|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 MASS SPECTROMETRIC IMAGING OF LARGE PEPTIDES BY MALDI LIT MS n By DANIEL PATRICK MAGPARANGALAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010
2 2010 Daniel P. Magparangalan
3 To my parents
4 ACKNOWLEDGMENTS There are m any people to whom I owe a debt of gratitude. First I want to thank my advisor, Rick, for accepting me into his research group a nd teaching me how to be a great scientist. The environment Rick fostered allowed me to learn the many facets of mass spectrometry, in addition to skills necessary for a good scientist: how to question and critically think abo ut a problem. And it goes without saying that without the members of the Yost group, both past and present made this experience worth the journey. First is my former manager at Mallinckrodt, David Berberich (a former Yostie) who introduced me to the art of mas s spectrometry Then all of the current and graduated students made the years a joy as thank Tim and Dodge for all the mass spectrometry knowledge you imparted to me over the years. I can only hope I have paid that debt back to the group by helping the current graduate students. push this project to fruition. A lso I would like to thank Mari, Jae, Ed George (recently at) and Jae at Thermo. Their help and support were instrumental in getting this project (and the instrument) on its way. Finally thank you all my friends, family and finally my fiance Shawnee Without your love an d support, it would have been a difficult time indeed.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURE S ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 Background and Significance ................................ ................................ ................. 13 MSI Workflow ................................ ................................ ................................ .......... 15 Post dissection Tissue/Organ Preparation and Storage ................................ ... 16 Tissue Sectioning and Mounting ................................ ................................ ...... 16 Tissue Section Preparation, MALDI Matrix Selection, and Deposition ............. 18 Tissue preparation ................................ ................................ ..................... 18 MALDI matrix selection ................................ ................................ .............. 19 MALDI matrix application methods ................................ ............................ 19 Spatial Resolution: Relationship between Laser Spot Size and Raster Step Size ................................ ................................ ................................ ............... 20 MSI Instrumentation ................................ ................................ ................................ 21 Microprobe Ionization Sources ................................ ................................ ......... 21 Matrix assisted laser desorption/ionization (MALDI) ................................ .. 22 Desorption electrospray ionization (DESI) ................................ ................. 24 Emerging ionization sources ................................ ................................ ...... 25 Mass Analyzers ................................ ................................ ................................ 26 Time of flight (ToF) mass analyzers ................................ .......................... 26 Linear ion trap (LIT) mass analyzers ................................ .......................... 27 Triple quadrupole and other hybrid quadrupole mass analyzers ................ 28 High resolution mass analyzers ................................ ................................ 29 Linear Ion Trap Mass Spectrometry ................................ ................................ ........ 30 Ion Trap Theory and Fundamentals ................................ ................................ 30 Linear Ion Trap (LIT) Description ................................ ................................ ...... 31 MALDI Source for the LIT ................................ ................................ ................. 31 Peptide Fragmentation and Identification ................................ ................................ 32 Amyloid Peptides ................................ ................................ ................................ 33 2 ANALYSIS OF LARGE PEPTIDES BY MALDI USING A LINEAR QUADRUPOLE ION TRAP WITH MASS RANGE EXTENSION ............................ 47 Overview ................................ ................................ ................................ ................. 47 Mass Range Extension on a Linear Ion Trap ................................ .......................... 48
6 Experimental Methods ................................ ................................ ............................ 50 Results and Discussion ................................ ................................ ........................... 52 Full s can Mass Spectra of High MW peptides ................................ .................. 52 MS n Experiments ................................ ................................ .............................. 5 3 Isotopically Labeled Standards ................................ ................................ ......... 54 Conclusions ................................ ................................ ................................ ............ 56 3 IN SITU ANALYSIS OF INTACT LARGE PEPTIDES BY MALDI LIT MS N ............. 62 Introduction ................................ ................................ ................................ ............. 62 Experimental Methods ................................ ................................ ............................ 65 Chemicals and Materials ................................ ................................ .................. 65 Animals ................................ ................................ ................................ ............. 65 Tissue Preparation ................................ ................................ ........................... 66 Instrumentation ................................ ................................ ................................ 67 Data Processing ................................ ................................ ............................... 67 Optical Microscopy ................................ ................................ ........................... 68 Results and Discussion ................................ ................................ ........................... 68 Effects of Sample Preparation on MS Spectra ................................ ................. 68 MSI of Intact A Peptides using a Mass Range Extension ............................... 69 Histological versus MSI Image Resolution of Individual Plaques ..................... 72 Conclusions ................................ ................................ ................................ ............ 74 4 DIGESTION OF LARGE PEPTIDES IN SITU PRIOR TO MSI STUDIES ............... 83 Introduction ................................ ................................ ................................ ............. 83 Experimental Methods ................................ ................................ ............................ 85 Chemical s and Materials ................................ ................................ .................. 85 Animals ................................ ................................ ................................ ............. 86 Tissue Preparation ................................ ................................ ........................... 86 Enzymatic Digestion ................................ ................................ ......................... 87 MALDI Matrix Application ................................ ................................ ................. 87 Inst rumentation ................................ ................................ ................................ 88 Data Processing ................................ ................................ ............................... 88 Results and Discussion ................................ ................................ ........................... 88 Effects of MALDI Matrix Choice for MS Analysis of Enzymatically digested Tissue Sections ................................ ................................ ............................. 88 In situ Tryptic Digestion of A Peptide Standards ................................ ............ 90 In situ Tryptic Digestion of Tg2576 Brain Tissue ................................ .............. 92 Reduction of Analyte Migration within Tissue during in situ Peptide Digestion ................................ ................................ ................................ ....... 94 Conclusions ................................ ................................ ................................ ............ 95 5 CONCLUSIONS AND FUTURE WORK ................................ ............................... 107
