Spatial analysis of renal acetaminophen metabolism and its modulation by 4-methylpyrazole with DESI mass spectrometry imaging

Abstract

Acute kidney injury (AKI) is a common complication in acetaminophen (APAP) overdose patients and can negatively impact prognosis. Unfortunately, N-acetylcysteine, which is the standard of care for the treatment of APAP hepatotoxicity does not prevent APAP-induced AKI. We have previously demonstrated the renal metabolism of APAP and identified fomepizole (4-methylpyrazole, 4MP) as a therapeutic option to prevent APAP-induced nephrotoxicity. However, the kidney has several functionally distinct regions, and the dose-dependent effects of APAP on renal response and regional specificity of APAP metabolism are unknown. These aspects were examined in this study using C57BL/6J mice treated with 300–1200 mg/kg APAP and mass spectrometry imaging (MSI) to provide spatial cues relevant to APAP metabolism and the effects of 4MP. We find that renal APAP metabolism and generation of the nonoxidative (APAP-GLUC and APAP-SULF) and oxidative metabolites (APAP-GSH, APAP-CYS, and APAP-NAC) were dose-dependently increased in the kidney. This was recapitulated on MSI which revealed that APAP overdose causes an accumulation of APAP and APAP GLUC in the inner medulla and APAP-CYS in the outer medulla of the kidney. APAP-GSH, APAP-NAC, and APAP-SULF were localized mainly to the outer medulla and the cortex where CYP2E1 expression was evident. Interestingly, APAP also induced a redistribution of reduced GSH, with an increase in oxidized GSH within the kidney cortex. 4MP ameliorated these region-specific variations in the formation of APAP metabolites in renal tissue sections. In conclusion, APAP metabolism has a distinct regional distribution within the kidney, the understanding of which provides insight into downstream mechanisms of APAP-induced nephrotoxicity.

An acetaminophen (APAP) overdose is the most frequent cause of acute liver failure (ALF) in the United States (Fisher and Curry, 2019; Jaeschke, 2015). Though APAP is well recognized as a hepatotoxicant, strong clinical evidence demonstrates that an APAP overdose can also induce nephrotoxicity and acute kidney injury (AKI) leading to renal insufficiency (Kwok et al., 2004) and renal failure (Blakely and McDonald, 1995; Hengy et al., 2009) in children (Le Vaillant et al., 2013; Ozkaya et al., 2010), adolescents (Boutis and Shannon, 2001), and adults (Saleem and Iftikhar, 2019). The fact that nephrotoxicity can even develop in APAP overdose patients independent of liver failure also suggests that APAP can have direct effects on the kidney (Eguia and Materson, 1997; Jeffery and Lafferty, 1981). Despite the recognition of APAP-induced nephrotoxicity, key mechanisms involved in renal cell death are unknown, hampering the development of targeted interventions to counteract them. This is relevant because data from mice suggest that N-acetylcysteine (NAC), the only clinically approved antidote to treat APAP-induced hepatotoxicity, is less effective for the prevention of nephrotoxicity than for treatment of liver injury after APAP overdose (Slitt et al., 2004). This highlights the need for insight into mechanisms of APAP-induced nephrotoxicity to develop targeted therapeutics.

Though predominantly metabolized by glucuronidation and sulfation, APAP can also be metabolized by cytochrome P450 2E1 (CYP2E1), to form the highly reactive electrophile N-acetyl-p-benzoquinone imine (NAPQI) in a process termed metabolic activation (McGill and Jaeschke, 2013; Prescott, 1980). Like the liver, the kidney is protected from the harmful effects of NAPQI formed at therapeutic doses of APAP by thiol-scavenging (Hoivik et al., 1995; Stern et al., 2005a), where it binds glutathione (GSH) to form APAP-GSH (Newton et al., 1982; Newton et al., 1986). Further metabolism of APAP-GSH to APAP-cysteine (APAP-CYS) is catalyzed by γ-glutamyl transferase (γGT), and dipeptidase (DP) in the kidney, which are both highly expressed on the luminal or brush-border plasma membrane of various renal epithelial cells (Lash, 2005). APAP-CYS can then be reabsorbed into renal tubular cells in the cortex where it is N-acetylated to form APAP-N-acetylcysteine (APAP-NAC), also known as mercapturic acid, which ultimately undergoes renal excretion (Akakpo et al., 2020b; Lash, 1994). However, after an APAP overdose, excessive NAPQI production leads to extensive GSH depletion (Akakpo et al., 2020b; Hoivik et al., 1995; Slitt et al., 2005), which induces covalent binding of NAPQI to renal proteins to ultimately trigger acute nephrotoxicity (Akakpo et al., 2020b; Emeigh Hart et al., 1991; Hart et al., 1994). Although analysis of APAP metabolism in biofluids or whole kidney homogenates provides significant insights into the disposition of APAP, spatial information is lacking. This is relevant because the kidney is made up of distinct functional compartments and regional variations in the formation and abundance of APAP metabolite could influence cellular response to excessive NAPQI formation, and subsequent nephrotoxicity.

Recently, we have demonstrated that desorption electrospray ionization mass spectrometry imaging (DESI-MSI) can successfully detect and generate ion images for APAP and its metabolites under ambient air, without chemical labeling or prior coating of liver tissue (Akakpo et al., 2022) reducing chemical interference and perturbation of small molecule localization on tissue (Wiseman et al., 2008). We also demonstrated the functional significance of these findings where the repurposed drug 4-methypyrazole (4MP) substantially influenced the topographical distribution of APAP and its metabolites in the liver (Akakpo et al., 2022). Importantly, our recent evidence also suggests that 4MP is likely a promising therapeutic option to prevent severe APAP nephrotoxicity (Akakpo et al., 2020b). Here, we demonstrate that DESI-MSI provides sensitive mapping of APAP metabolism to spatially distinct regions of the kidney and reveals the influence of 4MP on the topographical distribution of renal APAP metabolism and disposition.

Materials and methods

 

 

Animals and experimental design

For this study, 8–12-week-old male C57BL/6J mice were purchased from Jackson Laboratories (Bar Harbor, Maine). The mice were housed in a temperature-controlled room (14-h light/10-h dark cycle) with ad libitum access to food and water. Fifteen hours prior to in vivo experiments, food was removed, and mice were treated intraperitoneally (i.p.) with the indicated APAP doses (Sigma-Aldrich, St Louis, Missouri) dissolved in saline. Control mice were treated the same way using the vehicle solution (saline). The mice were then sacrificed 30 min later by cervical dislocation under isoflurane anesthesia. Blood was drawn from the caudal vena cava into heparinized syringes and was centrifuged at 18 000×g for 3 min to obtain plasma. Kidneys were snap-frozen in liquid nitrogen and stored at −80°C. Experiments were performed following the National Research Council for the Care and Use of Laboratory Animal Guidelines. All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Kansas Medical Center.

