturbances were observed in glycolysis, ATP metabolism, TCA cycle and fatty acids metabolism in the process of renal fibrosis. Rao et al.32 investigated the lipidome in early ischemia reperfusion-related acute kidney injury in mice by using sequential window acquisition of all theoretical spectra-MS and MALDI-MSI. They found two potential lipid markers, namely phosphatidylcholine (38:1) and phosphatidylethanolamine (42:3). Up-regulated phosphatidylcholine (38:1) occurred in proximal tubules following ischemia reperfusion. In our study, metabolic fingerprints were obtained from the analysis of LC-MS/MS combined with MALDI-MSI for the investigation of significant variations in specific regions of kidney, which might open the way to explore whether the exposure to EDCs might result in morphological variation and lipidome disturbance in mice for risk assessment. Materials and methods Mice model and BPS administration. Briefly, female BALB/c nude mice aged 4-weeks-old were obtained from the Chinese University of Hong Kong (CUHK) and were maintained in sterile individually ventilated cages (IVC) under the following conditions: 20 ± 2 oC of temperature, 45 ± 10% of relative humidity with a 12-h light/dark cycle in each day. Female nude mice were randomly divided into three groups (n=6 for each) including control group (olive oil), 10 µg BPS group (10 µg /kg body weight (wb)/ day) and 100 µg BPS group (100 µg /kg wb/ day). All mice were treated by gastric infusion method with a dosage of BPS in olive oil once daily. The body weights of mice were measured every week, and the physical state was monitored every 4 days. After 56 days consecutive administration of olive 6


Page 6 of 33

Page 7 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


oil or BPS, the mice were sacrificed and the kidney samples were collected for lipidomics analysis by using LC-MS/MS and MALDI-MSI. Histological analysis of kidney with BPS exposure. Sections (14 µM) were prepared and stained with haematoxylin and eosin (H & E) staining to assess kidney injury following BPS exposure. Histological changes of kidney was determined by Leica DM 2500 microscope (Leica, Wetzlar, Germany) (H&E, 40×magnification). Metabolite extraction from kidney tissues. Kidney tissues were dissected from control (olive oil), 10 and 100 µg BPS group and were washed with ice-cold PBS, quick-frozen with liquid nitrogen. Kidney tissues were stored at -80 oC prior to metabolites extraction. Fifty mg of samples were selected near the maximum cross-section area of kidney and were homogenized using a Polytron PT2100 homogenizer (Kinematica, Lucerne, Switzerland) in 600 µL of ice-cold methanol and 150 µL of water, and then 450 µL of chloroform was added, homogenate was vortexed for 30 min at the 3,400 rpm on ice, 150 µL water was added to promote phase separation. The homogenate was again vortexed for 1 min and allowed to equilibrate at room temperature for 5 min. The mixture was centrifuged at 12,000 g for 15 min at -6 oC. The bottom layers were collected and transferred to another 1.5 mL-tube and dried in a Max-Up (NB-504CIR) IR vacuum concentrator (N-Biotek, GyeongGi-Do, Korea) at 4 oC. LC-MS/MS-based lipidomics, data processing, determination of cytokine secretion and gene expression. The experiments were carried out based on the procedures described in Zhao et al.33 Briefly, a 2.1 × 100 mm Hypersil Gold 1.9 µm C18 column (Thermo Fisher Scientific Inc., Waltham, MA, USA) was used for the separation of lipids. Mobile phase A contained 7



acetonitrile/water (60:40) with 10 mM NH4HCO3 and 0.1% formic acid. The mobile phase B contained IPA/ ACN (90:10) with 10 mM NH4HCO3 and 0.1% formic acid. The gradient program was as follows: 0.0-30.0 min from 32% B to 90% B, 30.0-32.0 min to 100% B and kept for 3.0 min, 35.0-35.1 min to 32% B and 35.1-40.0 min kept for 32% B. An Orbitrap Fusion Tribird MS system (Thermo Fisher Scientific Inc., Waltham, MA, USA) equipped with a heated electrospray ionization (HESI) source was used (Figure S-1 and S-2). For obtaining comprehensive metabolite

coverage, ESI-MS was operated in both positive and negative ionization modes. The precursor ions were fragmented following the full scan acquisition. In our study, we choose the alternating full, MS2 and MS3 scan mode to realize data analysis of lipid profiling. Thus, the fragments of eluted lipids were detected after full, MS2 and MS3 scans without limiting to predefined precursor ion and neutral loss ion. We examined the MS conditions in both positive and negative ionization modes, which included the normalized collision energy (NCE), data-dependent high-energy collision dissociation (dd HCD)-MS2 acquisition mode, maximum ion injection time (Table S-1). Sufficient data points, along with relative abundance of precursor and product ions, were produced in MS2 and MS3 acquisitions. LC separation conditions were optimized for better separation efficiency, reduced analysis time and better repeatability of retention time from polar to nonpolar lipids. The raw data of LC-MS/MS were processed by using LipidSearch software for automated extraction, identification and relative quantitation of the lipid compounds (Figures S-2). Base on the polar head groups and fatty acids chains, [M+H]+ and [M+NH4]+ ions in positive ionization mode as well as [M-H]- and [M+HCOO]- ions in negative ionization mode were extracted and identified by neutral loss and precursor ion scan with a specific algorithm. The 8