7 LIST OF REFERENCES ................................ ................................ ............................. 111 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 120
8 LIST OF TABLES Table page 3 1 Observed amyloid peptides in situ during MSI experiments ............................ 77 4 1 MWs of tryptic fragments originating from A 1 40 ................................ ............ 98
9 LIST OF FIGURES Figure page 1 1 Example of two MS images of phosphatidylcholine (PC) 16:0/16:0 and PC 18:0/18:1 within the same brain section. ................................ ............................ 36 1 2 Workflow of an example MSI experiment. ................................ .......................... 37 1 3 Chemical structures of three common MALDI matrix molecules. ....................... 38 1 4 Illustration of variou s sampling techniques for MSI ................................ ............. 39 1 5 Mathieu stability diagram ................................ ................................ .................... 40 1 6 Sc hematic of a 2D linear ion trap ................................ ................................ ........ 41 1 7 Schematic of MALDI LTQ system ................................ ................................ ....... 41 1 8 Current instrument set up of the MALDI source region. ................................ ...... 42 1 9 Nomenclature for peptide fragments genera ted by tandem mass spectrometry ................................ ................................ ................................ ....... 43 1 10 Illustration of process that converts APP to amyloid beta plaques. .................... 44 1 11 Histological stain of amyloid plaques in a 30 month old Tg2576 mouse. ......... 45 1 12 Cleavage of APP forms amyloid beta fragments of v arying lengths. ................. 46 2 1 Full scan mass spectrum of A 1 40. ................................ ................................ 58 2 2 MS/MS spectra of the [M+H]+ ions of A 1 40 and A 1 42 ............................. 59 2 3 MS 3 and MS 4 experiments for the A 1 40 ion. ................................ ................ 60 2 4 MS and MS 2 mass spectra of a 1:1 mixture of A 1 40 and uniformly 15 N labeled analog ................................ ................................ ................................ .... 61 3 1 Ef fect of tissue washing on a Tg2576 mouse brain section. ............................... 75 3 2 MSI images from the anterior portion of a sagitt al Tg2576 mouse brain section. ................................ ................................ ................................ ............... 76 3 3 MSI images a sagittal Tg2576 mouse brain section. ................................ .......... 78 3 4 Comparison of MS 2 fragmentation of A 1 40 from tissue and an A 1 40 standard ................................ ................................ ................................ ............. 79
10 3 5 MS n image maps of the fragmentation of endogenous A 1 40 ....................... 80 3 6 MS image of A 1 40 ion with three regions expanded ................................ ..... 81 3 7 Effect of step size on an MSI experiment. ................................ .......................... 82 4 1 Image o f trypsin digestion apparatus. ................................ ................................ 96 4 2 In silico digestion of A 1 40 using ProteinProspector ................................ ....... 97 4 3 Comparison of MALDI matrices for analysis of A tryptic fragments .................. 99 4 4 MALDI MS spectrum of a tryptic digestion of A 1 40 atop tissu e using DHB as a MALDI matrix ................................ ................................ ............................ 100 4 5 Mass spectrum of a series of tryptic digestion atop tissue. ............................... 101 4 6 Comparison of mass spectra from two in situ digestion experiments. .............. 102 4 7 MS 2 and MS 3 spectra of the [A 29 40+Na] + ion, m/z 1108.0. ......................... 103 4 8 Comparison of washing a tissue section befor e and after tryptic digestion. ..... 104 4 9 MS 2 and MS 3 images of A 29 40 after digestion of Tg2576 mice brain sections. ................................ ................................ ................................ ........... 105 4 10 ................................ ......... 106
11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MASS SPECTROMETRIC IMAGING OF LARGE PEPTIDES BY MALDI LIT MS n By Daniel Patrick Magparangalan December 2010 Chair: Richard A. Yost Major: Chemistry Mass spectrometric imaging (MSI) is a power ful analytical tool that probes a tissue surface to determine analyte distribution throughout the tissue section. After sampling a tissue se ction at spatially defined positions and col lecting a mass spectrum at each position, three dimensional chemical image maps are co nstructed by plotting each MSI instruments utilize a time of flight mass spectrometer for these experiments. However these instruments are incapable of performing multiple stages of mass spe ctrometry (MS n ). This limitation results in limited structural elu cidation capabilities and lower analyte selectivity Linear ion traps (LIT) are capable of performing MS n ; however, commercial ion traps have a limited upper m/z limit ( m/z 4000). This work describes the extension of the upper mass limit of a n LIT past m/z 4000 for the analysis of amyloid (A ) peptides (molecular weights ranging from 4100 to 4600 Da) for both in vitro and in vivo analysis These peptides have been associ ated with dementia cases Disease. Initial work demonstrated the ability to extend the mass range of the instrument past m/z 4000.
12 A peptides were analyzed in situ using two different strategies Intact peptide analysis was achieved after performing a series of washes that reduced endogenous interfering ions (particularly lipids). The A peptides were located in the cortex and hippocampus regions of the brain. The c omparison of the mass spectral image of A 1 40 in a Tg2576 mouse brain section to a serial section stained with Thioflavin S indicated a good correlation between the two images. Images of the 1 38 and 1 42 fragments also illustrated localization in the cortex and hippocampus regions of the brain In addition, MS 2 and MS 3 images of selected product fragments of the A 1 40 peptide fragment demon strated similar localization. Spatial r esolution was improved by using smaller step size at a cost of longer analysis times. In addition to examining int act A peptides, these peptides were digested to form smaller molecules whose molecular weights were well within the mass range of an unmodified LIT. Various sample preparation strategies were explored and the digested fragment A 29 40 was identified by using MS 3 on the tissue section. However, analyte migration was noted after the tryptic digestion A strategy for reducing analyte migration d uring tissue digestion and the results are presented in this work.
13 CHAPTER 1 INTRODUCTION Background and S igni ficance Peptides and proteins play a major role within the body as messengers, transporters or building blocks that form the structural backbone of cells and tissues  A greater abundance of a specific biomolecule versus health state levels could be the first symptom for a patient that is not exhibiting any symptoms. For example, increased levels of certain amyloid (A ) fragments are suspected to play a major role in the progre The levels of certain A peptides build up over a long period of time; however, the disease may not present any clinical symptoms until treatment is futile. Thus t he analysis of endogenous proteins and peptides is impo rtant for the study of many disease states and determination of new biomarkers. Peptide and protein analysis requires several steps of sample preparation. Analysis of endogenous proteins and peptides are typically performed by homogenizing a tissue samp le extracting the peptides, performing a chromatographic separation, and then analyzing the peptides by mass spectrometry. Based upon the mass spectrometer platform, a digestion step may be necessary as some mass spectrometers have a limited upper molecu lar weight detection limit. During the sample preparation process, however, al l spatial information is lost An alternative analytical method, fluorescence microscopy can localize of specific proteins  ; h owever, this imaging approach lack s the chemical information offered by mass spectrometry and requires tagging the protein of interest A complementary technique to established imaging techniques is mass spectrometric imaging (MSI) [3 5] MSI employs a microscopically focused io nization
