RESEARCH PAPER

Evaluation of lipid coverage and high spatial resolutionMALDI-imaging capabilities of oversampling combined with laserpost-ionisation

Andrew P. Bowman1& Jeroen F. J. Bogie2

& Jerome J. A. Hendriks2 & Mansour Haidar2 & Mikhail Belov3 &

Ron M. A. Heeren1& Shane R. Ellis1

Received: 9 August 2019 /Revised: 4 November 2019 /Accepted: 18 November 2019# The Author(s) 2019

AbstractMatrix-assisted laser desorption/ionisation-mass spectrometry imaging (MALDI-MSI) is a powerful technique for visualising thespatial locations of lipids in biological tissues. However, a major challenge in interpreting the biological significance of local lipidcompositions and distributions detected using MALDI-MSI is the difficulty in associating spectra with cellular lipid metabolismwithin the tissue. By-and-large this is due to the typically limited spatial resolution of MALDI-MSI (30–100 μm) meaningindividual spectra represent the average spectrum acquired from multiple adjacent cells, each potentially possessing a uniquelipid composition and biological function. The use of oversampling is one promising approach to decrease the sampling area andimprove the spatial resolution in MALDI-MSI, but it can suffer from a dramatically decreased sensitivity. In this work weovercome these challenges through the coupling of oversampling MALDI-MSI with laser post-ionisation (MALDI-2). Wedemonstrate the ability to acquire rich lipid spectra from pixels as small as 6 μm, equivalent to or smaller than the size of typicalmammalian cells. Coupled with an approach for automated lipid identification, it is shown that MALDI-2 combined withoversampling at 6 μm pixel size can detect up to three times more lipids and many more lipid classes than even conventionalMALDI at 20 μm resolution in the positive-ion mode. Applying this to mouse kidney and human brain tissue containing activemultiple sclerosis lesions, where 74 and 147 unique lipids are identified, respectively, the localisation of lipid signals to individualtubuli within the kidney and lipid droplets with lesion-specific macrophages is demonstrated.

Keywords Mass spectrometry imaging .MALDI . Lipids . Kidney . Brain .Multiple sclerosis

Introduction

Matrix-assisted laser desorption/ionisation-mass spectrometryimaging (MALDI-MSI) is a powerful method for visualisingthe spatial distributions of lipids throughout biological tissues[1–3]. Its versatility for mapping changing lipid compositionswithin tissues has been demonstrated in a variety of applica-tions, including oncology [4–6], bacterial infections [7, 8] andliver disease [9, 10], amongst many others. The heterogeneouslipid compositions observed withMSI are ultimately the resultof cellular-level lipid metabolism occurring within the cellsconstituting the tissue sample. However, to date, detailed in-terpretation of lipid MSI data in terms of cellular-level lipidmetabolism has been a major challenge, due to limitations inboth technology and informatics. One of the most significantlimitations has been the relatively low spatial resolution ofconventional methods. Typical experiments are performed at

Published in the topical collection Current Progress in Lipidomics withguest editors Michal Holčapek, Gerhard Liebisch, and Kim Ekroos.

Electronic supplementary material The online version of this article(https://doi.org/10.1007/s00216-019-02290-3) contains supplementarymaterial, which is available to authorized users.

* Shane R. Elliss.ellis@maastrichtuniversity.nl

1 Maastricht MultiModal Molecular Imaging (M4I) Institute, Divisionof Imaging Mass Spectrometry, Maastricht University,Universiteitssingel 50, 6629, ER Maastricht, The Netherlands

2 Department of Immunology and Biochemistry, Biomedical ResearchInstitute, Hasselt University, 3590 Diepenbeek, Belgium

3 Spectroglyph LLC, Kennewick, WA 99338, USA

https://doi.org/10.1007/s00216-019-02290-3Analytical and Bioanalytical Chemistry (2020) 412:2277–2289

/Published online: 26 December 2019

pixel sizes of ~ 30–100 μm, significantly larger than mostmammalian cells. As a result, each pixel (spectrum) representsthe averaged lipid profile acquired from multiple adjacentcells, each potentially possessing distinct metabolic hallmarksand biological functions. This precludes measuring a lipidprofile that is reflective of any individual cell, or even singlecell type, within the tissue.

The main limitation of pixel size in MALDI-MSI is thearea of desorption/ionisation on the tissue surface, that, in turn,is dependent on the laser spot size. Several groups have re-ported optical modifications to commercial MALDI-MSI ionsources that have reduced the laser spot size and, thus, achiev-able pixel sizes, down to < 10 μm and at best ~ 1 μm [11–15].For example, Kompauer et al. recently reported a modifiedatmospheric pressure MALDI-MSI source capable ofreaching 1.4 μm pixel size and imaging lipids within singlecells [12]. A drawback of these approaches for reducing laserspot size is the requirement of modifications to instrumentoptics and the relatively short depth of focus of narrowly fo-cussed lasers. The latter issue can potentially render the ap-proach sensitive to sample topology or imperfect flatness ofthe sample stage, which can lead to changing laser fluence andionisation efficiency across the sample surface. Although re-cent developments in autofocussing methods to compensatefor sample topography can help ameliorate these effects, theseare not yet widely used [16].

An alternative method to improve spatial resolution ofMALDI-MSI is through the oversampling approach [17].In oversampling, once all material at given sampling posi-tion has been desorbed/ablated the sample stage is movedby a distance smaller than the laser spot size. As a result,only part of the laser beam is used for desorption/ionisationand pixel sizes smaller than the spot size can be achieved.The primary advantage of oversampling is it can beemployed on most commercial MALDI systems withouthardware modifications, so long as the stage is capable ofperforming sufficiently small steps. Examples of MALDI-MSI in the oversampling mode include imaging ofglycosphingolipids in spleen tissue from a Gaucher diseasemodel at a pixel size of 15 μm [18], phospholipids andsulfatides in brain and lung tissue down to 10 μm pixel size[19] and human colon tissue at a pixel size of ~ 5 μm [20].High-throughput imaging using continuous raster-mode ac-quisitions can also generate ions under oversampling-likeconditions, where only the edge of the laser beam is gener-ating signal [21], although it has been suggested that severeoversampling conditions can lead to reduced sensitivity inraster-modeMSI [22]. In addition toMALDI, oversamplinghas also been utilised in IR-MALDESI experiments where a10-μm pixel size was achieved for cholesterol imaging fromhuman cervical tissue, although the authors noted that atsuch pixel sizes a dramatic decrease in sensitivity and lipidcoverage was observed [23].

A challenge with oversampling using conventionalMALDI lasers with Gaussian-like intensity profiles is thatonly the edges of the laser spot, where the fluence is lowest,is available for desorption/ionisation. This can lead to condi-tions where, although matrix and analyte are desorbed fromthe surface, analyte molecules are not efficiently ionised. Sucheffects can be particularly significant in the conventionalpixel-by-pixel acquisition mode [24]. An innovative optionto overcome this decreased ionisation efficiency is throughthe use of laser post-ionisation combined with MALDI-MSI(so-called MALDI-2). MALDI-2 has been demonstrated toenable an up to two order of magnitude increase in sensitivityfor lipids and other molecular classes from biological tissues[25–27]. Recently, MALDI-2 has been combined with trans-mission mode MSI to enable spatial resolutions as low as600 nm under oversampling conditions [28], but it has notyet been evaluated for oversampling using conventionalfront-side MALDI laser introduction.