Page 8 of 33

Page 9 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


retention time window and accurate peak areas were used to align and quantify the precursor ions. NCE and m-score from LipidSearch software were matched with each other at the appropriate fragmentation. The comparative analysis was performed on the kidney tissues in control, 10 and 100 µg BPS group by T-test statistics. At the same time, the data were imported into SIMCA-P software for multivariate pattern recognition analysis. The results of comparative analysis from LipidSearch and SIMCA-P were verified and complemented to each other. MALDI-MSI. Kidney was fixed by solid glue on the cutting stage, and was sectioned at 14 µm thickness using a CryoStar Nx70 cryostat (Thermo Fisher Scientific, Walldorf, Germany) at -20 °C, and thaw-mounted onto indium tin oxide (ITO) coated glass slides. The ITO-slide was dried in a vacuum desiccator for 20 min before matrix spraying. Solutions of N-(1-naphthyl)-ethylenediamine dihydrochloride (NEDC) for matrix and mixture standard solutions containing small molecules were prepared as described in Wang et al.34 The matrix was sprayed onto the sections mounted onto ITO-slides using an automatic matrix sprayer (ImagePrep, Bruker Daltonics, Billerica, MA) with the spraying protocol of Wang et al.34 MALDI-MSI was carried out as described in Wang et al.34 and Liu et al.31 with some modifications. In brief, the experiments were performed on an rapifleXTM MALDI TissuetyperTM (Bruker Daltonics) equipped with a smartbeamTM 3D laser in the single mode. The mass spectra data were acquired at a mass range of m/z 200-1,100 in the negative reflector ion mode by averaging signal from 1,000 shots at 3.0 × 2810 volts detector gain and 30% laser power. The selection of negative ionization mode was related to the selection of matrix. NEDC was found to optimize homogeneous crystallization in negative ion mode, 9



which achieved batch stability via automatic deposition and shorter drying time to reduce matrix diffusion and sample delocalization. NEDC matrix worked exclusively in negative ion mode with abundant peaks and high spatial resolution.34 The other parameters were optimized and fixed during the whole experiments, including a reflector voltage of 20.84 kV, a lens voltage of 11.00 kV, an ion source voltage of 20 kV, a pulsed ion extraction time of 100 ns and the matrix suppression of m/z 270. The spatial resolution for MALDI-MSI was acquired at 50 µm. The external mass calibration was performed before data acquisition. Four lipids ([PE (38:4)-H]-, 766.537 Da; [PI (38:4)-H]-, 885.549 Da; [ST (d18:1/22:0)-H]-, 878.602 Da; [ST (d18:1/24:0)-H]-, 906.633 Da) were selected as internal mass calibration in the linear mode to monitor the MSI performances. MALDI-MSI raw data and all ion images were first opened in flexImaging 5.0 software (Bruker Daltonics) and then analyzed, calibrated and imported into SCiLS Lab 2016a software (Bruker Daltonics). The standard processing pipeline was used for statistical analysis of MSI data, including segmentation, weak denoising, normalization of total ion count and probabilistic latent semantic analysis (pLSA). Three-dimensional (3D) reconstruction of mice kidney. MSI data of all consecutive sections were imported into SCiLS Lab 2016a software. For creation of 3D project, the first of sections was placed in the center of the visual field. And then, each of sections was positioned over the first section by adjusting the variation of the X- or Y-coordinate. For the location of Z-coordinate, the thickness of section (µm) and the distance of serial sections were entered from the first section to the last section. The 3D reconstruction of kidney was started after the compensation of translations and rotations in X-, Y- and Z-coordinate. The 10


Page 10 of 33

Page 11 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


comparative results from LipidSearch, SIMCA-P and SCiLS Lab 2016a software served the purpose to complement and verify each analysis. Results and discussion Assessment of BPS-induced nephrotoxicity. BPS dosage was adjusted for body weight daily. The 10 µg and 100 µg/kg bw/day dose was selected because it appeared to induce clear adverse effects in our previous studies (such as hepatic damages 35) and tumor proliferation in breast cancer xenografts [Figure S-3]. The exposure route of BPA or BPS, was reported to occur mostly via food and beverages from polycarbonate plastics. As oral BPA or BPS undergoes extensive presystemic elimination whereby glucuronidation accounts to more than 90%, the activity of the metabolites is important for risk assessment. Dermal absorption of BPA or BPS is limited. The dermal contact area for potentially BPA- or BPS-containing materials is small, and dermal penetration of BPA or BPS is limited.

