Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2025 Aug 26;97(35):19001–19008. doi: 10.1021/acs.analchem.5c01767

To Spray or To Sublimate: Considerations for the Matrix Application of Three Common Positive Ion Mode Matrices

Ian Spears †, Gabriella Lagdameo †, Alyson Black ‡, Dalton R Brown †, Cole C Johnson †, Tae-Hun Hahm †, Wanyue Wang †, Alain Creissen ‡, Kristine Glunde †,§,∥, Caitlin M Tressler †,§,*
PMCID: PMC12424017  PMID: 40855949

Abstract

Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) is a powerful tool for imaging lipids, proteins, N-glycans, metabolites, and drugs in situ. A key step in MALDI MSI is matrix deposition, which must balance the extraction of molecules from the surface of the tissue without inducing excessive analyte delocalization. Further, MALDI MSI spatial resolution is limited by the size of the matrix crystals formed during matrix deposition. Minimizing crystal size is paramount to obtaining detailed images and refined data. Sublimation has been shown to be a reliable method of depositing matrix while minimizing crystal size. Since sublimation does not use any solvents to deposit the matrix, it is also theorized to minimize analyte delocalization. However, these same solvents are thought to aid in analyte extraction, which may limit detection by sublimation. In this study, we compared crystal size, analyte delocalization, and signal intensity differences when using a commercial automatic pneumatic sprayer and a sublimator for lipid imaging analyses. We tested three matrices: α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxyacetophenone (DHAP), and 2,5-dihydroxybenzoic acid (DHB) on mouse brain, heart, and kidney. Our results revealed that crystal size depended upon the matrix and whether the crystals formed on or off tissue. Our data also showed that only a subset of the lipid spectrum was susceptible to increased delocalization from spraying and decreased signal intensity from sublimation.

graphic file with name ac5c01767_0009.jpg

graphic file with name ac5c01767_0007.jpg

Introduction

Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) is a powerful tool for visualizing in situ the spatial distribution of lipids, − proteins, − N-glycans, − metabolites, ,− neurotransmitters, and drugs. ,− A key step in MALDI MSI is matrix deposition. The matrix acts as a proton donor or acceptor to aid in the desorption and ionization of analytes. In this step, researchers must balance maximizing analyte extraction from tissues and minimizing analyte delocalization. While multiple methods of matrix deposition have been put forward, automatic pneumatic spraying , and sublimation − are the two most popular methods due to their high reproducibility. Improvements in MALDI MSI technology have allowed for smaller pixel size and better spatial resolution capability from the mass spectrometers. A next-level advancement for MALDI imaging is to achieve single cell and subcellular resolution. However, there are still challenges that must be overcome before this is a widespread application, especially in matrix deposition.

One of the primary limiting factors for the spatial resolution that can be achieved with MALDI MSI is the size of the matrix crystals. The size of a matrix crystal must be smaller than the size of the pixel to not compromise the spatial resolution. Therefore, a matrix deposition method that creates smaller crystals is desirable because it would allow for the generation of more detailed images. Current commercial sublimators have been shown to produce crystal sizes smaller than those of commercial sprayers. In previous studies using sprayers with strategic optimization of parameters, researchers were able to obtain 2,5-dihydroxybenzoic acid (DHB) crystals in the range of 5–25 μm. Researchers using sublimators have been able to obtain DHB crystals less than 1 μm in width.

In addition to producing smaller crystal sizes, sublimation is believed to minimize delocalization because it does not require solvent to apply matrix to the sample. Lipids are particularly susceptible to delocalization due to their relatively low molecular weight and solubility in organic solvents. However, it is believed that solvents used in spraying help extract analytes from the tissue’s surface, resulting in greater signal intensity compared to sublimation. This balance between delocalization and signal intensity presents a decision point in the sample preparation technique for most MALDI MSI experiments.

In this study, we compared matrix crystal size, analyte delocalization, and lipid signal intensity differences when using a commercial automatic sprayer and a commercial sublimator. We tested three commonly used MALDI matrices: α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxyacetophenone (DHAP), and 2,5-dihydroxybenzoic acid (DHB) on mouse brain, heart, and kidney. Additionally, we directly compared the spatial resolution capabilities of the two matrix deposition methods when MALDI MSI experiments were run with small laser and raster sizes.

Methods

Materials

Trifluoracetic acid, CHCA, DHAP, DHB, acetone, acetonitrile, ethanol, and methanol were purchased from Sigma-Aldrich (St. Louis, MO). All solvents used were HPLC grade or higher.

