Investigation of Xenobiotics Metabolism In <em>Salix alba</em> Leaves via Mass Spectrometry Imaging

Claire Villette; Lo&#239;c Maurer; Dimitri Heintz

doi:10.3791/61011

JoVE Journal > Environment

Medio ambiente

Investigation of Xenobiotics Metabolism In Salix alba Leaves via Mass Spectrometry Imaging

Published: June 15, 2020

doi:

10.3791/61011

Claire Villette¹, Loïc Maurer^1,2, Dimitri Heintz¹

¹Plant Imaging and Mass Spectrometry (PIMS), Institut de biologie moléculaire des plantes, CNRS,Université de Strasbourg, ²Département mécanique,ICube Laboratoire des sciences de l’ingénieur, de l’informatique et de l’imagerie, UNISTRA/CNRS/ENGEES/INSA

Summary

This method uses mass spectrometry imaging (MSI) to understand metabolic processes in S. alba leaves when exposed to xenobiotics. The method allows the spatial localization of compounds of interest and their predicted metabolites within specific, intact tissues.

Abstract

The method presented uses mass spectrometry imaging (MSI) to establish the metabolic profile of S. alba leaves when exposed to xenobiotics. Using a non-targeted approach, plant metabolites and xenobiotics of interest are identified and localized in plant tissues to uncover specific distribution patterns. Then, in silico prediction of potential metabolites (i.e., catabolites and conjugates) from the identified xenobiotics is performed. When a xenobiotic metabolite is located in the tissue, the type of enzyme involved in its alteration by the plant is recorded. These results were used to describe different types of biological reactions occurring in S. alba leaves in response to xenobiotic accumulation in the leaves. The metabolites were predicted in two generations, allowing the documentation of successive biological reactions to transform xenobiotics in the leaf tissues.

Introduction

Xenobiotics are widely distributed around the world due to human activities. Some of these compounds are water-soluble and absorbed by soil¹, and enter the food chain when they accumulate in plant tissues²^,³^,⁴. The plants are eaten by insects and herbivores, which are prey to other organisms. The intake of some xenobiotics and their impact on a plant’s health have been described⁵^,⁶^,⁷^,⁸, but only recently at a tissue level⁹. Therefore, it is still unclear where or how the metabolism of xenobiotics occurs, or if specific plant metabolites are correlated to xenobiotic accumulation in specific tissues¹⁰. Moreover, most research has overlooked the metabolism of xenobiotics and their metabolites in plants, so little is known about these reactions in plant tissues.

Proposed here is a method to investigate enzymatic reactions in biological samples that can be associated to the tissue localization of substrates and products of the reactions. The method can draw the complete metabolic profile of a biological sample in one experiment, as the analysis is non-targeted and can be investigated using custom lists of analytes of interest. Provided is a list of candidates tracked in the original dataset. If one or several analytes of interest are noted in the sample, the specific tissue localization can provide important information on the related biological processes. The analytes of interest can then be modified in silico using relevant biological laws to search for possible products/metabolites. The list of metabolites obtained is then used to analyze the original data by identifying the enzymes involved and localizing the reactions in the tissues, thus helping to understand the occurring metabolic processes. No other method provides information on the types of reactions occurring in the biological samples, the localization of the compounds of interest, and their related metabolites. This method can be used on any type of biological material once fresh and intact tissues are available and the compounds of interest can be ionized. The proposed protocol was published in Villette et al.¹² and is detailed here for use by the scientific community.

