Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Aug 1.
Published in final edited form as: J Am Soc Mass Spectrom. 2008 Apr 8;19(8):1069–1077. doi: 10.1016/j.jasms.2008.03.016

Enhancement of Protein Sensitivity for MALDI Imaging Mass Spectrometry after Chemical Treatment of Tissue Sections

Erin H Seeley 1, Stacey R Oppenheimer 1,2, Deming Mi 1, Pierre Chaurand 1, Richard M Caprioli 1
PMCID: PMC2582528  NIHMSID: NIHMS65679  PMID: 18472274

Abstract

MALDI Imaging Mass Spectrometry (IMS) has become a valuable tool for the investigation of the content and distribution of molecular species in tissue specimens. Numerous methodological improvements have been made to optimize tissue section preparation and matrix deposition protocols, as well as MS data acquisition and processing. In particular for proteomic analyzes, washing the tissue sections prior to matrix deposition has proven useful to improve spectral qualities by increasing ion yields and the number of signals observed. We systematically explore here the effects of several solvent combinations for washing tissue sections. To minimize experimental variability, all of the measurements were performed on serial sections cut from a single mouse liver tissue block. Several other key steps of the process such as matrix deposition, MS data acquisition and processing have also been automated or standardized. To assess efficacy, after each washing procedure the total ion current and number of peaks were counted from the resulting protein profiles. These results were correlated to on-tissue measurements obtained for lipids. Using similar approaches, several selected washing procedures were also tested for their ability to extend the lifetime as well as revive previously cut tissue sections. The effects of these washes on automated matrix deposition and crystallization behavior as well as their ability to preserve tissue histology were also studied. Finally, in a full scale IMS study, these washing procedures were tested on a human renal cell carcinoma biopsy.

Keywords: Imaging mass spectrometry, MALDI, Time-of-Flight, Tissue, Washing

Introduction

Imaging mass spectrometry (IMS) is becoming a key technology for the investigation of the molecular content of thin tissue sections in direct spatial correlation with the underlying histology [1–4]. An analysis by MALDI MS performed directly on a tissue section allows detection of a wide range of endogenous molecular species such as lipids [5,6], peptides [7–10] and proteins [1,3,11] as well as administered pharmaceuticals [12–15]. Typically, several hundred distinct protein mass signals are observed in a single analysis in a mass-to-charge (m/z) range up to 70,000, although the technology is capable of measurements exceeding 200,000 [16]. In IMS, the homogeneous deposition of the MALDI matrix followed by a systematic acquisition of data allows reconstruction of hundreds of distinct ion signal intensity maps, or images, for virtually all of the different molecular species observed [1,3]. IMS has been used to study a wide variety animal and plant systems [17]. Particular emphasis has been given to the study of disease tissues from both the biological and clinical perspectives [18–20]. Of particular interest is the fact that the expression of subsets of protein signals could be correlated to the progression of disease as well as patient outcome.

Over the last decade, numerous methodological and technical developments have emerged to improve upon sample preparation for protein analyses. Efforts in sample preparation have the goal of increasing the quality and reproducibility of the mass spectrometric data. Schwartz et al. have first defined a set of practical guidelines for the proper handling of tissue sections including tissue storage, sectioning and mounting for peptide and protein analyses [21]. In the same article, other parameters such as the choice and concentration of matrix, solvent composition, matrix deposition methodologies, and instrumental parameters for optimal mass spectrometric data acquisition are discussed [21]. The influence of the tissue thickness (between 2 and 40 μm) has also been studied [22]. The authors concluded that the overall peak intensities and the total number of observed peaks drastically increased as the sections became thinner. Several MALDI compatible tissue staining protocols combined with the use of conductive indium-tin oxide coated glass slides for tissue mounting have been developed to directly link the observed histological features with the recovered mass spectrometric data [23]. Several tissue rinsing procedures prior to matrix application have also been examined [3,24].

Matrix deposition strategies and procedures aimed towards proteomic analyses have dramatically evolved. The first tissue imaging experiments were performed after manually spreading one large drop of matrix over the tissue section [1,25]. Although in some instances this procedure is successful, this matrix deposition approach often leads to a significant degree of analyte delocalization. Matrix deposition by pneumatic spray coating is largely employed for homogeneous deposition [11,21]. This approach has the disadvantage of in some cases significantly degrading the quality of the mass spectrometric data and to some extent delocalizing the analytes [16]. More recently, the development of technologies for automated dispensing of picoliter size droplets of matrix on the sections into independent spots of approximately 100–200 μm diameter have circumvented the signal quality and delocalization aspects but limit the imaging resolution to the center-to-center distance between adjacent spots [3,26,27]. Because of the high spotting accuracy of the automated dispensing approaches, a histology-directed matrix deposition strategy to target precise areas of the tissue sections has been developed [4]. Further, these spotters have also been used to first print enzymatic proteases prior to IMS [28,29]. This approach allows for direct on-tissue identification of proteins in spatially discrete regions using enzymatic digestion followed by IMS and MALDI MS/MS analyses of the resulting peptides. This strategy has been successfully employed to analyze the proteomic content of sections from formalin fixed paraffin embedded tissue specimens [30].

Finally, MS instrumentation for the acquisition of IMS data as well as post-acquisition data processing approaches has significantly progressed. Historically, IMS data were first collected on MALDI TOF mass spectrometers equipped with compact nitrogen lasers (337 nm). Additionally, an iris can been installed across the laser beam pathway to control the irradiated tissue surface [1]. Homebuilt MALDI-TOF MS systems with laser beam dimensions on target of 10 μm or smaller have since been developed [31,32]. Another homebuilt position sensitive mass spectrometer has been described capable of generating images with a lateral resolution of better than 5 μm [33]. Early commercially available MALDI TOF systems allowed laser repetition rates of a few Hertz. Current systems with enhanced electronics and improved acquisition systems are now equipped with solid state Nd:YAG lasers (355 nm) functioning at repetition rates as high as 200 to 300 Hz [25,34]. IMS has also been successfully performed on other types of commercially available MS instruments namely QqTOF, ion traps, and more recently FT-ICR systems [7,35,36]. Whereas early software programs for IMS data acquisition and visualization were written “in-house” [31,37], today almost every commercially available MALDI-based mass spectrometer is “imaging capable” and can be equipped with software for automated data acquisition, image reconstruction and visualization. IMS data pre-processing (background subtraction and noise estimation, peak alignments, normalization) and post-processing (statistical analyses) have also been developed [38–41].