Sample Preparation

Indium tin oxide (ITO) coated slides (Delta Technologies, Loveland, CO) were checked for conductivity to confirm their orientation. Slides were then washed for 10 min in hexanes and 10 additional minutes in ethanol, both with sonication. Cryosectioning was performed on a Leica CM1860 UV Cryostat (Leica Biosystems, Deer Park, IL) (Figure ). All organs were cryosectioned at 10 μm thickness at −20 °C. Each slide contained two brain sections, two heart sections, and two kidney sections. All sections were sagittal sections of the tissue. Slides were stored in a −80 °C freezer until matrix deposition and were desiccated at room temperature prior to matrix application. Each sublimation or spray replicate was done individually (i.e., one slide was sprayed or sublimated at a time). Several matrix deposition parameters, such as spray/sublimation temperature, solvent composition, etc., can affect the crystal size. Established optimized protocols were used to minimize the crystal size. Matrix deposition parameters can be found in the Supporting Information (Tables S1 and S2).

1.

1

Open in a new tab

Experimental workflow. Brain, heart, and kidney samples were cryosectioned onto ITO-coated slides. Matrix was then applied through either spraying or sublimation. Matrix crystal sizes were measured on an optical slide scanner, and then MALDI MSI was performed on a Bruker rapifleX or timsTOF fleX instrument depending on the matrix.

Slide Weighing

Slides were weighed using a Pioneer PX85 balance (Ohaus Corporation, Parsippany, NJ). Slides were weighed before and after matrix deposition. A two-tailed paired t-test using Microsoft Excel (Microsoft, Redmond, WA) was calculated to determine if the mass of a matrix was significantly different for sublimated and sprayed samples. Significance was determined depending on whether p < 0.05.

Polarized Light Microscopy Images

Crystal images were measured on a VS200 Slide Scanner (Olympus Life Science, Waltham, MA; Figure ). Images were taken with a 40× objective and a light polarization of 95°. For each slide that was imaged, five individual crystal measurements were taken in each of the following regions: off-tissue matrix, brain, heart, and kidney. Crystal measurements were made in the Olympus OlyVIA 4.1.1 software. Length was measured across the longest cross-section of each crystal to account for irregularities in crystal shape. Crystals were measured randomly from five different points of view to account for tissue or structural differences. A two-tailed paired t-test using Microsoft Excel (Microsoft, Redmond, WA) was performed to determine if crystal size differed significantly for sublimated and sprayed samples. Significance was determined by p < 0.05.

MALDI Imaging

MALDI imaging of DHB-coated samples was carried out on a rapifleX MALDI ToF/ToF instrument (Bruker, Billerica, MA) (Figure ). Samples were imaged in positive ion mode with a 50 μm spatial resolution, 50 μm raster size, 50 μm M5 small laser spot size, and 200 shots per pixel. MALDI imaging of CHCA- and DHAP-coated samples was acquired on a timsTOF fleX MALDI-2 instrument (Bruker, Billerica, MA; Figure ). Samples were imaged in positive ion mode with a 50 μm spatial resolution, 50 μm raster size, 50 μm single laser spot size, and 200 shots per pixel. For each slide, four regions of interest (ROIs) were drawn: brain, heart, kidney, and a square for off-tissue matrix. All tissue ROIs were sagittal sections. Regions were drawn to surround the entire tissue with additional space on the slide to analyze delocalization. For each experiment, a sublimated slide and a sprayed slide were run together in a single imaging experiment. A total of three replicates was completed for each matrix.

High Spatial Resolution MALDI Imaging

High spatial resolution MALDI imaging of DHB-coated samples was performed on a rapifleX MALDI ToF/ToF instrument (Bruker, Billerica, MA). CHCA- and DHAP-coated samples were imaged on a timsTOF fleX MALDI-2 instrument (Bruker, Billerica, MA). Sagittal whole brain sections were imaged in positive ion mode with a 10 μm spatial resolution, 10 μm raster size, 10 μm single laser spot size, and 200 shots per pixel. Sagittal cerebellum sections were imaged in positive ion mode with a 5 μm spatial resolution, 5 μm raster size, 5 μm single laser spot size, and 200 shots per pixel. Our threshold is 5 μm because this is the minimum laser size possible for the Bruker rapifleX and timsTOF fleX MALDI-2.

Data Analysis

Following imaging, data were imported into SCiLS Lab (Version 11.01.14,623). Data were normalized by Total Ion Count (TIC). A Receiver Operating Characteristic (ROC) analysis was used to determine lipids that had significantly greater intensities in one matrix deposition method over the other. An Area Under the Curve (AUC) value greater than 0.75 or less than 0.25 was considered significant. All organs of the same type and matrix deposition method were combined into a single ROI for each ROC analysis (i.e., all three sublimated brains were combined into one region). Analyte delocalization was identified in the MALDI imaging data by a visual inspection of individual ion images. Identifications were made using tandem MS experiments, the details of which can be found in the Supporting Information (Tables S15–S23, Figures S11–S43).

Results

Matrix Weights

Slides were weighed before and after matrix deposition to determine the amount of matrix deposited on the slide in milligrams. Two-tailed paired t testing revealed that the weight of matrix deposited was not significantly different between spraying and sublimation for all three matrices. Results from matrix weight determination are shown in Figure . Mass measurements and t-test results can be found in Supporting Information in Tables S3–S6. These results demonstrate that differences in the MALDI imaging data between spray and sublimation are not due to differing amounts of matrix deposited between the two methods of matrix application.

