Visualization of acetylcholine distribution in central nervous system tissue sections by tandem imaging mass spectrometry
- 1.8k Downloads
Metabolite distribution imaging via imaging mass spectrometry (IMS) is an increasingly utilized tool in the field of neurochemistry. As most previous IMS studies analyzed the relative abundances of larger metabolite species, it is important to expand its application to smaller molecules, such as neurotransmitters. This study aimed to develop an IMS application to visualize neurotransmitter distribution in central nervous system tissue sections. Here, we raise two technical problems that must be resolved to achieve neurotransmitter imaging: (1) the lower concentrations of bioactive molecules, compared with those of membrane lipids, require higher sensitivity and/or signal-to-noise (S/N) ratios in signal detection, and (2) the molecular turnover of the neurotransmitters is rapid; thus, tissue preparation procedures should be performed carefully to minimize postmortem changes. We first evaluated intrinsic sensitivity and matrix interference using Matrix Assisted Laser Desorption/Ionization (MALDI) mass spectrometry (MS) to detect six neurotransmitters and chose acetylcholine (ACh) as a model for study. Next, we examined both single MS imaging and MS/MS imaging for ACh and found that via an ion transition from m/z 146 to m/z 87 in MS/MS imaging, ACh could be visualized with a high S/N ratio. Furthermore, we found that in situ freezing method of brain samples improved IMS data quality in terms of the number of effective pixels and the image contrast (i.e., the sensitivity and dynamic range). Therefore, by addressing the aforementioned problems, we demonstrated the tissue distribution of ACh, the most suitable molecular specimen for positive ion detection by IMS, to reveal its localization in central nervous system tissues.
KeywordsImaging mass spectrometry Neurotransmitter Acetylcholine MS MS/MS Imaging IMS
Central nervous system
Imaging mass spectrometry
Indium tin oxide
Limit of detection
Matrix-assisted laser desorption/ionization
Region of interest
Total ion current
Imaging mass spectrometry (IMS), a mass spectrometry (MS)-based molecular imaging technique, is gaining greater popularity as a means of visualizing the distribution of molecular ions in tissue sections and cultured cells [1, 2]. The most characteristic feature of this molecular imaging technique is an MS-based detection principle that has wide versatility, allowing the analysis of many types of analyte molecules, particularly using matrix-assisted laser desorption/ionization (MALDI)-IMS. Consequently, this unique approach provides a novel opportunity to visualize diverse types of molecules directly on tissue surfaces, from small compounds to much heavier biopolymers, which in some cases can be visualized simultaneously [3, 4, 5]. MALDI-IMS has been successfully applied for the localization imaging of large structural proteins , neuropeptides , various types of membrane lipids [8, 9, 10, 11], and energy-related secondary metabolites  in the brains of healthy and diseased mouse models [13, 14].
This emerging imaging technique was initially developed as a tool for protein imaging [15, 16, 17], and most of the early reports on MALDI-IMS described it as a protein or peptide research tool [13, 18, 19]. Conversely, research on detecting and imaging small metabolite molecules has rapidly been expanding [11, 20, 21]. Currently, researchers do not have an established imaging technology for diverse metabolite species, therefore leading to the emergence of IMS as a tool for metabolite imaging. Additionally, the ability of IMS to simultaneously image many types of metabolites is also important because it can be used to visualize the molecular “conversion” of upstream metabolites into downstream metabolites at specific tissue locations in a two-dimensional manner .
Despite the promising capabilities of MALDI-IMS, this technique still faces several critical problems regarding its practical use in neurochemical research. One concern is related to its quantitative ability. As the matrix compound and other endogenous isobaric molecules often share nominal mass with the endogenous metabolites of interest, researchers cannot simply interpret the obtained ion intensity as being the analyte concentration. Both peak assignment to specific compounds by mass and careful structural validation by MS/MS or MS/MS imaging are necessary . The other concern is insufficient sensitivity for trace amounts of molecules. This is due to limitations regarding the sample purification process causing a severe ion suppression effect, whereas molecular separation techniques such as gas chromatography, liquid chromatography, and capillary electrophoresis have been traditionally utilized in MS .