In this article, we explore tissue section rinsing protocols to improve IMS protein signal sensitivity. The action of washing or rinsing the sections prior to matrix application and MALDI IMS analyses is performed to remove contaminants such as tissue embedding media as well as endogenous molecular species, namely lipids or biological salts, that may affect protein desorption/ionization efficiency. The use of organic solvents to extract lipids from blood and tissues has been developed in the early to mid 20th century [42]. In particular Folch and colleagues, and Bligh and Dyer have respectively developed and optimized mixtures of chloroform-methanol for the rapid extraction of lipids from animal tissues [43–45]. Lin et al. tested a mixture of ethyl acetate and ethyl alcohol as an alternative to the Folch reagent for extracting animal lipids [46]. Numerous other solvents and mixtures of solvents for lipid extractions have also been reported [47,48]. The lipid removal potency of non-crosslinking chemical fixatives applied to tissue sections have also been reported [49]. When using ethanol, it was estimated that ~75% of the lipid content was extracted from rat liver and brain sections. Some organic solvents such as alcohols, chloroform or acetone or mixtures of these are also used to precipitate proteins from solutions [46,50]. These same solvents are also classically used in histotechnologies to dehydrate, fix and generally preserved tissues [51]. For example ethanol and xylene are routinely used to dehydrate sections during several staining procedures or prior to laser capture microdissection (LCM) of individual cells [52]. When thin tissue sections are subjected to these solvents prior to imunohistochemistry, LCM or IMS, it is presumed that most proteins will precipitate in situ and stay localized within the cellular substructures.

We have observed early on that submerging sections in ethanol-based saturated matrix solutions to generate small seeding crystals did not negatively affect the IMS signal quality [21]. Similarly, no significant signal quality deterioration was observed during the development of the MALDI compatible staining procedures [23]. We have recently established an ethanol-based tissue rinsing procedure prior to the IMS analyses of proteins which not only gives cleaner IMS data but also seems to better preserve the tissue sections over time [3]. Others have also studied several tissue rinsing procedures. Lemaire et al. studied the efficacy of chloroform, hexane, toluene, acetone and xylene to remove lipids from the sections [24]. In conjunction with lipid removal, they observed an increase in overall peptide and protein detection in the m/z range from 5,000 to 30,000 as well as increased signal reproducibility. Using a very systematic analytical approach including automated matrix dispensing technology in combination with automated MS data acquisition and processing, we examine the effect of numerous solvent systems on IMS analyses of fresh frozen mammalian tissue sections. Using sections from a single mouse liver tissue specimen, parameters such as the number of peaks and overall protein (m/z from 2,000 to 25,000) and lipid (m/z range from 500 to 2000) total ion currents (TIC) are compared across solvent systems. For some of the solvent systems investigated, a study on the stability and recoverability of rinsed sections as a function of time is also examined. Finally, using a human renal cell carcinoma biopsy containing different histologies, IMS measurements were performed to assess individual signal variability as well as eventual delocalization behaviors.

Materials and Methods

Ethanol, methanol, acetonitrile, acetone, acetic acid, isopropanol, and hexane were purchased from Fisher Scientific (Suwanee, GA). Trifluoroacetic acid and xylene were purchased from Acros (Morris Plains, NJ). Chloroform, tert-butyl methyl ester, sinapinic acid, and 2,5-dihydrodyacetophenone were purchased from Sigma-Aldrich (Milwaukee, WI). Toluene was purchased from JT Baker (Phillipsburg, NJ).

A control mouse liver was serially sectioned at 12 μm thickness on a Leica CM1900 cryostat (Leica Microsystems GmbH, Wetzlar, Germany) and thaw-mounted onto gold-coated MALDI targets. For most analyses one section was collected per plate. Plates were placed in a desiccator to dry for 20 min prior to washing. The sections were washed by submerging the plates in 50 mL of solvent placed in a glass Petri dish and gently swirling. Tissue washes were for 30 s per solution as follows:

  1. 2 × 100% acetone (acetone)

  2. 70% ethanol followed by 95% ethanol (ethanol)

  3. 70% isopropanol followed by 95% isopropanol (isopropanol)

  4. 70% methanol followed by 95% methanol (methanol)

  5. 70% ethanol followed by 90% ethanol/9% acetic acid (acetic acid)

  6. 2 × 100% chloroform (chloroform)

  7. 2 × 100% toluene (toluene)

  8. 2 × 100% hexane (hexane)

  9. 2 × 100% xylene (xylene)

  10. 2 × 100% water (water)

  11. 2 × 100% tert-butyl methyl ether (t-BME)

  12. No wash (control)

After washing, plates were tilted to maximize solvent removal and were returned to the desiccator to dry prior to matrix deposition.

Sinapinic acid (25 mg/mL in 50/0.1 (v/v) acetonitrile/trifluoroacetic acid) was used as the matrix for protein analysis, while 2,6-dihydroxyacetophenone (DHA, 10 mg/mL 60/40 (v/v) ethanol/water) was used as the matrix for phospholipid analysis. Discrete matrix spots printed in 5 × 5 arrays (25 spots per tissue section, ~250 μm diameter), were deposited using an acoustic robotic microspotter (Labcyte, Sunnyvale, CA) [26]. Care was taken to deposit matrix to approximately the same location on each section while avoiding obvious blood vessels.

