Article Cite This: Anal. Chem. 2018, 90, 729−736
pubs.acs.org/ac
N‑Phenyl-2-naphthylamine as a Novel MALDI Matrix for Analysis and in Situ Imaging of Small Molecules Huihui Liu,† Yueming Zhou,†,‡ Jiyun Wang,† Caiqiao Xiong,† Jinjuan Xue,† Lingpeng Zhan,† and Zongxiu Nie*,†,§ †
Key Laboratory of Analytical Chemistry for Living Biosystems, Institute of Chemistry, Chinese Academy of Sciences, and Beijing National Laboratory for Molecular Sciences, Beijing 100190, China ‡ College of Chemistry, Biology and Material Sciences, East China University of Technology, Nanchang 330013, China § National Center for Mass Spectrometry in Beijing, Beijing 100190, China S Supporting Information *
ABSTRACT: Due to its strong ultraviolet absorption, low background interference in the small molecular range, and salt tolerance capacity, N-phenyl-2-naphthylamine (PNA) was developed as a novel matrix in the present study for analysis and imaging of small molecules by matrix-assisted laser desorption/ionization mass spectrometry timeof-fight (MALDI-TOF MS). The newly developed matrix displayed good performance in analysis of a wide range of small-molecule metabolites including free fatty acids, amino acids, peptides, antioxidants, and phospholipids. In addition, PNA-assisted LDI MS imaging of small molecules in brain tissue of rats subjected to middle cerebral artery occlusion (MCAO) revealed unique distributions and changes of 89 small-molecule metabolites including amino acids, antioxidants, free fatty acids, phospholipids, and sphingolipids in brain tissue 24 h postsurgery. Fifty-nine of the altered metabolites were identified, and all the changed metabolites were subject to relative quantitation and statistical analysis. The newly developed matrix has great potential application in the field of biomedical research.
S
However, intensive background interferences generated by conventional organic matrices such as α-cyano-4-hydroxycinnamic acid (CCA) and 2, 5-dihydroxybenzoic acid (DHB) would suppress the ionization and interfere with the detection of the small-molecule analytes (typically m/z < 500), hindering the application of MALDI-MS to the analysis of molecules in the low-mass range. Furthermore, the salt-tolerance ability of matrices is also an important factor which affects its application in complex biosamples. Therefore, the search for new matrices that generate lower abundance background signal, are effective in tolerance of salt, and possess high sensitivity at the same time is of great significance for acquiring high-quality MALDI MS signals. To circumvent this problem, many novel promising matrices, such as porous silicon,10 nanomaterials,11−14 small molecule organic matrices,15−17 and organic salts5,8,18 have sprung up in recent years. These matrices displayed good application prospect in analysis/in situ imaging of small molecules. Ischemic stroke is one of the leading cause of death and a major cause of disability. Furthermore, pathophysiological mechanisms of cerebral ischemia are extremely complicated, and elucidation of altered metabolic pathways and networks
mall-molecule metabolites such as metabolic intermediates, signaling molecules, and secondary metabolites play significant roles in cellular architecture, energy storage, and signal transduction.1,2 Changes of metabolites have been reported to be related to various diseases such as cancers,3 cardiovascular disease,4 and neurological disease.5 Analysis of small metabolites is very important to discover biomarkers and elucidate pathophysiological mechanisms of diseases. Among the many analytical techniques, nuclear magnetic resonance (NMR) and mass spectrometry (MS) have been most widely used for analysis of small molecule metabolites. Matrix-assisted laser desorption/ionization (MALDI) MS, a soft ionization technique introduced by Karas et al. in 1985,6 has become established as a robust analytical tool for mass spectrometric analysis of wide variety of biomolecules including proteins and small-molecule metabolites. Furthermore, MALDI MS imaging (MALDI MSI) possessed unique features including being free of labeling, high sensitivity, high throughout, molecule-specificity, and the capacity for in situ localizing a wide range of biomolecules simultaneously from a tissue specimen in one single run. The technique was introduced by Caprioli7 and has emerged as a remarkable new technology among the analytical techniques for investigation of metabolic alterations under pathological conditions.5,8,9 The detection of molecules in MALDI analysis is believed to be to a large extent dependent on the choice of matrix. © 2017 American Chemical Society
Received: July 12, 2017 Accepted: November 27, 2017 Published: November 27, 2017 729
DOI: 10.1021/acs.analchem.7b02710 Anal. Chem. 2018, 90, 729−736
Article
Analytical Chemistry
suggest that the newly developed matrix has potential application in in situ MSI of small-molecule metabolites and in the field of biomedical research.
