Sample preparation of bone tissue for MALDI-MSI for forensic and (pre)clinical applications

In the past decades, matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) has been applied to a broad range of biological samples, e.g., forensics and preclinical samples. The use of MALDI-MSI for the analysis of bone tissue has been limited due to the insulating properties of the material but more importantly the absence of a proper sample preparation protocol for undecalcified bone tissue. Undecalcified sections are preferred to retain sample integrity as much as possible or to study the tissue-bone bio interface in particular. Here, we optimized the sample preparation protocol of undecalcified bone samples, aimed at both targeted and untargeted applications for forensic and preclinical applications, respectively. Different concentrations of gelatin and carboxymethyl cellulose (CMC) were tested as embedding materials. The composition of 20% gelatin and 7.5% CMC showed to support the tissue best while sectioning. Bone tissue has to be sectioned with a tungsten carbide knife in a longitudinal fashion, while the sections need to be supported with double-sided tapes to maintain the morphology of the tissue. The developed sectioning method was shown to be applicable on rat and mouse as well as human bone samples. Targeted (methadone and EDDP) as well as untargeted (unknown lipids) detection was demonstrated. DHB proved to be the most suitable matrix for the detection of methadone and EDDP in positive ion mode. The limit of detection (LOD) is estimated to approximately 50 pg/spot on bone tissue. The protocol was successfully applied to detect the presence of methadone and EDDP in a dosed rat femur and a dosed human clavicle. The best matrices for the untargeted detection of unknown lipids in mouse hind legs in positive ion mode were CHCA and DHB based on the number of tissue-specific peaks and signal-to-noise ratios. The developed and optimized sample preparation method, applicable on animal and human bones, opens the door for future forensic and (pre)clinical investigations. Electronic supplementary material The online version of this article (10.1007/s00216-020-02920-1) contains supplementary material, which is available to authorized users.


Index
: Overview of MALDI-MSI and similar techniques studies performed on bone tissue. Table S2: Comparison of CHCA, DAN, DHA, DHB, and norharmane in positive ion mode using mice bone for the untargeted application.    Information S1: Untargeted application for lipids in negative ion mode using mice hind legs. Table S4: Comparison of DAN, NEDC, and norharmane in negative ion mode using mice bone for the untargeted application.     Information is provided about the sample, sample preparation   (pretreatment, embedding material, sectioning specifications, and washing), MSI specification, and results Abbreviations: APS-MALDI-MSI = atmospheric pressure scanning microprobe matrix-assisted laser desorption/ionization mass spectrometry imaging, CMC = carboxymethyl cellulose, IR-MALDESI-MSI = infrared matrix assisted laser desorption electrospray ionization mass spectrometry imaging, MALDI-MSI = matrix-assisted laser desorption/ionization mass spectrometry imaging, MSI = mass spectrometry imaging, PFA = paraformaldehyde, SIMS = secondary ion mass spectrometry, TCA = trichloroacetic acid, μXRF = micro X-ray fluorescence

Information S1: Untargeted application for lipids in negative ion mode using mice hind legs
For the selection of matrices to test for their suitability for the detection of lipids in bone and bone marrow in negative ion mode, DAN, NEDC, and norharmane were compared. NEDC was included in this analysis although it is more often used for the detection of metabolites than for lipids, as multiple molecules in the lower mass range were seen during initial analysis.
In negative ion mode, different matrices allowed for the detection of different ions from the bone and bone marrow and most m/z values were only presented in one measurement (see Fig. S5). For DAN and norharmane, it was only possible to obtain specific signals for bone marrow, while for NEDC, it was possible to observe specific signals from both, bone and bone marrow. NEDC had the highest number of m/z values specific for bone or bone marrow, of which most of the m/z values were obtained from the bone (see Table S4). However, these specific m/z values are only in the lower mass range (< m/z 500). DAN resulted in the highest number of specific m/z values for bone marrow.
Furthermore, NEDC has higher signal intensities for bone and relatively high S/N values for bone marrow (see Table S4), while the intensities and S/N values for bone marrow are higher in DAN and norharmane. DAN showed the highest S/N values for bone marrow. In negative ion mode, DAN and NEDC have the most interference of matrix peaks with the tissue-related peaks, while there is some interference for norharmane (see mass spectra in Fig. S6).
Of the compared matrices in negative ion mode, DAN caused an unwanted reaction with the doublesided tape. For the matrices tested in negative ion mode, it was possible to create distribution images up to m/z 500 (see Fig. S5). For DAN and norharmane, a few specific m/z values for bone marrow could be found above m/z 500, but not for NEDC. With limited or no detection of lipids in bone and bone marrow, major lipid classes can be missed, for example, certain subclasses phospholipids that ionize better in negative ion mode [6]. Although these subclasses have been shown to play a role in bone health [7]. No specific m/z values for bone tissue could be obtained with DAN and norharmane. Based on specific m/z values, NEDC is the only matrix for which specific m/z values for bone as well as bone marrow could be obtained. However, the background signal and interference of this matrix is high (see Fig. S6). In negative ion mode, none of the tested matrices provided the desired result in terms of detection of m/z values specific for bone and bone marrow in combination with low interference of the matrix with the signal from the tissue. Therefore, further testing of the best matrix in negative ion mode is necessary.