In this work we evaluate for the first time the combinationof oversampling MALDI-MSI employing conventional front-side laser introduction with MALDI-2 for the imaging oflipids on a high-resolution Orbitrap mass spectrometer.Exploiting the increased ionisation efficiency enabled byMALDI-2, we demonstrate the ability to generate rich lipidsignals from pixel sizes as low as 6 μm from an original laserspot size of ~ 15 μm. In addition, using an automated lipididentification workflow, we have studied the types and num-bers of lipid species that can be detected using MALDI andMALDI-2 in both conventional and oversampling imagingmodes. This provides what is to date the most comprehensiveoverview of lipid detection usingMALDI-2. The utility of thismethod for high content and high spatial resolution lipid im-aging using 6 μm pixel sizes is demonstrated using rat liver,mouse kidney and human brain tissue containing active mul-tiple sclerosis lesions, where localisation of lipid signal toindividual cellular-level features is found. This unique combi-nation of highmass accuracy, highmass resolving power, highspatial resolution and enhanced sensitivity provides an excit-ing method to study lipid metabolism at the cellular levelwithin heterogeneous and complex tissue sections.

Methods

Materials

Isopropanol (LC-MS grade), ethanol (LC-MS grade), 2,5-dihydroxybenzoic acid (DHB, ≥ 99.9% purity), 2,5-dihydroxyacetophenone (DHA) and water (LC-MS grade)were purchased from Sigma Aldrich (Zwijndrecht,The Netherlands) and used without further purification.Haematoxylin (Merck, Darmstadt, Germany) and eosin Y(J.T. Baker, Center Valley, PA, USA) were used under

Bowman A.P. et al.2278

standard laboratory protocols. Indium tin oxide (ITO)-coatedglass slides were purchased from Delta Technologies(Loveland, USA).

Biological samples

Healthy rat liver was obtained from Maastricht University inaccordance with protocols approved by the Animal Care andUse Committee (DEC number 2014-120) and was from thesame animal used in a recent study [29]. Rats were providedad libitum access to water and regular chow. One mouse kid-ney was collected from an LDL R−/− mice from a protocolapproved by the Animal Care and Use Committee (DECnumber 2014-069) at Maastricht University. Mice were pro-vided ad libitum access water and regular chow. Animals werehoused and cared for at the Central Animal Facility ofMaastricht University according to local standards. 12-μm-thick sections were prepared using a cryo-microtome (Leica,Nussloch, Germany) at – 20 °C and thaw-mounted on ITO-coated glass slides. Sections were stored at − 80 °C until anal-ysis which occurred approximately 6 months after sectioning.Human brain tissue with activemultiple sclerosis (MS) lesionswere obtained from the Netherlands Brain Bank (NBB,Amsterdam, The Netherlands). The Netherlands Brain Bankreceived permission to perform autopsies for the use of tissueand for access to medical records for research purposes fromthe Ethical Committee of the VU University Medical Center,Amsterdam, The Netherlands. The experiment protocols andmethods used for analysing activeMS lesions were conductedwith the approval of the Netherlands Brain Bank and theMedical Ethical Committee Hasselt University and carriedout according to institutional guidelines. Brain tissue wasstored at − 80 °C until sectioning; 10-μm-thick sections wereprepared using a cryo-microtome and were thaw-mounted on-to standard glass slides, transported to Maastricht Universityon dry ice and then again stored at − 80 °C until matrix appli-cation and MSI analysis. The time between sectioning andanalysis was less than 1 month.

Sample preparation

Matrix application was performed via sublimation [30].Sublimation was performed under the following conditions:40 mg of DHB dissolved in isopropanol and sublimed for4 min at 160 °C at a pressure of < 4 × 10−5 bar. Samples thatwere coated in DHA followed slightly different conditions:40 mg of DHA dissolved in acetone and sublimed for 4 minat 140 °C at a pressure of < 4 × 10−5 bar. Samples were thenrecrystallised in a lab-made apparatus containing 1 mL of0.5% ethanol in water, at 50 °C for 90 s.

Haematoxylin and eosin (H&E) staining was performed onkidney sections after MALDI imaging. Matrix-coated tissuesections were cleaned of remaining MALDI matrix by

immersion in 100% ethanol for 20 s. A standard H&E proto-col was then used (95% EtOH, 70% EtOH, H2O for 30 s each;haematoxylin for 3min; H2O, 70%EtOH, 95%EtOH each for30 s; eosin for 1 min; 95% and 100% EtOH for 30 s each;xylene for 2 min). High-resolution optical images of stainedtissues were generated using a Mirax Desk scanner(3DHistech, Budapest, Hungary). Unfixed cryosections of hu-man brain tissue were stained with 0.3% Oil Red O (ORO,Sigma) for 10 min to visualise neutral lipids (cholesterol es-ters). Counterstaining of cell nuclei was done usinghaematoxylin incubation. Analysis was carried out using aLeica DM 2000 LED microscope and ImageJ software.

CD68 immunostaining and analysis was performed onfixed cryosections of human brain tissue using the antibodyanti-CD68 (1:100, cat. #14-0688, Invitrogen) and a speciescompat ib le Alexa647 secondary ant ibody (Li feTechnologies, A21247) as described previously [31]. To labelthe myelinated areas, the immune-stained sections were sub-sequently incubated for 30 min at RT with 2 μM Bodipy®493/503 solution (ThermoFisher Scientific, D3922) diluted inPBS. Analysis was then carried out using a Nikon eclipse 80imicroscope.

Mass spectrometry instrumentation

All MSI experiments were performed on an Orbitrap Elitemass spectrometer (Thermo Fisher Scientific GmbH,Bremen, Germany) coupled to a reduced pressure ESI/MALDI ion source (Spectroglyph LLC, Kennewick, WA,USA). Further details on the ion source can be found in[32]. The 349-nm MALDI laser (Spectra Physics, MountainView, CA, USA) was operated at a repetition rate of 100 Hzand pulse energy of ~ 1.0 μJ. The laser was focussed to a spotsize of ~ 15 × 12 μm as determined by the size of ablationcraters in a matrix layer (Electronic supplementary material(ESM) Fig. S1). Laser post-ionisation (MALDI-2) was per-formed as previously described for the same experimentalsetup [26]. Briefly, post-ionisation was achieved using awavelength tuneable optical parametric oscillator laser system(Ekspla NT-230, Vilnius, Lithuania). The MALDI-2 laser wasoperated at 260 nm with a post-attenuation pulse energy of500 μJ. Using adjustable mirrors and a right-angled prism thebeamwas guided to be parallel with and ~ 250–400 μm abovethe sample surface where it intersected the desorbed plumegenerated by the MALDI laser pulse. OPO laser emissionwas synchronised with the MALDI laser using a digitalpulse/delay generator (DG645, Stanford Research Systems,Sunnyvale, USA). Emission from the OPO laser occurred 20μs after each MALDI laser pulse. The mass spectrometer wasoperated in positive-ion mode using an ion injection time of250 ms, automatic gain control (AGC) turned off and a massrange of 350–2000.