A time-course experiment consisting of two parts was carried out on some of the washes from the initial experiment. In the first, five plates were washed the same day they were sectioned with ethanol, isopropanol, water, or acetic acid. Washed tissue was returned to the desiccator to dry and stored until the day of analysis. Tissue was spotted and analyzed on Days 1, 2, 4, 7, and 14 after sectioning. In the second part, the plates were not immediately washed, but left in the desiccator and “revived” by washing on Days 2, 4, 7, and 14 after sectioning. This was accomplished by subjecting the plates to the same 4 washes listed above. Two controls were included in the time-course study. In the first, tissue was not washed, left in the desiccator, and spotted with matrix on the day of analysis. For the other control, sections were not washed, but spotted with matrix the day they were sectioned and then left in the desiccator until mass spectral analysis was conducted on a later date.

All of the MALDI mass spectrometric measurements were performed on a Bruker Autoflex II (Billerica, MA) equipped with a SmartBeam laser (Nd:YAG, 355 nm) and an electrostatic reflector. Protein mass spectra were acquired from each spot in an automated fashion in linear positive ion mode. The nominal accelerating voltage was 20 kV with an extraction voltage of 18.65 kV, an Einzel lens voltage of 6.00 kV and a delay time of 350 ns. Delayed extraction parameters were optimized for optimal resolution at 12 kDa. Lipid spectra were acquired in both positive and negative ion reflector mode. Positive ion mode lipid spectra were acquired with an accelerating voltage of 19 kV, an extraction voltage of 17 kV, a delay time of 100 ns, a lens voltage of 8 kV, and a reflector voltage of 20 kV. Negative ion mode spectra were collected using an accelerating voltage of −19 kV, an extraction voltage of −17 kV, a delay time of 100 ns, a lens voltage of −8.2 kV, and a reflector voltage of −20 kV. For both positive and negative ion mode lipid spectra, delayed extraction parameters were optimized for optimal resolution at 800 Da. All spectra were the sum of 500 individual laser shots collected in 50 shot increments within each matrix spot. The laser position was moved within the spot between each incremental step.

Spectra were preprocessed using software from Biodesix, Inc. (Steamboat Springs, CO) which included baseline correction and alignment to common peaks and averages generated for each wash type. Evaluation of results included comparison of TIC, number of peaks, and intensity of background. Spectra were also visually examined for presence or absence of signals.

Mass spectral images were acquired from sections of human clear cell renal cell carcinoma samples that had been subjected to each of the washes used in the time course study. All sections were mounted and simultaneously imaged on a single MALDI target. Sinapinic acid was applied in a 55 × 12 grid across each section with a lateral resolution of 300 μm using a Portrait 630 Multireagent Spotter (Labcyte, Sunnyvale, CA). Discrete matrix spots were deposited in fly-by mode with 20 passes of 3 drops each. Protein spectra were acquired on a Bruker Autoflex II in linear mode geometry as described above using the same instrumental parameters. Each pixel was a composite of 200 laser shots acquired in 50 shot increments from 4 positions within a single matrix spot. Images were visualized using BioMap software (Novartis, Basel, Switzerland).

Photomicrographs of the tissue sections after matrix deposition were obtained at 4× and 10× magnification using an Olympus BX50 microscope (Olympus America Inc., Center Valley, PA) equipped with a Micropublisher 3.3 RTV digital camera (Q Imaging, Surrey, BC, Canada). Hematoxylin and eosin (H&E) section staining was performed according to traditional protocols. Magnified photomicrographs of the H&E stained tissue sections were obtained after scanning the slides using a Mirax Desk microscope slide scanner (Zeiss, Thornwood, NY) at a pixel resolution of 0.23 μm. Digital photomicrographs were exported using the Mirax Viewer software.

Results and Discussion

Sample preparation is of the utmost importance when carrying out direct tissue analysis by IMS. Care must be taken to ensure that the sample is handled properly from the time it is removed from the organism up through mass spectral analysis. Tissues should be snap frozen immediately after resection to prevent proteolytic degradation and kept frozen at −80°C. Sections should be thoroughly thawed when mounted onto a MALDI target as residual moisture can lead to freezer burn. Fixation or washing can greatly improve protein signal quality and prolong the viable life of the tissue. Matrix deposition is important in determining the subset of proteins that are extracted from the tissue and detected [3]. Carefully selection of acquisition parameters on the mass spectrometer is necessary to optimize for the mass range in which the signals of interest fall.

The solvents used in this study were selected based on their known efficacy for lipid solubilization, tissue preservation or have been previously reported for washing tissue sections prior to analyses by mass spectrometry. A systematic study was undertaken in which 11 different tissue washing approaches were examined for their influence on protein and lipid signal. From the results of this, 4 of these washes were used to assess their ability to preserve the tissue for protein analysis on a later date and for the possibility that tissue could be “revived” by washing several days after sectioning. Finally, protein images were carried out on serial sections of a human clear cell renal cell carcinoma to visually assess the degree of protein loss induced by the washing techniques and the extent of protein delocalization.

The first part of the study involved exploring 11 different tissue washing approaches for their effect on protein and lipid signals. To minimize variability coming from the sample, all of the experiments were performed on serial sections originating from a single mouse liver lobe. Liver was used because of its relative homogeneity over a large volume. Sections of control mouse liver were thaw mounted onto gold-coated MALDI targets, subjected to the 11 wash methods listed above and spotted with sinapinic acid for protein analysis. To monitor intra-sample variability, each tissue section was spotted with 5 × 5 matrix spot array and analyzed by mass spectrometry. Subsequent to data collection, the 25 resulting spectra were baseline corrected, aligned, and peaks selected. The average traces from the processed spectra from each wash as well as the non-washed control are displayed in Figure 1A. While a fair majority of the signals observed are very similar for all of the washes (e.g. Figure 1B), there are some notable differences. For example, m/z 9912 which was the base peak observed after most of the washes is completely devoid from the water average spectrum (Figure 1D). Additionally, the signal at m/z 13777 is only readily apparent in the tissue sections washed with water, chloroform, and acetic acid (Figure 1E). Figure 1C displays a pair of signals that show an interesting trend; 2 signals are observed at m/z 7928 and 7936 which appear to be specific to the type of wash used. The signal at m/z 7928 shows a strong preference for chloroform and acetic acid whereas m/z 7936 shows a strong preference for methanol and isopropanol. The other washes tend to show one or the other of the signals with hexane showing equal distribution of both. The protein spectra were also evaluated for TIC, background, noise, and number of observed peaks. Plots showing TIC and number of peaks as a function of wash type are shown in Figure 2A and B, respectively. The different washes were ordered with increasing protein TIC. The TIC was found lowest for acetone and highest for isopropanol. A simple linear fit was drawn to guide the eye. Interestingly, when comparing all washes the increase in the number of peaks observed was found proportional to the increase in TIC. This behavior was also found roughly true for background and noise (not shown). The total number of peaks tracked ranged from 85 for the acetone washed tissues to 129 for the methanol washed tissue.