underlying cerebral ischemia is of great significance. Middle cerebral artery occlusion (MCAO) is a widely recognized and well-established animal model for investigation of cerebral ischemia.19 Previous studies using MSI combined with capillary electrophoresis/mass spectrometry revealed spatiotemporal changes in adenylates and NADH in a mouse MCAO model.20 ATP5i, COX6C, and UMP-CMP kinase were profiled and identified to be involved in brain ischemia using MALDI MSI.21 Furthermore, several studies have been carried out in recent years using MALDI MSI for investigating changes of small-molecule metabolites in rat brain following MCAO. Studies by Koizumi et al. performed by MALDI-MSI revealed the production of lyso-phosphatidylcholine in the injured ischemic rat brain.22 Whitehead et al. examined the spatial profile and changes of ganglioside species in the mouse brain following ischemia using MALDI MSI.23 Another study by Shanta confirmed changes in phospholipids such as lysophosphatidylcholine, phosphatidylcholine, phosphatidylethanolamine, and sphingomyelin by MALDI IMS.24 Miyawaki et al. detected dynamic changes of phosphatidylcholine in rat hippocampal CA1 after transient global ischemia using imaging mass spectrometry.25 Nielsen et al. also revealed biomarker lipids for phagocytosis and signaling during focal cerebral ischemia.26 MALDI-MSI with 9-aminoacridine (9-AA) visualized changes of metabolites involved in glycosis, TCA cycle, glutamate−glutamine cycle, and malate−aspartate shuttle in response to pathological perturbation.15,16 Previous study by our group revealed changes of metal ions, antioxidants, phospholipids, and metabolites involved in glycosis, TCA cycle, glutamate−glutamine cycle, malate−aspartate shuttle, and nucleotide metabolism during the ischemic period.5 These previous studies provided abundant information, helping us better understand the spatial information and changes of metabolites during the process of ischemia. Nevertheless, pathophysiological mechanisms of ischemic stroke are very complicated. Metabolites in biological systems always contain a wide spectrum of chemically diverse compounds such as amino acids, organic acids, lipids, carbohydrates, nucleotides, and so on. It is very challenging to detect them all in a single run even up to now. Our ability to prevent and cure stroke remains limited up to now. Thus, MALDI matrix development still is a challenging task up to now. In the present study, a commercially available matrix, Nphenyl-2-naphthylamine (PNA), was developed as a novel MALDI matrix for analysis of small-molecule endogenous metabolites. In order to more systematically and comprehensively investigate metabolic alterations underlying ischemia, in situ MALDI imaging of small molecules on brain tissues of rats subjected to MCAO was carried out. The newly developed matrix possessed unique features, namely, low cost, strong ultraviolet absorption, and fewer background signals (typically in the mass range m/z < 500). PNA-assisted LDI MSI of small molecules in brain tissues of a MCAO rat model permitted us to visualize the spatial distribution and alteration of a broad range of small-molecule metabolites including amino acids, fatty acids, peptide, phospholipids, and sphingolipids, as well as antioxidants and other small molecule metabolites simultaneously. Fifty-nine of the altered metabolites were identified, and all the changed metabolites underwent relative quantitation and statistical analysis. Furthermore, the imaging experiments carried out in 48 h demonstrated very little change in the metabolites detection, suggesting low volatility and high chemical stability of PNA under high vacuum. These findings
■