Evaluation of lipid coverage and high spatial resolution MALDI-imaging capabilities of oversampling... 2279

Data acquisition

Line scans of rat liver Rat liver tissue was used for comparisonof MALDI and MALDI-2 data generated at different stepsizes. Three sets of alternate MALDI andMALDI-2 line scanswere acquired in a single experiment (i.e. a line of MALDIdata followed by a line of MALDI-2 data) and performed at20, 15, 10, 8, 6 and 4 μm step sizes for both DHB- and DHA-coated tissues. As the first row in each acquisition does notresult in oversampling in the vertical direction, one extra rowwas acquired at the start of each acquisition but not used fordata analysis. This ensured each line scan was acquired underrepresentative oversampling conditions. Line scans were per-formed using a mass resolution setting of 240,000 (FWHM@m/z 400) giving a total scan time of 1.06 s/scan.

Imaging of mouse kidney and human brain tissue Humanbrain tissue containing active multiple sclerosis lesions andmouse kidney tissues were coated in DHB matrix andanalysed using MALDI-2-MSI at a step size of 6 μm and amass resolution of 120,000 (FWHM @ m/z 400), giving atotal scan time of 0.67 s/scan. As with the line scans, the firstrow of each dataset was removed so that all rows were ac-quired under the same oversampling conditions. To supportthe identifications of several unexpected compounds, ion trapMS/MS spectra were acquired from one human brain and onekidney tissue section using the DDA-imaging method [33].

Data analysis and lipid identification

All data image visualisation and data analysis was performedusing LipostarMSI (Molecular Horizon Srl, Bettona, Italy).Prior to import, all proprietary Thermo Fisher .raw data wasconverted into imzML [34]. This was done by first convertingraw data into mzML using msconvert (ProteoWizard) [35].Using the in-built converter of LipostarMSI, the mzmL filewas then combined with the positioning file created by theMALDI/ESI injector to generate a profile mode imzML file.Lipid identification within LipostarMSI was performed withreference to the LIPIDMAPS database (.sdf format) [36] andwas based on accurate m/z matching using a tolerance of ± 2ppm. All lipid identifications are therefore reported to thesum-composition level. Phospholipids, sphingolipids and ste-rols were considered for identification. Note that in the case ofsterols, many isomeric species are possible and we thus groupall m/z matches to a general “sterol” group, while for etherphospholipids, identifications containing isomeric acyl andalkenyl linkages were grouped into a general ether sub-group (e.g. PE–O and PC–O). In the case of MALDI analysisof liver tissue, potassiated species dominated the spectra,while MALDI-2 spectra were dominated by protonated spe-cies, consistent with prior observations comparing MALDIand MALDI-2 [25, 26]. Therefore, to avoid the occurrence

of the same lipid being detected as multiple adducts andcounting as multiple identifications, only [M+K]+ ions ofphospholipids and sphingolipids were considered forMALDI data, and only [M+H]+ ions considered forMALDI-2 data. Sterols were searched for as [M+H-H2O]

+

ions for both MALDI and MALDI-2, apart from cholesterolesters (CE) for the analysis of human brain tissue (see below).Several sub-classes unlikely to be observed in positive-ionmode data (e.g. sulfatides and cardiolipins) were removedfrom the search list and only identifications corresponding toeven-number acyl/ether chains were considered to limit falsepositives. Further sample-specific parameters are providedbelow.

Line scans of rat liver Each line scan was converted to imzMLfile consisting of 75 pixels (spectra) using a dummy .xmlposition file. This resulted in three MALDI and threeMALDI-2 datasets per step size and matrix type. While nolower limit was set for peak intensity for peak picking duringdata import, the minimum peak frequency was set to 50%(meaning that peaks had to appear in at least half of the pixelsin any given line) with a tolerance for peak alignment of 3ppm. In this way, only reproducible signals were consideredand very low abundance peaks close to the detection limit and/or corresponding to random electronic noise were discarded.All scans were recalibrated using up to three peaks, [choles-terol-H2O+H]

+, [PE(38:4)+H]+, and [PC(34:2)+K]+, duringdata import. The final ID list was then manually curated andseveral seemingly spurious identifications were removed.Single-scan noise values were taken from Xcalibur QualBrowser 2.3 (Thermo Fisher Scientific GmbH, Bremen,Germany).

Imaging of mouse kidney and human brain tissue Import ofimzML files to LipostarMSI for the kidney and brain sampleswas performed using the following parameters: intensitythreshold of 1% of base peak, peak alignment tolerance of 3ppm, peak detection frequency of 2% and a minimum spatialchaos value of 0.7 (with a value of 1 corresponding to highimage structure and a value of 0 to a random (structureless)spatial distribution). Identification followed an identical pro-cess outlined above, apart from [M+K]+ ions of CEs alsobeing considered in the brain data. These were added follow-ing manual interrogation of the raw data and the observationof several localised and abundant potassiated CE signals. Inaddition to manual curation of identified lipid species asoutlined for liver line scans above, we also only kept identifi-cations that displayed tissue-specific distributions. Several an-tioxidant species observed as radical cations from kidney tis-sue were manually annotated but not counted in the final iden-tification lists as these identifications were not automated(LipostarMSI does not yet support radical cations as a searchclass).

Bowman A.P. et al.2280

MSI data was visualised following total ion current normal-isation and applying hotspot removal (high quantile 99%).

Scanning electron microscopy

Scanning electron microscopy images of matrix-coated kid-ney tissue after DHB sublimation and recrystallisation wereacquired using a Philips XL30 microscope.