For this reason, most IMS studies, including those of the author’s group, have reported the analyses of abundant metabolite species, particularly membrane-constituting lipids. However, it is important to expand the application of IMS to much smaller amount of molecules, such as neurotransmitters. The goal of this study was to develop an IMS application to visualize neurotransmitter distribution in the central nervous system (CNS). We raise two technical problems that must be resolved to achieve neurotransmitter imaging: (1) because there are lower concentrations of the bioactive molecules than there are of membrane lipids, we should achieve higher sensitivity and/or selectivity in MS detection, and (2) the molecular turnover of the neurotransmitters is fast; thus, we also should pay attention to the tissue preparation procedure to minimize postmortem changes. To verify these problems, we examined three animal organs using fixation techniques, namely, in situ freezing (ISF)  and a conventional fixation with decapitation at different times, and we found that brain samples subjected to ISF exhibited improved IMS data quality in terms of the number of effective pixels and the image contrast, i.e., improved sensitivity and dynamic range.
Acetylcholine (ACh), γ-aminobutyric acid (GABA), glutamate, dopamine, and serotonin were obtained from Sigma-Aldrich (St. Louis, MO, USA). Norepinephrine was purchased from Wako Pure Chemical Industries Ltd. (Osaka, Japan). α-Cyano-4-hydroxycinnamic acid (CHCA) and 2,5-dihydroxybenzoic acid (DHB) were obtained from Bruker Daltonics (Leipzig, Germany). Male C57BL/6J mice were purchased from SLC (Hamamatsu, Japan).
Examination of the intrinsic sensitivity using reference standards
To evaluate the efficiency of ionization and interference by the matrix, serially diluted (1.0 × 10−6 to 1.0 mg/mL) reference compounds were spotted on a stainless steel plate using DHB and CHCA as matrices. MS detection was performed via laser scanning of the spotted areas on the target plate. Analyte ion intensities and chemical noise from the matrices were calculated from the region of interest (ROI) drawn on each sample spot. The x-intercept values were calculated from the intercepts of the semi-logarithmic graphs that were plotted with the log values of the compound concentrations and the signal intensities.
Sample preparation of tissue sections for IMS analyses
All animals received humane care in accordance with the Japanese Association for Laboratory Animal Science Guidelines, and all of the animal experiments were approved by the Animal Experimentation Committee of Kansai Medical University. For postmortem freezing, mice were deeply anesthetized with pentobarbital to relieve suffering. After decapitation, we removed the brains of the animals within 30 s and the spinal cord within 2 min, and the tissues were frozen in powder dry ice. Frozen tissues were stored at −80 °C until sectioning. For ISF, animals were deeply anesthetized with diethyl ether and the head skin trimmed as described previously . The tip of the head was dipped into liquid nitrogen, with great care taken not to immerse the nose. Frozen brains were dissected with a surgical knife in a refrigerated box at −30 °C. For tissue sectioning and matrix coating for IMS analyses, frozen brains or spinal cords were prepared as 10-μm cryosections on ITO-coated glass slides (Bruker Daltonics) at −20 °C. The sections on the slide glasses were put in a compact desiccator and then brought out, not to be surrounded by frost and not to be degraded by the hydrolytic enzymes. Then, 50 mg/mL DHB in 70 % methanol and 0.1 % trifluoroacetic acid was uniformly sprayed over the samples using a Procon Boy FWA Platinum 0.2-mm caliber airbrush (Mr. Hobby, Tokyo, Japan). The samples were subjected to laser scanning for MALDI-IMS or MS/MS.
Imaging mass spectrometry
We adopted previously described procedures for MALDI-IMS [20, 25], with some modifications. Chemical compounds and tissue sections were analyzed using a MALDI time-of-flight (TOF)/TOF-type instrument, the Ultraflex II (Bruker Daltonics), and a linear ion trap MALDI LTQ XL™ mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). Data were acquired in the positive ion mode using an external calibration method. Calibration compounds which are the aforementioned six neurotransmitters were deposited on the surfaces of the sample support materials to minimize mass shift. MS data by TOF/TOF instrument Ultraflex in the MS mode and 100 laser beam shots were delivered to each data point (Figs. 2 and 3). LTQ XL linear ion trap mass spectrometer was also used for MS measurement (Fig. 4). The intervals between data points were 100 μm (Figs. 4, 5, 6, and 7a, b) and 50 μm (Fig. 7c). MS/MS imaging was performed using an LTQ instrument in the MS/MS mode (Figs. 5 and 6) and Ultraflex in “LIFT” MS/MS mode (Fig. 7). In the MS/MS operation, the data acquisition conditions (i.e., the laser power, collision energy, and number of laser irradiations) were optimized to obtain product ion mass spectra with high signal-to-noise (S/N) ratios for the fragment peaks.