14 source to simultaneously desorb and ionize molecules at a spatially defined point on the tissue ; the ions are then transferred to the mass analyzer and a mass spectrum is generated for that point (similar to a pixel) The tissue sample is then r astered undernea th the microprobe and a mass spectrum is collected at each defined interval. Images are generated by selecting an ion of interest and creating a position specific two dimensional ion map typically with a color gradient used to indicate t he third dimension ( ion signal intensity ) Figure 1 1 illustrates the localization of two different endogenous lipids in a coronal rat brain section. These images demonstrate the ability of the mass spectrometer to distinguish two unique molecules in the tissue section and produce spatial information for the user. A significant advantage of MSI over other imaging techniques is molecular selectivity without the need for fluorescently or radio labeled compounds especially when performed in conjunction wit h tandem mass spectrometry or high resolution accurate mass analysis. Although fluorescently/radio labeled compounds offer excellent selectivity, the detection is based on the la bel For example, a pharmaceutical company may be interested in the distribution of a peptide that treats cancer. However, if the fluorescently or radio labeled peptide is metabolized within the body, detection of the metabolite would be impossible if cleav ed from the radio label, and the labeled peptide could not be differentiated from a metabolite that retained the label. The ability to probe discrete sections of tissue for peptides, proteins, and exogenous compounds is a powerful investigative tool for di seased state studies. Compound specific images generated from MSI offer insights into peptide distribution with little sample preparation and high molecular selectivity However, MSI does suffer
15 from a few drawbacks. MSI lacks a separation step that eli minate s a portion of the clutter that inhabits a tissue section (e.g., heme, salts, etc.) Second, MSI is limited to in situ studies. Because experiments cannot be performed on live organisms, the e ffects of a drug compound on the peptide levels in a br ain section cannot be determined in real time as when in vivo studies are performed. Despite these two limitations, MSI offers a complementary analytical technique to established methods to ascertain the biological functions of peptides and proteins withi n disease state tissues A typical workflow for MSI employing ionization by matrix assisted laser desorption/ionization (MALD I) is illustrated in Figure 1 2 Tissue samples are sectioned at a specific thickness and mounted onto a sample stage, after whi ch a MALDI matrix is applied. In the source region, a microprobe interrogates a series of spatially defined spots on the tissue, where the size of the spot is most often defined by the diameter of the focused laser beam (typically 50 200 m). The mass s pectrum is then collected for each spot and saved for future data processing, resulting in mass to charge ( m/z ) specific ion maps illustrating the distribution of the analyte within the tissue  MSI Workflow MSI requires significantly less sample preparation than other imaging techniques such as autoradiography or fluorescence techniques; however, optimizing sample preparation and instrument al parameters is important to generating quality mass spectrometric images. For example, poor tissue h andling can lead to analyte degradation due to enzymatic processes within the tissue section  Several topics such as sample handling and sectioning, matrix selection, and MSI analysis modes will be discussed in the following sections.
16 Post dissect ion Tissue /O rgan P reparation and S torage It is important to immediately preserve the tissue and whole or gans after sacrificing the animal to reduce any post mortem enzymatic processes that occur. Even three minutes is sufficient time for low level peptide s to degrade beyond the point of detection  When surgically removing tissue and whole organs from th e host animal, care must be taken to preserve the original shape of the tis sue, lest spatial information be lost. After excision of the tissue, the tissue may be loosely wrapped in aluminum foil prior to immersion in liquid nitrogen for 30 to 60 seconds. Quick submersion is not suggested, as tissue may undergo tissue cracking and degradation. In addition, direct submersion is discouraged to prevent sample adhesion to the walls of the Dewar. Finally, freshly dissected tissue should not be immediately pla ced in small plastic tubing; otherwise the tissue may mold to conform to the shape of the tube when frozen. Whole tissues and organs can be stored at 80 C for at least a year with little sample degradation  Tissue Sectioning and Mounting Although histological sectioning is typically performed on formalin fixed, paraffin embedded tissue, fresh frozen sectioning of the tissue is the preferred method for MSI experiments. Use of paraffin or OCT (optimal cutting temperature medium ) for tissue mounting to the sample stage can cause ion suppression during MSI data collection  To avoid contamination of the tissue sections from the embedding medium, one can deposit several drops of deionized water to the sample stage and then immediately position the tissue onto the drops, thereby freeze mounting the tissue to the sample stage  Fresh frozen sectioning of tissue thus avoids additional sample preparation steps.
17 Tissue sec tioning is performed in a cryostat with a microtome. The cryostat is held at a temperature ranging from 20 C to 30 C, depending on the tissue type. Tearing of the tissue usually occurs at warmer temperatures; thus, lowering the cryostat temperature should improve tissue section quality. In addition, tissue section thickness can affect the amount of analyte that may be extracted from the tissue (and in turn analyte response) during MALDI experiments because a molecule must travel a longer distan ce wi th in thicker tissue sections (although thicker sections may also provide more analyte)  Tissue sections are transferred to the sample plate (or microscope glass slide) by picking up the tissue with a set of forceps or using an artistic brus h  Additionally, samples may be transferred by placing the plate or microscope slide directly on top of the tissue section and allowing the tissue to naturally adhere to the sam ple plate surface. For time of flight mass spectrometry experiments, a conductive surface must be used to avoid inaccuracies in mass measurement due to surface charging [11, 12] Becau se time of flight mass spectrometers measure the flight time of an ion, any unaccounted changes in kinetic energy of the ion can change its velocity and cause errors in mass analysis. However, o ther mass analyzers (such as ion traps) do not have this limi tation and can use either conductive or non conductive surfaces; thus, conventional glass microscope slides are an option as well as stainless steel MALDI sample plates  Tools used for the transfer and mounting of the tissue should be kept within the cryostat to avoid warming of the tissue. After the tissue section has been transferred to the microscope slide the tissue is gently heated (e.g., placing a finger agai nst the back of the microscope slide), then
18 placed back into the cryostat to thaw mount the tissue to the sample plate surface. After thaw mounting, tissue sections should be stored at 80 C until analysis to prevent any enzymatic processes from occurrin g. Tissue Section Preparation, MALDI Matrix Selection, and Deposition Tissue sample preparation is the last step prior to insertion into the MSI instrument. This section will focus on the selection and application of the MALDI matrix to the tissue section. Although a single general purpose protocol will often generate adeq uate MSI results, ideally each MSI experiment should be individually optimized to tune the experiment for in situ peptide analysis from various tissue types. Tissue preparation Prior to matrix application, the tissue may undergo washing followed by drying in a desiccator to remove excess water. The washing step may be performed to remove endogenous salts that can potentially inhibit MALDI matrix crystal formation. In addition, washing t he tissue can remove some endogenous compounds (e.g., lipids) and imp rove signal to noise. For example, a tissue section can be rinsed (or immersed) in 80% to 100% ethanol for 30 seconds  However, care must be taken to limit the immersion, as analytes may migrate or worse, wash away. As a final step prior to matrix application, the tissue section is placed in a desiccator or under vacuum to remove any excess w ater (typically from 30 minute s to an hour). Additional water in the tissue sample may result in un even matrix solvent evaporation, which may lead to uneven co crystallization of the matrix and analyte. Uneven co crystallization may that lead to regions of artificially higher signal.