Results and discussion

Lipid coverage with oversampling coupledwith MALDI-2 from liver tissue

Data for both MALDI and MALDI-2 spectra were generatedusing stage step sizes of 20 μm (no oversampling) and 6 μm(significant oversampling) from line scans of liver tissue tocompare the number and types of lipid species detected. Thespot size of the MALDI laser on the tissue was ~ 15 μm ×11 μm (ESM Fig. S1). As expected, at smaller step sizes thetotal ion current and overall signal-to-noise (S/N) decrease forboth MALDI and MALDI-2 data (ESM Fig. S2). When usinga 4-μm step size, lipid signals were low and unstable; thus,4-μm data was not explored further. Figure 1 a shows repre-sentative MALDI spectra from rat liver tissue coated withDHB matrix using a step size of 20 μm (10 consecutive scansaveraged with single-scan S/N values indicated, m/z 600–1000 shown). Full mass range spectra can be found in theESM Figs. S3 and S4. As typically observed, MALDI spectraare dominated by phosphat idylchol ine (PC) andsphingomyelin (SM) lipids, with their potassiated adducts be-ing most abundant. By contrast, MALDI-2 spectra generatedunder analogous conditions (Fig. 1a, bottom) resulted in botha ca. 200-fold increase in base peak intensity and dramaticincreases in signal for a variety of lipid species, while lipid-related signals also became dominated by [M+H]+ ions, con-sistent with prior observations [25, 26]. The benefit ofMALDI-2 is observed to hold even under oversampling con-ditions. Figure 1 b (top) shows representative MALDI spectraacquired using a 6-μm step size that results in significantoversampling. While the overall spectrum resembles thatshown in the top panel of Fig. 1 a, the signal intensity isreduced by 10-fold with many peaks no longer being detected.However, data acquired using a 6-μm step size and MALDI-2still produces high S/N spectra with base peak intensity ~ 3-fold higher than those generated by conventional MALDI at20 μm (Fig. 1b, bottom). Using a 6-μm step size, the S/N ofthe base peak corresponding to the phosphatidylethanolamine(PE), [PE(36:4)+H]+, is 136 using MALDI-2, whereas usingthe same step size, the corresponding lipid detected with con-ventional MALDI, observed as the [M+K]+ ion, has a single-scan S/N of < 1. For the abundant PC(34:2) lipid, the signal

intensity for the protonated species acquired at 6 μm step sizewas comparable to that measured for the abundant potassiatedspecies with MALDI at 20 μm step size. It is noteworthy forthese data acquired at 6μm step size that individual spectra aretaken from a tissue area equivalent to or smaller than the sizeof a typical mammalian cell (10–100 μm), thereby providingdata that is reflective of cellular-level lipid compositions with-in the tissue, although we acknowledge that for smaller cellsthere is a reasonable probability of collecting ion signal fromtwo adjacent cells in a single pixel.

MALDI-2 oversampling was also evaluated using 2,5DHA matrix (ESM Fig. S5). Using a 20-μm step size,MALDI and MALDI-2 spectra were similar to those obtainedusing DHB. Interestingly though, with an increasing extent ofoversampling, many of the MALDI-2-specific signals ob-served with DHB yielded lower relative intensities usingDHA. For example, at 6 μm step sizes using MALDI-2, pro-tonated PC signals yielded the highest signal intensities (ratherthan PE lipids), while a relative increase in the abundance ofpotassiated PC lipids compared to their protonated forms wasalso observed. In contrast, the overall spectral profile at dif-ferent step sizes remained relatively consistent using DHB(Fig. 1a, b). The origin of this effect is unclear, but appearsto suggest a shift of DHA ionisation properties underoversampling conditions when usingMALDI-2.We speculatethis effect is related to (i) the higher volatility of DHA,resulting in a larger desorption area upon irradiation by theMALDI laser. Upon oversampling, this could result in matrixfurther from the centre of the laser spot being desorbed leadingto a plume of lower density in which MALDI-2 can occur;and/or (ii) possible differences in matrix morphology and dif-fusion rates of lipids from the tissue with lipids closer to thesurface being more selectively desorbed by the edge of thelaser beam. Different diffusion rates of lipids have recentlybeen reported in sublimed matrix coatings using secondaryion mass spectrometry [37]. Thorough investigation of thiseffect lies outside the scope of this work, but provides anexciting avenue to pursue that could provide insight into theMALDI-2 mechanism.

We next evaluated the breadth of lipid coverage that can beobtained in such high spatial resolution MSI experimentsusing DHB. Peaks used for automated identification were de-fined during data import as having a peak detection frequency≥ 50% (equivalent to being detected in half or more scans/pixels within a 3-ppm tolerance) across each 75 pixel linescan. Figure 1 c shows the number of lipid species detectedwithin 2 ppm of their theoretical m/z for both MALDI andMALDI-2 at step sizes of 20 μm (no oversampling), 10 μm(moderate oversampl ing) and 6 μm (signif icantoversampling). The full list of tentatively identified speciesand their abundances is provided in ESM Table S1.ConventionalMALDI detected primarily PC lipids along withseveral abundant SM and PE lipids as [M+K]+ ions. In total,

Evaluation of lipid coverage and high spatial resolution MALDI-imaging capabilities of oversampling... 2281

24, 19 and 11 unique sum-composition lipid species weredetected with MALDI at step sizes of 20, 10 and 6 μm, re-spectively. Dramatically more lipid species were detectedusing MALDI-2, with 149, 117 and 66 lipid species beingdetected as [M+H]+ ions ([M+H-H2O]

+ for sterols) at 20, 10and 6 μm step sizes, respectively. While PE and PE–O specieswere amongst the biggest beneficiaries of MALDI-2, it alsoenabled the detection of a diverse array of lipid classes notobserved with conventional MALDI. For example, MALDI-2could detect both phosphat idyl inosi to l (PI) andphosphatidylserine (PS) lipid species as [M+H]+ ions. These

lipids are often only observable in negative-ion modeMALDIanalysis thereby demonstrating the added lipid coverage of-fered by MALDI-2. Supporting these observations is the factthat the most intense protonated species observed for each,PI(38:4) and PS(36:1), have been shown to be the two mostabundant species of each class in liver using LC-MS/MS [38].Despite the expected drop in the number of detected lipidswith decreasing pixel size, the above data demonstrates thatrich lipid MSI data covering many different lipid species canbe generated from pixel sizes as low as 6 μm, with a ~ 3-fold

Fig. 1 MALDI (top, red trace) andMALDI-2 (bottom, blue trace) spectraacquired from rat liver tissue coated in DHB matrix using stage step sizesof a 20 μm (no oversampling) and b 6 μm (with oversampling). Eachspectrum is the average of 10 consecutive scans. c The number of detect-ed lipid species across different lipid classes observed using MALDI andMALDI-2 using 20, 10 and 6 μm line scans across liver tissue. Peaksused for identification had a detection frequency ≥ 50% across the line

scans consisting of 75 pixels (equivalent to being detected in half or moreindividual scans). [M+K]+ ions were considered for MALDI identifica-tion and [M+H]+ ions considered for MALDI-2 identifications, with theexception for sterols which were identified in both cases as [M+H-H2O]

+

ions. Error bars represent ± 1 standard deviation across three replicate linescans. b.p. = base peak

Bowman A.P. et al.2282

increase in the number of detected lipid species compared toconventional MALDI using a 20-μm pixel size.

High spatial resolution imaging of lipids in kidneytissue

Mouse kidney tissue was used to evaluate the high-resolutionimaging capabilities of MALDI-2 oversampling. A ~ 4-mm2

area of DHB-coated tissue section was analysed using a 6-μmstep resulting in an image consisting of 108,558 pixels. Usingscanning electron microscopy, the DHB crystal sizes werefound to be significantly smaller than 6 μm (ESM Fig. S6).The average spectrum is shown in Fig. 2 a and demonstratesthe detection of a variety of lipid species across the m/z 350–2000mass range. Using the described approach for automatedlipid identification (see “Methods”), 74 unique sum-composition lipid species were identified as [M+H]+ ions([M+H-H2O]

+ ions for sterols) from this dataset. The distribu-tion of these identifications across the different lipid sub-classes is shown in Fig. 2 b. As observed for the liver dataabove, PC, PE and PE–O lipids constituted the majority ofdetected lipid species, and in total, 12 different lipid sub-classes could be identified and imaged. The full list of identi-fied lipids is provided in ESM Table S2.

