Spatial Analysis of Phosphatidylinositol Molecular Species in Pork Chop Tissues Using Matrix-assisted Laser Desorption/ionization-Mass Spectrometry Imaging.

Matrix-assisted laser desorption/ionization-mass spectrometry imaging (MALDI-MSI) is a powerful technique for visualizing lipids in biological tissues. Phosphatidylinositol (PI), a phospholipid in pork, is a major source of inositol in animal-derived foods believed to be protective against diseases related to pregnancy and cancer. However, the distribution of PI molecular species in pork is not well understood. Here, we performed MALDI-MSI analysis to investigate the distribution and composition of PI molecular species in pork chop comprising Longissimus thoracis et lumborum muscle (loin), intermuscular fat tissue, transparent tissue, and spinalis muscle. Twelve diacyl-PI molecular species were identified using liquid chromatography-electrospray ionization-tandem mass spectrometry (MS/MS) and MALDI-MS/MS analysis and visualized using MALDI-MSI. Spinalis muscle had the highest amount of identified PI molecular species, followed by loin, transparent tissue, and intermuscular fat tissue. The diacyl-PI molecular species containing hexadecadienoic, oleic, linoleic and eicosadienoic acids at the sn-2 position were mainly abundant in the loin and spinalis muscle, whereas those containing mead, arachidonic, docosatetraenoic, and docosapentaenoic acids at the sn-2 position were mainly abundant in both muscles as well as transparent tissues. Notably, the balance of PI molecular species differed among the tissues depending on fatty acid compositions at the sn-2 position. These results suggested that MALDI-MSI is a promising tool for assessing the association between individual pork tissues and the protective effects of PI molecular species against diseases related to pregnancy and cancer. To the best of our knowledge, this is the first report showing tissue-specific distributions of PI molecular species in pork chop using MALDI-MSI.

microscopic resolution without the need for antibodies, staining, or complicated preliminary procedures 10 14 .
Meat is an important source of lipids, proteins, vitamins, and micronutrients 15,16 . Pork is one of the most consumed meats worldwide. Phospholipids are major components of meat that affect its qualities including nutritional and health-related properties 1,15,16 . Previously, we identified phosphatidylcholine and sphingomyelin molecular species as the major phospholipids in pork chop and visualized them using LC-ESI-MS/MS and MALDI-MSI 12,13 . Phosphatidylinositol PI , a major phospholipid in meat, is a component of the cell membrane bilayer and functions as a reservoir of polyunsaturated fatty acids PUFAs , predominantly arachidonic acid 20:4 . PUFAs have various health-promoting effects in individuals suffering from chronic diseases 17 . PIs are also a major source of animal food-derived inositol. Also known as a pseudo-vitamin vitamin Bh or B8 , inositol has been suggested to be protective against diseases related to pregnancy and cancer 18 21 . However, the distribution of PI molecular species in meat is not well known.
In the present study, we aimed to identify PI molecular species in pork chop using LC-ESI-MS/MS and MALDI-MS/ MS analysis. Subsequently, we visualized and compared the amounts of PI molecular species in four different tissues and revealed characteristic distribution patterns of PI molecular species in the four different pork tissues.

Materials
Water, chloroform, methanol, acetonitrile, isopropanol, and ammonium formate were purchased from FUJIFILM Wako Pure Chemical Corp. Tokyo, Japan , and 9-aminoacridine 9AA was purchased from Sigma Aldrich St Louis, MO, USA . Alpha-cyano-4-hydroxycinnamic acid CHCA was purchased from Tokyo Kasei Kogyo Co., Ltd. Tokyo, Japan . A peptide calibration standard containing angiotensin II was purchased from Bruker Corp. Billerica, MA, USA . All reagents and solvents used in the study were of analytical grade.

Pork samples
Pork chops 6-month-old female crossbred pigs, Duroc Danish Landrace Yorkshire were purchased from a local supermarket one day after slaughter. The samples were stored in a 80 freezeruntil needed.

Preparation of pork sections
Pork sections were prepared as described in our previous study 12,13 . Briefly, consecutive 10-μm sections containing four different tissues; namely, Longissimus thoracis et lumborum muscle loin , intermuscular fat tissue, trans-parent tissue, and spinalis muscle, were prepared using a cryostat CRYOCUT CM1860; Leica Microsystems, Wetzlar, Germany . The frozen sections were mounted on indium tin oxide ITO -coated glass slides 100Ω/m 2 ; Matsunami Glass, Osaka, Japan , which were then placed in 50 mL centrifuge tubes containing silica gel and preserved at 80 until MALDI-MSI.