36, 37

Thus, we selected

gastric infusion as exposure route of BPS in our experiments. Morphological alterations induced by control (olive oil), 10 and 100 µg BPS were shown in Figure 1. Control group displayed normal status of renal tubules and glomeruli (Figure 1A, 1B and 1C). In contrast, distinct morphological alterations were observed in the morphological assessment on H & E-stained kidney sections, showing the main kidney damage features in tubules and glomeruli (Figure 1D, 1E, 1G and 1H) (such as glomerular necrosis, cloudy swelling in proximal convoluted tubules and distal tubular) during 10 and 100 µg BPS exposure with respect to control group (Figure 1A, 1B and 1C). These damage features presented in renal cortex and medulla particularly. In renal pelvis region, interstitial widening, thinning and collapsing of tubular epithelium were examined (Figure 1F and 1I) and compared with control group. The 11



results shown in Figure 1 demonstrated that BPS induced significant kidney damage that located in renal cortex, medulla and pelvis. As a mediator of immune function and initiator of renal damage, inflammatory cytokines plays central role in chronic and acute diseases.38-40 Herein, we found that levels of cytokines in kidney increased significantly in 10 and 100 µg BPS group compared with control group. Treatment with 10 and 100 µg BPS varified the secretion and transcript levels of immune-related cytokines (Figure S-4). In 100 µg BPS group, the secretion of pro-inflammatory cytokines IL-6, IL-1β and TNF-α increased at 1.7, 1.2 and 2.3-fold, respectively. The secretion of anti-inflammatory cytokines IL-10 and TGF-β decreased at 0.8- and 0.7-fold, respectively (Figure S-4A). Moreover, the transcript levels of pro-inflammatory cytokines increased, whereas the transcript levels of anti-inflammatory cytokines declined in 10 and 100 µg BPS group compared to control group (Figure S-4B). Therefore, the BPS exposure induced secretion of pro-inflammatory cytokines in mouse kidney, which might subsequently result in kidney damage. Lipidomics analysis of BPS effect on kidney metabolic profiling. Renal injury may induce the change in renal lipidome.32 Lipid contents and inflammatory signaling are closely linked and mutually influenced in renal diseases. Lipids change inflammatory responses by up- or down-regulating lipid-related enzymes. Meanwhile, inflammatory signals modulate lipid-related metabolism and pathway positively or negatively. To better unstandard the role of lipid metabolism in toxicitical mechanisms of BPS and inflammation, kidney tissues were analyzed by LC-MS/MS. The lipids were eluted in the order of glycerophospholipids (GPs; 3-20min) (phosphatidylcholine [PC], phosphatidylethanolamines [PE], phosphatidylserine 12


Page 12 of 33

Page 13 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[PS], phosphatidylglycerols [PG] and phosphatidylinositols [PI]), sphingolipids (SPs; 10-19 min) (sphingomyelin [SM], ceramide [Cer] and sphingosine [So]) and glycerolipids (GLs; 12-25 min) (diacylglycerols [DAG] and triacylglycerol [TAG]), with the decrease of hydrophobicity (Figure S-1). As shown in Figure 2A and 2B, the partial least squares discriminant analysis (PLS-DA) score plots for results from the analyses in positive and negative ionization modes revealed that 10 and 100 µg BPS group vs control groups showed very distinct separation on lipidome (R2Y = 0.993 and Q2 = 0.916 for positive ionization mode; R2Y = 0.991 and Q2 = 0.920 for negative ionization mode). The identified metabololites that contibuted to the metabolic differentiation were selected according to the threshold of variable importance (VIP) value >1.5 by using SIEVE, LipidSearch and SIMCA-P software. The potential markers were identified and shown in Table S-2 and Figure S-5. The heatmap chart (Figure 2E) also displayed remakable variation of lipid metabolism in 10 and 100 µg BPS group compared to the control groups. Abundance of lipid metabolite displayed significant up-regulation in SM, Cer, So, PS, PG and TAG as well as remarkable down-regulation in PI, PE, PC and DAG in the kidney tissues of 10 and 100 µg BPS-treated mice. The obtained results indicated that the lipidome of kidney was perturbed significantly after the BPS exposures. Because lipids are markers for the integrity of cell membrane structure and cell phenotype, we then determined whether any alterations in relative composition of subclasses of lipidome occurred in the kidney tissues of 10 and 100 µg BPS-treated mice (Figure 2C and 2D). Composition of nine subclasses, including SM, Cer, PC, PE, PG, PI, PS, DAG and TAG, were changed remarkablely in the 100 µg BPS group. The results showed that the 13