2.

2

Open in a new tab

Weight of matrix deposited on slides. A two-tailed t-test was used to compare matrix weights, and no significant difference was found (p < 0.05). P-values for CHCA, DHAP, and DHB were 0.364, 0.113, and 0.707, respectively. Data are shown as mean and standard deviation. n = 3.

Crystal Sizes

Crystal measurements were measured both on- and off-tissue for all experimental conditions by using polarized light microscopy. Example crystal images can be found in Figure S1. Individual crystal sizes were measured using a point-to-point line tool in the slide scanner software (Tables S7–S14). Five distinct crystals were measured per tissue sample preparation under a 40× objective lens. Three replicates were done for each tissue sample and off-tissue crystal measurement event. No significant difference was found in crystal size between tissue types for the same matrix deposition method, except for DHB sublimated brains compared with DHB sublimated kidneys. We identified a crystal measurement from the brain (2.63 μm) that is bringing the average up and likely causing a significant difference. More testing is needed to determine how the tissue type affects crystal formation. Two-tailed paired t testing revealed that CHCA crystals were significantly smaller when sublimated as compared to sprayed on-tissue but had no significant difference when deposited off-tissue (Tables S7, S8, S13, and S14). DHAP crystals were found to have no significant size difference between spraying and sublimation both on- and off-tissue. DHB crystals were significantly smaller from sublimation, both on- and off-tissue, as compared to spraying. Graphs of the crystal sizes are shown in Figure . These results demonstrate that sublimation does not automatically produce smaller crystals across all matrices compared to spraying. Under the reported conditions, both sublimated CHCA and DHB produced smaller crystals on the tissue. DHAP, however, produced the same-sized crystals when sprayed and sublimated. For all conditions except for sprayed DHB, all crystal sizes were well below the 5 μm cutoff needed for MALDI imaging at 5 μm spatial resolution.

3.

3

Open in a new tab

(A). Off-tissue crystal sizes as measured by polarized light microscopy. A two-tailed paired t-test showed that the sizes of DHB off-tissue crystals were significantly different when the matrices were sublimated versus sprayed. Data are shown as mean and standard deviation. n = 3. **p = 1.28 × 10–5. (B). On-tissue crystal sizes as measured by polarized light microscopy. Data for all three tissue types, i.e., mouse brain, heart, and kidney, were pooled. A two-tailed paired t-test showed that the crystal sizes of all matrices except DHAP were significantly different when the matrix was sublimated rather than sprayed. Data are shown as mean and standard deviation. n = 3. †p = 4.97 × 10–11, **p = 3.84 × 10–17.

MALDI Imaging Spectra

Analysis of the average spectra revealed that across the different matrices and organs, spraying and sublimating had comparable mass ranges (Figures S2–S4). For both deposition methods, we see analytes detected across the entire spectrum for all three matrices. Notably, none of the spectra for any of the three matrices showed a consistent decrease in signal intensity from sublimation as compared to that from spraying of the same matrix. Our hypothesis based on the literature was that sublimation would lead to decreased signal intensity. This was not observed for CHCA, DHAP, or DHB in positive ion mode, as evident from these data. Consistent overall signal intensity across the spectra was observed for all tissues, matrices, and sample preparations. The most pronounced differences between spraying and sublimation can be observed in the matrix background (Figures S2A, S3A, and S4A). For CHCA (Figure S2A) and DHAP (Figure S3A), sublimated matrix produced greater background signal in the lower end of the mass range of the spectrum (<400 m/z) while spraying produced matrix background in the higher mass range of the spectrum (>600 m/z). These data indicate that spraying may produce larger matrix ion clusters, leading to the increase in signal in the higher mass range of the spectrum.

Signal Intensity ROC Analyses

To begin quantitatively assessing signal intensity differences between sublimation and spraying, receiver operating characteristic (ROC) analysis was performed with sublimated ROIs as class 1 and sprayed ROIs as class 2 for each tissue type. All detected peaks were used for this analysis, not just identified lipids. Several lipid peaks had significantly greater signal intensity when matrix was sprayed rather than sublimated (AUC <0.25) or when matrix was sublimated rather than sprayed (AUC >0.75). The number of peaks with greater signal intensity for either deposition method is listed in Table , broken down by matrix and tissue type. In general, the matrix noise peaks are the most different between sublimated and sprayed ROIs, rather than the organ lipid analyte peaks. This is consistent with observations of the average spectra. We hypothesize that these differences result from different matrix clusters and adducts being formed by each matrix deposition method. We expected that there would be more analytes in the sprayed rather than sublimated group (AUC <0.25) if spraying always produces more intense peaks than sublimation. However, our peak intensity data do not support this hypothesis.