Matrix vapor deposition
Matrix vapor deposition/recrystallization was performed as previously described 8 , with minor modifications. An ITO-coated glass slide loaded with a frozen section was taken out of the freezer and dried in a vacuum desiccator for 30 min. Matrix vapor deposition was performed using a vacuum deposition system SVC-700TMSG, Sanyu Electron Co., Ltd., Tokyo, Japan . The glass slide was attached to the sample holder using adhesive tape, and 9AA powder 60 mg was added on the matrix holder. The distance between the sample holder and the matrix holder was maintained at 8 cm. After the vacuum pressure in the chamber reached 5 10 3 Pa, 9AA was heated from 240 to 300 for approximately 30 min, until the entire powder had sublimed.

MALDI-MSI and MS/MS analysis
MALDI-MSI analysis was performed as described in our previous study 7,8 . The 9AA-coated sections were applied to a MALDI-time of flight TOF/TOF instrument UltrafleXtreme, Bruker equipped with a 355-nm Nd:YAG laser at a repetition rate of 2000 Hz. Data were collected in negative ion mode reflector mode at a step size of 100 μm. The laser diameter was set to medium size. The m/z values ranging from 850 to 920 were measured. Calibration was performed externally before the measurements using the exact m/z of both CHCA M H ions m/z 188.03532 and angiotensin II M H ions m/z 1044.52725 . All spectra were collected automatically. Normalization was performed based on the total ion current using FlexImaging 4.1 software Bruker . Two-dimensional ion-density maps were also prepared using this software.
Three sections prepared from the same pork chop attached on the same ITO-coated glass slide were measured, and detection intensities in the loin, intermuscular fat tissue, transparent tissue, and spinalis muscle were obtained using the region of interest function of FlexImaging 4.1 software. The mean detection intensities of the PI molecular species identified in the four tissues were compared.
MALDI-MS/MS analyses were performed in collision-induced dissociation LIFT MS/MS mode, and the selected precursor and product ions were acquired. The MS/MS window was set to the precursor ion m/z 6 Da. The MS/ MS spectra were analyzed using FlexAnalysis 3.4 software Bruker .

Preparation of total lipid extracts
Consecutive 10-μm pork chop sections were collected total 100 mg in 1.5 mL micro tubes. Loin 161 mg and spinalis muscle 133 mg were separated from the pork chop with a knife, and put in 2.0 mL micro tubes. Total lipid extract was prepared according to the Bligh and Dyer method 22 . After drying in a vacuum centrifugal evaporator CVE-2200, Eyela Tokyo Rikakikai Co. Ltd., Tokyo, Japan , lipids were dissolved in 1 mL methanol:isopropanol 9:1, v/ v and subjected to LC-ESI-MS and MS/MS analysis.

LC-ESI-MS, and MS/MS analysis
One microliter of the total lipid extract was injected into a Waters Acquity UPLC connected to a Waters Synapt XS quadrupole-TOF mass spectrometer equipped with an ESI ion source Waters Corp., Milford, MA . A Waters ACQUITY UPLC BEH C18 reversed-phase column 2.1 100 mm; 1.7-μm particle size was used at 40 . The mobile phase comprised solvent A acetonitrile:water 9:1, v/v and solvent B isopropanol , both of which contained 10 mM ammonium formate. The LC flow rate was 0.4 mL/min. The LC gradient conditions were as follows: 15 -48 B from 0 to 20 min, 90 B from 20 to 25 min, and 15 B from 25 to 28 min. The MS instrument was set to negative ion mode. Each total lipid extract was measured three times, and base peak intensity chromatogram of each exact m/z value 10 ppm was obtained using MassLynx software Waters . Relative intensities were calculated from their peak areas, and shown as mean value standard deviation n 3 . To compare the total PI concentrations between loin and spinalis muscle, the mean detection intensities of total diacyl-PI molecular species were shown as per 1 mg weight.
MS/MS spectra were collected at the exact m/z values of PI molecular species ions approximately 1 Da as the precursor ions using the same instrument. The trap and transfer collision energies were set at 45 and 2 eV, respectively. Data analysis was performed using MassLynx software Waters .

Statistical analyses
In Figs. 3a and 3b, the MALDI-MSI data are expressed as mean values standard error n 3 . The data were compared using one-way ANOVA, and post-hoc analysis was conducted using the Tukey s test. Statistical analyses were performed using the four tissues for each PI molecular species. Mean values with different letters indicate significant differences p 0.05 . In Fig. 3c, asterisks * in the LC-ESI-MS data indicate significant differences p 0.05 determined by Student s t-test.