EXPERIMENTAL SECTION Chemicals and Reagents. N-Phenyl-2-naphthylamine (PNA), 1,5-diaminonaphthalene (1,5-DAN), and 1,8-bis (dimethylamino)naphthalene (DMAN) for matrix preparation, 2,3,5-triphenyltetrazolium chloride (TTC) for tissue staining, and chloral hydrate for anesthesia were purchased from SigmaAldrich (St. Louis, MO). Standards including aspartate, taurine, glutamate, glutamine, N-acetylaspartate, ascorbic acid, hypoxanthine, O-phosphoethanolamine, 4-hydroxyphenylpyruvic acid, hydroxyphenylacetylglycine, glutathione, amino acids standards (see the Supporting Information), fatty acids standards (lauric acid, pentadecanoic acid, palmitate sodium, heptadecanoic acid, stearic acid, linolenic acid, linoleic acid, oleic acid, arachidonic acid, and docosahexaenoic acid), and peptide standards (see the Supporting Information) were also purchased from Sigma-Aldrich (St. Louis, MO). Sample Preparation. Preparation of Matrixes and Standard Solutions. Matrix solution of DMAN was prepared at 10 mg/mL in methanol. 1,5-DAN solution was freshly prepared at a concentration of 10 mg/mL in 0.2% formic acid in water/acetonitrile (3:7, v/v). PNA was dissolved in acetonitrile to obtain a final concentration of 10 mg/mL. Fatty Acids. Stock solutions of fatty acids (lauric acid, pentadecanoic acid, palmitate sodium, heptadecanoic acid, stearic acid, linolenic acid, linoleic acid, oleic acid, arachidonic acid, and docosahexaenoic acid) were prepared in methanol at 10 mM, and a fatty acid standard mixture was made with the concentration of each analyte being 1 mM. Amino Acids and Peptides. Standard solution of glutamate (Glu) was prepared in deionized water containing 0.2 M HCl with a concentration of 1 mM. Standard solution of proline (Pro) was prepared in methanol with a concentration of 1 mM. Standard solutions of phenylalanine (Phe), tyrosine (Tyr), and tryptophan (Try) were prepared in deionized water containing 0.5 M NaOH at 1 mM with a concentration of 1 mM. Standard solutions of other amino acids (Gly, Ala, Val, Leu, Ile, Ser, Thr, Cys, Met, Asn, Gln, Asp, Lys, Arg, His) and peptides (Gly-Ala, Gly-Asp, Gly-His, GSH, Gly-Leu-Tyr) were prepared in deionized water at 10 mM and diluted with deionized water to the desired concentration. A peptide standard mixture containing Gly-Tyr, Val-Tyr-Val, leucine enkephalin, and methionine enkephalin acetate with the concentration of each analyte being 0.5 mg/mL was prepared in deionized water and diluted to desired concentration with deionized water. Endogenous Metabolites. Standard solutions of taurine, ascorbic acid, N-acetylaspartate, 4-hydroxyphenylpyruvic acid, and hydroxyphenylacetylglycine were prepared in deionized water. Hypoxanthine was prepared in deionized water containing 0.5 M NaOH. O-Phosphoethanolamine was prepared in methanol. Standard mixture containing these endogenous metabolites was made with the concentration of each analyte being 1 mM. Optical Absorption Analysis. Ultraviolet−visible (UV− vis) absorption spectra of N-phenyl-2-naphthylamine solution was analyzed using a TU-1900 double-beam UV−vis spectrophotometer (Purkinje General Instrument Co. LTD, Beijing, China) in a 1 cm path length cuvette at room temperature. The wavelength scanning range was from 200 to 400 nm. 730
DOI: 10.1021/acs.analchem.7b02710 Anal. Chem. 2018, 90, 729−736
Article
Analytical Chemistry Animals. Three male Kunming mice (20−22 g) and 10 Sprague−Dawley rats (250−260 g) were provided by the Laboratory Animal Center, Academy of Military Medical Sciences. All animals were kept in an environmentally controlled room (20−24 °C, 40−70% humidity) with a 12 h light/dark cycle. Standard diet and water were provided ad libitum to the rats. The animal experiments were performed according to the NIH Guide for the Care and Use of Laboratory Animals (National Institutes of Health Publication, No. 3040-2, revised 1999, Bethesda, MD) and were approved by the Animal Care and Use Committee of the Chinese Academy of Sciences. Establishment of the MCAO Model. Animals (n = 10) were divided into two groups with five rats in sham-operation group and another five rats in MCAO group. Establishment of the MCAO model and neurological evaluation was carried out as described previously.27 The brain tissue sections were stained with TTC (Supporting Information) in order to identify regions of necrosis. Preparation of Plasma Samples. Blood samples from healthy Kunming mice were obtained via angular vein and placed in heparinized Eppendorf tubes. Plasma was separated after centrifuging at 4000 × g for 10 min. A portion of 50 μL of plasma sample was transferred into a 1.5 mL tube followed by precipitation of proteins with 3 volumes of acetonitrile. The mixture was vortexed for 10 s and stored at −20 °C for 18 h. After the mixture was centrifuged at 15 000 g for 60 min at 4 °C, the supernatant was transferred, and 500 μL of hexane and 500 μL of water were added. The mixture