compositions of fatty acid chains had remarkable difference between control and BPS-treated group. We further investigated the remodelling of GPs and calculated the change of composition of fatty acid chain (Figure 2D). We found that ether-linked residues (PC and PE) and poly-unsaturated fatty acids (PUFA)-acyl chains of PC, PE, PG and PI were enriched in 100 µg BPS group with 1.2- to 2.2-fold increase, while saturated fatty acids (SFA) and mono-unsaturated fatty acids (MUFA-acyl) prone were deficient in kidney with 0.7- to 0.9-fold reduction, compared to the control group (Figure 2D). PUFA, ether-linked PC and PE played an important role in cell signaling and membrane structure. They included docosahexaenoic acid (22:6), eicosapentaenoic acid (20:5) and arachidonic acid (20:4) that are critical substrates of leukotrienes and prostaglandins.41,42 Our results showed that BPS induced abnormal accumulation of lipid metabolism within kidney, which might result in the instability of cellular membrane and turnover, causing generation of kidney injury. Metabolic pathway of BPS-induced nephrotoxicity in mice. According to the LipidSearch software and KEGG pathway database, we constructed a metabolic pathway network of BPS-induced nephrotoxicity in mice (Figure 3). As shown in Figure 3, Table S-3 and Table S-4, BPS exposure un-regulated the mRNA level of ceramidase (CDase), sphingosine kinase types 2 (SphK2) and acid sphingomyelinase (SMase) in Cer metabolism pathway (L1) and SM hydrolysis (L2). It was demonstrated that lipid networks changed in the cellular sigaling transduction and membrance remodeling after the BPS exposure. The key step in Cer metabolism pathway (L1) is the production of sphingosine 1-phosphate (S1P) from Cer and So, which are catalyzed by CDase and SphK2. SphK2 restricts cell growth and induces apoptosis by the release of cytosolic free calcium and 14


Page 14 of 33

Page 15 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


cytochrome c.43,44 Our results indicated that BPS enhanced the transcription of CDase and SphK2 (Table S-3) for a secretion of pro-inflammatory cytokines in Cer metabolism pathway (L1). In addition, Cer formation might be induced by BPS exposure via SMase (Table S-3), regulating the variation of abundance of SM and So. Moreover, abnormalities of generation and metabolism of Cer in cancer cells is intimately linked with a variety of diseases because Cer can affect signaling responses to stress, generate inflammation and regulate cancer cell cycle and survival.45 We speculated that BPS-induced variation of Cer metabolism in signaling pathway as a second messenger to initiate the inflammation via disturbing the secretion and transcript levels of pro-/anti-inflammatory cytokines and abundance of SM, Cer and So, resulting in nephrotoxicity in mice. Our results of lipidomics analysis on BPS-induced toxicity were consistent with the ionizing radiation-induced Cer functions.46,47 GPs are considered as the major structural lipids and also the precursors of lipid mediators. Upon the BPS exposure, GPs undergo reprogramming of de novo biosynthesis via cross-regulation of PS synthesis pathway (L3), Pemt methylation pathway (L4) and remodeling of FAs (L5) by the rearrangement of substituent groups from PC and PE (Figure 3). In addition, BPS could induce the accumulation of TAG by the synergistic regulation of essential enzymes (diacylglycerol acyltransferase, Dgat, Table S-3) and lipid metabolites (TAG, Figure 3). Upon the BPS exposure, we speculated that GPs and GLs could cause the disruption of biological membrane homeostasis. The pathway perturbations might alter lipid affinity as substrates of enzymes and change toxic responses.48-51 MALDI-MSI confirmation of lipid metabolites for BPS nephrotoxicity. Upon the BPS toxicity evaluation from the morphological assessment, LC-MS and multivariate statistical 15



analysis, kidney sections from 10 and 100 µg BPS-treated mice were analyzed by using MALDI-MSI to investigate the variation of spatial distribution and relative quantitation of GPs (PC, PE, PS, PG and PI), SPs (SM, Cer and So) and GLs (DAG and TAG) in comparison with those from control mice. Figure 4 presented three-cluster pLSA score plots from MALDI-MSI analysis of the whole kidney tissue, renal cortex, medulla and pelvis regions, respectively, according to the structural difference of kidney tissue and spatial distribution properties of BPS nephrotoxicity. The spectra from 10 µg BPS group clustered were more close to the control group than 100 µg BPS group in the whole kidney (Figure 4A) and renal medulla (Figure 4C), suggesting BPS-induced change in lipid metabolism was concentration dependent. In the profiled spectra, distinctive GPs, SPs and GLs were identified by pLSA with four components at 95% confidence intervals. To study more precisely the spatial distribution of lipid markers in renal cortex, medulla, pelvis and the whole kidney sections, we performed a Venn diagram analysis to compare the differential distribution of lipid species (Figure S-6A). In BPS exposure group, a total of 10 lipids were found significantly altered by using segmentation and pLSA, which distributed in the regions of renal cortex, medulla and pelvis. In the renal cortex from the 10 and 100 µg BPS compared to control group, SM (d22:0/20:4), TAG (16:0/14:0/16:0), PS (18:0/22:6) and PG (16:0/16:0) intensities increased significantly, whereas decreased abundances of DAG (18:0/22:6), PE (20:1/20:4) and PI (16:1/18:1) were found (Figure S-6). One Cer and three GPs species localized in renal medulla were detected with up-regulated Cer (d18:2/24:1) and PG (16:0/16:0) and down-regulated PC (20:0e/22:6) and PI (16:1/18:1). In addition, down-regulated PE (20:1/20:4) and up-regulated So (d18:0), 16