Figure 3 shows an optical image of the post-MSI H&E-stained tissue section (Fig. 3a) and an overlay of three identi-fied lipid species (Fig. 3b, [PC(38:6)+H]+ (green),[PE(O-40:8)+H]+ (blue) and [PE(O-36:5)+H]+ (pink)).Using the distributions of these three ion signals, a numberof different tissue regions can be localised at the cellular level.[PE(O-36:5)+H]+ is localised to the inner medulla, innerstripe, glomeruli and the interstitium regions of the kidney.Both [PE(O-40:8)+H]+ and [PC(38:6)+H]+ are specific tothe kidney tubuli, with [PC(O-40:8+H]+ being more abundantin tubuli contained within the outer stripe of the medulla and[PC(38:6)+H]+ being more abundant in tubuli contained with-in the cortex. The high specificity for these lipid signals forhistologically different tissue regions can be seen in thezoomedMSI and H&E data shown in Fig. 3 c–f. The outlinedarea of Fig. 3 d (white dotted line) highlights tubuli-specificlipid signals corresponding to the tubular regions outlined inthe H&E-stained tissue with excellent spatial specificity (Fig.3f). This strong agreement between the MSI and histologicaldata confirms the high spatial resolution enabled by both theoversampling method and that the employed sample prepara-tion using sublimation minimises analyte delocalisation.Combined, this data demonstrates the ability to achieve pixel

Fig. 2 a Average spectrumacquired from mouse kidneytissue usingMALDI-2 and a pixelsize of 6 μm betweenm/z 350 and1000. b Number of automaticallyidentified lipid species frommouse kidney tissue. Lipids wereidentified as [M+H]+ ions ([M+H-H2O]

+ for sterols) using an m/ztolerance of 2 ppm

Evaluation of lipid coverage and high spatial resolution MALDI-imaging capabilities of oversampling... 2283

sizes on the scale of cellular-level features within tissues whilestill being able to detect and identify numerous lipid species.

In addition to the [M+H]+ and [M+H-H2O]+ ions automat-

ically annotated, we also observed a variety of radical cationscorresponding to different lipid-soluble antioxidants withinthe mouse kidney tissue. These species are generated by directabsorption of the 260-nm MALDI-2 laser light via a [1 + 1]resonance enhancedmultiphoton ionisation (REMPI) process.In particular, we observed the radical cation of vitamin E atm/z 430.3808 which exhibited elevated signal within the innermedulla (Fig. 4, red). We also observed both the oxidised andreduced forms of coenzyme Q9 (m/z 792.6059 and m/z794.6211) and coenzyme Q10 (m/z 860.6689 and m/z

862.6848) as shown in the Fig. 4 spectrum. These constitutedabundant signals with the reduced coenzyme Q9 correspond-ing to the base peak in the spectrum when using a 6-μm stepsize (Fig. 2a). MS/MS spectra supporting the identification ofthese radical species are provided in ESM Figs. S7 and S8.The high signals for these species can possibly be explainedby a higher efficiency REMPI process that is decoupled fromthe MALDI-based ionisation processes. All four of these ionsignals exhibited similar distributions and were observedthroughout most of the tissue with the distribution of the re-duced form of coenzyme Q9 shown in blue in Fig. 4. Whilethe protonated forms of oxidised and reduced coenzyme Q9and Q10 have been detected using MALDI-MSI from brain

Fig. 3 a Optical image of thepost-MSI H&E-stained tissuesection. b Ion distribution imagesof [PC(38:6)+H]+ (green),[PE(O-40:8)]+ (blue) and[PE(O-36:5)+H]+ (pink)throughout mouse kidney tissueacquired using MALDI-2 and apixel size of 6 μm. c, d Selectedenlarged regions of the MSI data.The corresponding H&E imagesof these enlarged regions areshown in e and f. All MSI data isvisualised using total ion currentnormalisation and hotspotremoval (99% quantile)

Fig. 4 Ion distribution images ofm/z 794.6211 ([coenzyme Q9]+•,blue) and m/z 430.3808 ([vitaminE]+•, red) throughout mousekidney tissue. The correspondingMALDI-2 mass spectrumshowing the detection of both theoxidised and reduced forms ofcoenzyme Q9 and coenzyme Q10is shown below

Bowman A.P. et al.2284

tissue [39], to our knowledge, this is the first report of theirdetection in renal tissue with MSI.

Lipid imaging of human brain tissue containing activemultiple sclerosis lesions

Finally, the lipid imaging capabilities of MALDI-2 coupledwith oversampling were evaluated using active human multi-ple sclerosis tissue where the high spatial resolution enabledspecific lipid accumulations to be visualised within the tissuelesions. During multiple sclerosis, an autoimmune response isdirected against the lipid-rich myelin sheath surroundingaxons. Myelin is broken down and cleared by phagocyteswhich causes failure of axonal conduction, and dependingon the affected region, disease symptoms such as impairedmuscle control, balance, vision and speech.Myelin processingwithin the phagocytes leads to the release of lipid mediatorsthat direct the function of the phagocytes and thereby lesionprogression and resolution [40–42]. More generally, lipidshave been identified to play important roles in multiple scle-rosis [43–46], but the precise processes and mechanism bywhich this occurs, along with the specific functions of indi-vidual lipid molecules, remain unknown.

Within the imaging dataset acquired from an ~ 4.1-mm2

area of tissue and consisting of 114,263 pixels, 147 uniquesum-composition lipid species were identified (see “Methods”for identification details). Each of these lipid species revealedtissue-specific distributions throughout the brain tissue. Themean spectrum from this dataset is provided in Fig. 5 a, whilethe distribution of the lipids across the different classes isshown in Fig. 5 b. Again PC and PE lipids contribute thelargest fraction of identified species. Other species with sig-nificant contributions to the identified list were sterols, PS andhexosylceramides (HexCer) species. The detection of a num-ber of glycosphingolipids is consistent with their known highabundance and diversity in brain tissue [47]. The full list ofidentified species is provided in ESM Table S3.