was then vortexed for 30 s. After centrifugation at 1000 g for 10 min, the hexane layer was pipetted out and dried under nitrogen gas. Tissue Dissecting. Sprague−Dawley rats 24 h post MCAO were euthanized with chloral hydrate (350 mg/kg, i.p.), and then brain tissues were removed and snap-frozen in liquid nitrogen. Brain tissues from normal Kunming mice were dissected and flash-frozen with the same method. All tissues were stored at −80 °C until further preparation. Tissue Sectioning. To avoid signal suppression of embedding medium, tissues were fixed on the cutting stage atop a drop of saline. All tissues were sliced to a thickness of 10 μm using a Leica CM1950 cryostat (Leica Microsystems GmbH, Wetzlar, Germany) at −17 °C and thaw- mounted onto conductive indium tin oxide (ITO)-coated glass slides. The glass slides were then dried a vacuum desiccator for approximately 1 h before matrix application. MALDI Analysis. Evaluation of New Matrix Candidate. For MALDI analysis, the samples were prepared using the conventional dried-droplet method as follows: the sample solution was mixed with the matrix solution at ration of 1:1, and 1 μL of the resulting mixture was then pipetted on the MALDI target plate, following by drying under a stream of nitrogen gas at room temperature. Mass Spectrometry Imaging. For MSI, N-phenyl-2naphthylamine dissolved in acetonitrile at a concentration of 5 mg/mL was uniformly sprayed onto the tissue sections mounted onto ITO-coated conductive glass slides using an automatic matrix sprayer (ImagePrep, Bruker Daltonics). MALDI analysis was carried out on an Ultraflextreme MALDI-TOF/TOF MS (Bruker Daltonics, Billerica, MA) equipped with a smartbeam Nd: YAG 355 nm laser. The analyzer was operated in negative reflectron mode, and mass spectra were recorded with a pulsed ion extraction time of 130 ns, an accelerating voltage of 20.00 kV, an extraction voltage of
17.70 kV, a lens voltage of 6.4 kV, and a reflector voltage of 21.1 kV. Laser spot size was set at medium focus (∼50 μm laser spot diameter), and laser power was optimized prior to each run and then fixed for the whole experiment. The mass spectra data were acquired over a mass range of m/z 0−1000 Da. Mass calibration was achieved with external standards prior to data acquisition. The spatial resolution was set to 10 and 100 μm for brain tissue sections from mice for evaluation of the acceptable spatial resolution of MS imaging with this matrix. When applying to brain tissue sections of rat subjected to MCAO, the spatial resolution was set to 200 μm. The imaging data for each array position consists of 200 laser shots. Regions of interest (ROIs) were manually defined in the imaging software using both the optical image and MSI data image. MALDI mass spectra were normalized with the total ion current (TIC), and the signal intensity of each imaging data was represented as the normalized intensity. Further Detailed Structural Confirmation. Further detailed structural confirmation was performed on the Ultraflextreme MALDI-TOF/TOF MS in the LIFT mode and Fourier transform ion cyclotron resonance (FT-ICR) MS. Detailed information is provided in Supporting Information. Metabolites were identified or tentatively predicted by comparing their MS or MS/MS data with those of the standards or referring to databases (METLIN, http://metlin.scripps.edu/; MassBank, http://www.massbank.jp/; Human Metabolome Database, http://www.hmdb.ca/; and LIPID MAPS, http://www. lipidmaps.org/). Relative Quantitation and Statistical Analysis. Relative changes of metabolites in the ischemic hemisphere compared with contralateral hemisphere (n = 5) were quantitated. Regions of interest were defined, and the corresponding average intensity was acquired. To minimize the individual differences, average intensity of metabolites in the ischemic hemisphere was divided by corresponding average intensity of the metabolites in the contralateral hemisphere on every individual section. Two-tailed Student’s t test was performed to compare the intensity ratio of metabolites between control and model group. Furthermore, in order to correct t test results by false discovery rate (FDR), the Benjamini−Hochberg (BH) procedure was used to recalculate adjusted p-values in R.28 pValues ≤0.05 were considered statistically significant.