1. Number of Peaks with AUC under 0.25 (Increased in Sprayed) and above 0.75 (Increased in Sublimated), by Tissue Type.

matrix region AUC <0.25 AUC >0.75
CHCA off-tissue matrix 897 232
brain 272 214
heart 213 123
kidney 172 70
DHAP off-tissue matrix 536 271
brain 24 9
heart 89 15
kidney 66 14
DHB off-tissue matrix 56 143
brain 17 8
heart 29 14
kidney 29 10
Open in a new tab

Side-by-side ion image comparisons of samples prepared via either sublimation or spraying are compiled in Figures S5–S7. ROC and intensity box plots were included to quantify the differences in signal intensity for each shown m/z value.

For all matrices, there was no overall trend regarding which deposition method resulted in a more intense signal across all ions. Signal intensity differences between matrix deposition methods were ion specific. Example images can be found in the Supporting Information. As demonstrated in Figures S5–S7, entirely different m/z’s are found between the sprayed and sublimated samples.

Sprayed CHCA tissues had an average of 73,719 total peaks. Sublimated CHCA tissues had an average of 71,272 total peaks. Sprayed DHAP tissues had an average of 74,667 total peaks. Sublimated DHAP tissues had an average of 72,171 total peaks. Sprayed DHB tissues had an average of 84,433 total peaks. Sublimated DHB tissues had an average of 75,193 total peaks. These results may indicate that spraying can detect more analytes; however, this is not a perfect measure because it may also include matrix clusters.

Delocalization

In general, we observed more delocalization when the matrix was sprayed than when the matrix was sublimated (Figures S8–S10). This observation held true across all three matrices examined. We observed more intense delocalization as well as more dispersion of analyte farther off tissue. Notably, sublimation does not result in zero delocalization, as can be seen in Figures S8A and S10A.

High Spatial Resolution MALDI Imaging

To further assess the impact of the matrix deposition method on image quality, we performed high spatial resolution imaging, again comparing all 3 matrices via sublimation or spraying. We chose brain tissue for these experiments because of its highly structured tissue anatomy and morphology. Whole mouse brain sections were imaged at 10 μm pixel size and mouse cerebellum sections were imaged at 5 μm pixel size for CHCA and DHAP. High spatial resolution imaging experiments revealed comparable spatial resolution between sublimation and spraying for 10 and 5 μm pixel size when using CHCA (Figure ) and DHAP (Figure ). The structural details of both brains are clearly identifiable. Notably, gray and white matter differences are observable in the 10 and 5 μm data, with more dramatic differences observed by sublimation. In the 10 μm images, we can discern the hippocampus and the corpus callosum. These data are consistent with crystal image measurements, which indicated the crystal sizes were below 5 μm with both matrix deposition methods. However, both matrices seemed to produce higher signal when sublimated versus sprayed at this imaging resolution, CHCA in particular. When using DHB, 10 μm imaging showed significantly better resolution with sublimation compared to spraying (Figure ). Hippocampus and corpus callosum structural boundaries are not easily discerned in the sprayed image. Additionally, the ion intensity map of the sprayed image appears to reflect the arrangement of the crystals on top of the tissue rather than the substructures of the tissue itself. This finding is consistent with our crystal size data, as the sprayed DHB crystal size was greater than 10 μm, leading to poor image quality and reduced spatial resolution. While 10 μm imaging was performed on DHB as a proof of concept that spraying resulted in compromised spatial resolution, we did not perform 5 μm imaging.

4.

4

Open in a new tab

High spatial resolution MALDI imaging for sagittal whole brain and cerebellum images sections using the CHCA matrix.

5.

5

Open in a new tab

High spatial resolution MALDI imaging for sagittal whole brain and cerebellum images sections using the DHAP matrix.

6.

6

Open in a new tab

High spatial resolution MALDI imaging for sagittal whole brain and cerebellum images sections using the DHB matrix.

Discussion

In this study, we have systematically investigated the differences in the MALDI MSI matrix deposition methods of spraying versus sublimation to begin to address the persistent knowledge gap in sample preparation for MALDI imaging. To date, there are only a handful of publications that address the differences in MALDI imaging data resulting from matrix application methods. − We compared crystal size, analyte delocalization, and analyte signal intensity when using a commercial automatic pneumatic sprayer and a commercial sublimator. We tested three commonly used positive ion mode matrices: CHCA, DHAP, and DHB on three tissue types, i.e., mouse brain, heart, and kidneys. CHCA and DHB crystals were significantly smaller for on-tissue sublimation, but only DHB was significantly smaller off-tissue for sublimation. DHAP had no significant difference in the crystal size between the two deposition methods. Our results also showed that while analyte delocalization was increased in some cases when using a sprayer, there were several analytes that had comparable localization when using a sprayer and a sublimator. Likewise, while the results showed an increased signal intensity for several sprayed analytes, they also showed an increased signal intensity for several sublimated analytes. Together, these findings present some novel results for the commonly held paradigm for the impact of solvent in matrix application methods.