Figure 1.

Figure 1

Open in a new tab

Average MALDI-TOF MS protein profiles directly acquired from serial mouse liver tissue sections not washed or washed with different solvent systems (see text for details).

Figure 2.

Figure 2

Open in a new tab

A) TIC variations recorded from the MALDI-TOF MS protein profiles acquired from serial mouse liver tissue sections not washed or washed with different solvent systems in the m/z range from 500 to 1100 (lipid component) and m/z range from 2000 to 25000 (protein component). B) Number of peak variations as a function of the same washes for the protein component.

The washed and control mouse liver sections were also spotted with 5 × 5 arrays of 2,6-dihydroxyacetophenone and analyzed for lipid content in both positive and negative ion reflector mode. The resulting average traces for the positive mode analysis are shown in Supplemental Figure 1. All of the alcohol-based washes showed spectra practically devoid of discernable peaks and had the lowest calculated TIC. A similar trend was observed for the negative mode lipid analysis (not shown). Lipid TIC values (integrated between m/z 500 and 1100) were plotted on the same graph and in the order as previously obtained for protein TIC in Figure 2A. Again a simple linear fit was drawn to guide the eye. Interestingly, the lipid TIC trend is roughly opposite that observed from the protein analysis, indicating that the removal of lipids from the tissue enhances the observation of protein signals. This can be tentatively explained by 1) the disruption of the cell membranes and removal of the lipids allows better access to intracellular proteins [24], and 2) since both lipids and proteins compete for charge, if one species is removed the measured abundance of the other should increase. Taken together, these results lead to the conclusion that the alcohols as well as acetic acid and water washes were the most beneficial to protein mass spectral analysis from liver tissue sections and warranted further investigation.

Four washes (ethanol, isopropanol, acetic acid and water) were further evaluated for their ability to extend the viable life of tissue sections for direct tissue analysis by MALDI MS. Serial sections cut from the same mouse liver were washed the day they were sectioned and placed into the desiccator until analysis. As a control, a second set of tissue sections was left unwashed. Tissue sections from each wash were again spotted with 25 matrix spots in a 5 × 5 array and mass spectra acquired on each of Day 1 (the day the tissue was sectioned), Day 2, Day 4, Day 7, and Day 14 post sectioning. Spectra were baseline corrected, aligned, and averaged per day and wash type. Again, the TIC was evaluated and plotted as a function of time for each of the washes as seen in Supplemental Figure 2. To emphasize variations with time, relative TIC yields were plotted with respect to Day 1. Only the unwashed tissue showed a drop off in signal intensity of ~65% on Day 4 and later in contrast to all of the washed tissues which showed no signs of TIC decrease with respect to Day 1 over the time course of the study. Number of peaks was also considered in the analysis, but proved not to be an accurate representation of the process because tissue degradation with time resulted in an increase in the number of peaks observed in the low mass region (m/z < 4000) which bias these results (see below). Washing with isopropanol appeared to give the best tissue preservation with the Day 7 spectrum practically indistinguishable from the Day 1 spectrum as is shown in Figure 3A. The Day 14 isopropanol spectrum showed signs of decreased signal resolution across the studied mass range. When looking in detail at the protein signal from the unwashed tissue, these rapidly deteriorated over time (Supplemental Figure 2 and Figure 3B). Most of the signals showed a decrease in intensity as a function of time while some completely disappeared. For example, the protein signal detected at m/z 12131 (most likely cytochome c) is only clearly observed up to Day 4. Another example is shown in the Insets beside Figure 3A and B showing the same set of signals centered around m/z 6600 (arrows in Figure 3A and B), which remained fairly constant in the isopropanol washed tissue but decreased dramatically in the unwashed tissue. The unwashed tissue also showed an increase in the signals observed in the low mass region of the spectrum (m/z < 4,000) likely as the result of protein degradation. It is concluded that the lifetime of the sections can be significantly extended, up to 14 days, after washing with the 4 chosen solvent systems.

Figure 3.

Figure 3

Open in a new tab

Average MALDI-TOF MS protein profiles acquired from serial mouse liver tissue sections, A) washed with isopropanol or B) not washed on Day 1 and spotted with matrix and analyzed on subsequent days. The insets present detailed peak behaviors in the m/z range from 6450 to 6750 for each set of sections.

These same 4 washes were assessed for their ability to “revive” tissue that had been allowed to age in a desiccator after sectioning. In this case, the tissue was sectioned on Day 1 and was not washed until the day it was spotted and analyzed (Day 1, 2, 4, 7, or 14). Example spectra are shown in Supplemental Figure 3 for Days 2 and 7. While the Day 2 spectra are fairly comparable to the unwashed control, the Day 7 revived showed considerable degradation especially resulting in an increase in the detection of low mass signals. Figure 4 shows a selected set of peaks from the isopropanol washed and the water washed tissue over the full time course of the experiment. A clear loss of both signal intensity and resolution started to be observed after Day 2. From a practical standpoint, it is concluded that tissue sections can be confidently revived by washing one day after sectioning (Day 2) and successfully analyzed without any major noticeable sample alterations.

Figure 4.

Figure 4

Open in a new tab

Evolution of average MALDI-TOF MS protein profiles for a selected mass range as a function of time (days) for liver tissue sections cut on Day 1 but washed with isopropanol or water and analyzed on subsequent days.