Biochemical assays

Blood urea nitrogen (BUN) was measured using a QuantiChrom Urea Assay kit from BioAssay Systems (Hayward, California) as per manufacturer’s instructions to assess the extent of kidney injury. Total tissue glutathione (GSH + GSSG) levels were measured using a modified Tietze assay (Jaeschke and Mitchell, 1990).

APAP-CYS protein adduct analysis

APAP protein adducts were measured as described in detail previously (Akakpo et al., 2022). Briefly, a 10% homogenate of kidney tissue was prepared in homogenizing buffer (25 mM 4-2-hydroxyethyl-1-piperazineethanesulfonic acid [HEPES], pH 7.5, 5 mM ethylenediaminetetraacetic acid [EDTA], 2 mM dithiothreitol, 0.1% CHAPS, 1 μl/ml pepstatin, 0.1 μl/ml leupeptin and 0.1 μl/ml aprotinin). Samples were then spun at 16 000×g for 5 min at 4°C. The resulting supernatant was filtered through Bio-Spin 6 columns (Bio-Rad, Hercules, California) to remove free cysteine residues and low-molecular weight metabolite conjugates that could interfere with the detection of adducts. Then, an 8 U/ml Streptomyces griseus (Sigma-Aldrich, St Louis, Missouri) solution was mixed with the filtrate-containing proteins with APAP bound to cysteine residues in a 1:1 ratio. The mixture was digested at 50°C for 15 h and filtered through an ultra-free-MC centrifugal filter of 0.22 µm pore size (Millipore Sigma, Burlington, Massachusetts). The APAP-Cys standard was purchased from Toronto Research Company (Ontario, California) and APAP-Cys derived from digested proteins were quantified by high-pressure liquid chromatography using a Coularray electrochemical detector (ESABiosciences, Chelmsford, Massachusetts).

LC-MS/MS analysis

LC-MS/MS analysis was conducted using a Waters Acquity Ultra-Performance Liquid Chromatography (UPLC) system equipped with a Xevo XS triple quadrupole mass spectrometer (Waters, Milford, Massachusetts) as described previously (Akakpo et al., 2018). Parent APAP, APAP-glucuronide (APAP-GLUC), APAP-sulfate (APAP-SULF), APAP-glutathione (APAP-GSH), free APAP-CYS, and APAP-NAC standards were purchased from Toronto Research Chemicals (Toronto, Canada). Briefly, calibration standards, samples and quality control (QC) standards (low: 1 µM; high: 25 µM) were prepared in drug-free mouse plasma and tissue homogenate. A 9-point calibration curve was prepared from 0.25 to 75 μM. About 20 μl of plasma and tissue homogenate samples were mixed with 100% methanol and the internal standard 4-acetaminophen-d3 sulfate (APAP-d3), briefly vortexed and then incubated on ice for 10 min before centrifugation (20 000×g for 20 min at 4°C) to remove precipitated proteins. The supernatant was transferred into new 2 ml glass vials (Waters). Before analysis, the system cleanliness was ensured by injecting 5 µl of a blank sample containing only the sample matrix. Then, system suitability was tested by injecting the low QC sample consisting of a mixture of all the standards and the internal standards. This was followed by the analysis of the standard samples and study samples. After analysis of the samples, 5 μl of a second set of calibration samples were injected from low to high concentrations. The lower limits of quantification were determined to be 0.25 μM for APAP-GSH; 0.125 μM for APAP-SULF and APAP-GLUC; 0.063 μM for APAP-CYS; and 0.025 μM for APAP-NAC. Analyte levels below the limits of quantification were considered zero.

DESI-MSI of the kidney sections

Snap-frozen mouse kidney tissues were cryo-sectioned at 12 μm thickness and thaw-mounted on glass slides before DESI-MSI analysis, which was performed on a high-resolution quadrupole time-of-flight (TOF) SYNAPT HDMS XS high mass resolution mass spectrometer (Waters, Milford, Massachusetts) equipped with the DESI 2D XS source which included a high-performance sprayer providing high spatial resolution (Akakpo et al., 2018). General DESI acquisition parameters for the SYNAPT HDMS XS MS were MS acquisition mode: positive; MS acquisition rate: 0.5–1 s; TOF MS mode: sensitivity; sampling cone voltage: 25 V. MS source temperature: 150°C; Step wave RF: 100; Ion guide RF: 150; source sprayer voltage: 0.6 kV; source sprayer gas pressure: 15 psi; DESI solvent: 98:2 methanol/water with 0.01% (v) formic acid; DESI solvent flow rate: 2 µl/min. Pixel sizes: 200 µm or 50 μm. To correct for any potential m/z drift during MSI experiments, Leucine enkephalin was used as a lock mass after addition to the DESI sprayer solvent. Data acquisition was set up in the High-Definition Imaging (HDI) version 1.5 software and acquired using Masslynx software version 4.2 (Waters). Data processing, visualization, regions of interest generation, and advanced analysis were all performed in HDI. The imaging data were processed using a target list, with an extraction window of 0.01 Da. Putative metabolite and lipid identification was made based on accurate mass searches in Lipidmaps (Schmelzer et al., 2007), Metlin (Smith et al., 2005), and the Human Metabolome Database (HMDB) (Wishart et al., 2013).

Histology and immunostaining

Kidney sections were stained with hematoxylin and eosin to highlight morphological features on the kidney on the Leica Autostainer XL. Immunostaining was performed on sample replicates after DESI MSI analysis by fixing cryosections with 3% formaldehyde. This was followed by blocking with 5% normal goat serum and overnight incubation with 1:200 diluted primary rabbit antibodies for cytochrome P450 2E1 (Abcam rabbit polyclonal antibody, cat. no. ab28146) at 4°C. The next day, sections were washed in phosphate-buffered saline (PBS) and SignalStain Boost Detection Reagent (Cell Signaling Technology, Rabbit no. 8114) was applied for 320 min before adding SignalStain DAB Chromogen (cat. no. 8059, Cell Signaling Technology, Danvers, Massachusetts). As per the manufacturer’s instructions, samples were counterstained with hematoxylin and imaged with a Zeiss Axioimager microscope (Carl Zeiss AG, Jena, Germany). Finally, multimodal ion images and stained images were superimposed using ImageJ (http://rsb.info.nih.gov/ij/) and HDI. Immunofluorescence staining was performed using cryosections prepared after DESI-MSI. Sections were blocked with 3% BSA and stained with anti-SGLT2 (cat. no. sc11415, Santa Cruz, Dallas, Texas) overnight at 4^oC. The next day, sections were washed in PBS before application of the secondary Alexa Fluor 594-conjugated goat anti-rabbit antibody (cat. no. A11037, Life Technologies, Eugene, Oregon), followed by imaging on a Nikon Ti-2 inverted fluorescence microscope.