Protocol

1. Sample preparation

Obtain the biological sample and either keep it fresh and intact (e.g., do not force it into a tube) or freeze it. The proposed protocol applies to any type of solid biological sample (i.e., plant, animal, or human tissues) to localize compounds in specific tissues.
Cool down a cryomicrotome to -20 °C. Keep the sample holder and the blade at the same temperature.
If necessary, embed the object in M1 embedding medium to preserve it during cutting.
1. Pour some matrix in a plastic mold placed in the cryomicrotome chamber. Rapidly add the sample and pour some more embedding medium to cover it. Maintain the sample in the center of the mold as the matrix solidifies while cooling down.
  NOTE: Embedding is not necessary for all biological objects. Homogenous objects such as mouse brains do not need embedding medium and can be cut frozen.
Place the embedded or frozen sample on the cryomicrotome holder and cut it with a sharp blade. A thickness of 5–30 µm is good for plant samples, which are difficult to cut due to their heterogeneity. Adapt the cutting thickness and temperature to the sample, making several cuts to find the best conditions. In the example provided, the sample was cut at -20 °C. Consider keeping the sample under 0 °C to avoid sample degradation.
NOTE: The use of a binocular microscope placed next to the cryomicrotome can help determine the quality of the slices. The tissues must stay intact.
Carefully move the slice to an ITO-coated slide using forceps or a small paintbrush, then place a finger under the slide to warm up and dry the sample. Keep the slide in the cryomicrotome chamber until all samples are ready. Take the slide out of the chamber slowly to avoid heat shock.
NOTE: To check which side of the slide is ITO-coated, place a drop of water on the slide at room temperature (RT). The drop will stay round on the ITO-coated side or flatten on the uncoated side.
Draw marks on the slide using a thin marker pen or correction fluid and scan it with a high-resolution scanner before matrix deposition. The marks will be used to determine the sample's exact position on the slide when it is in the mass spectrometer. The best way to obtain precise points of reference is to draw a cross.

2. Matrix deposition

Prepare the MALDI matrix: weigh 70 mg of α-cyano-4-hydroxycinnamic acid (HCCA) matrix and dilute it in 10 mL of a water and methanol solution (50:50) with 0.2% TFA. Sonicate the matrix for 10 mins at RT. There may be extra solid matrix after sonication.
NOTE: Different types of MALDI matrices can be used depending on the analytes of interest.
Clean the matrix deposition robot with 100% methanol.
Using methanol and a precision wipe, clean the slide, holding the samples without touching them and without removing the marks. Add a clean coverslip over the slide in an area without sample. That part of the slide is then placed over the optical detector to track matrix deposition.
NOTE: To optically track matrix thickness, the slide and coverslip must be very clean.
Use the robot for matrix deposition: place only 6 mL of the matrix solution in the reservoir, and add 2 mL of 100% methanol for a total volume of 8 mL.
NOTE: Recording of matrix thickness along deposition is automatic and can be recovered using a USB stick. The development of a spraying method is not detailed here.

3. Data acquisition

Place 1 µL of the matrix on a MALDI plate to calibrate the mass spectrometer using the matrix peaks as references. The development of an acquisition method on the mass spectrometer is not detailed here.
1. Insert the MALDI plate in the source, click “Load Target” in ftmsControl software, and wait for the plate to be loaded.
2. Choose the position of the matrix spot by clicking on the spot in the image representing the MALDI plate.
3. Indicate the sample name and the folder in the “Sample Info” tab of the software.
4. Start acquisition by clicking “Acquisition”.
5. Move the plate slightly during the acquisition using the mouse pointer on the “MALDI video” tab so that the laser points at different spots.
6. When the acquisition is finished, go to the “Calibration” tab, choose HCCA calibration list in Quadratic Mode and click Automatic. The global calibration result is indicated in the “Calibration Plot” window and should be less than 0.2 ppm for a SolariX XR 7T.
7. If the calibration is good, click “Accept” and save the method.
Once the matrix deposition on the samples is finished, place the slide in a slide adapter and retrieve the position of the marks using a plastic cover.
In flexImaging software, set up a new imaging run using the first window that opens with the software.
1. Name the imaging run, select the Result Directory, and click Next.
2. Indicate the raster size (i.e., user-defined measurement region), choose the method to be used (calibrated at section 3.1), and click Next.
3. Load the optical image of the slide obtained at step 1.6 after scanning the slide and click Next.
4. The image opens in a wider window. Use the marks on the slide to teach the position of the slides: in ftmsControl, place the target of the MALDI video window on the exact position of a mark, then come back to felxImaging and click on the exact same point on the optical image. Repeat this for three independent points. The plastic cover bearing the marks is placed on a MALDI plate to recover its position and facilitate teaching.
  NOTE: If the marks were made with a marker, adding white correction fluid at the back of the slide can make them stand out during teaching.
Draw the regions of interest in the samples in the flexImaging software using the “Add Measurement Regions” tools.
Save the imaging run and an “AutoXecute Sequence” if several samples are to be analyzed.
Launch a sequence using “AutoXecute Batch Runner” if several samples are analyzed.
NOTE: Synchronize the data regularly to a secure location.