MSI of the human brain tissue using a 6-μm pixel sizeyielded both high contrast and spatially specific signals formany lipid species, thereby allowing visualisation of finestructural features only 10–20 μm in size. Histopathologicalanalysis was performed on a multiple sclerosis lesion acquiredfrom the same patient revealing this lesion to contain abundantmacrophages and microglia (CD68 staining, Fig. 6a). Thesemacrophages show a foamy appearance and are filled withesterified cholesterol and other neutral lipids as demonstratedwith a Bodipy (Fig. 6a) and ORO (Fig. 6b) staining. The highsingle pixel spectral quality and rich lipid signals achievedfrom only 6 μm pixels is demonstrated in Fig. 6 f-i that showthe corresponding spectra obtained from the individual pixelsindicated by the white arrows in Fig. 6 c. The overlaid distri-butions of three lipid ions, [Chol+H-H2O]+ (pink),[HexCer(d36:2)+H]+ (blue) and the cholesterol ester (CE)

[CE(16:0)+K]+ (green), are shown in Fig. 6 c and clearlycorrelate with distinct tissue regions. Two enlarged regionsof the MSI data are shown in Fig. 6 d and e to further highlightthese specific distributions. Analogous distributions were alsoobserved in brain tissue acquired from a second multiple scle-rosis patient (ESM Fig. S9). Signal for [HexCer(d36:2)]+ waslocalised to the myelin surrounding the multiple sclerosis le-sion and was virtually absent within the lesion. Signal for theCE species [CE(18:1)+K]+ is localised in areas only 10–20μm wide and corresponded to the lipid accumulations withinthe phagocytes as shown in the ORO staining (Fig. 6a, b). Asimilar distribution is also observed for the potassium adductof CE(16:0) (ESM Fig. S10a). Interestingly, the signal for[Chol+H-H2O]

+, likely representing both free cholesteroland in-source fragmentation of CEs, localised to a wider re-gion adjacent to the centres of the phagocyte lipid accumula-tions. Also, abundant higher m/z signals observed at m/z1302.2069 and m/z 1276.1906 assigned based on accuratemass to the protonated CE dimer species ([2CE(18:1)+H]+

(− 0.3 ppm error) and [CE(18:1)+CE(16:0)+H]+ (− 1.0 ppmerror) yielded similar distributions as observed for [Chol+H-H2O]

+ (ESM Fig. S10b and S10c). Evidence for the formationof [2M+H]+ ions of CE was obtained upon analysing aCE(18:0) standard using MALDI-2 that yielded an analogous[2CE(18:0)+H]+ dimer (ESM Fig. S11). Despite the CE-related signals in the regions around the phagocytes, there islittle signal in the ORO staining within these regions. OROstaining is widely thought to be specific for neutral lipid drop-lets (e.g. those containing CE and triacylglyceride lipids)and does not strongly stain other cellular/tissue regions out-side these regions that contain other lipid species. We there-fore speculate that the lack of ORO staining in the regionsadjacent to the phagocytes is caused by the presence of otherlipid species. For example, two PC–O species, ([PC(O-40:7)+H]+ and [PC(O-40:6)+H]+, were also found specifically in theregions adjacent to the phagocytes (ESM Fig. S10). We alsonote that the mechanism leading to alkali adducted CE speciesbeing observed in the phagocytes and protonated-related sig-n a l s s u r r o u n d i n g t h e s e r e g i o n s i s u n known .Nonetheless, these results suggest specific functions of bothfree cholesterol and CEs in multiple sclerosis and, in particu-lar, phagocyte metabolism. Accumulation of cholesterol inmacrophages activates the nuclear liver-X-receptors that mod-ulate cellular lipid metabolism and the production of inflam-matory mediators [41]. Interestingly, aging was shown to pro-mote cholesterol accumulation in phagocytes which stimu-lates the inflammatory activity of these immune cells and im-pairs their reparative properties [48, 49]. More generally, thelocalisation of certain lipid species to phagocytes within thebrain tissue confirms the high, cellular-level, spatial resolutionthat can be achieved with this method, while still obtainingrich lipid spectra. This broad lipid coverage and high spatialresolution enabled MALDI-2 coupled with oversampling can

Evaluation of lipid coverage and high spatial resolution MALDI-imaging capabilities of oversampling... 2285

provide a new tool to study in-depth the lipid-specific alter-ations of different cell populations within multiple sclerosisprogression as well as other neurodegenerative diseases.

Conclusions

In this work, we have demonstrated that MALDI-2 combinedwith an oversampling acquisition approach is capable of bothgenerating rich lipid spectra from tissue and imaging lipids atpixel sizes as low as 6 μm, without modifications to the opticsor stage of the commercially available ion source. MALDI-2data generated from 6 μm pixels was capable of detectingthree times more lipid species than conventional MALDI ac-quired without oversampling at 20 μm pixel size from rat livertissue. From kidney and brain tissue imaging experiments, wecould identify 74 and 147 unique lipid species, in addition tovisualising their spatial distributions within the tissue.Importantly, lipid spectra obtained from individual pixels rep-resent an area equivalent to or less than the size of a typicalmammalian cell. This is an important capability that willgreatly enhance the ability to interpret lipid MSI data in terms

of cellular-level lipid metabolism occurring within biologicaltissues, while preserving the context of the cell within thetissue microenvironment. For example, data acquired frommouse kidney clearly enables the localisation of lipid signalsto individual tubuli, while data from human multiple sclerosistissue enables lipid accumulations within lesion-specific mac-rophages to be visualised. The combination of information-rich spectra and cellular-level spatial resolutions provides apowerful approach to study spatial and cell-type-specific al-terations in lipid metabolism within many different diseasetypes.

Fig. 5 a Average spectrumacquired from human multiplesclerosis brain tissue usingMALDI-2 and a pixel size of6 μm between m/z 350 and 2000.b Number of automaticallyidentified lipid species fromhuman multiple sclerosis braintissue. Lipids were identified as[M+H]+ ions ([M+H-H2O]

+ forsterols and [M+K]+ forcholesterol esters) using an m/ztolerance of 2 ppm

Fig. 6 a CD68 (macrophages, purple) and Bodipy (myelin/neutral lipids,green) immunostaining and b Oil Red O staining of human brain tissueslices acquired from the same patient used to collect the MSI data shownin c–e with several tissue regions indicated. c Ion distributions images ofm/z 689.5636 ([CE(18:1)+K]+, green), m/z 726.5882 ([HexCer(d36:2)+H]+, blue) and m/z 369.3517 ([Chol+H-H2O]

+, pink) acquired usingMALDI-2 and a pixel size of 6 μm from human multiple sclerosis braintissue. d, e Selected enlarged regions of the MSI data shown in c. f–iSingle pixel spectra acquired from the regions indicated by the whitearrows in c. All MSI data is visualised using the total ion current normal-isation and hotspot removal (99% quantile). NAWM = normal appearingwhite matter

b

Bowman A.P. et al.2286

Evaluation of lipid coverage and high spatial resolution MALDI-imaging capabilities of oversampling... 2287

From a technology standpoint, this work provides the mostcomprehensive overview of the lipid detection capabilities ofMALDI-2 reported to date, demonstrating the detection ofmany more lipid classes than possible with positive-ion modeMALDI. Moreover, the coupling of this approach with a newautomated lipid identification workflow utilising accurate m/zmeasurements enables the rich lipidomics information ac-quired in such experiments to be readily exploited.

Acknowledgements We are grateful to both Dr. Carine Peutz-Kootstraand Dr. Marion Gijbels (Maastricht University Medical Centre) for theirhelp with kidney histology and both Dr. Sara Tortorella (MolecularHorizon Srl, Italy) and Paolo Tiberi (Molecular Discovery Ltd, UK) fortheir help and support with LipostarMSI software.