Matrix Density

Previous studies comparing matrix deposition methods did not control for matrix density. − To date, we do not have a good understanding of the full extent of the matrix density in MALDI MSI experiments. We know that there is a direct correlation between matrix density and signal intensity; however, more matrix is not always better for optimal detection of analytes in MALDI imaging. Increased matrix density corresponds to an increased matrix signal, which can drown out the analyte signal. This can be detrimental, especially in the case of smaller-molecular-weight compounds such as metabolites that are similar in m/z values to the matrix background. In this study, we implemented weighing slides before and after matrix deposition to ensure that a comparable amount of matrix was applied to each slide, regardless of the matrix deposition type. While we did not optimize for the amount of matrix deposited, we did control for it, which allows our results to be directly comparable in terms of signal intensity observed from the experiments. This experiment is key in allowing us to conclude that sublimation does not inherently result in lower signal intensity when compared to spraying. It is possible that differences in crystal size could lead to varied optimal matrix density. This could be due to the way crystals are irradiated by the MALDI laser and how smaller and larger sizes impact that process. Further studies could explore matrix density optimization for both matrix deposition methods.

Crystal Size

Each of the three matrix types showed different trends in crystal size between spraying and sublimation. In one previous study, crystal sizes of CHCA and DHB matrix were qualitatively compared following their deposition by airbrushing, spraying, and sublimation. In this study, Gemperline et al. observed that airbrushing produced large crystals, but they did not quantify the size of the crystals from spraying versus sublimation. In our study, we sought to ensure that crystals were below 5 μm in size to obtain the minimum spatial resolution possible on our commercial MALDI imaging instrumentation. This approach required quantifying the crystal size using polarized light microscopy. In the case of CHCA, our results showed significantly smaller sublimated crystal size for matrix spraying versus sublimation only on-tissue but not off-tissue. This finding indicates that molecular interactions between the CHCA matrix, tissue biomolecules, and the solvent used in spraying promote larger crystal growth. It is possible that in the presence of a nucleating site, such as tissue, it is thermodynamically favorable for CHCA to cocrystallize with the tissue analytes and existing crystals as solvent evaporates. This would cause the observed phenomenon that sprayed CHCA crystal size is significantly larger on-tissue compared to CHCA sublimation. Crystal size is often proportional to growth rate and inversely proportional to nucleation rate. It is possible that the sublimation of CHCA is occurring so rapidly that crystals are not given as much time to grow as they are when sprayed. The fact that the nucleation rate impedes crystal size may explain why this phenomenon only occurs on tissue since it gave crystals access to more nucleation sites.

DHAP produced the same crystal sizes both on- and off-tissue by both methods of matrix deposition, unlike CHCA or DHB. This may be the case because the DHAP matrix may form smaller crystals than other matrices based on its intrinsic crystallization behavior. This may in turn lead to matrix crystallization in which the crystal size does not change on- or off-tissue.

DHB was the only matrix that showed dramatic reduction in crystal size with sublimation as compared to spraying. Spray deposition of DHB crystals resulted in the largest size and the largest variability in crystal size of all matrices tested. The dramatic reduction in both average value and standard deviation for crystal size achieved by DHB sublimation versus spraying exemplifies the consistent minimal crystal size hypothesized to be achievable by sublimation.

We observed that matrix deposition on- and off-tissue played a crucial role in crystal formation. Interestingly, for both CHCA and DHB matrices, we detected differently sized crystals when on- and off-tissue crystal formation was compared. This is a critical observation when considering the crystal size in relation to spatial resolution. Based on our findings, it is highly recommended to measure on-tissue crystal size for accurately assessing matrix crystal size for MALDI imaging experiments.

Previous studies have shown that lower salt concentrations may minimize crystal size. Further studies could explore if a cold ammonium formate wash to remove excess salt could achieve an even smaller crystal size.

ROC Analyses for Signal Intensity and Delocalization

In this study, we used ROC analyses to compare the differences between the sprayed and sublimated samples. Previous studies have counted the number of peaks in the spectra to determine the number of analytes in the sample. This does not, however, indicate how many peaks are different under the two conditions. The use of ROC analyses in our study demonstrated that different analytes are being ionized depending on the matrix application method utilized. This is also notable in the matrix blank region off-tissue where we observed many different peaks by ROC analysis when comparing spraying versus sublimation. These data indicate that different matrix clusters and impurities are introduced when depositing matrix by sublimation versus spraying, which then ionize and produce signals in the MALDI imaging experiment. Furthermore, for on-tissue MALDI imaging experiments, we observed different tissue analytes are being extracted from the tissues based on the matrix application method. An example of this is how m/z 912.458 had greater signal intensity in the brain with CHCA sublimation compared to CHCA spraying, but m/z 788.619 had greater signal intensity in the brain with CHCA spraying compared to CHCA sublimation. We conclude from these data that the analyte or analytes of interest should be considered when choosing a matrix deposition method.