4. Data processing

Import the raw data to the visualization software (SCiLS Lab) and create a dataset. This is done in two separate steps in this software.
1. Use the “Batch Importer” tool and select the raw data, indicate the target directory, and click on “Import”.
2. After import, several datasets can be combined for further analysis. To combine datasets, use the “File|New|Dataset” tool.
3. Select the type of instrument used for acquisition, add the imported datasets by clicking “+”, and arrange the images by clicking and dragging the objects.
4. Check the mass range settings or modify if needed by indicating the range of interest. Click Next to see the import summary and launch the import.
Visualize the m/z of interest in the different tissues of a sample and/or in different samples. Simply click on the spectra to select the m/z of interest or type the value in the m/z box.
1. Perform statistical tests if needed to search for colocalized or discriminative values between different tissues and/or different samples to determine the m/z of interest. The tools are available in the “Tool” menu and will not be described in this protocol.
Export the m/z of interest (e.g., colocalized, discriminative compounds) as a .csv file from the Object tab. The whole dataset can also be exported if it is not too big (i.e., maximum 60,000 lines). Click on the “Export” icon on the right side of the region of interest indicated by a red arrow.
Import the .csv file to create a new dataset in the annotation software (Metaboscape). Click on “Projects | Import CSV project”. Be aware that only the exact m/z will be considered, and the isotopic profile is lost.
NOTE: The 5.0 software version allows direct import of data from the visualization software, which enables a more accurate annotation, because the isotopic profile can be considered.
Annotate with custom-made analyte lists, which can be derived from publicly available databases. A template is given by the software to create analyte lists.
Use the prediction software (Metabolite Predict) to perform in silico prediction of the metabolites of the annotated compounds. The developed formula of the compound of interest is needed, either drawn by the user in the software or imported as a .mol file. Then the method is a simple step-by-step protocol.
Recover the list of metabolites, create an analyte list, and use it in the annotation software to annotate the raw data with predicted metabolites. Alternatively, if the list of metabolites is short, manually search for it in step 4.8.
Visualize the tissue distribution of the metabolites in the raw data in the visualization software.
Recover the names of the enzymes involved in the metabolic processes in the prediction software by right-clicking on the predicted metabolite of interest.

Representative Results

This protocol was applied to fresh leaves sampled from a S. alba tree exposed to xenobiotics in the environment. The process is depicted in Figure 1. The first step is to prepare thin slices of the sample of interest. Plant samples are often more difficult to cut than animal samples, as the tissues are heterogeneous and can contain water and/or air. This difficulty is handled using embedding medium, which forms a homogeneous block around the sample. The matrix deposition is facilitated by the use of a robot, avoiding hand manipulation and assuring reproducible results. The MALDI matrix layer thickness is followed during the entire process and can be recorded. Data acquisition requires learning to handle a high-resolution mass spectrometer and to adapt the method to the type of samples and compounds being investigated. Raw data are imported into a visualization software to search for the compounds of interest and display their tissue localization. Discriminative or colocalized compounds can be found using statistical tools available in the software. At this step, only the exact m/z of the compounds are known. The compounds of interest are then exported to the annotation software, which can compare the exact m/z with a custom list of compounds of interest defined by the user. If the exact m/z matches the m/z of a compound of interest, it is annotated. In the context of a metabolic profile investigation, the annotated compounds are selected for in silico prediction of metabolites. The types of biological reaction rules used to generate metabolites are easily chosen by the user, as well as the number of generations over which the metabolites are predicted. The list of predicted metabolites can be used in the annotation software to search for matches between raw data exact m/z and predicted metabolites m/z (Figure 1). The annotated metabolites can be searched for in the visualization software to obtain their tissue localization (Figure 2). The enzymes involved in the metabolism of the original compounds of interest can be recovered to draw the metabolic reactions occurring in the biological sample (Figure 3).