We reconstructed the lipid and neurotransmitter images of interest from the IMS data. Image reconstruction from signals in the spectra was performed using FlexImaging (Bruker Daltonics) and ImageQuest (Thermo Fisher Scientific) software. As the ionization efficiency could vary depending on the matrix–analyte co-crystallization conditions and their sublimation during measurement [19, 25], the absolute intensities of the mass spectra were normalized to the same value of total ion current for the lipid distribution imaging performed by LTQ instrument (Figs. 4, 5, and 6), and this normalization did not apply the normalization to the MS/MS imaging dataset.
Results and discussion
Evaluation of the intrinsic sensitivity of six neurotransmitters
Summary of the detection sensitivity of MALDI-IMS for the six neurotransmitters using DHB and CHCA as matrices
MS/MS ion transition improved the S/N ratio for detecting ACh-derived signals
Distribution imaging of ACh in mouse spinal cord sections
Conversely, peak clusters derived from phospholipids were clearly observed at a mass range of 700 < m/z < 900 in the same section, and their characteristic distribution patterns were useful for understanding the anatomical features of the tissue sections. For example, because the glycolipid galactosylceramide counts myelin sheaths as one of its major lipid components, the ion at m/z 850 derived from galactosylceramide was highly localized in the white matter region of the spinal cord. In addition, a phosphatidylcholine (PC) molecular species observed at m/z 772 [10, 38], PC (diacyl-16:0/16:0, K+ adduct), was localized in the gray matter region at m/z 772, especially in the dorsal spinal horn.
Tandem MS imaging of ACh in mouse spinal cord sections
We noticed high-intensity ACh signals in both the ventral and dorsal horns (Fig. 5c). ACh is an important modulator of motor and sensory processing, especially at the spinal level, at which point pain-related nociceptive stimuli enter the CNS, which integrates the stimuli and relays them to the brain. There are large cholinergic motor neurons in the spinal anterior horn ; thus, it is reasonable that ACh was abundantly detected in ACh-producing cells containing choline acetyltransferase (ChAT), the ACh-synthesizing enzyme . The source of ACh in the sensory spinal cord has not been clearly established; however, the presence of a dense plexus of cholinergic fibers has been reported in the spinal lamina II  and III  with the use of anti-ChAT immunolabeling. In addition to immunohistochemical findings, exogenous ChAT was found in the ventral and dorsal horns in ChAT promoter-driven EGFP-expressing mice .
We directly detected the localization of ACh using IMS, whereas conventional methods such as immunohistochemistry indirectly described the localization of ACh sites of action with its receptors, transporters, and synthetic or catabolic enzymes. ISH data of the mouse spinal cord from the Allen Brain Atlas (http://www.brain-map.org/) demonstrate that the distribution of IMS images is reasonable compared with the localization of mRNA for ACh-related molecules such as ChAT, acetylcholinesterase (AChE), nicotinic receptor (nAChR), muscarinic receptor (mAChR), vesicular acetylcholine transporter (VAChT), and choline transporter (ChT). Many previous results matching our IMS data also match those of ISH and/or immunohistochemistry for ACh-related molecules including ChAT [42, 44, 46], AChE , nAChR , and VAChT [46, 49, 50]. Through its action on spinal cholinergic receptors, endogenous ACh participates not only in motor action but also in the setting of nociceptive thresholds and in the effect of clinically relevant analgesics; thus, the direct detection of ACh in the spinal cord will be useful for assessments such as evaluations of therapeutic strategies.
