The tissue images and MS data presented in this note were provided by Dr. Reid Groseclose and Dr. Steve Castellino from the Department of Bio-Imaging of GSK, 709 Swedeland Road, King of Prussia, PA 19406, USA.

Application & Background

On average, only one in ten drug candidates that enters clinical phase testing is approved for human treatment (1). While many factors contribute to the 90% failure rate of potential drugs in clinical trials, safety and efficacy are most often cited as the reasons drug candidates are abandoned. Thus, there is an obvious benefit to improved nonclinical drug analytical analyses for in vitro and in vivo studies (1). One such methodology that is emerging in pre-clinical pharmaceutical studies is matrix-assisted laser desorption/ionization (MALDI) imaging mass spectrometry (IMS) (2). IMS allows for direct spatial correlation of histological findings and analytical molecular data. By preserving the location of analytes at a high spatial resolution, MALDI IMS can allow a better understanding of the molecular and cellular processes underlying the pathologic findings in pre-clinical drug development that could enable scientists to better anticipate and predict human risk and response to a candidate pharmaceutical (2).

In the present study, we aimed to investigate potential age sensitivity in dabrafenib-induced renal toxicity (2). Dabrafenib (DAB) is a competitive ATP-competitive inhibitor of RAF kinase activity that has been approved for use in adults with tumors with a BRAF V600E mutation (2). However, early studies demonstrated adverse kidney effects in juvenile rats in response to DAB, including tubular deposits and dilation, cortical cysts, and tubular basophilia, although investigation into the mechanism of action of these adverse effects went beyond the scope of the original study (3). Here, we aim to use MALDI IMS to analyze the distribution of DAB and its metabolites in the kidney tissue of juvenile rats and to determine the molecular composition of the renal deposits (2).

Experimental

Study Design

All animal procedures were conducted in an American Association for the Accreditation of Laboratory Animal Care accredited facility at GlaxoSmithKine (GSK) in accordance with GSK policies on the care, welfare, and treatment of laboratory animals, and they were reviewed and approved by GSK's Institutional Animal Care and Use Committee as appropriate. Ten litters of 10 male rats were assigned to receive either the vehicle only or the vehicle + DAB.2 Five different treatment periods and two different doses administered by oral gavage were assessed (Table 1). Groups 5 and 10 were included to match the conditions of the original juvenile toxicity study (3). These rats were dosed at 10 mg/kg/day from post-natal day (PND) 7 to PND 21 and then increased to 20 mg/kg/day from PND 22 to PND 35. For each group, termination occurred approximately 24 hours after the last dose (2).

 
Table 1.  Treatment conditions of 10 experimental groups.

Table 1. Treatment conditions of 10 experimental groups.

 

Sample Preparation

At necropsy, the left kidney was bisected transversely at the hilus and kept frozen until MALDI IMS analysis.2 Thin sections (6μm) from the midline of the kidney tissues were collected in a cryostat at -20°C and mounted onto ITO coated glass slides. Prior to matrix deposition, all tissue sections were scanned at high magnification (20x-40x) using an Aperio Scanscope CS. For a subset of animals from each group, kidney tissue homogenates were prepared from serial sections adjacent to those collected for MSI and used for LC-MS analysis (2).

Matrix Application

DHB was applied to the slides at a concentration (C) of 50 mg/mL (in 50:50 methanol:water) using the HTX TM-Sprayer (2). The heated spray allowed high flow rate deposition for maximum extraction of analytes, while minimizing spatial delocalization. The total sample preparation spraying took 12 minutes per ITO slide (25x75 mm).2 The slides were coated using the following parameters:

 
Table 1.  Spray parameters.

Table 1. Spray parameters.

 

Mass spectrometry analysis

LC-MS quantification was performed using a Thermo Orbitrap XL.2 MS imaging experiments were performed using a Bruker Solarix 7T Fourier transform-ion cyclotron resonance mass spectrometer (FT-ICR MS). Images were acquired at spatial resolutions ranging from 10 to 100 μm. Mass spectra were acquired both in full scan mode (m/z 80-1000) and continuous accumulation of selected ions (CASI) mode for the mass range of m/z 460-620 in order to enhance sensitivity for DAB-related material. In order to estimate tissue concentration of DAB-related material by MALDI IMS, a tissue mimetic model was used (2). A tissue mimetic model consists of homogenate tissue cores spiked with known concentrations of the analyte of interest (4). These cores are then analyzed under the same conditions as the tissue of interest so the analyte quantity in the tissue can be estimated from comparing the ion intensity between the calibration curve of the mimetic tissue model and the MALDI images of the samples.4 All ion images were generated using FlexImaging v4.0 software from the raw data (2).