19 MALDI matrix selection Two properties are necessary for performing optimal MALDI experiments: 1) the matrix should absorb at the wavelength of the laser to limit in source fragmentation and 2) the matrix should co crystallize with the target analyte so that the analyte is desorbed into the gas phase upon ablation of the matrix. UV absorbing compounds that are commonly used with MALDI, and in turn with MSI, are 2,5 dihydroxybenzoic acid ( DHB ) cyano 4 h ydroxycinnamic acid ( CHCA ), and 3,5 dimethoxy 4 hydroxycinnamic acid or sina pinic acid (SA)  Figure 1 3 illustrates the chemical structure of these three compounds. Selection of a specific MALDI matrix is often based upon the analyte of interest. SA is typically used for proteins and large peptides, CHCA is typically used for smaller molecules, and DHB is used as a general purpose matrix. Ho wever, a drawback to using all matrices is the abundance of low m/z matrix and cluster ions. Alternative higher molecular weight matrices such as porphyrins  also absorb in the UV region but produce less interference in the low m/z region  Other options include the use of ionic matrices, which have been reported to have the advantage of conti nuous matrix coverage over a tissue section [15, 16] MALDI matrix applicatio n methods Crystal size and coating uniformity play a larger role for MSI applications than standard MALDI experim ents. Spatial resolution (the effective pixel dimension of an image) is limited not just by the raster step size of the tissue and the diamete r of the laser spot, but also by the size of individual MALD I matrix crystals Thus DHB, with its elongated crystals, potentially offers a poorer spatial resolution than CHCA or SA, which have small spherical crystals. In addition, coating uniformity wil l affect the variability of the ion signal across the tissue An excess of matrix in one location can potentially lead
20 to an unexpected enhancement or suppression of the ion signal. Uniform matrix deposition should reduce any matrix related variations in signal. MALDI matrix coating uniformity and crystal size can be optimized by the choice of matrix the MALDI matrix solvent solution, and the MALDI matrix application technique. Although dried droplet matrix application requires the least amount of equip ment (a pipettor), it is by far the most irreproducible and time consuming of the matrix deposition techniques. Other c ommon matrix application techniques for MSI found in the litera ture include pneumatic spraying  inkjet printing  acoustic matrix depositi on  sublimation  and sol vent free matrix dry coating  Of these methods, pneumatic spraying is by far the most common technique. Pneumatic spraying can be performed with a TLC sprayer, a Meinhard nebulizer or even an artistic airbrush, as each can produce even, homogenous layers of small matrix crystals across the entire tis sue section  Commercial devices can further increase matrix crystal reproducibility by automating the matrix coating pro cess. Spatial Resolution: Relationship between Laser Spot Size and Raster Step Size Selection of a specific MSI sampling methodology is based upon the size of the tissue sample and information needed. Figure 1 4 illustrates the various MSI image collection strategies available. Typically, the raster step size in an imaging experiment is set equivalent to the spot size of the laser. The spatial resolution is then limited by th e spot size of the laser (assuming that the MALDI matrix crystals are not larger than the spot size). To increase the spatial resolution, the raster step size can b e decreased During such an oversampling experiment, the raster step size is reduced to s maller than the diameter of the laser spot. This has a disadvantage, however of increasing the
21 analysis time. However, the increase in analysis time can be mitigated by using a more directed approach of interrogating a smaller region of the tissue. I n some cases, the raster step size is increased to be greater than the diameter of the laser spot. Whole body tissue sections commonly utilize such an approach due to the large area that must be examined. For example, Khatib Shahidi, et al.  used a raster step size of 500 m for analyzing whole body rat tissue sections. A lthough analysis time decreases by in increasing the raster step spatial resolution is reduced. MSI Instrumentation Two primary components are necessary for MSI, 1) a microprobe ( i.e., a microscopically focused ionization source to generate ions at a spat ially defined point in the tissue section), and 2) a mass analyzer. In the following sections, various microprobe ion sources and mass analyzers typically associated with MSI are presented. In additi on, sample preparation and post acquisition data proces sing techniques are discussed. Microprobe Ionization S ources Three aspects that must be considered when selecting an ionization source are ionization efficiency, spatial resolution, and sampling rate. In general, it is important to optimize ionization eff iciency to generate the greatest number of ions from a given spot without causing in source fragmentation The size of the microprobe spot defines the spatial resolution, sensitivity, and overall image quality of the MS image. Finally, the sampling rate w ill affect the overall analysis time for each tissue section. Optimization of each parameter is important; however, the interconnectivity of the parameters needs to be considered. For example, decreasing the spot size to increase spatial resolution will result in less analyte ablated, thus increasing the detection limit for
22 the instrument (provided that the laser fluence remains constant) Furthermore, a decrease in step size results in an increase in total analysis time (for example given identical sca n areas, a 50% decrease of step size would result in a four times increase in analysis time). The three most common sources for MSI are matrix assisted laser desorption/ionization (MALDI) secondary ion mass spectrometry (SIMS) and desorption electrospray ionization (DESI). Although SIMS offers spatial resolution superior to the other two sources, SIMS does not produce intact ions for peptides in the mass range of interest in this study. For this reason, SIMS will not be discussed here (although more inf ormation can be found in a review paper by Solon et al. )  The following sections will focus on the two ionization sources (MAL DI and DESI) most commonly used for MSI as well as emerging ionization sources. Matrix a ssisted l aser d esorption/ i onization (MALDI) The most commonly used ionization source for MSI is matrix assisted laser desorption/ionization (MALDI). MALDI is a soft ionization technique that employs a laser to produce predominantly singly charged ions [24, 25] The key to MALDI is the application of a matrix to the sample. Pri or to the introduction of MALDI laser desorption of intact ana lytes from tiss ue was difficult due to the harsh ionization proce ss that caused a high degree of fragmentation high energies needed for desorption of the molecules into the gas phase. Karas and Hillenkamp found that addition of a large excess of an organi c matrix that absorbs at the laser wavelength greatly increased ionization efficiency and as well as the likelihood of forming intact molecular ion species. With this discovery, the analysis of high molecular weight biomolecules (e.g., peptides and protei ns) became the most common application for MALDI