Page 16 of 33

Page 17 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TAG (16:0/14:0/16:0) and PS (18:0/22:6) were observed in renal pelvis (Figure S-6A). Three of the most prominent SPs in renal cortex, medulla and pelvis were detected at m/z 807.64±0.15, 644.61±0.15 and 300.30±0.15, corresponding to [SM (d22:0/20:4)-H]-, [Cer (d18:2/24:1)-H]- and [So (d18:0)-H]-, respectively. The identified SPs might serve as markers for implicating activation of inflammatory responses and metabolic disturbance of SPs (Figure 3).52, 53 The box plots were established and confirmed that normalized intensity of three ions were higher in the kidney tissues of 10 and 100 µg BPS group than those control group (Figures S-6B, S-6C and S-6D). The results of MALDI-MSI indicated that kidney tissues of BPS-treated mice have significant injury regions (such as renal cortex, medulla and pelvis), which was consistent with the results of morphological assessment (Figure 1) and cytokine detection (Figure S-4). Above all, the degree of lipid variation and accumulation in renal cortex was higher than those in renal medulla and pelvis, implicating that lipids in renal cortex were more sensitive to BPS exposure than renal medulla and pelvis. It was reported that superoxide activity might be activated by nicotinamide adenine dinucleotide phosphate (NADPH) oxidase to promote reactive oxygen species (ROS) production directly in renal cortex, resulting in nephrotoxicity.54 In addition, we also analyzed BPS in renal cortex, medulla and pelvis regions by UPLC-MS/MS to elucidate the relationship between BPS accumulation and lipid abundance. The highest concentration of BPS was found in renal cortex from 10 (accounted for 0.81-1.21% of the dosed concentrations of BPS; n=6; p< 0.05) and 100 µg BPS group (accounted for 0.91-1.49%). This suggested that BPS induced cortex damage through BPS accumulations and lipid variations with high specificity of spatial distribution. 17



2D and 3D-imaging for kidney heterogeneous induced by BPS. Environmental pollutants critically affect the molecular characteristics of tissues and cause heterogeneous growth. After the analysis of differential metabolites profiling and the MALDI-MSI determination, DAG (18:0/22:6) and SM (d22:0/20:4) in renal cortex, Cer (d18:2/24:1) and PC (20:0e/22:6) in renal medulla and So (d18:0) in renal pelvis were screened and applied as markers to reveal the variation of renal specific regions with the targeted markers upon the 10 and 100 µg BPS exposure (Figure S-6A). Subsequently, serial sections of kidney tissues in control, 10 µg and 100 µg BPS-treated mice were prepared at an equal distance, enabling the use of 3D reconstruction and establishment of multivariate statistical approaches rendering to discover specific localization patterns and heterogeneous sub-structure. Briefly, serial sections were analyzed using MALDI-MSI (Figures 5 and S-7) with a 14 µm thickness and a 70 µm Z-spacing (step-size). To assess tissue heterogeneity induced by BPS, a 3D model of kidney was assembled using a standard processing pipeline of 3D reconstruction (Figure S-2). We combined the ion abundance analysis of lipid markers with spatial segmentation analysis to identify the heterogeneous-related substructure in the whole kidney. Comparing with control group, spatial segmentation and component analysis of MSI in BPS-treated groups showed that four specific classes with four colors (blue, green, yellow and red) were co-localized in (-) Z-coordinate [DAG (18:0/22:6)], renal cortex [SM (d22:0/20:4)], renal medulla [Cer (d18:2/24:1)] and renal pelvis [So (d18:0)] followed by the (+) Z-coordinate, respectively (Figure S-7). As shown in Figure 5 and Figure S-7, two regions of interest were allowed a more detailed analysis of spatial substructure inside the kidney including the renal cortex region (100 µg BPS treated group in Figure 5 (a) and green segments in Figure S-7; 18