Results

Histological Findings

Rats exposed to DAB only pre-weaning (Groups 6, 7 and 10) were found to have a high incidence of tubular deposits, while groups dosed only post-weaning (Groups 8 and 9) had no incidence of tubular deposits. Although dosed for a longer period of time, Group 10 had a similar density of tubular deposits to Groups 6 and 7, suggesting the cessation of tubular deposit formation around PND 22, despite continued DAB treatment. No tubular deposits were noted in the control animals (Figure 1).

Figure 1.  Optical scan of kidney tissue sections from juvenile rats: (a) Group 6 Rat 4, (b) Group 10 Rat 2, (c) Group 1 Rat 1.

Figure 1. Optical scan of kidney tissue sections from juvenile rats: (a) Group 6 Rat 4, (b) Group 10 Rat 2, (c) Group 1 Rat 1.

Quantification of DAB Metabolites by LC-MS and Mimetic Tissue Model

The youngest rats of Group 6 were found to have the highest concentration of DAB metabolites present in the kidney by LCMS analysis. Of the three metabolites of DAB, carboxy- Dabrafenib (CDAB) was found to be the predominant species in the kidney of the Group 6 (PND 7-14) rats. By comparing the tissue mimetic model to the average ion signal detected by MALDI IMS for an analyte in a tissue section, the estimated the concentration of CDAB in for two rats in Group 6 were 8,000 ng/g and 6,400 ng/g, respectively (Figure 2). LC-MS of tissue homogenates found the CDAB concentration in the kidneys of these two rats to be 11,600 ng/g and 6,100 ng/g, respectively. These data support the tissue mimetic model as a reliable method to quantify analytes in MALDI imaging, especially when considering that MALDI IMS results are generated from a single tissue section (<1 mg) while LC-MS results are generated on half-kidney extracts (>200 mg). In addition, the differences could be partially attributed to regional concentration differences of analytes in the kidney.

 
 
Figure 2.  Optical scans of kidney tissue sections from PND 7-13 juvenile rats (a) Group 6 Rat 2 and (b) Group 6 Rat 3 analyzed by MALDI MSI in CASI mode ( m/z  460-620) at 100 μm spatial resolution. (c) and (d) Respective ion images for CDAB ( m/z  508.1083) and quantity (ng/g of tissue) predicted using tissue mimetic model. (e) Pre-analysis optical scan of tissue mimetic model cores with spiked concentration of CDAB labeled for each (f) ion image for CDAB ( m/z  508.1083) from tissue model analyzed under the same conditions as the kidney tissue sections.

Figure 2. Optical scans of kidney tissue sections from PND 7-13 juvenile rats (a) Group 6 Rat 2 and (b) Group 6 Rat 3 analyzed by MALDI MSI in CASI mode (m/z 460-620) at 100 μm spatial resolution. (c) and (d) Respective ion images for CDAB (m/z 508.1083) and quantity (ng/g of tissue) predicted using tissue mimetic model. (e) Pre-analysis optical scan of tissue mimetic model cores with spiked concentration of CDAB labeled for each (f) ion image for CDAB (m/z 508.1083) from tissue model analyzed under the same conditions as the kidney tissue sections.

Analysis of Renal Distribution of DAB Metabolites by MALDI IMS

Low-resolution MALDI IMS (100 μm) of kidney tissues demonstrated that CDAB was most concentrated near the tubular deposits found in the inner medulla and pelvic regions of the kidney of the youngest rats (Figure 3). Higher resolution MALDI IMS of the medulla of a Group 6 rat shows that CDAB is specifically localized to the lumen of the damaged collecting ducts, but no DAB metabolites were found directly from the tubular deposits themselves (Figure 3).