Additionally, a third experiment was undertaken in which the tissue was not washed, but was spotted the day it was sectioned and then placed in a desiccator for analysis on a later day. The theory behind this is that the tissue will be locally washed by the matrix solution and proteins extracted from the tissue and embedded into the matrix crystals will remain stable. However, as is seen in Supplemental Figure 4, the effects of aging cannot be overcome. A decrease in overall intensity is observed over time (inset) and some signals disappear completely as the matrix spots aged. It is therefore concluded that for maximum ion yield recovery and optimal spectrum quality, MALDI MS analyses of unwashed tissues should be performed as soon as possible after matrix deposition.

For each wash, in addition to the intensity and number of signals observed, it is important to consider matrix deposition behavior and histological integrity of the tissue for sections may be destined to be stained post MS analyses. This was monitored by examining the regularity of matrix spot arrays, the crystal formation within a spot, as well as histology on a stained tissue section. Crystal architecture was explored by examining the matrix spots deposited on tissue. Figure 5A shows photomicrographs of 5 × 5 grids of printed matrix spots on tissues previously subjected to each of the 4 selected washing procedures and the control unwashed tissue. Unwashed, isopropanol, and acetic acid showed very regular arrays with no spots merging. Ethanol showed a somewhat irregular array where several spots have merged. For water, the array observed was clearly misaligned with multiple spots merging due to delocalization and larger spot size. When examining the crystals from single spots (Figure 5B), unwashed, ethanol, and isopropanol looked very similar with long spindly crystals. Crystal coverage on the acetic acid washed tissue was sparser and the water washed showed smaller, more round crystals.

Figure 5.

Figure 5

Open in a new tab

Magnified photomicrographs of liver tissue sections non-washed and washed with isopropanol, ethanol, acetic acid or water after matrix spot array automated deposition (A, B), and H&E staining (C, D). See text for details.

To examine the histology, serial liver sections were collected on glass microscope slides and subjected to the 4 selected tissue washes prior to standard hematoxylin and eosin (H&E) staining. As a control, a fifth section was stained without prior washing. The stained sections were then evaluated for the effects of the washes. Representative pictures demonstrating the histology are shown at an equivalent 4× (Figure 5C) and 40× (Figure 5D) magnification. Ethanol and isopropanol were found to be nearly indistinguishable from the unwashed control. Cell membranes, nuclei, as well as nucleoli could clearly be distinguished within the sections. The acetic acid wash had some disruptive effect on the tissue. While membranes and nuclei could still be observed, the cells appeared blurred and the nuclei did not stain as intensely as those in the other alcohol based washes. Histological analysis on this tissue section would be compromised as compared to control. The water wash completely destroyed the histology of the tissue section. Holes were observed within the section and cells could no longer be distinguished, nor could any other features present within the tissue such as the duct visible in the upper left D panel of the other washes. It is concluded that the alcohols (ethanol and isopropanol) are the most conducive to histological staining of fresh frozen tissue sections.

Finally, an imaging experiment was carried out to visualize the effects of washing on the tissue with regard to signals observed as well as delocalization of proteins (Figure 6). For this experiment, a human clear cell renal cell carcinoma section was used as it showed different histological regions as opposed to liver tissue which is very homogeneous. The collected sections contained areas of tumor, stroma, normal kidney cortex, and a highly vascularized region within the cortex (Figure 6A). Five serial sections of the tissue block were collected on one MALDI target plate with one section at each of the edges and one in the center (Figure 6B). By submerging only part of the plate in the wash solutions, each of the sections at the edge of the plate was subjected to one of each of the ethanol, isopropanol, acetic acid, and water washes. The tissue section at the center of the plate served as the unwashed control. Each of the sections was spotted approximately over the same area (Figure 6A) with a 55 × 12 spot array with a lateral spacing of 300 μm that transversed areas of tumor, stroma, cortex, and vascularization. After IMS data acquisition, as was the case in the profiling experiments, a number of differences in the proteins observed were seen in the resulting images. Figure 6C–H summarizes the IMS results from this experiment. Of note, the protein signals detected at m/z 2758, 5075, and 13804 were not observed in the unwashed tissue; these only appear after washing. For all of the monitored signals, the alcohol based washes (ethanol and isopropanol) showed very similar protein distribution and intensities. Several ion density images were clearly affected by some of the washes. Washing with acetic acid resulted in the decrease or loss of some signals such as the ones detected at m/z 2758 and 3336. As observed in the profiling studies, the water wash had the largest impact on the proteins observed and resulted in the complete loss of some signals such as the one detected at m/z 4965 (likely to correspond to Thymosin β4), one of the most intense signals observed in the sections subjected to the other washes. While one of the aims with tissue washing is the removal of hemoglobin from residual blood in the sample, the images obtained for the alpha chain (m/z 15127) indicate that this is not the case. In fact, all of the washes showed enhanced hemoglobin signal as compared to the unwashed tissue. Based on the sum of results from the IMS experiment, protein delocalization across the sections did not appear to be a problem with any of the studied washes. It is believed that considering the volume of solvent used and the fact that the sections were fully submerged, proteins soluble in the wash solutions were completely removed from the surface of the tissue and did not redeposit at another location on the section or the target plate in any significant amounts.

Figure 6.

Figure 6

Open in a new tab

IMS analyses of serial sections cut from a clear cell renal cell carcinoma resection not washed or after washing with isopropanol, ethanol, acetic acid or water. A) Photomicrograph of a serial H&E stained section detailing the histology of the biopsy. Areas of normal kidney cortex, stroma and tumor are visible. The area boxed in red was analyzed. B) Photomicrograph of the sections used for IMS after washing and automated matrix deposition. The sections on the edge of the target were washed whereas the center section was left unwashed. C–H) Ion density maps for selected protein signals from the non-washed and washed sections. HOAc: acetic acid; EtOH: ethanol; IPA: isopropanol.