Statistical analysis

All statistical analyses were performed with SPSS Statistics 25 (IBM Co., Armonk, New York), and differences with p < .05 were considered statistically significant. The student’s 2-tailed t-test was used to perform statistical analysis between 2 groups. In addition, statistical analysis between multiple groups was performed with a 1-way analysis of variance (ANOVA), followed by the Student-Newman-Keul’s test. Finally, the Kruskal-Wallis test (nonparametric ANOVA) and Dunn’s Multiple Comparison Test were performed for data that were not normally distributed.

Results

 

APAP metabolites show a dose-dependent increase in plasma within 30 min of an overdose

The current studies were focused on the characterization of the dose dependency of APAP metabolism in the kidney and its spatial differentiation within the organ. Thus, mice were treated with 300, 600, 900, or 1200 mg APAP/kg body weight, and initial experiments measured metabolites in plasma and kidney at 30 min, which is the time point of peak APAP metabolism (Akakpo et al., 2020b). As seen in Figure 1, all the APAP metabolites could be detected at the 30-min time point in plasma. A dose-dependent elevation in parent APAP was evident between 300 and 900 mg/kg dose, with significant elevations between successive doses, which plateaus by the 1200 mg/kg dose (Figure 1A). APAP-GLUC, which is a predominant metabolite, showed a dose-dependent increase from 300 to 1200 mg/kg APAP (Figure 1B), with APAP-SULF levels also showing a similar dose-dependent increase in plasma (Figure 1C). Interestingly, the oxidative metabolites APAP-GSH, APAP-NAC, and APAP-CYS in plasma showed a very different trend compared with APAP-GLUC and APAP-SULF. Though APAP-GSH levels did increase significantly up to the 600 mg/kg APAP dose, there was a substantial elevation subsequently at the 900 and 1200 mg/kg dose (Figure 1D). APAP-NAC was elevated at the 300 mg/kg dose, further increasing significantly at the 600 and then substantially at the 900 mg/kg dose, with a slight decrease subsequently at 1200 mg/kg APAP (Figure 1E). Interestingly, the APAP-CYS dose-dependently increased from 300 to 600 mg/kg with substantial elevation at 900 and then further increased at the 1200 mg/kg dose of APAP (Figure 1F).

Figure 1.

Dose-dependent change in APAP and its metabolites in circulation after APAP overdose. Fasted male C57BL/6J mice received different doses of APAP (300–1200 mg/kg) or 10 ml/kg saline and were sacrificed 0.5 h later. Plasma concentrations of the parent compound APAP (A), the nonoxidative metabolites APAP-glucuronide (B) and APAP-sulfate (C), the oxidative metabolites APAP-GSH (D), APAP-NAC (E), and APAP-cysteine (CYS) (F) were measured by LC-MS/MS. Data represent means ± SEM of n = 4 animals per group. *p < .05 (compared with control), **p < .05 (compared with 300 mg/kg), ***p < .05 (compared with 600 mg/kg), and ****p < .05 (compared with 900 mg/kg).

Although renal metabolism of APAP and generation of the oxidative metabolite is evident, GSH depletion and circulating protein adducts are predominantly seen only at the highest APAP dose

The plasma measurements indicated a dose-dependent increase in APAP-GLUC and sulfate within 30 min after an APAP overdose, with oxidative metabolites being predominantly elevated at the higher doses of 900 and 1200 mg/kg. The next series of experiments evaluated various APAP metabolites within the kidney along this concentration curve to test whether renal metabolism to the glucuronide and sulfate forms can be saturated at sufficiently high APAP concentrations. In the kidney homogenate, the concentration of the parent APAP compound showed a significant dose-dependent increase between 300 and 1200 mg/kg, though the increase was somewhat blunted between the 900 and 1200 doses (Figure 2A). Renal APAP-GLUC concentrations showed a significant increase from 300 to 600 mg/kg APAP, which did not change at the 900 mg/kg dose. However, a substantial and significant increase was then evident at the 1200 mg/kg APAP dose when compared with the 900 mg/kg dose (Figure 2B). APAP-SULF levels increased in a dose-dependent manner up to 900 mg/kg but plateaued subsequently at 1200 doses (Figure 2C). Consistent with our previous report (Akakpo et al., 2020b), only the oxidative metabolite APAP-CYS was detected within the kidney, with a small but significant increase between 300 and 600 mg/kg, a plateau at the 900 mg/kg dose, with substantial elevation subsequently at the 1200 mg/kg dose (Figure 2D). Notably, APAP-GSH and APAP-NAC were below the limit of quantification in the kidney. Thus, although sulfation in the kidney seems to saturate at the higher APAP doses, glucuronidation seems to be capable of significant upregulation to conjugate APAP.

Figure 2.

Dose-dependent change in concentration of APAP metabolites, GSH, and protein adducts 30 min after APAP treatment. Fasted male C57BL/6J mice received different doses of APAP treatment from 300 to 1200 mg/kg or 10 ml/kg saline and were sacrificed 0.5 h later. Renal concentrations of the parent compound APAP (A), the nonoxidative metabolites APAP-glucuronide (B), and APAP-sulfate (C), as well as the oxidative metabolites APAP-CYS (D), were measured by LC-MS/MS. Concentrations of all metabolites are expressed as nmol/mg protein. Renal GSH levels (E). The concentration of protein-derived APAP-CYS adducts in tissue homogenate (F) and plasma (G) were measured by HPLC-ECD. Data represent means ± SEM of n = 4 animals per group. *p < .05 (compared with control), **p < .05 (compared with 300 mg/kg), ***p < .05 (compared with 600 mg/kg), and ****p < .05 (compared with 900 mg/kg).

The next series of experiments evaluated glutathione depletion and APAP adduct formation in the kidney with the escalating doses. As seen in Figure 2E, unlike in the liver (Akakpo et al., 2018) at this very early time point, APAP did not deplete renal GSH levels in the whole kidney tissue homogenate at the 300 and 600 mg/kg doses, indicating that renal NAPQI formation at these doses was minimal and could be efficiently scavenged by glutathione stores (Figure 2E). A significant decrease in renal GSH is only evident at the 900 mg/kg dose, which further dropped by the 1200 mg/kg dose of APAP (Figure 2E), indicating that NAPQI generation is beginning to overwhelm renal glutathione stores at these higher doses. APAP-CYS protein adducts in the kidney, indicative of NAPQI-protein binding, show a dose-dependent increase from 300 to 900 mg/kg (Figure 2F), and the significant increase in adducts between 900 and 1200mg/kg compared with the 300 and 600 mg/kg dose suggests that the rate of NAPQI protein adduct formation is accelerated once significant GSH depletion occurs. Notably, despite elevation in renal adduct levels even at the lower doses, substantial plasma adduct levels were only detected at the 1200 mg/kg dose (Figure 2G). As expected at this very early time point, no elevations in plasma BUN or apparent histological changes were evident in the kidney tissue (Supplementary Figs. 1A–F).