Figure 3.  (a) Optical scan of kidney tissue sections from Group 6 Rat 2 analyzed by MALDI IMS in CASI mode ( m/z  460–620) at 25 μm spatial resolution. (b) Ion image for CDAB ( m/z  508.1083). (c) Serial H&amp;E. (d) 10× magnification of outlined region in optical scan. (e) Magnified view of CDAB ion image coregistered with optical scan. (f) Magnified view of CDAB ion image co-registered with H&amp;E. (g) Histopathology annotated H&amp;E

Figure 3. (a) Optical scan of kidney tissue sections from Group 6 Rat 2 analyzed by MALDI IMS in CASI mode (m/z 460–620) at 25 μm spatial resolution. (b) Ion image for CDAB (m/z 508.1083). (c) Serial H&E. (d) 10× magnification of outlined region in optical scan. (e) Magnified view of CDAB ion image coregistered with optical scan. (f) Magnified view of CDAB ion image co-registered with H&E. (g) Histopathology annotated H&E

Analysis of Molecular Composition of Tubular Deposits by MALDI MSI

In order to elucidate the mechanism by which pre-weaning exposure to DAB contributes to the formation of tubular deposits, it was necessary to analyze the composition of these deposits. Specifically, since no DAB metabolites were found directly to comprise the tubular deposits that formed after pre-weaning exposure to DAB, the molecular process by which these tubular deposits are formed could provide critical insight into potentially age-sensitive drug toxicities. High resolution MALDI MSI revealed that the tubular deposits were clusters of several DHB-calcium compounds (Figure 4).

Figure 4.  (a) Optical scan of kidney tissue sections from Group 6 Rat 4 analyzed by MALDI MSI in fullscan mode ( m/z  100–1000) at 10 μm spatial resolution. (b) Ion image for [3DHB – H + Ca]+ (m/z 501.0339) in green, and glycerophosphocholine [M + K]+ ( m/z  296.0659) in red. (c) Ion image for [2DHB – 2H + Ca]•+ ( m/z  345.9999). (d) Ion image for [2DHB – O – 2H + Ca]•+ ( m/z  330.0047).

Figure 4. (a) Optical scan of kidney tissue sections from Group 6 Rat 4 analyzed by MALDI MSI in fullscan mode (m/z 100–1000) at 10 μm spatial resolution. (b) Ion image for [3DHB – H + Ca]+ (m/z 501.0339) in green, and glycerophosphocholine [M + K]+ (m/z 296.0659) in red. (c) Ion image for [2DHB – 2H + Ca]•+ (m/z 345.9999). (d) Ion image for [2DHB – O – 2H + Ca]•+ (m/z 330.0047).

Conclusions

MALDI MSI is an emerging analytical technique that can help toxicologists associate pathological findings with molecular biology (2). By understanding the mechanism of action of a drug’s toxicity, safety risks to humans can be better assessed before the clinical phase of drug development. In the present study, age sensitive nephrotoxicity to BRAF inhibitor DAB was demonstrated in pre-weaned juvenile rats. While it was known that younger rats have less efficient biliary clearance than older rats, these MALDI MSI data provided several insights into the potential mechanism of age-related renal toxicity of DAB. First, it was confirmed that the histologic nephrotoxicity was spatially associated with higher concentrations of DAB metabolites. Second, it demonstrated that the tubular deposits are not themselves comprised of DAB or its metabolites, but calcium salts. Third, it was found that CDAB primarily localized to the lumen of damaged collecting ducts, which are the primary site of calcium regulation in the kidney, suggesting a potential causal link between the pathologic observation of tubular deposits and DAB pre-weaning exposure (2).

 
 

REFERENCES

(1) Van Norman, G. A. (2016). Drugs, Devices, and the FDA: Part 1: An Overview of Approval Processes for Drugs. JACC: Basic to Translational Science, 1(3), 170–179. https://doi.org/10.1016/j.jacbts.2016.03.002

(2) Groseclose, M. R., Laffan, S. B., Frazier, K. S., Hughes-Earle, A., & Castellino, S. (2015). Imaging MS in toxicology: An investigation of juvenile rat nephrotoxicity associated with dabrafenib administration. Journal of the American Society for Mass Spectrometry, 26(6), 887–898. https://doi.org/10.1007/s13361-015-1103-4

(3) Laffan, et al. Manuscript in preparation.

(4) Groseclose, M. R., & Castellino, S. (2013). A mimetic tissue model for the quantification of drug distributions by MALDI imaging mass spectrometry. Analytical Chemistry, 85(21), 10099–10106. https://doi.org/10.1021/ac400892z