Conclusions

The results presented here represent an important first step in the optimization of tissue washing protocols prior to direct MALDI MS analysis of proteins. In this study, we found that alcohol based washes of sections, in particular isopropanol, to be most effective for protein analysis when considering MS signal quality, matrix deposition regularity, and preservation and histological integrity of the tissue. These solvents were also found to effectively remove lipids from the tissue, the molecules most likely to significantly compete with proteins for ionization. A strong correlation was found between the effectiveness of the wash in removing lipids from the section and the detection of optimal protein signals. The fact that alcohol-based washes allow preservation of the sections for protein analysis for several days is very important if it becomes necessary to ship samples to another institution. This event may happen when working with a pathologist offsite for histology-directed analysis where it may take several days to annotate sections for areas to analyze. However, there are still many avenues that need to be explored in the realm of tissue washing. A detailed analysis was carried out here for liver tissue. Other than for the alcohol and acid-based washes which contained some percentage of water, other solvent combinations were not explored. Although alcohol-based washing procedures tested for liver were also found appropriate for the study of a clear cell renal cell carcinoma resection, it is possible that other solvents or mixtures of solvents may be found optimal for other organs. Optimization of tissue preparation is an ongoing process with the ultimate goal being high quality, reproducible mass spectral data.

Supplementary Material

01. Supplemental Figures.

Figure S1, Average MALDI-TOF MS lipid profiles directly acquired from serial mouse liver tissue sections not washed or washed with different solvent systems (see text for details).

Figure S2, Relative variations of the TIC as a function of time (days) recorded from the MALDI-TOF MS protein profiles acquired from serial mouse liver tissue sections washed with ethanol, isopropanol, acetic acid, water or not washed. The sections were cut and washed on Day 1 but spotted with matrix and analyzed on subsequent days.

Figure S3, Average MALDI-TOF MS protein profiles acquired from serial mouse liver tissue sections cut on Day 1 but washed with isopropanol, ethanol, acetic acid or water and spotted with matrix and analyzed on Day 2 and 7.

Figure S4, Protein profiles from unwashed sections spotted on Day 1 and analyzed on subsequent days. The inset presents the relative evolution of the TIC as a function of time (days).