Mass spectrometry imaging allows the mapping of parent APAP and APAP-GLUC to distinct spatial compartments within the kidney

Although the data so far clearly demonstrates the rapid generation of APAP metabolites within the renal tissue, it does not provide information regarding the compartmentalization of this metabolism. Though the renal proximal tubules in the cortex are suggested to be the predominant regions of APAP metabolism due to the abundance of CYP2E1 (Cummings et al., 1999; Hart et al., 1995; Lohr et al., 1998), the distribution of metabolites in the renal cortex and medulla is not well characterized. Now that we have established the characteristics of APAP metabolism in the whole kidney, further experiments were focused on determining their spatial localization. Toward this goal, we employed DESI-MSI, which allows the identification of metabolites on tissue sections. Because spatial registration of APAP metabolites to the various renal compartments was critical to the study, we stained kidney tissue sections for the glucose transporter SGLT2 which is expressed exclusively in the proximal tubular epithelial cells in the cortex of the kidney, allowing annotation of the kidney cortex (Figure 3A). We next confirmed localization by identifying specific endogenous phosphatidylcholine (PC) species of varying chain lengths and saturation of fatty acids that are exclusively present in different regions of the kidney. These include the cortex (36:3), the outer medulla (38:5), the inner medulla (36:4) and the papilla (34:1), the pelvis being devoid of any signal (Figs. 3B–E). Merging of the PC signal allows definition of several renal areas (Figure 3F), which can be used to define the regional specificity of APAP metabolites. Provided that APAP and its major nonoxidative metabolite, APAP-GLUC, were the most abundant analytes in the whole kidney (Figure 2), initial experiments attempted to determine whether APAP and APAP-GLUC could be sensitively detected at high spatial and mass resolution on the DESI XS-SYNAPT XS MSI platform. Kidney sections from animals treated with 900 or 1200 mg/kg APAP were analyzed, and ion images were generated for APAP and APAP-GLUC at a resolution of 200 µm. These experiments revealed that the signal is specific, as highlighted by its complete absence in controls that were not treated with APAP (Supplementary Figure 2A). APAP and APAP-GLUC ions signals were clearly detectable at the dose of 900 mg/kg, with much higher signal intensities for the 1200 mg/kg dosed tissue (Supplementary Figure 2A). The spatial distinction was also noted, with APAP and APAP GLUC ion signals being much higher in distinct regions of the kidney 30 min after APAP overdose, as indicated by the averaged ion intensities represented in the mass spectra displayed in Supplementary Figure 2B. The data also indicated a concentration-dependent increase in the ion signal of APAP, which replicates that seen in the measurement of the parent compound by LC-MS/MS (Figure 2A). Next, we examined if this sensitivity in detection is maintained at a higher resolution of data acquisition. Hence, we performed DESI-XS MSI analysis of APAP and APAP-GLUC on the 900 mg/kg dosed tissue at 50 µm resolution. The higher 50 µm spatial resolution enhanced the detection of APAP and APAP-GLUC, revealing their presence in distinct compartments within the kidney (Supplementary Figure 3A). Interestingly, the sodiated, potassiated, and protonated APAP species showed differential ion intensities and spatial distribution within the kidney cortex (Supplementary Figure 3A), with more of the sodiated form in the inner medulla compared with the other 2 species. Moreover, potassiated and protonated species were more abundant in the cortex, whereas APAP-GLUC was almost exclusively in the inner medulla. The specificity of ion signal to APAP and its metabolites was further confirmed by examining the distribution of lipid species such as PC (38:5), which showed a distinctly different tissue distribution, being absent in the inner medulla, but abundant in the outer medulla and the medullary rays separating the cortex from the outer medulla (Supplementary Figure 3B). Merging the APAP, APAP-GLUC, and PC images clearly illustrates the higher intensity signal for parent APAP in the cortex and inner medulla (Supplementary Figure 3B), whereas APAP-GLUC is mainly localized exclusively to the inner medulla. This is expected because renal filtrate concentrates in the inner medulla and could be contributing to the higher signal.

Figure 3.

Immunostaining for renal SGLT2 and DESI-MSI analysis of phospholipids for regional mapping in the kidney. Fasted male C57BL/6J mice received 600 mg/kg APAP or 10 ml/kg saline and were sacrificed 0.5 h later. Kidney sections after DESI-MSI were immunostained for SGLT2, a proximal tubular cell marker in the renal cortex (A). DESI-MSI ion images for regional-specific phosphatidyl choline (PC) species; PC (36:3) (m/z 806.5637) [M + Na]+; PC(34:1) (m/z 760.5800) [M + Na]+; PC (38:5) (m/z 792.5868) [M + Na]+; and PC(36:4) (m/z 782.5640) [M + H]+ at 50 µm resolution (B–E). Ion images were then merged to display different renal compartments (F).

DESI XS-MSI at 50 µm resolution allows spatial mapping of all relevant APAP metabolites in the kidney after a severe overdose

The data so far showed that APAP and APAP-GLUC were detectable at high mass and spatial resolution by DESIXS-MSI from kidney sections of mice administered a 900 mg/kg overdose and showed clear and unique spatial distribution in the different regions of the kidney. Yet, the primary aim of this study was to determine whether all APAP metabolites could be detected after a moderate (300 mg/kg) or severe (600 mg/kg) overdose where the outcome of kidney injury is very different. Mice treated with 300 mg/kg typically only show upregulation of renal stress markers such as NGAL without functional deficits, whereas those with a severe overdose show histological changes as well as alterations in BUN (Akakpo et al., 2020b). Hence, the next series of experiments examined the spatial distribution of not only APAP and APAP-GLUC but also APAP-SULF, APAP-GSH, APAP-CYS, and APAP-NAC on kidney tissue sections at 50 µm pixel size. Figure 4 demonstrates that APAP and its metabolites can be mapped onto the kidney tissue section with great sensitivity after the severe 600 mg/kg APAP overdose. We detected relatively high ion signals for APAP in the cortex and the outer medulla (Figure 4A). However, significantly higher ion signal intensity was detected for APAP and APAP-GLUC in the inner medulla suggesting that the analytes concentrated in these regions for excretion (Figs. 4A and 4B). This is consistent with our analysis of APAP metabolism in human urine from our volunteer trial where parent APAP and APAP-GLUC were the most abundant (Kang et al., 2020). Interestingly, APAP-SULF showed a different distribution, being present throughout the cortex and outer medulla, without significant accumulation in the inner medulla (Figure 4C). Notably, the oxidative metabolite APAP-CYS was found to be exclusively concentrated in the outer medulla and the medullary rays (Figure 4E), with the APAP-CYS ion signal being almost undetectable in the inner medulla. Moreover, though other oxidative metabolites such as APAP-GSH and APAP-NAC were not detected by LC-MS/MS of kidney homogenates, DESI-MSI showed a detectable signal. APAP-GSH and APAP-NAC were found distributed mainly in the cortex and outer medulla, again with a complete absence in the inner medulla (Figs. 4D and 4F). It is important to note that the signal for APAP and APAP-CYS was predominantly protonated, whereas that for APAP-GLUC was mainly sodiated (Supplementary Figure 3). Thus, DESI-MSI on the DESI XS-SYNAPT XS platform allows the detection of all relevant APAP metabolites at varying levels of abundance.