Mass spectra of pork chop sections
Portions of the pork chops Fig. 1a that contained four different tissues; namely, loin, intermuscular fat tissue, transparent tissue, and spinalis muscle Fig. 1b , were selected for analysis. We used 9AA as a matrix because it is commonly used for the analysis of PI molecular species using negative ion mode of MALDI-MSI 14 . Vapor deposition was used as the matrix coating method because it forms uniform and fine matrix crystals on the sections. Under experimental conditions, PI molecular species are generally detected as M H ions and odd-numbered peaks in the m/z rage from 800 to 950 in animal tissues 14 . Several peaks with odd numbers were observed in the mass spectra of the four different tissues Figs. 1c-1f , indicating the detection of PI molecular species in the pork chop sections. In addition, the mass spectra patterns of the different four tissues appeared different from one another, implying that the concentrations of PI molecular species among these tissues were dissimilar.

Identification of PI molecular species in pork chop
sections To identify the PI molecular species detected in pork chops by MALDI-MSI, total lipid extracts were analyzed using LC-ESI-MS/MS in negative ion mode. PI comprises a glycerol backbone with fatty acids at the sn-1 and sn-2 positions and an inositol phosphate head group at the sn-3 position Fig. 1g . In general, the fatty acids at the sn-1 position are palmitic 16:0 , stearic 18:0 , and oleic 18:1 acids, whereas those at the sn-2 position are PUFAs 23      of total PI molecular species. These results were similar to that of LC-ESI-MS analysis, suggesting that MALDI-MSI can analyze the relative abundance of PI molecular species in pork chop with the equivalent accuracy as LC-ESI-MS analysis. and spinalis muscle, their mean detection intensities obtained by MALDI-MSI analysis were combined. The total content of diacyl-PI molecular species of both muscle tissues was significantly higher than those of other two tissues Fig. 3b . In addition, the content of diacyl-PI molecular species in spinalis muscle was 1.24 times significantly higher than that in loin. To verify the result of MAL-DI-MSI analysis, the mean detection intensities of diacyl-PI molecular species between loin and spinalis muscle were compared using LC-ESI-MS analysis. The content of diacyl-PI molecular species in spinalis muscle was 1.35 times significantly higher than that in loin Fig. 3c . This result was similar to that of MALDI-MSI analysis, suggesting that the accuracy of MALDI-MSI analysis for both muscle tissues was equivalent to that of LC-ESI-MS analysis.

Distribution and composition of diacyl-PI molecular species in pork sections
In the present study, we identified 12 PI molecular species in four different pork chop tissues using LC-ESI-MS/MS and MALDI-MS/MS analysis. In addition, we revealed characteristic distribution patterns of PI molecular species in the four tissues of pork chop.
There are nine possible stereoisomers of inositol, and PI molecular species generally have myo-inositols 18 . It has been suggested that myo-inositol deficiency causes cranial neural tube defects NTDs , such as anencephaly and spina bifida in several rodent models 18 . Myo-inositol has been studied in humans for a range of conditions and appears to be safe for use in human pregnancy. In cohort studies from Italy and the United Kingdom, supplementation of myoinositol with folic acid to woman showed preventive effects against NTDs 18 . Previous studies have reported the role of myo-inositol in the management of polycystic ovary syndrome PCOS in women as well as in the prevention and treatment of gestational diabetes mellitus GDM 19,20 .
The anticancer effect of myo-inositol has also been demonstrated in various experimental models 21 . In the present study, we found that PI molecular species in pork chop tissues were most abundant in spinalis muscle, followed by loin, transparent tissue, and intermuscular fat tissue Figs. 3b, 3c , respectively. Further analysis using animal and cell culture models are needed to assess the protective effects of each tissue on diseases such as NTDs, PCOS, GDM, and cancer.
Other studies have reported various bioactive molecules associated with both health-promoting effects and unhealthiness in pork 15 17, 24 . Pork is commonly consumed in processed forms such as steak, ham, and sausage. Therefore, to assess the health-related properties of each pork tissue, spatial analysis of PI molecular species and other bioactive molecules in various types of raw and processed pork is necessary.

Conclusion
We demonstrated that the combination of LC-ESI-MS and MALDI-MSI analysis is useful for the spatial analysis of PI molecular species in pork chop tissues. The analytical approach used in this study as well as the spatial information of PI molecular species provided here would contribute to our knowledge on the beneficial properties, including health benefits, of pork and other meats.