23 The MALDI matrix typically employed is a small organic acid that co crystallizes with the analyte and absorbs light at the wavelength of the laser. As the majority of the energy is absorbed by the organic matrix, sample fragmentation is minimized. MALDI matrices typically used are 2,5 dihydroxybenzoic acid (DHB) c yano 4 hydroxycinnamic acid (CHCA), and s inapinic acid (SA). The selection of a specific MALDI matrix is based on the tissue type and analyte of interest. Further discussion of MALDI matrix selection can be found in the tissue preparation section During the laser ablation of the matrix surface, charged analytes (i.e., ions) are generated and then transferred to the mass spectrometer. The singly charged ions produced by MALDI are preferred as the mass spectrum is cleaner and easier to interpret (as opposed to an ESI source that can generate a variety of multiply charged species). This is important for tissue analysis as there is a wide abundance of endogenous molecules MALDI experiments typically use a pulsed UV laser ( e.g., N 2 laser with a 337 nm wavelength) as the ionization source. The laser energy and th e number of laser shots required for a MALDI experiment are optimized with respect to MALDI matrix selection and tissue type for each experiment. For example, DHB typically requires more laser  Although the repetition rate of the laser does not play a direct role in the image quality, higher repetition rates increase sample throughput, with optimal rates employed near the scan rate of the mass analyzer. This is important when performing experiments on larger tissue sections, such as MSI over whole body tissue sections. For example depending upon the rastering pattern and repetition rate of the laser,
24 whole body tissue experiments can take upwards of 5 days (~115,000 pixels over the entire tissue section at a lateral resolution of 250 m with an acquisition rate of ~4 s /pixel )  Desorption e lectrospray i onization (DESI) A recent alternative method to MALDI is desorption electrospray ionization (DESI) [27, 28] an atmospheric pressure surface sampling technique that, unlike MALDI, does not requi re a matrix Charged droplets are emit ted from the electrospray source and strike the tissue surface at an angle, thus desorbing compounds as ions that are then transferred into the mass spectr omet er. The tissue surface is rastered beneath the DESI emitter  The primary advantages of the DESI technique are that there is little sample preparation required (no MALDI matrix application) and experiments are performed a t atmospheric pressure. A further advantage is that specimen size is not limited, so subdivision of whole body tissue sections is unnecessary for DESI MS experiment. In contrast, a MALDI MSI experiment may require the division of a whole body section int o pieces that will fit onto the MALDI target plate. This may require computer stitching programs that can degrade image quality. A disadvantage of DESI is that many parameters must be optimized prior to a DESI experiment (multiple spray angles and sample distance from both the ESI emitter and to the mass spectrometer)  Also, the sampling rate must be optimized to balance length of analysis time vs. signal response. In addition the image quality is much lower than MALDI MSI experim ents due to the poor spatial resolution (250 m)  Finally inaccurate localization of desorbed ions is a possibility due to the DESI process. For example during a scan performed from left to right on a surface, it was
25 noted that analytes were washed toward the right  The washing affect is due to the impact of the DESI plume onto the surface as well as analyte surface interaction, which can result in an MSI representation that is shifted in the direction of the DESI spray  Despite these challenges, DESI has potential for increased use within the MSI field. Emerging i onization s ources An advantage of atmosp heric pressure ionization sources for surface sample analysis is the ability to sample a surface without moving it into the vacuum region, which is usually small t o limit the volume to be evacuated for a standard MALDI target plate, typically ~8 x ~13 cm. Another alternative atmospheric pressure source amendable to MSI experiments is a liquid microjunction s urface sampling probe (LMJ SSP) [33, 34] This probe consists of an inner and outer sampling tube that forms a microjunction at the surface of the tissue section as solvent is delivered to the surface and then drawn away toward the mass spectrometer. Thus, the effective resolution of this probe is the diameter of the sampling tube (~500 m) Thi s technique has been used to sample various organs in a whole body tissue sections; however, no MS images have been published using this technique [33, 34] A disadvantage of this technique is that the sampling solvent will bias against analytes not compatible with the solvent. As a consequence, incorrect assumptions may be drawn if ion maps of two analytes are compared. Another alternative microprobe source is n anostructure initiator mass spectrom etry (NIMS)  Although it is not an atmosphe ric pressure technique, this technique has a similar advantage as SIMS in that no matrix application step is necessary. But unlike SIMS, NIMS is a soft ionization technique; thus, analytes are less likely to undergo fragmentation during the ionization pro cess. And with a similar lateral
26 spatial resolution (150 m), NIMS has the potential to generate image maps comparable to MALDI MSI. This technique however is still relatively new; thus, only one application of NIMS to drugs and metabolites has appeared  Mass A nalyzers Several mass analyzers are in use for MSI experiments. The most common mass analyzer for MSI found in the literature is the time of flight (ToF) mass analyzer  However, other mass analyzers offer their own advantages that should not be discounted. For experiments examining compound s in a matrix as complex as tissue, tandem mass spectrometry and/or high resolution mass spectrometry is almost a necessity. This is due to lack of clean up separation steps prior to mass analysis, such as extraction and LC prior to MS for removal of interfering compounds. Therefore, linear ion traps (LIT)  Fourier transform ion cyclotron resonance (FT ICR)  o rbitrap  QqToF  and ToF/ToF  instruments are effective when performing MSI experiments for peptide studies in tissue. Time of flight (ToF) mass analyzers ToF mass analyzers offer rea sonably high mass accuracy (< 20 ppm), mass resolution (> 10,000) and a large m/z window (up to m/z 10,000 in reflectron mode and up to m/z sample throughput when paired with a high repetition rate laser. In addition, the pulsed nature of the MALDI laser is well suited for use with a ToF, as a complete mass spectrum can be recorded for each laser pulse. However, these advantages are balanced by an increased number of laser shots necessary to obtain an adequate signal t o noise ratio due to the spot to spot variability noted in MALDI experiments
27 (typically 80 to over 700 las er shots per spot in MS mode and 400 to 8000 laser shots in MS/MS mode ) [37, 40] A second drawback of the ToF mass analyzer is the limited structural elucidation capabilities and limited ability to distinguish between analyte and interfering species at the same nominal m/z (isobaric species). Most MSI experiments employ a single ToF analyzer, and thus provide only molecular weight information ( no structural information ). At most, a ToF instrument can provide two stages of mass spectrometry by coupling a se cond ToF analyzer to form a ToF/ToF  or adding a quadrupole mass filter to form a QqToF [22, 37 41 44] QqToF and ToF/ToF instruments offer higher analyte selectivity than stand alone ToFs because they provide collision induced dissociation (CID) for monitoring product ions in MS/MS. However in some cases, MS 3 (or even further stages) ma between isobaric species  In this case a n ion trap is necessary. Linear i on t rap (LIT) mass analyzers Although it lacks the wide mass range of a ToF mass analyzer, an ion trap (either a 3 D quadrupole ion trap or a linear ion trap) offers advanced structural elucidation and analyte selectivity by providing two or more stages of tandem mass spectr ometry (MS/MS and MS n ) capabilities. Some of the earliest MALDI MSI experiments were performed on a 3 D quadrupole ion trap, studying the levels of pacl itaxil spiked into rat liver sections  Selecting an ion of interest and inducing fragmentation enables the identification of an analyte or differentiation from a potentially interfering peak  Positive identification is especially important in MALDI analysis as there are typically many interfering MALDI matrix related peaks after matrix application This overlap makes distinguis hing analyte from matrix difficult or impossible, par ticularly at trace