Figure 4.

DESI-MS analysis for the detection of APAP and its nonoxidative and oxidative metabolites in the kidney at 50 µM resolution after APAP overdose. Fasted male C57BL/6J mice received 600 mg/kg APAP or 10 ml/kg saline and were sacrificed 0.5 h later. DESI ion images of APAP ([m/z 152.0712] [M + H]+; [m/z 174.0531] [M+Na]+, [m/z 190.0270] [M + K]+) (A), APAP-GLUC ([m/z 366.0591] [M + H]+; [m/z 350.0852] [M+Na]+, [m/z 328.1032] [M + K]+) (B), APAP-SULF ([m/z 232.0280] [M + H]+; [m/z 254.0099] [M+Na]+, [m/z 269.9839] [M + K]+) (C), APAP-CYS ([m/z 271.0766] [M + H]+; [m/z 293.0857] [M+Na]+, [m/z 3309.0362] [M + K]+) (D), APAP-GSH ([m/z 457.1393] [M + H]+; [m/z 479.1213] [M+Na]+, [m/z 495.0952] [M + K]+) (E), APAP-NAC ([m/z 313.0858] [M + H]+; [m/z 335.0678] [M+Na]+, [m/z 351.0417] [M + K]+) (F).

DESI-MSI allows visualization of 4MP effects on APAP metabolism

The next series of experiments compared renal signals for APAP metabolites after the 600 mg/kg severe APAP overdose to those after a moderate overdose (300 mg/kg), which does not produce functional renal deficits, but induces stress markers such as NGAL (Akakpo et al., 2020b). To further confirm the relevance of these signal intensities and to enhance the translational application of our findings, we compared the data to those from mice that had been concurrently treated with 4-methyl pyrazole (4MP). This is an FDA-approved antidote for ethylene glycol and methanol poisoning, which we have shown to provide significant protection against APAP hepatotoxicity (Akakpo et al., 2018, 2019, 2020a, 2021) and nephrotoxicity (Akakpo et al., 2020b). DESI ion images demonstrate discrete signals with variable abundance throughout the tissue, with distinct regional specificity (Figs. 5 and 6). The parent compound APAP, and all of its metabolites, were readily detected on kidney sections from mice treated with 300 mg/kg APAP, with significant elevation evident at the higher dose of 600 mg/kg (Figure 5B). Consistent with the LC-MS/MS data, the ion intensities of the metabolites APAP-SULF, APAP-GSH, and APAP-NAC were much lower than APAP and APAP-GLUC (Figs. 5 and 6). Interestingly, animals treated with 4MP along with 600 mg/kg APAP showed significant elevations in the parent APAP in all regions of the kidney (Figure 5C), which would be expected because 4MP inhibits cytochrome P450-mediated metabolism of APAP (Akakpo et al., 2018). The relevance of this inhibition is also evident in the concomitant disappearance of the oxidative metabolites APAP-GSH, APAP-CYS, and APAP-NAC in 4MP-treated mice (Figure 6C). Together, the data indicate that 4MP induced a robust inhibition of cytochrome P450-mediated generation of the oxidative metabolite in the kidney, the spatial distribution of which is detectable by DESI-MSI.

Figure 5.

Visualization of the effect of 4MP on the intensity of APAP and its nonoxidative metabolites after a moderate or severe APAP overdose. Fasted male C57BL/6J mice received 300 (A) or 600 (B) mg/kg APAP along with 50 mg/kg 4MP (C), or 10 ml/kg saline, and were then sacrificed 0.5 h later. DESI-MSI ion images of APAP (m/z 152.0712) [M + H]+; APAP (m/z 174.0531) [M+Na]+; APAP-NAC (m/z 313.0858) [M + H]+; APAP-GLUC (m/z 350.0852) [M + K]+ at 50 µm resolution were then obtained as described in the materials and methods section.

Figure 6.

Visualization of the effect of 4MP on the intensity of APAP and its oxidative metabolites after a moderate or severe APAP overdose. Fasted male C57BL/6J mice received 300 (A) or 600 (B) mg/kg APAP along with 50 mg/kg 4MP (C), or 10 ml/kg saline, and were sacrificed 0.5 h later. DESI ion images of APAP (m/z 152.0712) [M + H]+; APAP (m/z 174.0531) [M+Na]+; APAP-CYS (m/z 271.0766) [M + H]+; and APAP-NAC (m/z 313.0858) [M + H]+ were then acquired at a resolution of 50 µm.

APAP-CYS exclusively accumulates in the outer medulla of the kidney but does not overlap with APAP accumulation in the inner medulla

Next, we merged spatially distinct PC signals with those of the various metabolites to clearly illustrate the regional demarcation in APAP metabolite abundance. Parent APAP and APAP-GLUC species were distributed throughout the cortex as well as the outer and inner medulla with significant accumulation in the papilla (Figure 7A). Interestingly, the APAP-GLUC signal in the cortex was more discrete and intense than that of parent APAP (Figure 8B—white arrows showing merged signal in the cortex). In contrast, APAP-SULF, APAP-GSH, APAP-CYS, and APAP-NAC showed no signal in the medulla or papilla (Figs. 7 and 8). Although a minor signal for APAP-SULF could be detected in the outer medulla, and to a lesser extent the inner medulla, the majority of APAP-SULF was confined to the cortex (Figure 7). APAP-GSH was mainly localized to the outer medulla while also being present in the cortex and APAP-CYS was predominantly confined to the outer medulla and medullary rays at both 300 or 600 mg/kg APAP (Figure 8). The specificity and the almost exclusive accumulation of APAP-CYS to the outer medulla is consistent amongst all animals treated with 600 mg/kg APAP (Supplementary Figure 5). APAP-NAC was mapped to the cortex and to a lesser extent to the outer medulla while faint signals were also detected to the inner medulla (Figure 8).