Funding information This work has been made possible with the finan-cial support of the Dutch province of Limburg through the LINK pro-gram. A.P.B, R.M.A.H and S.R.E are grateful for funding from Interreg VEMR and the Netherlands Ministry of Economic Affairs within the“EURLIPIDS” project (EMR23). The work has also been supported bythe Flemish Fund for Scientific Research (FWO Vlaanderen) and specialresearch fund UHasselt (BOF).

Compliance with ethical standards

The experiment protocols and methods used for analysing active MSlesions were conducted with the approval of the Netherlands BrainBank and the Medical Ethical Committee of Hasselt University and car-ried out according to institutional guidelines.The experiment protocolsand methods used for analysing rat liver tissues and mouse kidney tissueswere conducted with the approval of the Animal Careand use CommitteeofMaastricht University. Animals were housed and cared for according tolocal standards.

Conflict of interest Mikhail Belov is the general manager ofSpectroglyph LLC, the supplier of the dual-funnel MALDI/ESIInjector™ interface. The remaining authors declare no conflict of interest.

Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing,adaptation, distribution and reproduction in any medium or format, aslong as you give appropriate credit to the original author(s) and thesource, provide a link to the Creative Commons licence, and indicate ifchanges weremade. The images or other third party material in this articleare included in the article's Creative Commons licence, unless indicatedotherwise in a credit line to the material. If material is not included in thearticle's Creative Commons licence and your intended use is notpermitted by statutory regulation or exceeds the permitted use, you willneed to obtain permission directly from the copyright holder. To view acopy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

References

1. Bowman AP, Heeren RMA, Ellis SR. Advances in mass spectrom-etry imaging enabling observation of localised lipid biochemistrywithin tissues. TrAC, Trends Anal Chem:in press,. 2018. https://doi.org/10.1016/j.trac.2018.1007.1012.

2. Gode D, Volmer DA. Lipid imaging by mass spectrometry—a re-view. Analyst. 2013;138(5):1289–315.

3. Murphy RC, Hankin JA, Barkley RM. Imaging of lipid species byMALDI mass spectrometry. J Lipid Res. 2009;50:S217–322.

4. Guo S, Wang Y, Zhou D, Li Z. Significantly increased monounsat-urated lipids relative to polyunsaturated lipids in six types of cancermicroenvironment are observed by mass spectrometry imaging. SciRep. 2014;4:5959.

5. Paine MRL, Liu J, Huang D, Ellis SR, Trede D, Kobarg JH, et al.Three-dimensional mass spectrometry imaging identifies lipidmarkers of medulloblastoma metastasis. Sci Rep. 2019;9(1):2205.

6. Wang X, Han J, Hardie DB, Yang J, Pan J, Borchers CH.Metabolomic profiling of prostate cancer by matrix assisted laserdesorption/ionization-Fourier transform ion cyclotron resonancemass spectrometry imaging using matrix coating assisted by anelectric field (MCAEF). Biochimica et Biophysica Acta (BBA) -Proteins and Proteomics. 2017;1865(7):755–67.

7. Scott AJ, Post JM, Lerner R, Ellis SR, Lieberman J, Shirey KA,et al. Host-based lipid inflammation drives pathogenesis inFrancisella infection. Proc Natl Acad Sci. 2017;114(47):12596–601.

8. Blanc L, Lenaerts A, Dartois V, Prideaux B. Visualization of my-cobacterial biomarkers and tuberculosis drugs in infected tissue byMALDI-MS imaging. Anal Chem. 2018;90(10):6275–82.

9. Ščupáková K, Soons Z, Ertaylan G, Pierzchalski KA, Eijkel GB,Ellis SR, et al. Spatial systems lipidomics reveals nonalcoholic fattyliver disease heterogeneity. Anal Chem. 2018;90(8):5130–8.

10. Hall Z, Bond NJ, Ashmore T, Sanders F, Ament Z, Wang X, et al.Lipid zonation and phospholipid remodeling in nonalcoholic fattyliver disease. Hepatology (Baltimore, Md). 2017;65(4):1165–80.

11. Kettling H, Vens-Cappell S, Soltwisch J, Pirkl A, Haier J, MüthingJ, et al. MALDI mass spectrometry imaging of bioactive lipids inmouse brain with a Synapt G2-S mass spectrometer operated atelevated pressure: improving the analytical sensitivity and the lat-eral resolution to ten micrometers. Anal Chem. 2014;86(15):7798–805.

12. Kompauer M, Heiles S, Spengler B. Atmospheric pressure MALDImass spectrometry imaging of tissues and cells at 1.4-mm lateralresolution. Nat Methods. 2017;14(1):90–6.

13. Feenstra AD, Dueñas ME, Lee YJ. Five micron high resolutionMALDI mass spectrometry imaging with simple, interchangeable,multi-resolution optical system. J Am Soc Mass Spectrom.2017;28(3):434–42.

14. Korte AR, Yandeau-Nelson MD, Nikolau BJ, Lee YJ. Subcellular-level resolution MALDI-MS imaging of maize leaf metabolites byMALDI-linear ion trap-Orbitrap mass spectrometer. Anal BioanalChem. 2015;407(8):2301–9.

15. Zavalin A, Todd EM, Rawhouser PD, Yang J, Norris JL, CaprioliRM. Direct imaging of single cells and tissue at sub-cellular spatialresolution using transmission geometry MALDI MS. J MassSpectrom. 2012;47(11):1473–81.

16. Kompauer M, Heiles S, Spengler B. Autofocusing MALDI massspectrometry imaging of tissue sections and 3D chemical topogra-phy of nonflat surfaces. Nat Methods. 2017;14:1156.

17. Jurchen J, Rubakhin S, Sweedler J. MALDI-MS imaging of fea-tures smaller than the size of the laser beam. J Am Soc MassSpectrom. 2005;16(10):1654–9.

18. Snel MF, Fuller M. High-spatial resolution matrix-assisted laserdesorption ionization imaging analysis of glucosylceramide inspleen sections from a mouse model of Gaucher disease. AnalChem. 2010;82(9):3664–70.

19. Sparvero LJ, Amoscato AA, Dixon CE, Long JB, Kochanek PM,Pitt BR, et al. Mapping of phospholipids by MALDI imaging(MALDI-MSI): realities and expectations. Chem Phys Lipids.2012;165(5):545–62.

20. Garate J, Fernández R, Lage S, Bestard-Escalas J, Lopez DH,Reigada R, et al. Imaging mass spectrometry increased resolutionusing 2-mercaptobenzothiazole and 2,5-diaminonaphtalene

Bowman A.P. et al.2288

matrices: application to lipid distribution in human colon. AnalBioanal Chem. 2015;407(16):4697–708.

21. Trim P, Djidja M-C, Atkinson S, Oakes K, Cole L, Anderson DG,et al. Introduction of a 20 kHz Nd:YVO4 laser into a hybrid quad-rupole time-of-flight mass spectrometer for MALDI-MS imaging.Anal Bioanal Chem. 2010;397(8):3409–19.

22. Spraggins J, Caprioli R. High-speed MALDI-TOF imaging massspectrometry: rapid ion image acquisition and considerations fornext generation instrumentation. J Am Soc Mass Spectrom.2011;22(6):1022–31.