Our data do support the general hypothesis that matrix spraying leads to more analyte delocalization than sublimation of the same matrix; however, sublimation did also result in some delocalization for some analytes. Sublimation does not entirely prevent delocalization; however, it can minimize it. Sublimation does not replace the need for optimization experiments to minimize the delocalization. Delocalization continues to be a challenge that must be overcome as we approach smaller spatial resolution and branch into different application spaces with different tissue and cell types and analytes of interest being examined.

High Spatial Resolution MALDI Imaging

High spatial resolution MALDI imaging (10 μm whole brain and 5 μm cerebellum) demonstrated that we can reach the limits of spatial resolution with both spraying and sublimation as the matrix deposition method. We demonstrated that all three matrices tested can result in high-quality MALDI imaging data at high spatial resolution at small pixel sizes when sublimated. Spraying CHCA and DHAP provided comparable high spatial resolution quality as compared to sublimation; however, spraying DHB showed a stark loss of spatial resolution at small pixel sizes. Maintaining good spatial resolution at small pixel sizes is crucial for eventually achieving MALDI imaging at single cell or subcellular resolution. Our results show that matrix deposition by sublimation of all tested matrices and by spraying for some matrices are highly reproducible and effective matrix deposition methods for achieving high spatial resolution at small pixel sizes.

Conclusions

The matrix application step of MALDI MSI sample preparation can have profound effects on the resulting imaging data. Yet optimizing this step is often overlooked, and this becomes especially relevant to high spatial resolution imaging experiments. To achieve images with high spatial resolution, uniform application of fine matrix crystals is necessary, with the matrix crystal size being kept below the intended pixel size of imaging. Typically, the size of matrix crystals is thought to be influenced by the presence of solvent in spray-based applications, and sublimation is therefore considered a gold standard for producing ultrasmall matrix crystals.

In this study, we examined the hypothesis that sublimating matrix for MALDI MSI leads to less analyte delocalization and lower signal intensity than spraying for DHB, CHCA, and DHAP. Our results demonstrate that this hypothesis is partially incorrect. While spraying generally led to more delocalization, matrix spraying and sublimation resulted in the ionization of different matrix clusters as well as different tissue analytes. Our results showed that signal intensity was not dependent on the matrix deposition method but rather the analyte being observed. These results are bolstered by our careful control of the amount of matrix deposited on to the sample. These data, taken together, show that the analyte or analyte class of interest and the desired spatial resolution should dictate the sample preparation method used.

Supplementary Material

ac5c01767_si_001.pdf (2.8MB, pdf)

Acknowledgments

We would like to acknowledge the Johns Hopkins Applied Imaging Mass Spectrometry (AIMS) Core facility where all experiments were performed. We would like to acknowledge NIH S10 OD030500, which funded the Bruker timsTOF fleX MALDI-2 instrument. We would also like to acknowledge the Department of Defense (Grant Number W81XWH22C0047) and the NIH (R01NS136102) for funding.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.5c01767.

  • Sublimation, spray, and MS/MS protocols, slide weights and statistics, crystal images, measurements, and statistics, representative average spectra, delocalization images, identifications, and annotated MS/MS spectra (PDF)

The authors declare the following competing financial interest(s): Alyson Black and Alain Creissen work for HTX Technologies which are the makers of both the sublimate and the sprayer used in this study.