28 levels An additional advantage is that interme diate pressure MALDI LIT (~70 mT orr) requires fewer laser shots per pixel (~10 laser shots or less)  This both decreases analysis time and reduces the amount of matrix ablated from the tissue surface, w hich in turn increases the number of times that a tissue sample may be analyzed. Ion traps are not without drawbacks. Most commercial systems have an upper m/z limit of 4000 (although here we have modified the linear ion trap to scan up to m/z 5500 )  This upper m/z cutoff could preclude detection of peptide/protein based therapeutic drugs or detection of proteins affected by the administration of a drug. In this case, on tissue enzymatic digestion is an option  However, most drug molecules have molecular weights of less than 1000 Da, and thus are well within the m/z range of the linear ion trap instrument. In addition to the m/z limitation, space charging of ions within the trap can cause mass shifts in the m ass spectrum. Although using an LIT as opposed to a 3 D trap largely alleviates the space charging issues  highly abundant background compounds (such as those found during tissue imaging experiments) may impact analyte identification. Despite these challenges, ion traps are a remarkably useful tool for mass spectrometric images because of the ease with which they perform tandem mass spectrometry ( MS/MS and MS n ) Triple quadrupole and other hybrid quadrupole mass anal yzers Although triple quadrupole (QqQ) mass analyzers are widely utilized for quantitation QqQ instruments are not typically coupled to a MALDI source. An early laser microprobe system employed laser desorption on a QqQ system  The strength of a QqQ instrument is the use of MS/MS to perform selected reaction monitoring (SRM) and/or multiple reaction monitoring (MRM) experiments. In particular,
29 MRM scans on a QqQ instrument allow for high selectivity for detection of a compound with 100% duty cycle. Coupling a high repetition rate laser (1kHz) to a QqQ instrument increases sample throughput [49, 50] Whereas there are few MALDI QqQ instruments, hybri d quadrupole instruments have been employed for MSI experiments, for example a QqLIT [51, 52] Here, a linear ion trap replaces Q 3 to serve as either a mass filter or an ion trap. This is key because tandem mass spectrometry is almost a necessity for peptide analysis due to the ab undance of interfering ions in tissue and the need for structural elucidation The QqToF instrument, mentioned previously, is another example of a quadrupole h ybrid instrument. High resolution mass analyzers An alternative method to positively identify an analyte is accurate mass analysis with a high resolution mass analyzer (mass resolution > 100,00 0). Two options are to use an o rbitrap [9, 53] or an FT ICR  instrument. Both mass analyzers offer superior mass accuracy (< 2 ppm) and mass resolution, when compared to an ion trap or ToF. The higher mass accuracy and mass resolution often allows the user to distinguish between analyte and background peaks at the same nominal mass  This is of particular importance with peptide studies, due to the number of isobaric endogenous species in tissue One drawback to using an instrument with high mass accuracy is the low sample throughput for the instrument. Sample throughput for MSI experiments can be defined as the number of pixels (i.e., spatially defined data points) per unit time. ToF instruments typically provide spatially defined mas s spectra at a rate of 30 50 pixels/min  with ion traps having a similar throughput (60 70 pixels/min)  In
30 contrast, four pixel s/min is the norm for an FT ICR experiment. Thus for a profiling experiment, it could take eight hours, or even more at higher spatial resolution, to obtain FT ICR mass spectrometric data over an entire tissue section even for a modest 1700 pixels  An o rbitrap experiment requires much less time than an FT ICR (25 pixels/min, especially if the scanned mass range is narrowed); however, the sample throughput is still not comparable to either a ToF or an LIT [9 ] Linear Ion Trap Mass Spectrometry Ion Trap Theory and Fundamentals The linear ion trap is a complex mass analyzer that relies upon oscillating radio frequency (RF) pulses to trap ions injected into the trap. The potential of an ion to be trapped wit hin the RF field is illustrated by the Mathieu stability diagram (Figure 1 5)  This figure illustrates the solutions to the reduced M athieu equation (Equations 1 1 and 1 2) [56 58] a x = a y = 8 eU/mr 0 2 2 ( Equation 1 1 ) q x = 2q y = 4 eV /m r 0 2 2 ( Equation 1 2 ) Where a and q describe dimensionless parameters that describe regions where an ion is stable within the ion trap V is the zero to peak RF peak voltage, U describes the DC potential placed on th e quadrupole rods, is the drive frequency of the RF voltage, r 0 is the internal radio of the LIT, and m is the ion mass. After trapping, ions are cooled to the center of the trap using a buffer gas (typically helium). These ion s are stored within the t rap until they reach a boundary and are ejected from the trap. To simplify the DC power supplies
31 are typically eliminated. Thus the a x term is set to zero and thus ions fall along the q x axis. Initial commercial ion traps utilized the right hand edge of the ion trap to eject ions (q eject of 0.908). This however led to poor mass resolution. In addition, use of a q eject of 0.908 limited the ion trap to an upper m/z limit of ~650. Later discoveries det ermined that a supplementary voltage applied across the end caps of a three dimensional ion trap created a hole in the RF field  Ions scanned along the q x line wo uld leave the trap upon reaching that hole. This type of scan (mass selective instability scan with resonant ejection ) dramatically improved mass resolution and increased the upper mass limit of an ion trap to m/z 2000 when a q eject of 0.88 is employed. L inear Ion Trap (LIT) Description Improvements of trapping design lead to the development of the 2D ion trap. The 2D ion trap consists of three sections, (Figure 1 6) [ 47] Ions enter through the front section of the ion trap and then are trapped within the center section. During ion trapping and isolation, the front and back sections of the LIT are held at a higher positive voltage (for positive ions) or negative v oltage (for negative ions) to reduce ion leakage through the axial ends of the trap. During ejection, ions a re ejected through two slits in the center section rods and are detected by two electron multipliers. Compared to a 3D trap, the 2D trap holds mor e ions due to the stretched geometry in the z direction. Also more ions are detected as ions are ejected from both sides of the trap (radially). MALDI Source for the LIT A conventio nal ESI LIT has two sets of qua d r upole rods in front of the LIT to guide ions from the high pr essure source region to the low pressure trapping region.