23. Nazari M, Muddiman DC. Cellular-level mass spectrometry imag-ing using infrared matrix-assisted laser desorption electrospray ion-ization (IR-MALDESI) by oversampling. Anal Bioanal Chem.2015;407(8):2265–71.

24. Wiegelmann M, Dreisewerd K, Soltwisch J. Influence of the laserspot size, focal beam profile, and tissue type on the lipid signalsobtained byMALDI-MS imaging in oversamplingmode. J Am SocMass Spectrom. 2016;27(12):1952–64.

25. Soltwisch J, Kettling H, Vens-Cappell S, Wiegelmann M, MüthingJ, Dreisewerd K. Mass spectrometry imaging with laser-inducedpostionization. Science. 2015;348(6231):211–5.

26. Ellis S, Soltwisch J, Paine MRL, Dreisewerd K, Heeren R. Laserpost-ionisation combined with a high resolving Orbitrap mass spec-trometer for enhanced MALDI-MS imaging of lipids. ChemCommun. 2017;53:7246–9.

27. Barré FPY, Paine MRL, Flinders B, Trevitt AJ, Kelly PD, Ait-Belkacem R, et al. Enhanced sensitivity using MALDI imagingcoupled with laser postionization (MALDI-2) for pharmaceuticalresearch. Anal Chem. 2019;91(16):10840–8.

28. Niehaus M, Soltwisch J, Belov ME, Dreisewerd K. Transmission-mode MALDI-2 mass spectrometry imaging of cells and tissues atsubcellular resolution. Nat Methods. 2019;16(9):925–31.

29. Flinders B, Huizing LRS, van Heerden M, Cuyckens F, NeumannUP, van der Laan LJW, et al. Cross-species molecular imaging ofbile salts and lipids in liver: identification of molecular structuralmarkers in health and disease. Anal Chem. 2018;90(20):11835–46.

30. Hankin JA, Barkley RM, Murphy RC. Sublimation as a method ofmatrix application for mass spectrometric imaging. J Am Soc MassSpectrom. 2007;18(9):1646–52.

31. Bogie JFJ, Mailleux J, Wouters E, Jorissen W, Grajchen E, VanmolJ, et al. Scavenger receptor collectin placenta 1 is a novel receptorinvolved in the uptake of myelin by phagocytes. Sci Rep. 2017;7:44794.

32. Belov ME, Ellis SR, Dilillo M, Paine MRL, Danielson WF,Anderson GA, et al. Design and performance of a novel interfacefor combined matrix-assisted laser desorption ionization at elevatedpressure and electrospray ionization with Orbitrap mass spectrom-etry. Anal Chem. 2017;89(14):7493–501.

33. Ellis SR, Paine MRL, Eijkel GB, Pauling JK, Husen P, JervelundMW, et al. Automated, parallel mass spectrometry imaging andstructural identification of lipids. Nat Methods. 2018;15:515–8.

34. Schramm T, Hester Z, Klinkert I, Both J-P, Heeren RMA, BrunelleA, et al. imzML—a common data format for the flexible exchangeand processing of mass spectrometry imaging data. J Proteome.2012;75(16):5106–10.

35. Chambers MC, Maclean B, Burke R, Amodei D, Ruderman DL,Neumann S, et al. A cross-platform toolkit for mass spectrometryand proteomics. Nat Biotechnol. 2012;30:918.

36. Sud M, Fahy E, Cotter D, Brown A, Dennis EA, Glass CK, et al.LMSD: LIPID MAPS structure database. Nucleic Acids Res.2007;35(Database issue):D527–32.

37. Van Nuffel S, Elie N, Yang E, Nouet J, Touboul D, Chaurand P,et al. Insights into the MALDI process after matrix deposition bysublimation using 3D ToF-SIMS imaging. Anal Chem. 2018;90(3):1907–14.

38. Ruiz-Jiménez J, Priego-Capote F,Mata-Granados JM, Quesada JM,Luque de Castro MD. Determination of the ubiquinol-10 andubiquinone-10 (coenzyme Q10) in human serum by liquid chroma-tography tandemmass spectrometry to evaluate the oxidative stress.J Chromatogr A. 2007;1175(2):242–8.

39. Tatsuta Y, Kasai K, Maruyama C, Hamano Y, Matsuo K, Katano H,et al. Imaging mass spectrometry analysis of ubiquinol localizationin the mouse brain following short-term administration. Sci Rep.2017;7(1):12990.

40. Grajchen E, Hendriks JJA, Bogie JFJ. The physiology of foamyphagocytes in multiple sclerosis. Acta Neuropathol Commun.2018;6(1):124.

41. Mailleux J, Vanmierlo T, Bogie JFJ, Wouters E, Lütjohann D,Hendriks JJA, et al. Active liver X receptor signaling in phagocytesin multiple sclerosis lesions. Mult Scler J. 2017;24(3):279–89.

42. Bogie JFJ, JorissenW,Mailleux J, Nijland PG, Zelcer N, VanmierloT, et al. Myelin alters the inflammatory phenotype of macrophagesby activating PPARs. Acta Neuropathol Commun. 2013;1(1):43.

43. Halmer R,Walter S, Faßbender K. Sphingolipids: important playersin multiple sclerosis. Cell Physiol Biochem. 2014;34(1):111–8.

44. Marbois BN, Faull KF, Fluharty AL, Raval-Fernandes S, RomeLH. Analysis of sulfatide from rat cerebellum andmultiple sclerosiswhite matter by negative ion electrospray mass spectrometry.Biochim Biophys Acta Mol Cell Biol Lipids. 2000;1484(1):59–70.

45. Kanter JL, Narayana S, Ho PP, Catz I, Warren KG, Sobel RA, et al.Lipid microarrays identify key mediators of autoimmune brain in-flammation. Nat Med. 2006;12(1):138–43.

46. Wheeler D, Bandaru VVR, Calabresi PA, Nath A, Haughey NJ. Adefect of sphingolipid metabolism modifies the properties of nor-mal appearing white matter in multiple sclerosis. Brain.2008;131(11):3092–102.

47. O'Brien JS, Rouser G. The fatty acid composition of brainsphingolipids: sphingomyelin, ceramide, cerebroside, and cerebro-side sulfate. J Lipid Res. 1964;5(3):339–42.

48. Cantuti-Castelvetri L, Fitzner D, Bosch-Queralt M, Weil M-T, SuM, Sen P, et al. Defective cholesterol clearance limits remyelinationin the aged central nervous system. Science. 2018;359(6376):684–8.

49. Bogie JFJ, Timmermans S, Huynh-Thu VA, Irrthum A, SmeetsHJM, Gustafsson J-Å, et al. Myelin-derived lipids modulate mac-rophage activity by liver X receptor activation. PLoS One.2012;7(9):e44998.

Publisher’s note Springer Nature remains neutral with regard to jurisdic-tional claims in published maps and institutional affiliations.

Evaluation of lipid coverage and high spatial resolution MALDI-imaging capabilities of oversampling... 2289