Figure 7.

DESI-MSI analysis of region-specific distribution of the nonoxidative metabolites of APAP in the kidney. Fasted male C57BL/6J mice received 600 mg/kg APAP or 10 ml/kg saline and were sacrificed 0.5 h later. (A) Ion images of APAP, APAP-GLUC, and APAP-SULF colocalized with PC (38:5). (B) Ion images of APAP, APAP-GLUC, and APAP-SULF colocalized with PC (38:5) and PC (36:3) at 50 µm resolution.

Figure 8.

DESI-MSI analysis of region-specific distribution of the oxidative metabolites of APAP in the kidney. Fasted male C57BL/6J mice received 600 mg/kg APAP or 10 ml/kg saline and were sacrificed 0.5 h later. (A) Ion images of APAP-GSH, APAP-CYS, and APAP-NAC colocalized with PC (38:5). (B) Ion images of APAP, APAP-GLUC, and APAP-SULF colocalized with PC (38:5) and PC (36:3) at 50 µm resolution.

Metabolite formation colocalized with the regional distribution of parent APAP

The data so far clearly demonstrates the unique spatial localization of various APAP metabolites within the kidney after an APAP overdose. Because all metabolites are derived from the parent APAP molecule, we next determined whether derivative metabolites colocalized with parent APAP and if inhibition of metabolism by 4MP influenced this relationship. Hence, we colocalized APAP-GLUC, as well as the oxidative metabolites APAP CYS and APAP-NAC ions with APAP ions in the kidney. As seen in Figure 9A, the APAP ion signal (blue) significantly overlapped with the APAP-GLUC (green) (teal in merged inset panel) and seems to be predominantly in the inner medulla though some signals were present in the cortex (Figure 9A). Interestingly, APAP-CYS showed significant overlap with the APAP signal in the outer medulla (Figure 9B—purple in merged panel), whereas ion signals were almost completely absent in the inner medulla where APAP was highly concentrated. The APAP-NAC ion signal (Figure 9C) is located in the cortex to some positions where an intense APAP signal is also seen resulting in a merged purple signal, whereas at other locations APAP-NAC signal (red) is distinct from the APAP (blue) (enlarged panel, Figure 9C). This is also seen with APAP-GLUC and APAP-CYS, indicating the dynamic nature of the process such that some microlocations may have already metabolized parent APAP completely, whereas others may have ongoing metabolism where both species would be detected (with merged signal). Further experiments then examined the effect of cotreatment with 4MP on the colocalization of APAP metabolites with the parent compound. 4MP treatment did not affect the localization of metabolite overlap with parent APAP in the medulla or cortex, though levels of oxidative metabolites APAP-CYS and APAP-NAC were significantly decreased (Figs. 9D–F) as expected.

Figure 9.

Spatial localization of parent APAP with APAP-GLUC, APAP-CYS, or APAP-NAC in the kidney by DESI-MSI analysis at 50 μm resolution after severe (600 mg/kg) APAP overdose ± 4MP. Fasted male C57BL/6J mice received 600 mg/kg alone (A–C) or with 50 mg/kg 4MP (D–F), followed by sacrifice 30 min later. Ion images were merged for APAP and APAP-GLUC (A and D), APAP and APAP-CYS (B and E), and APAP and APAP-NAC (C and F). (c = cortex).

Multimodal high-resolution DESI-XS MSI coupled with immunostaining reveals the dynamic nature of ongoing APAP metabolism in the kidney

To further increase the relevance of the colocalization data, we next evaluated whether the signal for parent APAP, as well as APAP-GLUC and the oxidized metabolite APAP-NAC, could be localized to the proximal tubular cells in the renal cortex after APAP treatment. To examine this, we first carried out DESI-XS MSI analysis after 300 or 600 mg/kg APAP treatment at 50 μm pixel size resolution on kidney sections and overlaid ion images on the corresponding sections stained for CYP2E1 (positive staining in the cortex). This experiment provides interesting data that illustrates the dynamic nature of APAP metabolism, and several distinct features are evident. In mice treated with 600 mg/kg APAP, some regions with mild CYP2E1 staining (Figure 10A) show elevated APAP-GLUC signal with low APAP and APAP-NAC (Figs. 10B–D, red circles) indicating APAP metabolism to APAP-GLUC. Other regions show some signal for APAP as well as APAP-GLUC indicating ongoing metabolism but again lack APAP-NAC (Figure 10, green circles). Interestingly, some regions with intense CYP2E1 staining show increased levels of the oxidative metabolite APAP-NAC, with the absence of APAP or APAP-GLUC (Figure 10, blue circles) possibly indicative of a shift to metabolism to reactive metabolite. Similar changes were also seen in mice treated with the moderate 300 mg/kg overdose as well (Supplementary Figure 6). Interestingly, 4MP treatment inhibited CYP2E1-mediated metabolism but did not affect the expression of CYP2E1 in the kidney as illustrated by the positive stained cells (Figure 11A). The increase in intensity of APAP and APAP-GLUC (Figs. 11B and 11C) combined with the almost complete absence in intensity of APAP-NAC (Figure 11D) suggest that the activity of CYP2E1 was inhibited. Taken together, the use of DESI-MSI coupled with CYP2E1 immunostaining provides unprecedented insight into the spatial distinctions of APAP metabolism in the kidney.

Figure 10.

Colocalization of DESI-MSI ion signal to proximal tubular cells after 600 mg/kg APAP treatment. Fasted male C57BL/6J mice received 600 mg/kg APAP or 10 ml/mg saline followed by sacrifice 30 min later. A representative photomicrograph of the cryosection from APAP-treated animal used for DESI-MSI analysis and then immunostained for CYP2E1 (A). DESI ion images of APAP, APAP-NAC, and APAP-GLUC at 50 µm resolution (B–D, extreme left). Representative images of APAP, APAP-NAC, and APAP-GLUC associated DESI-MSI ion signals superimposed onto the same kidney cryo-section immunostained for CYP2E1 (B–D, middle). Panels on extreme right are magnification of indicated areas from images in the middle column.

Figure 11.