References

  1. Jiang L., Chughtai K., Purvine S. O., Bhujwalla Z. M., Raman V., Pasa-Tolic L., Heeren R. M. A., Glunde K.. MALDI-Mass Spectrometric Imaging Revealing Hypoxia-Driven Lipids and Proteins in a Breast Tumor Model. Anal. Chem. 2015;87(12):5947–5956. doi: 10.1021/ac504503x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Muralidharan S., Shimobayashi M., Ji S., Burla B., Hall M. N., Wenk M. R., Torta F.. A reference map of sphingolipids in murine tissues. Cell Rep. 2021;35(11):109250. doi: 10.1016/j.celrep.2021.109250. [DOI] [PubMed] [Google Scholar]
  3. Stübiger G., Belgacem O.. Analysis of Lipids Using 2,4,6-Trihydroxyacetophenone as a Matrix for MALDI Mass Spectrometry. Anal. Chem. 2007;79(8):3206–3213. doi: 10.1021/ac062236c. [DOI] [PubMed] [Google Scholar]
  4. Aichler M., Walch A.. MALDI Imaging mass spectrometry: current frontiers and perspectives in pathology research and practice. Lab. Invest. 2015;95(4):422–431. doi: 10.1038/labinvest.2014.156. [DOI] [PubMed] [Google Scholar]
  5. Attia A., Schroeder K. A., Seeley E., Wilson K., Hammer N., Colvin D., Manier M. ., Nicklay J., Rose K., Gore J., Caprioli R., Skaar E.. Monitoring the Inflammatory Response to Infection through the Integration of MALDI IMS and MRI. Cell Host Microbe. 2012;11(6):664–673. doi: 10.1016/j.chom.2012.04.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Caprioli R. M., Farmer T. B., Gile J.. Molecular Imaging of Biological Samples: Localization of Peptides and Proteins Using MALDI-TOF MS. Anal. Chem. 1997;69(23):4751–4760. doi: 10.1021/ac970888i. [DOI] [PubMed] [Google Scholar]
  7. Angel P. M., Mehta A., Norris-Caneda K., Drake R. R.. MALDI Imaging Mass Spectrometry of N-glycans and Tryptic Peptides from the Same Formalin-Fixed, Paraffin-Embedded Tissue Section. Methods Mol. Biol. 2017;1788:225–241. doi: 10.1007/7651_2017_81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Drake R. R., Powers T. W., Norris-Caneda K., Mehta A. S., Angel P. M.. In Situ Imaging of N-Glycans by MALDI Imaging Mass Spectrometry of Fresh or Formalin-Fixed Paraffin-Embedded Tissue. Curr. Protoc Protein Sci. 2018;94(1):e68. doi: 10.1002/cpps.68. [DOI] [PubMed] [Google Scholar]
  9. Grgic A., Krestensen K. K., Heeren R. M. A.. Optimized protocol for MALDI MSI of N-glycans using an on-tissue digestion in fresh frozen tissue sections. Sci. Rep. 2023;13(1):2776. doi: 10.1038/s41598-023-29560-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Heijs B., Holst S., Briaire-de Bruijn I. H., van Pelt G. W., de Ru A. H., van Veelen P. A., Drake R. R., Mehta A. S., Mesker W. E., Tollenaar R. A., Bovée J. V. M. G., Wuhrer M., McDonnell L. A.. Multimodal Mass Spectrometry Imaging of N-Glycans and Proteins from the Same Tissue Section. Anal. Chem. 2016;88(15):7745–7753. doi: 10.1021/acs.analchem.6b01739. [DOI] [PubMed] [Google Scholar]
  11. Scupakova K. A. O., Balluff B., Ayyappan V., Tressler C. M., Cimino-Mathews A. M., White M. J., Argani P., Heeren R. M., Glunde K.. Clinical importance of high-mannose, fucosylated and complex N-glycans in metastatic breast cancer. JCI: Under review. 2021;6:e146945. doi: 10.1172/jci.insight.146945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Amstalden van Hove E. R., Blackwell T. R., Klinkert I., Eijkel G. B., Heeren R. M. A., Glunde K.. Multimodal mass spectrometric imaging of small molecules reveals distinct spatio-molecular signatures in differentially metastatic breast tumor models. Cancer Res. 2010;70(22):9012. doi: 10.1158/0008-5472.can-10-0360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Buck A., Ly A., Balluff B., Sun N., Gorzolka K., Feuchtinger A., Janssen K. P., Kuppen P. J., van de Velde C. J., Weirich G., Erlmeier F., Langer R., Aubele M., Zitzelsberger H., Aichler M., Walch A.. High-resolution MALDI-FT-ICR MS imaging for the analysis of metabolites from formalin-fixed, paraffin-embedded clinical tissue samples. J. Pathol. 2015;237(1):123–132. doi: 10.1002/path.4560. [DOI] [PubMed] [Google Scholar]
  14. Ly A., Buck A., Balluff B., Sun N., Gorzolka K., Feuchtinger A., Janssen K. P., Kuppen P. J., van de Velde C. J., Weirich G., Erlmeier F., Langer R., Aubele M., Zitzelsberger H., McDonnell L., Aichler M., Walch A.. High-mass-resolution MALDI mass spectrometry imaging of metabolites from formalin-fixed paraffin-embedded tissue. Nat. Protoc. 2016;11(8):1428–1443. doi: 10.1038/nprot.2016.081. [DOI] [PubMed] [Google Scholar]
  15. McLaughlin N., Bielinski T. M., Tressler C. M., Barton E., Glunde K., Stumpo K. A.. Pneumatically Sprayed Gold Nanoparticles for Mass Spectrometry Imaging of Neurotransmitters. J. Am. Soc. Mass Spectrom. 2020;31(12):2452–2461. doi: 10.1021/jasms.0c00156. [DOI] [PubMed] [Google Scholar]