Page 18 of 33

Page 19 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


control and 10 µg BPS treated group in Figure S-8) and the renal pelvis region (100 µg BPS treated group in Figure 5 (b) and yellow segments in Figure S-7; control and 10 µg BPS treated group in Figure S-8). The results demonstrated that two heterogeneous-related substructures (labeling a and b in Figure 5 and Figure S-8) and lipid markers [SM (d22:0/20:4) and Cer (d18:2/24:1)] might be used in novel specificity evaluation and early diagnosis for environmental pollutants-induced nephrotoxicity and related diseases (Videos name are “So-control”, “So-10-BPS”, “So-100-BPS”, “SM-control”, “SM-10-BPS” and “SM-100-BPS” in control, 10 µg and 100 µg BPS treated group, respectively). Conclusions An integrated method of LC-MS/MS-based lipidomics and MALDI-MSI was applied successfully for the determination of profile and distribution of lipids in kidney tissues from BPS-treated mice. For the first time, the combined approach was used to investigate BPS nephrotoxicity in mice, including inflammation generation, metabolic disturbance and variation of lipid spatial distribution. The combined analysis revealed significant metabolic disturbance and asymmetric distribution of specific lipid metabolites after BPS exposure. The obtained results indicated that metabolic profiling from LC-MS/MS and metabolite localization from MALDI-IMS analysis could directly reflect the nephrotoxicity caused by BPS, which provided novel insights into the relationship between the chemical exposure and toxicological mechanism on organisms. Analysis of lipidomics and spatial imaging might offer a deeper understanding of the critical role of metabolites reprogramming in environmental pollutants-induced nephrotoxicity. Supplementary materials 19



The online version of this article contains supplementary material (Figure S-1 to S-8, Table S-1 to S-4 and Videos “So-control”, “So-10-BPS”, “So-100-BPS”, “SM-control”, “SM-10-BPS” and “SM-100-BPS” in control, 10 µg and 100 µg BPS treated group, respectively), which is available to authorized users. Conflict statement The authors declare there is no conflict of interest in this manuscript. Acknowledgments The work was supported by the grants from the National Natural Science Foundation of China (grant number 21507106, 91543202), Hong Kong Research Grants Council-General Research Fund (grant number 1230195) and Hong Kong Baptist University Strategic Development Fund (grant number 15-1012-P04). We thank Bruker Daltonics for their help with data processing of MSI.

20


Page 20 of 33

Page 21 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References (1) Clark, E, Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons, New York, NY. 2012. (2) Liao, C.; Liu, F.; Kannan, K., Environ. Sci. Technol. 2012. 46, 6515-6522. (3) Yang, Y.; Lu, L.; Zhang, J.; Yang, Y.; Wu, Y.; Shao, B., J. Chromatogr. A. 2014, 1328, 26-34. (4) Liao, C.; Kannan, K. Arch. Environ. Contam. Toxicol. 2014, 67, 50-59. (5) Zhou, X.; Kramer, J.P.; Calafat, A.M.; Ye, X, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 2014, 944, 152-156. (6) Liao, C., Liu, F., Alomirah, H., Loi, V.D., Mohd, M.A., Moon, H.B., Environ. Sci. Technol. 2012, 46, 6860-6866. (7) Maffini, M. V.; Rubin, B. S.; Sonnenschein, C.; Soto, A. M., Molecular and Cellular Endocrinology. 2006, 255, 179-186. (8) Rezg, R.; El-Fazaa, S.; Gharbi, N.; Mornagui, B., Environment International. 2014, 64, 83-90. (9) Kataria, A.; Trasande, L.; Trachtman, H., Nature Reviews Nephrology, 2015, 11, 610-625. (10) Völkel, W.; Colnot, T.; Csanady, G.; Filser, J.; Dekant, W., Chem. Res. Toxicol. 2002,15,1281-1287. (11) Völkel, W.; Bittner, N.; Dekant, W., Drug Metabol. Dispos. 2005, 33, 1748-1757. (12) Sangai, N.P.; Verma, R.J.; Trivedi, M.H., Toxicol. Ind. Health, 2014, 30, 581-597. (13) Gowder, S.J., Curr. Mol. Pharmacol. 2013, 6,163-172. (14) Rochester, J. R.; Bolden, A. L., Environmental Health Perspectives. 2015, 123, 643-650.

21



(15)

Ciborowski,

M.;

Zbucka-Kretowska,

M.;

Bomba-Opon,

Page 22 of 33

D.;

Wielgos,

M.;