DESI-MSI analysis to spatially localize APAP, APAP-NAC, and APAP-GLUC to proximal tubular cells in kidney tissue sections after 600 mg/kg APAP treatment with 4MP. Fasted male C57BL/6J mice received 600 mg/kg APAP along with 50 mg/kg APAP or 10 ml/mg saline followed by sacrifice 30 min later. Representative photomicrograph of the cryosection used for DESI-MSI analysis and then immunostained for CYP2E1 (A). DESI ion images of APAP, APAP-NAC, and APAP-GLUC at 50 µm (B–D, extreme left). Representative images of APAP, APAP-NAC, and APAP-GLUC associated DESI-MSI signals superimposed onto the same kidney cryo-section immunostained for CYP2E1 (B–D-middle). Panels on extreme right are magnification of indicated areas from images in middle column.

GSH concentrates in the inner medulla while GSSG is produced in the cortex

Glutathione measurement in renal homogenate demonstrated that a slight depletion was evident after 600 mg/kg APAP (Figure 2E). To further localize the change in glutathione levels, DESI-MSI analysis was performed on the tissue section of mice treated with 600 mg/kg APAP alone, or with 50 mg/kg 4MP (Figure 12). In control tissue, the ion image shows that reduced GSH was detected in the cortex and outer medulla at lower levels than in the inner medulla where the majority of reduced GSH was concentrated (Figure 12A). After the 600 mg/kg APAP overdose, reduced GSH levels were decreased throughout the kidney though residual levels were still higher in the medulla compared with the cortex (Figure 12B). This regional difference is evident when overlayed with PC data in the inner medulla to demonstrate differences between the cortex and inner medulla (Figure 12D). This decrease was prevented by 4MP treatment (Figure 12C). Because oxidative stress is a feature of APAP-induced injury, we also examined whether glutathione disulfide (GSSG) also showed spatial restrictions after APAP. Although GSSG was not detected in control tissue (Figure 12E), it was exclusively mapped to the cortex of the kidney after APAP overdose (Figs. 12F and 12H) and this GSSG accumulation was also completely prevented by 4MP treatment (Figure 12G). Because oxidative metabolites such as APAP-GSH and APAP-NAC were predominantly seen in the cortex (Figs. 4E and 4F), this further indicates that GSH depletion in this compartment is more consequential.

Figure 12.

Reduced GSH and glutathione disulfide (GSSG) visualized by DESI-MSI in kidney after APAP overdose ± 4MP. Fasted male C57BL/6J mice received 600 mg/kg APAP, APAP + 50 mg/kg 4MP, or 10 ml/kg saline and were sacrificed 0.5 h later. Ion images were then obtained for reduced (m/z 308.0916) [M + H+] and oxidized (m/z 613.4766) [M + H+] glutathione at 50 µm resolution.

Discussion

 

APAP metabolism and reactions of protein adduct formation in the kidney are distinct from those in the liver

Although the main target of APAP toxicity is the liver, nephrotoxicity is also a clinical problem that is now being recognized. It is well known that significant differences exist in hepatic responses to a moderate (300 mg/kg) or severe (600 mg/kg) APAP overdose (Bhushan et al., 2014; Nguyen et al., 2023), but the effects of escalating APAP doses on nephrotoxicity are not characterized. Our earlier work showed that the kidneys likely metabolize APAP (Akakpo et al., 2020b), and we aimed to examine the dose-response and spatial distinctions in renal APAP metabolism in this study. Examination of circulating levels of APAP and its metabolites at the peak of metabolism, 30 min after the APAP overdose reveals that the nonoxidative metabolites, APAP-GLUC, and sulfate levels, are not saturated even at the highest APAP concentrations. Though the kidney has a significant influence on circulating metabolites (Rhee, 2018), measurement of APAP metabolites in the kidney revealed that sulfate levels are saturated by 600 mg/kg APAP, unlike glucuronides. This suggests that the increasing plasma levels of APAP sulfate were contributed by other sources such as the liver. Interestingly, glutathione depletion seems to occur at a much slower rate when measured in kidney homogenates when compared with the liver, where significant depletion is typically evident at 30 min post APAP (McGill et al., 2013). This is probably due to the extensive heterogeneity of cell types in the kidney compared with the liver, where the majority of cells are cytochrome P450 positive hepatocytes. The presence of CYP2E1 in the proximal tubular cells of the kidney (Lake et al., 2023) suggests that NAPQI is predominantly generated in these cells of the cortex and GSH depletion probably occurs selectively in these cells but is masked when measured in the whole kidney homogenate due to unaffected GSH stores in other kidney regions. These spatial nuances are clarified by DESI and will be discussed further below.

In addition to these spatial distinctions, differences in subcellular compartmentalization of NAPQI formation in the kidney versus the liver could also influence relative trends in protein adduct formation and glutathione depletion. We have demonstrated the absence of mitochondrial protein adducts in the kidney (Akakpo et al., 2020b) after an APAP overdose, and our recent studies indicate that NAPQI is predominantly generated in the ER in the kidney after an APAP overdose (Akakpo et al., 2023). Though glutathione S-transferase (GST) plays an important role in GSH scavenging of NAPQI, microsomal GST was found restricted to the glomerular and interstitial endothelium and collecting ducts deep in the medulla of the kidney (Harrison et al., 1989). Thus, the absence of GST at the subcellular location of NAPQI generation in proximal tubules could modulate efficacy of glutathione scavenging of NAPQI. A caveat to this, however, are findings from early studies (Coles et al., 1988) which indicate that the spontaneous reaction of NAPQI with GSH could predominate after an overdose, whereby regional variations in glutathione abundance are probably the main driver of relative differences in NAPQI scavenging and formation of protein adducts. These spatial nuances highlight the importance of understanding regional differences in renal APAP metabolism.

Mass spectrometry imaging allows the detection of regional variation in nonoxidative APAP metabolites

The mammalian kidney is a complex structure with multiple distinct compartments including the cortex, outer medulla, and inner medulla which also comprises the medullary rays (bundles of ducts) (Lake et al., 2023). These regions have various cell types such as renal tubular cells, glomerular cells, and interstitial cells with distinct metabolic functions (Lohr et al., 1998). Although enzymes involved in phase I and phase II reactions are predominantly localized to the proximal tubules of the kidney cortex (Knights et al., 2013; Lake et al., 2023), these cells are also the primary source of enzymes involved in oxidative reactions such as CYP2E1 (Cummings et al., 1999) and cysteine conjugate β-lyase (CCBL) (Lake et al., 2023). Among these, CYP2E1 has been shown to metabolize APAP, whereas CCBL was not involved (Emeigh Hart et al., 1996; Stern et al., 2005a,b). Advances in MSI provide the ability to investigate spatial variations in analytes directly within tissue regions. However, previous attempts using matrix-assisted laser desorption ionization (MALDI) coupled with atmospheric pressure (AP-MALDI) techniques have failed to localize all APAP metabolites. We have previously used DESI-MSI to detect both oxidative and nonoxidative APAP metabolites on liver sections with spatial annotation (Akakpo et al., 2022) and we now demonstrate that the ion intensities of APAP could be detected in several renal compartments including the cortex and outer and inner medulla. Further DESI-MSI analysis reveals a dose-dependent increase in metabolite signal intensity similar to LC-MS/MS data with kidney homogenate, attesting to the consistency between these 2 analytical methods. The significant elevation in APAP-GLUC signal compared with APAP-SULF is consistent with sulfotransferase enzymes rapidly saturating their activity in the kidney (Wong et al., 1986). The increased localization of APAP-GLUC to the inner medulla and APAP-SULF to the outer medulla and cortex aligns with the prevalence of UDP-glucuronosyltransferases and sulfotransferases in various regions of the kidney (Gaganis et al., 2007; Lake et al., 2023). These regional differences in intensity of the nonoxidative metabolites also likely reflect the dynamic nature of the simultaneous formation and accumulation of water-soluble metabolites down the concentration gradient in the kidney.