Distribution imaging of ACh in mouse sagittal brain sections
The ISF procedure protected endogenous ACh from postmortem degradation, resulting in improved MS/MS imaging sensitivity
Earlier studies described that the molecular turnover of ACh is rapid , and therefore, we should pay attention to the tissue preparation procedure to minimize postmortem changes. In this context, optimization of the organ sampling protocol has remained a critical issue because major degradation of the various metabolites could occur in tissues even within a second after respiratory arrest, especially concerning small molecules involved in signal transduction. As described below, we finally evaluated the extent of ACh postmortem changes among brain samples that were subjected to ISF, which ensures adequate perfusion until the arrival of the freezing front to avoid unnecessary autolysis, and postmortem freezing (decapitation before freezing). As a result, we found that ISF was the best method for ACh imaging.
Having observed the ACh-specific ion transitions of the brain section, distribution maps of the major fragment ions were then reconstructed and evaluated (Fig. 7b); the quality of these ACh-derived ion images was assessed among three groups of mice that were treated with different animal fixation methods, namely, ISF (right) and postmortem freezing with decapitation in which brain extraction was performed within 1 min (center) and 10 min of decapitation (left). As shown in the figure, the IMS data quality in terms of the number of effective pixels and the image contrast, i.e., sensitivity and dynamic range, was drastically improved in the ISF brain sample compared with that in the postmortem freezing-treated group, clearly demonstrating that the proper sample preparation technique is necessary for correct ACh imaging. Furthermore, increased sensitivity enables the distribution mapping of ACh at a higher spatial resolution. Figure 7c shows the expanded ACh distribution image of the hippocampus of ISF-treated brains, revealing a unique ACh distribution pattern even within hippocampal substructures.
Among the neurotransmitters, ACh was most sensitively detected using the positive ion mode of MALDI-MS. Based on this fundamental study, we successfully reconstructed ACh distribution images in the mouse brain and spinal cord from the scan data obtained from MS/MS ion transitions. Furthermore, the IMS data quality in terms of the number of effective pixels and the image contrast, i.e., the sensitivity and dynamic range, was drastically improved in ISF-treated brain samples, clearly demonstrating that the ISF sample preparation technique is necessary for precise ACh imaging.
The localization of ACh obtained in this study coincided agreeably with the expected localization based on the known distribution of the ACh-degrading enzyme, especially in the spinal cord and part of the brain. Therefore, we conclude that MS/MS-based IMS could be useful for neurotransmitter imaging and can be practically used in the field of neuroscience.
This study is supported by Research Grants for PRESTO and SENTAN from JST and a Grant-In-Aid for Young Scientists A from JSPS.
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.
- 2.Yang HJ, Ishizaki I, Sanada N, Zaima N, Sugiura Y, Yao I, Ikegami K, Setou M (2010) Detection of characteristic distributions of phospholipid head groups and fatty acids on neurite surface by time-of-flight secondary ion mass spectrometry. Med Mol Morphol 43(3):158–164. doi: 10.1007/s00795-009-0487-2 CrossRefGoogle Scholar
- 8.Sugiura Y, Shimma S, Konishi Y, Yamada MK, Setou M (2008) Imaging mass spectrometry technology and application on ganglioside study; visualization of age-dependent accumulation of C20-ganglioside molecular species in the mouse hippocampus. PLoS One 3(9):e3232. doi: 10.1371/journal.pone.0003232 CrossRefGoogle Scholar