■
RESULTS AND DISCUSSION Evaluation of the New Matrix Candidate. The interplay between the wavelength of the laser and the absorption profile of the matrix constitutes a crucial factor in MALDI-MS.29 Previous studies have shown that typically the best analytical results are obtained if the laser wavelength matches the UV absorption band of the matrix in the solid state well. Figure S1 shows the UV−vis absorption spectrum of 0.2 mmol/L of PNA in acetonitrile. The matrix has a molar extinction coefficient of ε = 3355 L/mol·cm at the laser excitation wavelength of 355 nm. The result suggested that the matrix has a relatively strong absorption at the applied laser wavelength, meeting one of the basic requirements of being a good MALDI matrix. The laser desorption/ionization (LDI) mass spectra of Nphenyl-2-naphthylamine acquired in negative and positive ion modes are presented in Figure S2. The negative ion spectrum exhibited prominent peak arising from [M − H]− at m/z 218, accompanied by low-abundance ions at m/z 285 ([2(M − phenyl) − H]−) and m/z 437 ([2M − H]−). The ion spectrum in positive mode showed prominent peak at m/z 219 and low731
DOI: 10.1021/acs.analchem.7b02710 Anal. Chem. 2018, 90, 729−736
Article
Analytical Chemistry
Figure 1. LDI-TOF MS spectra of (a) 1,5-DAN, (b) DMAN, and (c) PNA and MALDI-TOF MS spectra of fatty acids analyzed in negative ion mode using (d) 1,5-DAN, (e) DMAN, and (f) PNA as matrixes, respectively. 1, lauric acid; 2, pentadecanoic acid; 3, palmitate sodium; 4, heptadecanoic acid; 5, linolenic acid; 6, linoleic acid; 7, oleic acid; 8, stearic acid; 9, arachidonic acid; 10, docosahexaenoic acid. The amount of each analyte is 500 pmol.
Figure 2. MALDI-TOF MS spectra of a solution containing small-molecule endogenous metabolites in negative ion mode using PNA as matrix. 1, taurine; 2, aspartate; 3, hypoxanthine; 4, O-phosphoethanolamine; 5, glutamine; 6, glutamate; 7, N-acetylaspartate; 8, ascorbic acid; 9, 4hydroxyphenylpyruvic acid; 10, hydroxyphenylacetylglycine. The amount of each analyte is 500 pmol.
abundance ions at m/z 346 and m/z 437, corresponding to the PNA radical cation and matrix clusters. Based on the classical Brønsted−Lowry acid−base theory,30 the newly developed matrix might not be suitable to work in the positive mode. Furthermore, many small molecules such as peptides, drugs which are easily ionized in positive mode, were also attempted to be analyzed using the new matrix. But the results are unsatisfactory (data were not shown). The results indicated that PNA exhibited very clean mass spectrum and few matrixderived background signals especially at the m/z range of