APAP overdose causes APAP-CYS accumulation to the outer medulla, redistribution of GSH, and accumulation of GSSG in the renal cortex

In the case of oxidative metabolites, analysis of kidney homogenate by LC-MS/MS detected APAP-CYS, but not APAP-GSH and APAP-NAC, which is corroborated by earlier studies (Akakpo et al., 2018, 2020b; Jensen et al., 2004; Yin et al., 2001; Zhang et al., 2018), indicating the rapid enzymatic conversion of APAP-GSH to APAP-CYS through γGT and DP (Lash, 2005). Although AP-MALDI was also unable to detect APAP-GSH and APAP-NAC (Mamun et al., 2023), our DESI-MSI approach reveals that these oxidative metabolites were readily detectable, but at much lower abundance compared with APAP-CYS. Both oxidative metabolites were predominantly in the cortex, with APAP-GSH being lower than APAP-NAC, with some signal also seen toward the medulla. Thus, maintenance of tissue architecture seems to be essential in the measurement of APAP-GSH and APAP-NAC by preventing interference from γGT and DP activity seen in renal homogenates, highlighting the advantage of DESI-MSI for these measurements. The ability to overlay ion images with immunostaining for CYP2E1 allows the identification of metabolite alterations at the level of proximal tubular cells, which is an additional advantage of DESI-MSI. The formation of APAP-GSH in the cortex is probably due to GSH-mediated scavenging of NAPQI generated by CYP2E1 localized in the region. The activity of γGT present in the brush border of the renal proximal tubule (Lash, 2005, 2011) would then convert this to APAP-CYS. Interestingly, APAP CYS exclusively accumulated in the outer medulla while being absent in the other regions of the kidney. Although this was corroborated by AP-MALDI which mapped APAP-CYS to the medulla (Mamun et al., 2023), the signal was much broader and spread into the inner medulla, probably due to the extensive pressure applied on sections during processing as well as spraying of matrix (Mamun et al., 2023), which could delocalize APAP-CYS. This accumulation of APAP-CYS to the outer medulla is likely due to the counter-current multiplication mechanisms (Itoh et al., 2014; Kondo et al., 1992) which maintain the osmotic medullary gradient in the outer medullary tissue. The data also indicate that conversion of APAP-CYS to APAP-NAC is not immediate because APAP-CYS concentrations far exceed those of APAP-NAC at this very early time point after APAP. The localization of N-acetyl transferase, the enzyme converting APAP-CYS to APAP-NAC in the cortex (Lake et al., 2023) explains the predominant APAP-NAC signal in this region. Our DESI MSI data also revealed interesting nuances related to renal glutathione handling after APAP which clarifies findings from whole kidney homogenates. Reduced glutathione revealed localization throughout the different regions of the kidney, though significant concentration was evident in the inner medulla in controls, which is consistent with other recent studies using MSI (Liu et al., 2017; Wang et al., 2021). As mentioned earlier, differences in regional distribution could explain the relatively mild glutathione depletion seen when measured in the whole kidney homogenate because lower GSH levels in controls in the cortex (where NAPQI formation occurs) would result in rapid depletion and protein adduct formation compared with levels in the medulla which are still higher than cortex even after APAP overdose. This regional specificity of changes is supported by the increase in GSSG predominantly within the renal cortex after APAP. This clearly indicates the presence of oxidative stress and supports the localization of antioxidant enzymes to proximal tubules in the cortex (Mohandas et al., 1984). Interestingly, this increase in GSSG is not accompanied by altered total glutathione levels, suggesting that these early changes are efficiently handled by renal antioxidant enzymes, contributing to the slower depletion of total renal GSH when compared with the liver (Hoivik et al., 1995). The functional consequence of 4MP treatment, which prevents APAP-induced renal injury (Akakpo et al., 2020b) by inhibition of CYP2E1 (Akakpo et al., 2018) was also replicated by DESI-MSI which revealed accumulation of parent APAP and concurrent absence of oxidative metabolites. 4MP also maintained GSH concentrations in the inner medulla suggesting robust protection against APAP nephrotoxicity.

In conclusion, for the first time to our knowledge, we report the comprehensive detection and spatial annotation of all relevant APAP metabolites in the kidney early after an overdose. This spatial information from DESI-MSI corroborates quantitative data from LC-MS/MS of whole renal homogenates and reveals distinct compartmentalization of oxidative APAP metabolites as well as GSSG in the kidney. The multimodal use of DESI-MSI coupled with CYP2E1 immunostaining provides unprecedented insight into metabolite localization and reveals that the spatially distinct formation and disposition of oxidative and nonoxidative APAP metabolites is a dynamic process occurring in different regions of the kidney. The data also provide novel insights into the renal handling of APAP metabolites, demonstrating that regional variations in glutathione abundance influence handling of early oxidant stress. Additionally, the effect of interventions such as 4MP on APAP metabolism is also illustrated by the distinct increase in parent APAP accompanied by a decrease in oxidative metabolites, substantiating the relevance of DESI-MSI in obtaining biologically relevant spatial information. Further studies extending the time course to later events in APAP pathophysiology and analysis of the disposition of APAP metabolites are currently ongoing. These studies are projected to further provide clarifications into how the dynamic process involved in the formation and disposition of APAP metabolites relates to APAP pathophysiology.

Supplementary data

Supplementary data are available at Toxicological Sciences online.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (R01 DK102142, DK125465); National Institute of General Medicine (NIGMS) funded Liver Disease COBRE (P20 GM103549, P30 GM118247). Additional funding to J.Y.A. through a CTSA grant from NCATS awarded to the University of Kansas for Frontiers: University of Kansas Clinical and Translational Science Institute No. 5TL1TR002368 is acknowledged.