In this example, the drug telmisartan was identified in the plant leaves; it was distributed throughout the tissues. Telmisartan's metabolites were predicted and searched for in the raw data. The annotations showed that one first-generation (I) metabolite was detected in the internal tissues of the leaves and further degraded into second-generation (II) metabolites, which were localized in internal tissues or were more generally distributed in all leaf tissues (Figure 2). These results suggest an active metabolic reaction in the leaves to degrade telmisartan. The process was applied to several compounds of interest annotated in the leaves, and the enzymes involved in the reactions were recovered to investigate their role in the plant's response to xenobiotics accumulation. This gives an overview of the enzymes involved in xenobiotics metabolism in S. alba leaves (Figure 3).

Figure 1. General structure of the method. A fresh sample is cut and placed on an ITO-coated slide sprayed with the appropriate MALDI matrix. The MALDI acquisition provides raw data from which the localization within the tissue can be observed with the SCiLS Lab software. Metaboscape is used for annotation, and Metabolite Predict is used for metabolite (i.e., catabolites and conjugates) prediction. Please click here to view a larger version of this figure.

Figure 2. Example of the results obtained on S. alba leaves exposed to xenobiotics. Telmisartan was identified in the plant leaves and visualized in all the tissues. Telmisartan metabolites were predicted and annotated on the raw data to visualize their tissue localization. The first-generation (I) metabolite C₃₃H₃₂N₄O₃ was localized mainly in the internal tissues, while second-generation (II) metabolites were sometimes more generally distributed. This figure was adapted with permission from Villette et al.¹². Please click here to view a larger version of this figure.

Figure 3. Global enzymatic profile proposed for potential reactions occurring in S. alba leaves in response to xenobiotics exposure. The metabolite prediction and annotation on the object of interest suggested the potential enzymatic reactions responsible for the metabolism of the compound of interest. This figure was adapted with permission from Villette et al.¹². Please click here to view a larger version of this figure.

Discussion

The critical part of this protocol is the sample preparation: the sample must be soft and intact. Cutting is the most difficult part, as the temperature and thickness of the sample can vary depending on the type of sample studied. Animal tissues are usually homogeneous and easier to cut. Plant samples often incorporate different structures and therefore are more difficult to keep intact as the blade encounters soft, hard, or empty vascular tissues. It is highly recommended to use fresh tissues when working with plant samples to avoid the formation of ice in the hydrophilic tissues and their destruction. The slices must be moved gently when deposited on the ITO-coated slide. The MALDI matrix was slightly diluted to avoid the clogging of the spray sheet with matrix crystals, which can happen if the 2 mL of 100% methanol are not added at step 2.4.

This method offers an easy one-day sample preparation protocol that provides reproducible results due to the use of a robot for MALDI matrix deposition. The proposed protocol necessitates competence in tissue cutting and mass spectrometry imaging and only applies to compounds that can be ionized. However, it provides molecular identification without the need for labelling as used in immunohistochemistry¹¹. High sensitivity is achieved because the compounds are directly ionized in the tissues or cells, avoiding dilution effects produced by an extraction protocol¹². The samples are analyzed in a non-targeted way, which allows for a large-scale profiling of the endogenous or xenobiotics compounds in the samples. Therefore, biological responses to exogenous compounds can be followed. The in silico prediction of metabolites coupled to non-targeted analysis adds another dimension to classical mass spectrometry imaging, because metabolic reactions can be monitored without a priori knowledge of the exogenous compounds that will accumulate in the tissues. To date, only known compounds and a few metabolites have been followed with this method (e.g., drugs of interest fed to rats)¹³. With the proposed protocol, the original compounds and their metabolites can be localized within the tissues, and the biological responses to the accumulation of exogenous compounds and/or their metabolites can be followed.

This protocol does not only apply to the response of plants to xenobiotics, but can also be used to understand animal metabolism in response to drugs, to follow plant/fungi interactions, plant response to biotic or abiotic stresses, or to understand the evolution of diseases, revealing the metabolic processes in the tissues of interest.

Divulgaciones

The authors have nothing to disclose.

Acknowledgements

We thank Charles Pineau, Mélanie Lagarrigue and Régis Lavigne for their tips and tricks regarding sample preparation for MALDI imaging of plant samples.