32 Modifications of the first set of quadrupole rods (q00) were necessary to permit MALDI operation. Figure 1 7 illustrates the MALDI modification to the LIT. A hole was bored into one of the q00 rods to allow the laser beam to strike the sample plate. To accelerate the ion passage through q00, a set of auxiliary rods extract s ions produced at the sample surface and accelerate s them down toward q0. This is different from the ESI LIT where the pressure differential is sufficient to draw any ions toward the lower pressure region of the instrument. Initial versions of the LIT system in the lab utilized a fiber optic cable to transmit laser light to the sample plate. This how ever was found to be insufficient for experiments as it was necessary to perform constant alignment checks to achieve optimal laser transmission. The current laboratory setup utilizes a series of mirrors, neutral density filters and lenses to direct, focu s and attenuate the laser to the sample plate surface (Figure 1 8). The instrument is fitted with a 337 nm, 6 0 H z N 2 laser ( model MNL 100; LTB Lasertechnik Berlin GmbH ) The laser spot size for this instrument is approximately 100 m with an upper laser energy transmission of 70 J per pulse. However, laser energi es ranging from 2 J to 10 J are sufficient for most MSI experiments. Peptide Fragmentation and Identification Struc tural elucidation of peptides can be performed by fragmentation of a peptide ion and subsequently analyzing the resulting peptide fragment ions Fragmentation of the peptide ion into smaller fragments is commonly performed by collision induced dissociation (CID)  In a CID experiment, parent ions are isolated within the trap, after which activation energy is applied across the quadrupole rods at a frequency corresponding to the oscillation frequency of the ion. The amplitude of the applied
33 activation energy is enough to increase the amplitude of ion oscillation from the center of the trap, but not enough to eject it from t he trap. The ions will collide with the helium buffer gas, picking up enough internal energy to fragment and form characteristic product ions which are subsequently ejected from the trap to create a product ion MS/MS spectrum Figure 1 9 illustrates an exa mple of MS /MS fragmentation of a five amino acid residue peptide Fragmentation of a peptide generally takes place at one of the bonds along the peptide backbone (indicated by the dotted lines). After fragmentation, one or more of six types of fragme nt ions may be generated termed a, b, c, x, y, or z type ions. The most common ions formed from a low energy CID experiment are b and y type ions. In this example if a cleavage occurs at the dotted line, a protonated b 3 (Figure 1 9B) and y 2 (Figure 1 9 C) ion may result. The subscript number indicates the number of amino acid residues remaining in the peptide fragment. Multiple stages of mass spectrometry may be performed to glean further information on A lternatives to CID experim ents include elec tron capture dissociation (ECD) and electron transfer dissociation (ETD) [60, 61] These fragmentation e xperiments can offer further information of the peptide backbone, such as sites of phosphorylation or gly cosylation and typically generat e c and z type ions. Examination of the ions resulting from a tandem mass spectrometry experiment allows scientists to perform structural elucidation and identification of a peptide. Amyloid Peptides Amyloid (A ) peptides originate from the amyloid precurs or protein (APP). In the normal (non AD) brain APP is cleaved and eventually cleared away. However, it is
34 is affected and strands of A 1 42 are preferential ly formed  These fragments are aggregate more often than their 1 37, 1 38, and 1 40 isofo rms. Figure 1 10 illustrates the suggested formation of the amyloid plaques  Over a period of time, these fragments aggregate into soluble oligomers that eventually aggregate into insoluble pl aques (Figure 1 11 see chapter 3 for more information ). Figure 1 12 illustrates the potential fragments formed from APP and the enzym es involved with the cleavage  The A hypothesis suggests that these plaques are responsible for the progression of AD  Several methods are available for the analysis of A peptides. Thioflavin S is a common stain that stains the plaqu es green under a confocal fluorescence microscope as demonstrated in Figure 1 11  Selectivity for individual amyloi d isoforms can be achieved if a fluorescently tagged antibody is used  Chemical information can be obtained if tissue sections containing the plaques are first digested and then LC MS or MALDI MS is performed These techniques are adeq uate if spatial or chemical information is needed. However, histological methods lack any chemical information of the tissues surrounding the A plaques. Conversely, typical MS experiments lack spatial information due to the sample preparation steps nece ssary for MS analysis. MSI experiments on a ToF instrument have been demonstrated for A plaques [67 69] however ToF instruments lack MS n for peptide sequencing. To resolve these limitations, this dissertation will present work demonstrating the ability of a linear ion trap to analyze and locate the se high molecular weight peptides
35 within the tissue section. Several strategies will be offered for A peptide analysis that may be applied to other peptide studies.
36 Figure 1 1. Example of two MS images of p hosphatidylcholine (PC) 16:0/16:0 and PC 18:0/ 18:1 within the same brain section. T andem m ass spectrometry confirmed the identity of the two lipids.
37 Figure 1 2 Workfl ow of a n example MSI experiment. In this case (1) the brain is sectioned at a thi ckness of 10 m. Following sectioning the brain section is prepared (2) for MALDI MS analysis. Washing and digestion are two optional steps, performed if the method requir es. Finally, t he MALDI matrix application is performed prior to insertion into the mas s spectrometer. As the tissue section rasters (3) beneath the laser, a mass spectrum is collected at each spot (4) Following MS analysis, ions are selected for imagin g (5).
38 Figure 1 3. Chemical structures of three common MALDI matrix molecules.
39 Figure 1 4. Illustration of various sampling techniques for MSI. Typically the tissue is mov ed at a raster step size equivalent to the laser spot size, A. However for cases when a large area must be analyzed, undersampling may be necessary to avoid unacceptably long analysis times, B. Undersampling is any case where the raster step size is grea ter than the laser spot size. For cases where the molecule may reside in areas smaller than the spot size of the laser, oversampling may be useful In this case, the raster step size is decreased to less than that of the laser spot size. Oversampling ca n improve image resolution at the cost of dramatically increasing analysis times.
40 Figure 1 5 Mathieu stability diagram Ions injected inside the trap are trapped if the trapping field and ion characteristics lie within the shaded region of the dia gram Adapted from Douglas et al. 
41 Figure 1 6 Schematic of a 2D linear ion trap. Ions are trapped within the center section of the trap and then subsequently radially ejected through two slits in the center section Adapted from Schwartz et al.  Figure 1 7 Schematic of MALDI LTQ system. The box region indicates the area where the MALDI modification occurred Adapted from Garrett et al. 
42 Figure 1 8 Current instrument setup of the MALDI source region. A series of mirrors, neutral density (ND) filters and lenses guide and attenuate the laser beam to the samp le plate Adapted from Strupat et al. 
43 Figure 1 9. Nomenclature for peptide fragmen ts generated by tandem mass spectrometry. The top structure A, illustrates an example peptide with five amino acid residues (R 1 through R 5 ). If fragmentation occurs at the dotted line denoted by the arrow, a protonated b 3 B, or y 2 C, ion may result.
44 Figure 1 10 Illustration of process that converts APP to amyloid beta plaques Adapted from the National Institute on Aging 
45 Figure 1 11 Histological stain of amyloid plaques in a 30 month old Tg2576 mouse.
46 Figure 1 1 2 Cleavage of APP forms amyloid beta fragments of varying lengths A plaques primarily cons ist of A 1 40 and 1 42 peptides (indicated by the red box). Adapted from Ford et al. 