Acknowledgments

The authors would like to acknowledge Kristina Schwamborn, M.D. (Universitätsklinikum der RWTH Aachen, Germany) for her help with the histological evaluation of the H&E stained sections. The authors also acknowledge funding from the National Cancer Institute (grant # 5R33 CA116123-3) and the National Institutes of Health (grant # 5R01 GM058008-09).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Stoeckli M, Chaurand P, Hallahan DE, Caprioli RM. Imaging mass spectrometry: A new technology for the analysis of protein expression in mammalian tissues. Nat Med. 2001;7:493–496. doi: 10.1038/86573. [DOI] [PubMed] [Google Scholar]
  • 2.Chaurand P, Schwartz SA, Caprioli RM. Profiling and Imaging Proteins in Tissue Sections by Mass Spectrometry. Anal Chem. 2004;76:86A–93A. [PubMed] [Google Scholar]
  • 3.Chaurand P, Norris JL, Cornett DS, Mobley JA, Caprioli RM. New developments in profiling and imaging of proteins from tissue sections by MALDI mass spectrometry. J Proteome Res. 2006;5:2889–2900. doi: 10.1021/pr060346u. [DOI] [PubMed] [Google Scholar]
  • 4.Cornett DS, Mobley JA, Dias EC, Andersson M, Arteaga CL, Sanders ME, Caprioli RM. A novel histology-directed strategy for MALDI-MS tissue profiling that improves throughput and cellular specificity in human breast cancer. Mol Cell Proteomics. 2006;5:1975–1983. doi: 10.1074/mcp.M600119-MCP200. [DOI] [PubMed] [Google Scholar]
  • 5.Jackson SN, Wang HYJ, Woods AS. Direct profiling of lipid distribution in brain tissue using MALDI-TOFMS. Anal Chem. 2005;77:4523–4527. doi: 10.1021/ac050276v. [DOI] [PubMed] [Google Scholar]
  • 6.Garrett TJ, Prieto-Conaway MC, Kovtoun V, Bui H, Izgarian N, Stafford G, Yost RA. Imaging of small molecules in tissue sections with a new intermediate-pressure MALDI linear ion trap mass spectrometer. Int J Mass Spectrom. 2007;260:166–176. [Google Scholar]
  • 7.Verhaert P, Uttenweiler-Joseph S, de Vries M, Loboda A, Ens W, Standing KG. Matrix-assisted laser desorption/ionization quadrupole Time-of- Flight Mass Spectrometry: An elegant tool for peptidomics. Proteomics. 2001;1:118–131. doi: 10.1002/1615-9861(200101)1:1<118::AID-PROT118>3.0.CO;2-1. [DOI] [PubMed] [Google Scholar]
  • 8.Fournier I, Day R, Salzet M. Direct analysis of neuropeptides by in situ MALDI-TOF mass spectrometry in the rat brain. Neuroendocrinol Lett. 2003;24:9–14. [PubMed] [Google Scholar]
  • 9.Kruse R, Sweedler JV. Spatial profiling invertebrate ganglia using MALDI MS. J Am Soc Mass Spectrom. 2003;14:752–759. doi: 10.1016/S1044-0305(03)00288-5. [DOI] [PubMed] [Google Scholar]
  • 10.DeKeyser SS, Kutz-Naber KK, Schmidt JJ, Barrett-Wilt GA, Li LJ. Imaging mass spectrometry of neuropeptides in decapod crustacean neuronal tissues. J Proteome Res. 2007;6:1782–1791. doi: 10.1021/pr060603v. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Chaurand P, Schwartz SA, Caprioli RM. Imaging mass spectrometry: a new tool to investigate the spatial organization of peptides and proteins in mammalian tissue sections. Curr Opin Chem Biol. 2002;6:676–681. doi: 10.1016/s1367-5931(02)00370-8. [DOI] [PubMed] [Google Scholar]
  • 12.Reyzer ML, Hsieh Y, Ng K, Korfmacher WA, Caprioli RM. Direct analysis of drug candidates in tissue by matrix-assisted laser desorption/ionization mass spectrometry. J Mass Spectrom. 2003;38:1081–1092. doi: 10.1002/jms.525. [DOI] [PubMed] [Google Scholar]
  • 13.Reyzer ML, Caldwell RL, Dugger TC, Forbes JT, Ritter CA, Guix M, Arteaga CL, Caprioli RM. Early Changes in Protein Expression Detected by Mass Spectrometry Predict Tumor Response to Molecular Therapeutics. Cancer Res. 2004:64. doi: 10.1158/0008-5472.CAN-04-2231. [DOI] [PubMed] [Google Scholar]
  • 14.Khatib-Shahidi S, Andersson M, Herman JL, Gillespie TA, Caprioli RM. Direct molecular analysis of whole-body animal tissue sections by imaging MALDI mass spectrometry. Anal Chem. 2006;78:6448–6456. doi: 10.1021/ac060788p. [DOI] [PubMed] [Google Scholar]
  • 15.Stoeckli M, Staab D, Schweitzer A. Compound and metabolite distribution measured by MALDI mass spectrometric imaging in whole-body tissue sections. Int J Mass Spectrom. 2007;260:195–202. [Google Scholar]
  • 16.Chaurand P, Caprioli RM. Direct profiling and imaging of peptides and proteins from mammalian cells and tissue sections by mass spectrometry. Electrophoresis. 2002;23:3125–3135. doi: 10.1002/1522-2683(200209)23:18<3125::AID-ELPS3125>3.0.CO;2-#. [DOI] [PubMed] [Google Scholar]
  • 17.Cornett DS, Reyzer ML, Chaurand P, Caprioli RM. MALDI imaging mass spectrometry: molecular snapshots of biochemical systems. Nat Meth. 2007;4:828–833. doi: 10.1038/nmeth1094. [DOI] [PubMed] [Google Scholar]
  • 18.Chaurand P, Sanders ME, Jensen RA, Caprioli RM. Proteomics in Diagnostic Pathology: Profiling and Imaging Proteins Directly in Tissue Sections. Am J Pathol. 2004;165:1057–1068. doi: 10.1016/S0002-9440(10)63367-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Chaurand P, Schwartz SA, Caprioli RM. Assessing protein patterns in disease using imaging mass spectrometry. J Proteome Res. 2004;3:245–252. doi: 10.1021/pr0341282. [DOI] [PubMed] [Google Scholar]
  • 20.Chaurand P, Schwartz SA, Reyzer ML, Caprioli RA. Imaging mass spectrometry: Principles and potentials. Toxicologic Pathology. 2005;33:92–101. doi: 10.1080/01926230590881862. [DOI] [PubMed] [Google Scholar]
  • 21.Schwartz SA, Reyzer ML, Caprioli RM. Direct tissue analysis using matrix-assisted laser desorption/ionization mass spectrometry: practical aspects of sample preparation. J Mass Spectrom. 2003;38:699–708. doi: 10.1002/jms.505. [DOI] [PubMed] [Google Scholar]
  • 22.Sugiura Y, Shimma S, Setou M. Thin Sectioning Improves the Peak Intensity and Signal-to-Noise Ratio in Direct Tissue Mass Spectrometry. J Mass Spectrom Soc Jpn. 2006;54:45–48. [Google Scholar]
  • 23.Chaurand P, Schwartz SA, Billheimer D, Xu BJ, Crecelius A, Caprioli RM. Integrating histology and imaging mass spectrometry. Anal Chem. 2004;76:1145–1155. doi: 10.1021/ac0351264. [DOI] [PubMed] [Google Scholar]
  • 24.Lemaire R, Wisztorski M, Desmons A, Tabet JC, Day R, Salzet M, Fournier I. MALDI-MS direct tissue analysis of proteins: improving signal sensitivity using organic treatments. Anal Chem. 2006;78:7145–7153. doi: 10.1021/ac060565z. [DOI] [PubMed] [Google Scholar]
  • 25.Stoeckli M, Staab D, Staufenbiel M, Wiederhold KH, Signor L. Molecular imaging of amyloid beta peptides in mouse brain sections using mass spectrometry. Anal Biochem. 2002;311:33–39. doi: 10.1016/s0003-2697(02)00386-x. [DOI] [PubMed] [Google Scholar]
  • 26.Aerni HR, Cornett DS, Caprioli RM. Automated acoustic matrix deposition for MALDI sample preparation. Anal Chem. 2006;78:827–834. doi: 10.1021/ac051534r. [DOI] [PubMed] [Google Scholar]