Brawura-Biskupski-Samaha, R.; Pierzynski, P.; Szmitkowski, M.; Wolczynski, S.; Lipinska, D.; Citko, A.; Bauer, W.; Gorska, M.; Kretowski A., Prenatal Diagn. 2014, 34, 870-877. (16) Maiso, P.; Huynh, D.; Moschetta, M.; Sacco, A.; Aljawai, Y.; Mishima, Y.; Asara, J. M.; Roccaro, A. M.; Kimmelman, A. C.; Ghobrial, I. M., Cancer Res. 2015, 75, 2071- 2082. (17) Qiu, Y.; Cai, G.; Zhou, B.; Li, D.; Zhao, A.; Xie, G.; Li, H.; Cai, S.; Xie, D.; Huang, C.; Ge, W.; Zhou, Z.; Xu, L. X.; Jia, W.; Zheng, S.; Yen, Y.; Jia, W., Clin. Cancer Res. 2014, 20, 2136-2146. (18) Padberg, I.; Peter, E.; Gonzaĺez-Maldonado, S.; Witt, H.; Mueller, M.; Weis, T.; Bethan, B.; Liebenberg, V.; Wiemer, J.; Katus, H. A.; Rein, D.; Schatz, P., PLoS One 2014, 9, e85082. (19) Zhao, Y. Y.; Lin, R. C., Advances in Clinical Chemistry, 2014, 65,70-84. (20) McCarthy, N., Nature Reviews Cancer 2012, 12, doi: 10.1038/nrc3248. (21) Zhang, Z. H.; Vaziri, N. D.; Wei, F.; Cheng, X.L.; Bai, X.; Zhao, Y.Y. 2016., Sci. Rep. 2016, 6, 22151. http://dx.doi.org/ 10.1038/srep22151. (22) Stegemann, C.; Pechlaner, R.; Willeit, P.; Langley, S. R.; Mangino, M.; Mayr, U.; Menni, C.; Moayyeri, A.; Santer, P.; Rungger, G.; Spector, T. D.; Willeit, J.; Kiechl, S.; Mayr, M., Circulation 2014, 129, 1821-1831. (23) Han, J. C.; Yu, J.; Gao, Y. J., Pharm. Biol. 2014, 52, 1624-1628. (24) Dowd, S.R.; Bier, M.E.; Patton-Vogt, J. L., J. Biol. Chem. 2001, 276, 3756-3763. (25) Hannun, Y. A.; Luberto, C.; Argraves, K. M., Biochemistry. 2001, 40, 4893-4903. (26) Li, Z. Y.; Vance, D.E., J. Lipid Res., 2008, 49, 1187-1194.

22


Page 23 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(27) Pettus, B. J.; Chalfant, C. E.; Hannun, Y. A., Biochim. Biophys. Acta., 2002, 1585, 114-125. (28) Bartke, N.; Hannun, Y. A., J. Lipid Res. 2009, 50, S91-96. (29) Norris, J. L.; Caprioli, R. M., Chemical Reviews. 2013, 113, 2309-2342. (30) Moreno-Gordalizaa, E.; Esteban-Fernándezb, D.; Lázaroc, A.; Humanesc, B.; Aboulmagdb, S.; Tejedorc, A.; Linscheidb, M.W.; Gómez-Gómeza, M.M., Talanta, 2017, 164,16-26. (31) Liu, H. H.; Li, W.; He, Q.; Xue, J. J., Scientific Reports. 2017, 7, 41954. (32) Rao, S.; Walters, K. B.; Wilson, L.; Chen, B.; Bolisetty, S.; Graves, D.; Barnes, S.; Agarwal, A.; Kabarowski, J. H., Am. J. Physiol. Renal Physiol. 2016, 310, F1136-F1147. (33) Zhao, C., Tang, Z., Yan, J. C., Fang, J., Wang, H. L., Cai, Z. W., Science of the Total Environment. 2017, 592, 357-365. (34) Wang, J. N.; Qiu, S. L.; Chen, S. M.; Xiong, C. Q.; Liu, H. H.; Wang, J. Y.; Zhang, N.; Hou, J.; He, Q.; Nie. Z. X., Anal. Chem. 2015, 87, pp 422-430. (35) Zhao, C.; Xie, P. S., Yang, T., Wang, H. L., Chung, A. C. K., Cai, Z. W., Chinese Chem. Lett.2018, https://doi.org/10.1016/j.cclet.2018.01.034. (36) Hengstler, J. G.; Foth, H.;Gebel, T.; Kramer, P. J.; Lilienblum, W.; Schweinfurth, H.; Völkel, W.; Wollin, K.M.; Gundert-Remy, U., Crit. Rev. Toxicol. 2011, 41, 263-291. (37) Pal, S,; Sarkar, K.; Nath, P.P.; Mondal, M.; Khatun, A.; Paul, G., Toxicol. Rep. 2017, 4, 560-565. (38) Makris, K.; Kafkas, N., Adv. Clin. Chem. 2012, 58, 141-191.