- 9.Hayasaka T, Goto-Inoue N, Sugiura Y, Zaima N, Nakanishi H, Ohishi K, Nakanishi S, Naito T, Taguchi R, Setou M (2008) Matrix-assisted laser desorption/ionization quadrupole ion trap time-of-flight (MALDI-QIT-TOF)-based imaging mass spectrometry reveals a layered distribution of phospholipid molecular species in the mouse retina. Rapid Commun Mass Spectrom 22(21):3415–3426. doi: 10.1002/rcm.3751 CrossRefGoogle Scholar
- 11.Matsumoto J, Sugiura Y, Yuki D, Hayasaka T, Goto-Inoue N, Zaima N, Kunii Y, Wada A, Yang Q, Nishiura K, Akatsu H, Hori A, Hashizume Y, Yamamoto T, Ikemoto K, Setou M, Niwa S (2011) Abnormal phospholipids distribution in the prefrontal cortex from a patient with schizophrenia revealed by matrix-assisted laser desorption/ionization imaging mass spectrometry. Anal Bioanal Chem 400(7):1933–1943. doi: 10.1007/s00216-011-4909-3 CrossRefGoogle Scholar
- 20.Shimma S, Sugiura Y, Hayasaka T, Hoshikawa Y, Noda T, Setou M (2007) MALDI-based imaging mass spectrometry revealed abnormal distribution of phospholipids in colon cancer liver metastasis. J Chromatogr B Analyt Technol Biomed Life Sci 855(1):98–103. doi: 10.1016/j.jchromb.2007.02.037 CrossRefGoogle Scholar
- 21.Enomoto H, Sugiura Y, Setou M, Zaima N (2011) Visualization of phosphatidylcholine, lysophosphatidylcholine and sphingomyelin in mouse tongue body by matrix-assisted laser desorption/ionization imaging mass spectrometry. Anal Bioanal Chem 400(7):1913–1921. doi: 10.1007/s00216-011-4924-4 CrossRefGoogle Scholar
- 24.Hattori K, Kajimura M, Hishiki T, Nakanishi T, Kubo A, Nagahata Y, Ohmura M, Yachie-Kinoshita A, Matsuura T, Morikawa T, Nakamura T, Setou M, Suematsu M (2010) Paradoxical ATP elevation in ischemic penumbra revealed by quantitative imaging mass spectrometry. Antioxid Redox Signal 13(8):1157–1167CrossRefGoogle Scholar
- 26.Schikorski T, Stevens CF (1997) Quantitative ultrastructural analysis of hippocampal excitatory synapses. J Neurosci 17(15):5858–5867Google Scholar
- 28.Yao I, Iida J, Nishimura W, Hata Y (2002) Synaptic and nuclear localization of brain-enriched guanylate kinase-associated protein. J Neurosci 22(13):5354–5364Google Scholar
- 31.Miranda MI (2007) Changes in neurotransmitter extracellular levels during memory formation. In: Bermudez-Rattoni (ed) Neural plasticity and memory: from genes to brain imaging. Chapter 7. CRC, Boca RatonGoogle Scholar
- 32.Liu H, Sanuda-Pena MC, Harvey-White JD, Kalra S, Cohen SA (1998) Determination of submicromolar concentrations of neurotransmitter amino acids by fluorescence detection using a modification of the 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate method for amino acid analysis. J Chromatogr 828(1–2):383–395CrossRefGoogle Scholar
- 35.Koizumi S, Yamamoto S, Hayasaka T, Konishi Y, Yamaguchi-Okada M, Goto-Inoue N, Sugiura Y, Setou M, Namba H (2010) Imaging mass spectrometry revealed the production of lyso-phosphatidylcholine in the injured ischemic rat brain. Neuroscience 168(1):219–225. doi: 10.1016/j.neuroscience.2010.03.056 CrossRefGoogle Scholar
- 39.Carrozzo MM, Cannazza G, Pinetti D, Di Viesti V, Battisti U, Braghiroli D, Parenti C, Baraldi M (2010) Quantitative analysis of acetylcholine in rat brain microdialysates by liquid chromatography coupled with electrospray ionization tandem mass spectrometry. J Neurosci Methods 194(1):87–93CrossRefGoogle Scholar
- 45.Mesnage B, Gaillard S, Godin AG, Rodeau JL, Hammer M, Von Engelhardt J, Wiseman PW, De Koninck Y, Schlichter R, Cordero-Erausquin M (2011) Morphological and functional characterization of cholinergic interneurons in the dorsal horn of the mouse spinal cord. J Comp Neurol 519(16):3139–3158CrossRefGoogle Scholar
- 46.Ichikawa T, Ajiki K, Matsuura J, Misawa H (1997) Localization of two cholinergic markers, choline acetyltransferase and vesicular acetylcholine transporter in the central nervous system of the rat: in situ hybridization histochemistry and immunohistochemistry. J Chem Neuroanat 13(1):23–39CrossRefGoogle Scholar
- 54.Jenden DJ (ed) (1978) Cholinergic mechanisms and psychopharmacology. Plenum, New YorkGoogle Scholar