Materials

Cover slips	Bruker Daltonics	267942
Cryomicrotome	Thermo Scientific
Excel	Microsoft corporation
flexImaging	Bruker Daltonics
ftmsControl	Bruker Daltonics
GTX primescan	GX Microscopes
HCCA MALDI matrix	Bruker Daltonics	8201344
ImagePrep	Bruker Daltonics
ITO-coated slides	Bruker Daltonics	237001
M1-embedding matrix	ThermoScientific	1310
Metabolite Predict	Bruker Daltonics
Metaboscape	Bruker Daltonics
Methanol	Fisher Chemicals		No specific reference needed
MX 35 Ultra blades	Thermo Scientific	15835682
Plastic molds			No specific reference needed
SCiLS Lab	Bruker Daltonics
SolariX XR 7Tesla	Bruker Daltonics		The method proposed is not limited to this instrument
Spray sheets for ImagePrep	Bruker Daltonics	8261614
TFA	Sigma Aldrich		No specific reference needed

Referencias

Zhang, D., Gersberg, R. M., Ng, W. J., Tan, S. K. Removal of pharmaceuticals and personal care products in aquatic plant-based systems: A review. Environmental Pollution. 184, 620-639 (2014).
Adeel, M., Song, X., Wang, Y., Francis, D., Yang, Y. Environmental impact of estrogens on human, animal and plant life: A critical review. Environment International. 99, 107-119 (2017).
Prosser, R. S., Sibley, P. K. Human health risk assessment of pharmaceuticals and personal care products in plant tissue due to biosolids and manure amendments, and wastewater irrigation. Environment International. 75, 223-233 (2015).
Wang, J., et al. Application of biochar to soils may result in plant contamination and human cancer risk due to exposure of polycyclic aromatic hydrocarbons. Environment International. 121, 169-177 (2018).
Marsik, P., et al. Metabolism of ibuprofen in higher plants: A model Arabidopsis thaliana cell suspension culture system. Environmental Pollution. 220, 383-392 (2017).
He, Y., et al. Metabolism of ibuprofen by Phragmites australis: uptake and phytodegradation. Environmental Science and Technology. 51 (8), 4576-4584 (2017).
Huber, C., Bartha, B., Harpaintner, R., Schröder, P. Metabolism of acetaminophen (paracetamol) in plants-two independent pathways result in the formation of a glutathione and a glucose conjugate. Environmental Science and Pollution Research. 16 (2), 206-213 (2009).
Thomas, F., Cébron, A. Short-term rhizosphere effect on available carbon sources, phenanthrene degradation, and active microbiome in an aged-contaminated industrial soil. Frontiers in Microbiology. 7, 1-15 (2016).
Villette, C., et al. In situ localization of micropollutants and associated stress response in Populus nigra leaves. Environment International. 126, 523-532 (2019).
Sandermann, H. Plant metabolism of organic xenobiotics. Status and prospects of the ‘Green Liver’ concept. Plant Biotechnology and In Vitro Biology in the 21st Century. , 321-328 (1999).
Sula, B., Deveci, E., Özevren, H., Ekinci, C., Elbey, B. Immunohistochemical and histopathological changes in the skin of rats after administration of lead acetate. International Journal of Morphology. 34 (3), 918-922 (2016).
Villette, C., Maurer, L., Wanko, A., Heintz, D. Xenobiotics metabolization in Salix alba leaves uncovered by mass spectrometry imaging. Metabolomics. 15, 122 (2019).
Khatib-Shahidi, S., Andersson, M., Herman, J. L., Gillespie, T. A., Caprioli, R. M. Direct Molecular Analysis of Whole-Body Animal Tissue Sections by Imaging MALDI Mass Spectrometry. Analytical Chemistry. 78 (18), 6448-6456 (2006).

Play Video

PDF

DOI

DOWNLOAD MATERIALS LIST

Citar este artículo

Villette, C., Maurer, L., Heintz, D. Investigation of Xenobiotics Metabolism In Salix alba Leaves via Mass Spectrometry Imaging. J. Vis. Exp. (160), e61011, doi:10.3791/61011 (2020).