23



(39) Sigdel, T.K.; Klassen, R.B.; Sarwal, M. M., Adv. Clin. Chem. 2009, 47, 139-169. (40) Wei, Q.; Xiao, X.; Fogle, P.; Dong, Z., PLoS One. 2014, 9, e106647. (41) Tapiero, H.; Ba, G. N.; Couvreur, P.; Tew, K. D., Biomed Pharmacother. 2002, 56, 215-222. (42) Kremmyda, L. S.; Tvrzicka, E.; Stankova, B.; Zak, A., Univ. Palacky. Olomouc. Czech. Repub. 2011, 155, 195-218.  (43) Maceyka, M.; Sankala, H.; Hait, N. C.; Le Stunff, H.; Liu, H.; Toman, R.; Collier, C.; Zhang, M.; Satin, L. S.; Merrill, A. H.; Milstien, S.; Spiegel, S., J. Biol. Chem. 2005, 280, 37118-37129. (44) Zu Heringdorf, D. M.; Jakobs, K. H., Biochim. Biophys. Acta. 2007, 1768, 923-940. (45) Morad, S. A.; Cabot, M. C., Nat. Rev. Cancer. 2013, 13, 51-65. (46) Senchenkov, A.; Litvak, D. A.; Cabot, M. C., J. Natl. Cancer Inst. 2001, 93, 347-357. (47) Kolesnick, R.; Fuks, Z., Oncogene. 2003, 22, 5897-5906. (48) Calder, P. C.; Grimble, R. F., Eur. J. Clin. Nutr. 2002, 56, S14-S19. (49) Grimble, R. F., Proc. Nutr. Soc. 1998, 57, 535-542. (50) Miles, E. A.; Calder, P. C., Proc. Nutr. Soc. 1998, 57, 277-292. (51) Shindou, H.; Hishikawa, D.; Harayama, T.; Yuki, K.; Shimizu, T., J. Lipid Res. 2009, 50, S46-S51. (52) Glass, C. K.; Olefsky, J. M., Cell Metab. 2012, 15, 635-645. (53) Manabe, I., Circulation Journal. 2011, 75, 2739-2748.

24


Page 24 of 33

Page 25 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(54) Ratliff, B. B.; Abdulmahdi, W.; Pawar, R.; Wolin, M.S., Antioxid. Redox Signal. 2016, 25, 119-146.

25



for TOC only

26


Page 26 of 33

Page 27 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure legends Figure 1. Morphological analysis of BPS-induced nephrotoxicity in control group (A, B and

C), 10 µg BPS group (D, E and F) and 100 µg BPS group (G, H and I) group. A, D and G, in renal cortex; B, E and H, in renal medulla; C, F and I, in renal pelvis. Major morphological changes were labeled with black arrows. White boxes indicate the magnified areas in different regions of kidney section (H&E, 40×magnification). Figure 2. BPS-induced lipidomics variation on mice kidney tissues. (A and B) PLS-DA score

plots of kidney extract in the control, 10 µg BPS and 100 µg BPS group (n=6) in positive (A) and negative (B) ionization mode. (C) Relative composition of subclass lipids (lipid compositional reprogramming) % = amount of subclass/entire lipidome ×100% (n=6). BPS-treated/control, fold change of BPS-treated vs control group. (D) Composition of fatty acids in four GPs categories of 100 µg BPS group including PC, PE, PG and PI. We used LipidSearch software to identify the chain composition of lipids. Percentage of ether-linked FA, PUFA, MUFA and SFA chains were calculated based on identified acyl/alkyl/alkenyl chain composition of individual lipid species. (E) Heat map analysis of identified lipids after BPS exposure. The individual samples were represented in the vertical axis, and the identified lipids were represented in the horizontal axis. Up- and down-regulated lipids were represented in red and blue in 10 µg or 100 µg BPS group vs. control group. Figure 3. Metabolic network of BPS-induced nephrotoxicity in mice model. BPS exposure in

kidney can induce multiple toxicological effects and lipidome disturbances. Up- and down-regulated lipids and genes (italics) were represented in red and green, respectively. Representative images of BPS-treated kidney section in lipid metabolism pathway were 27



Page 28 of 33

displayed in negative reflector ion mode. CDase, Ceramidase; CDP, cytidine diphosphate (CDP); Cept, CDP-ethanolamine phosphotransferase; DAG, Diacylglycerols; Dgat, Diacylglycerol

acyltransferase;

LPG,

Lyso

phosphatidylglycerol;

LPI,

Lyso

phosphatidylinositol; PA, Phosphatidic acid; PC, Phosphatidylcholine; Pcyt2, Phosphate cytidylyltransferase 2; PE, Phosphatidylethanolamine; Pemt, Phosphatidylethanolamine N-methyltransferase; PG, Phosphatidylglycerol; PGP, Phophatidyl-glycerolphophate; PI, Phosphatidylinositol;

PS,

Phosphatidylserine;

SMase,

Sphingomyelinases;

SphK2,

Sphingosine kinase types 2; TAG, Triacylglycerols. Figure 4. pLSA score plots for MALDI-MSI profiles obtained from the whole kidney (A),

renal cortex (B), renal medulla (C) and renal pelvis (D) regions (n=6). Figure 5. MSI reconstructed kidney tissue exposure to 100 µg BPS group in a 3DI model. At

X- and Y-coordinate, each of serial kidney sections was analyzed by MALDI-MSI, generating a complete mass spectra dataset. Serial mass spectral images were aligned by translations and rotations at X-, Y- and Z-coordinate, yielding reconstructed kidney tissue.

28


Page 29 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1

29



Figure 2

30


Page 30 of 33

Page 31 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3

31



Figure 4

32


Page 32 of 33

Page 33 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5

33


MALDI-MS Imaging Reveals Asymmetric Spatial Distribution of Lipid

Recommend Documents