  16. Shimma S., Sugiura Y.. Effective Sample Preparations in Imaging Mass Spectrometry. Mass Spectrom. 2014;3(Special_Issue):S0029. doi: 10.5702/massspectrometry.s0029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Yalcin E. B., de la Monte S. M.. Review of Matrix-Assisted Laser Desorption Ionization-Imaging Mass Spectrometry for Lipid Biochemical Histopathology. J Histochem Cytochem. 2015;63(10):762–771. doi: 10.1369/0022155415596202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Tressler C. M., Wadsworth B., Carriero S., Dillman N., Crawford R., Hahm T.-H., Glunde K., Cadieux C. L.. Characterization of Humanized Mouse Model of Organophosphate Poisoning and Detection of Countermeasures via MALDI-MSI. Int. J. Mol. Sci. 2024;25:5624. doi: 10.3390/ijms25115624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Tressler C., Tilley S., Yang E., Donohue C., Barton E., Creissen A., Glunde K.. Factorial Design to Optimize Matrix Spraying Parameters for MALDI Mass Spectrometry Imaging. J. Am. Soc. Mass Spectrom. 2021;32(12):2728–2737. doi: 10.1021/jasms.1c00081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Tressler C. M., Ayyappan V., Nakuchima S., Yang E., Sonkar K., Tan Z., Glunde K.. A multimodal pipeline using NMR spectroscopy and MALDI-TOF mass spectrometry imaging from the same tissue sample. NMR in Biomedicine. 2023;36(4):e4770. doi: 10.1002/nbm.4770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hankin J. A., Barkley R. M., Murphy R. C.. Sublimation as a method of matrix application for mass spectrometric imaging. J. Am. Soc. Mass Spectrom. 2007;18(9):1646–1652. doi: 10.1016/j.jasms.2007.06.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Murphy R. C., Hankin J. A., Barkley R. M., Zemski Berry K. A.. MALDI imaging of lipids after matrix sublimation/deposition. Biochim. Biophys. Acta. 2011;1811(11):970–975. doi: 10.1016/j.bbalip.2011.04.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Shanmugaraj N., Rutten T., Svatoš A., Schnurbusch T., Mock H.-P.. Fast and Reproducible Matrix Deposition for MALDI Mass Spectrometry Imaging with Improved Glass Sublimation Setup. J. Am. Soc. Mass Spectrom. 2023;34(3):513–517. doi: 10.1021/jasms.2c00301. [DOI] [PubMed] [Google Scholar]
  24. Ščupáková K., Balluff B., Tressler C., Adelaja T., Heeren R. M. A., Glunde K., Ertaylan G.. Cellular resolution in clinical MALDI mass spectrometry imaging: the latest advancements and current challenges. Clin Chem Lab Med. 2020;58(6):914–929. doi: 10.1515/cclm-2019-0858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Good C. J., Neumann E. K., Butrico C. E., Cassat J. E., Caprioli R. M., Spraggins J. M.. High Spatial Resolution MALDI Imaging Mass Spectrometry of Fresh-Frozen Bone. Anal. Chem. 2022;94(7):3165–3172. doi: 10.1021/acs.analchem.1c04604. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Huizing L. R. S., Ellis S. R., Beulen B. W. A. M. M., Barré F. P. Y., Kwant P. B., Vreeken R. J., Heeren R. M. A.. Development and evaluation of matrix application techniques for high throughput mass spectrometry imaging of tissues in the clinic. Clin. Mass Spectrom. 2019;12:7–15. doi: 10.1016/j.clinms.2019.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Gemperline E., Rawson S., Li L.. Optimization and Comparison of Multiple MALDI Matrix Application Methods for Small Molecule Mass Spectrometric Imaging. Anal. Chem. 2014;86(20):10030–10035. doi: 10.1021/ac5028534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Sarretto T., Westerhausen M. T., McKinnon J. C., Bishop D. P., Ellis S. R.. Evaluation of combined workflows for multimodal mass spectrometry imaging of elements and lipids from the same tissue section. Anal. Bioanal. Chem. 2025;417(4):705–719. doi: 10.1007/s00216-024-05696-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Goodwin R. J. A.. Sample preparation for mass spectrometry imaging: Small mistakes can lead to big consequences. J. Proteomics. 2012;75(16):4893–4911. doi: 10.1016/j.jprot.2012.04.012. [DOI] [PubMed] [Google Scholar]
  30. Mielczarek P., Suder P., Kret P., Słowik T., Gibuła-Tarłowska E., Kotlińska J. H., Kotsan I., Bodzon-Kulakowska A.. Matrix-assisted laser desorption/ionization mass spectrometry imaging sample preparation using wet-interface matrix deposition for lipid analysis. Rapid Commun. Mass Spectrom. 2023;37(14):e9531. doi: 10.1002/rcm.9531. [DOI] [PubMed] [Google Scholar]
  31. Horstman E. M., Bertke J. A., Kim E. H., Gonzalez L. C., Zhang G. G. Z., Gong Y., Kenis P. J. A.. Crystallization and characterization of cocrystals of piroxicam and 2,5-dihydroxybenzoic acid. CrystEngComm. 2015;17(28):5299–5306. doi: 10.1039/C5CE00355E. [DOI] [Google Scholar]

Associated Data

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

Supplementary Materials

ac5c01767_si_001.pdf (2.8MB, pdf)

Articles from Analytical Chemistry are provided here courtesy of American Chemical Society

RESOURCES