  • 27.Sugiura Y, Shimma S, Setou M. Two-step matrix application technique to improve ionization efficiency for matrix-assisted laser desorption/ionization in imaging mass spectrometry. Anal Chem. 2006;78:8227–8235. doi: 10.1021/ac060974v. [DOI] [PubMed] [Google Scholar]
  • 28.Groseclose MR, Andersson M, Hardesty WM, Caprioli RM. Identification of proteins directly from tissue: in situ tryptic digestions coupled with imaging mass spectrometry. J Mass Spectrom. 2007;42:254–262. doi: 10.1002/jms.1177. [DOI] [PubMed] [Google Scholar]
  • 29.Shimma S, Furuta M, Ichimura K, Yoshida Y, Setou M. Direct MS/MS analysis in mammalian tissue sections using MALDI-QIT-TOFMS and chemical inkjet technology. Surface and Interface Analysis. 2006;38:1712–1714. [Google Scholar]
  • 30.Lemaire R, Desmons A, Tabet JC, Day R, Salzet M, Fournier I. Direct analysis and MALDI imaging of formalin-fixed, paraffin-embedded tissue sections. J Proteome Res. 2007;6:1295–1305. doi: 10.1021/pr060549i. [DOI] [PubMed] [Google Scholar]
  • 31.Spengler B, Hubert M. Scanning microprobe matrix-assisted laser desorption ionization (SMALDI) mass spectrometry: Instrumentation for sub-micrometer resolved LDI and MALDI surface analysis. J Am Soc Mass Spectrom. 2002;13:735–748. doi: 10.1016/S1044-0305(02)00376-8. [DOI] [PubMed] [Google Scholar]
  • 32.Chaurand P, Schriver KE, Caprioli RM. Instrument design and characterization for high resolution MALDI-MS imaging of tissue sections. J Mass Spectrom. 2007;42:476–489. doi: 10.1002/jms.1180. [DOI] [PubMed] [Google Scholar]
  • 33.Luxembourg SL, Mize TH, McDonnell LA, Heeren RMA. High-spatial resolution mass spectrometric imaging of peptide and protein distributions on a surface. Anal Chem. 2004;76:5339–5344. doi: 10.1021/ac049692q. [DOI] [PubMed] [Google Scholar]
  • 34.Holle A, Haase A, Kayser M, Höhndorf J. Optimizing UV laser focus profiles for improved MALDI performance. J Mass Spectrom. 2006;41:705–716. doi: 10.1002/jms.1041. [DOI] [PubMed] [Google Scholar]
  • 35.Verhaert PD, Conaway MCP, Pekar TM, Miller K. Neuropeptide imaging on an LTQ with vMALDI source: The complete ‘all-in-one’ peptidome analysis. Int J Mass Spectrom. 2007;260:177–184. [Google Scholar]
  • 36.Taban IM, Altelaar AFM, Van der Burgt YEM, McDonnell LA, Heeren RMA, Fuchser J, Baykut G. Imaging of peptides in the rat brain using MALDI-FTICR mass spectrometry. J Am Soc Mass Spectrom. 2007;18:145–151. doi: 10.1016/j.jasms.2006.09.017. [DOI] [PubMed] [Google Scholar]
  • 37.Stoeckli M, Farmer TB, Caprioli RM. Automated mass spectrometry imaging with a matrix-assisted laser desorption ionization time-of-flight instrument. J Am Soc Mass Spectrom. 1999;10:67–71. doi: 10.1016/S1044-0305(98)00126-3. [DOI] [PubMed] [Google Scholar]
  • 38.McCombie G, Staab D, Stoeckli M, Knochenmuss R. Spatial and spectral correlations in MALDI mass spectrometry images by clustering and multivariate analysis. Anal Chem. 2005;77:6118–6124. doi: 10.1021/ac051081q. [DOI] [PubMed] [Google Scholar]
  • 39.Norris JL, Cornett DS, Mobley JA, Andersson M, Seeley EH, Chaurand P, Caprioli RM. Processing MALDI mass spectra to improve mass spectral direct tissue analysis. Int J Mass Spectrom. 2007;260:212–221. doi: 10.1016/j.ijms.2006.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Klerk LA, Broersen A, Fletcher IW, van Liere R, Heeren RMA. Extended data analysis strategies for high resolution imaging MS: New methods to deal with extremely large image hyperspectral datasets. Int J Mass Spectrom. 2007;260:222–236. [Google Scholar]
  • 41.Klinkert I, McDonnell LA, Luxembourg SL, Altelaar AFM, Amstalden ER, Piersma SR, Heeren RMA. Tools and strategies for visualization of large image data sets in high-resolution imaging mass spectrometry. Rev Sci Instrum. 2007:78. doi: 10.1063/1.2737770. [DOI] [PubMed] [Google Scholar]
  • 42.Bloor WR. The determination of small amounts of lipid in bllod plasma. J Biol Chem. 1928;77:53–73. [Google Scholar]
  • 43.Folch J, Lees M, Stanley GHS. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem. 1957;226:497–509. [PubMed] [Google Scholar]
  • 44.Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol. 1959;37:911–917. doi: 10.1139/o59-099. [DOI] [PubMed] [Google Scholar]
  • 45.Iverson SJ, Lang SLC, Cooper MH. Comparison of the bligh and dyer and folch methods for total lipid determination in a broad range of marine tissue. Lipids. 2001;36:1283–1287. doi: 10.1007/s11745-001-0843-0. [DOI] [PubMed] [Google Scholar]
  • 46.Jiang L, He L, Fountoulakis M. Comparison of protein precipitation methods for sample preparation prior to proteomic analysis. J Chromato A. 2004;1023:317–320. doi: 10.1016/j.chroma.2003.10.029. [DOI] [PubMed] [Google Scholar]
  • 47.Christie WW. Preparation of lipid extracts from tissues. In: WCW, editor. Advances in lipid methodology - Two. Oily Press; Dundee: 1993. pp. 195–213. [Google Scholar]
  • 48.Hanahan DJ. A guide to phospholipid chemistry. Oxford University Press; 1997. [Google Scholar]
  • 49.Maneta-Peyret L, Compere P, Moreau P, Goffinet G, Cassagne C. Immunocytochemistry of lipids: Chemical fixatives have dramatic effects on the preservation of tissue lipids. Histochem J. 1999;31:541– 547. doi: 10.1023/a:1003844107756. [DOI] [PubMed] [Google Scholar]
  • 50.Wessel D, Flugge UI. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal Biochem. 1984;138:141–143. doi: 10.1016/0003-2697(84)90782-6. [DOI] [PubMed] [Google Scholar]
  • 51.Prophet EB, Mills B, Arrington JB, Sobin LH, editors. Laboratory Methods in Histotechnology. American Registry of Pathology; Washington, D.C.: 1994. [Google Scholar]
  • 52.EmmertBuck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang ZP, Goldstein SR, Weiss RA, Liotta LA. Laser capture microdissection. Science. 1996;274:998–1001. doi: 10.1126/science.274.5289.998. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01. Supplemental Figures.

Figure S1, Average MALDI-TOF MS lipid profiles directly acquired from serial mouse liver tissue sections not washed or washed with different solvent systems (see text for details).

Figure S2, Relative variations of the TIC as a function of time (days) recorded from the MALDI-TOF MS protein profiles acquired from serial mouse liver tissue sections washed with ethanol, isopropanol, acetic acid, water or not washed. The sections were cut and washed on Day 1 but spotted with matrix and analyzed on subsequent days.

Figure S3, Average MALDI-TOF MS protein profiles acquired from serial mouse liver tissue sections cut on Day 1 but washed with isopropanol, ethanol, acetic acid or water and spotted with matrix and analyzed on Day 2 and 7.

Figure S4, Protein profiles from unwashed sections spotted on Day 1 and analyzed on subsequent days. The inset presents the relative evolution of the TIC as a function of time (days).

NIHMS65679-supplement-01.ppt (2.9MB, ppt)

RESOURCES