47 CHAPTER 2 ANALYSIS OF LARGE PE PTIDES BY MALDI USIN G A LINEAR QUADRUPOL E ION TRAP WITH MASS RANGE EXTENSION Overview Analysis of large peptides can be used to discover or to monitor biomarkers for various diseases. For example, the levels of such peptides can determine the effectiveness of an experimental drug or the progress of a disease. Many mass spectrometric methods for monitoring these peptides use MALDI ToF instrument s due to their high molecular weights, although such instruments typically lack MS/MS or MS n capabilities. Here, the m/z range of a MALDI LIT instrument was extended to m/z 5500 for the MS or MS n analysis of large peptides. Instrument performance was exami ned using amyloid 1 40 and 1 42 (avg. MW 4330.8 and 4515.0, respectively), large peptides that comprise the bulk of neuritic plaques and are potential biomarkers for 1 40 was detected in the full scan mass spectrum wit h sufficient resolution to distinguish and match the expected isotopic pattern. The MS/MS product ion spectra of both peptides matched the expected fragmentation patterns; up to MS 4 experiments were performed to verify the identity of the peptides. These experiments clearly demonstrate the advantages of this approach, including MS n experiments for structural elucidation and simplified spectra due to singly charged parent ions, for large peptides. Reproduced in part with permission from Analytical Chemistry, Vol. 82 (3), Magparangalan, D. P., Garrett, T. G., Drexler, D. M., Yost, R. A. Pages 930 934. Copyright 2010 Analytical Chemistry
48 Mass Range Extension on a Linear Ion Trap There is much int erest in large peptides as biomarkers for various diseases. The [70 72] The analyses of these biomarkers are typically performed using mass spectrometry. Mass spectrometry offers accurate mass analysis and analyte fragmentation these peptides can have a high molecular weight, the ionization techniques typically emp loyed are electrospray ionization (ESI) and matrix assisted laser desorption/ionization (MALDI) [64, 70, 71, 73, 74] ESI generates multiply cha rged ions that are then transferred to the mass analyzer. Thus, large peptides and proteins can be detected using mass analyzers with a limited mass to charge ( m/z ) range. Conversely, MALDI predominately generates singly charged ions. The choice of ioni m/z cutoff. In addition to accurate mass analysis and enhanced mass resolution, time of flight (ToF) mass spectrometers have a high upper m/z cutoff (up to m/z ~ 10,000 in reflectron mode) that e nables the coupling of a MALDI source to observe high MW singly charged ions. Unfortunately, MS/MS capabilities for structural elucidation are limited with a ToF mass spectrometer; at best, it can perform MS 2 and that typically requires a ToF ToF tandem instrument. In comparison, linear ion trap (LIT) mass spectrometers provide multiple stages of MS (MS n ), which enhance selectivity, structural elucidation, and identification. However, an LIT has a lower m/z cutoff (typically m/z 2000 or 4000) compared to a ToF. When ESI sources are coupled to LITs, this is not typically a limitation, as analytes
49 introduced into the MS are multiply charged species whose m/z falls within the LIT m/z window. One disadvantage of multiple charging is increased complexity of the MS and MS n spectra. Here we evaluate a MALDI LIT MS with an extended m/z range. There are several techniques for extending the mass range of an LIT. The following equation gives the upper m/z limit, ( m/z ) max of an LIT operated in mass selective instability mode [ 56] : where V max is the zero to peak RF peak voltage level, r 0 is the internal radius of LIT, 0 is the drive frequency of the RF voltage, and q eject is the point on the stability diagram where the ion becomes unstable and is ejected. From equation 1, the m/z range can be extended in four ways: (i) increasing the maximum RF voltage, (ii) decreasing the trapping dimensions, (iii) decreasing the RF driv e frequency, or (iv) decreasing the q eject  For these studies, the q eject was lowered to achi eve a higher m/z cutoff ( m/z 5500). To test the effectiveness of the instrument modification for the analysis of large peptides, fragments of the amyloid peptide population were selected. Amyloid peptides (A ) comprise the bulk of neuritic plaques tha disease [76, 77] These peptides are produced by the sequential cleavage of the amyloid precursor protein by the aspartyl protease secretase and the presenilin dependent protease secretase  An analytical technique that can monitor the
50 l evels of such peptide fragments could be used for biomarker determination in [79, 80] This chapter describes the effect of lowering the qeject of the LIT mass spectrometer to facilitate analysis of singly charged amyloid peptide fragments. This approach allows for multiple stages of fragmentation (MSn) to be performed in order to enhance selectivity and structural elucidation. The use of MAL DI for ionization offers the potential for mass spectrometric imaging of these markers in intact tissue. Experimental Methods Sinapinic acid (SA, 3,5 dimethoxy 4 hydroxycinnamic acid) was purchased from Sigma Aldrich (St. Louis, MO) and used as the MALDI m atrix for all analyses. Ammonium hydroxide and HPLC grade acetonitrile and water were purchased from Fisher Scientific (Fairlawn, NJ). Trifluoroacetic acid (TFA) was purchased from Acros Organics (Morris Plains, NJ). The analytes [A 1 40 and A 1 42 a nd ProteoMass TM MALDI calibration kit were purchased from Sigma Aldrich. Uniformly 15 N labeled A 1 40 and 15 N labeled A 1 42 were purchased from rPeptide (Athens, GA). Analytes were prepared at a concentration of 100 g/mL in 1% ammonium hydroxide (aq) Sinapinic acid MALDI matrix solution was prepared at a concentration of 20 mg/mL in 50:50:0.1 acetonitrile:water:TFA (v:v:v). The dried droplet method was employed to deposit 1.5 L of sample analyte and 1.5 L of matrix onto the MALDI well plate. A F innigan LTQ linear ion trap fitted with an intermediate pressure vMALDI source (San Jose, CA) was used for all experiments. The MALDI source has a nitrogen laser (337 nm) with a frequency of 20 Hz and maximum energy of 250 J/pulse. For collection of ful l scan mas s spectra, a laser power of 12% was used so as to limit
51 space charging of the peptide ions within the trap (laser output for the vMALDI instrument is measured as % laser power) To increase the number of ions available for MS n experiments, the l aser power was increased to 25%; an isolation window of 12 AMU and a CID energy of 20 (arb) during the MS n experiments was required for adequate sensitivity at this extended mass range. A more detailed description of this instrument has been previously pu blished  Each MS scan consisted of 3 laser shots, and approximately 50 scans were averaged to generate a mass spectrum for the sample spot. The sample was autom atically moved in a circular raster pattern underneath the laser spot. Simulated spectra were generated using Qual Browser software (ver. 2.0.7) from Thermo Fisher. Expected fragment ions were determined using ProteinProspector  When scanning in normal mass mode (up to m/z 2000), the instrument has a q eject of 0.88  In high mass mode, the q eject is lowered to 0.44, w hich allows for analysis of ions of m/z 150 to 4000. In order to analyze ions beyond m/z 4000, the q eject was further lowered to 0.30. This allows for analysis of ions from m/z 300 to 5500. Although a decrease in q eject should allow for an upper mass limit of over 5800, the instrument software limits the mass range of the instrument to 5500. The q eject value was lowered through an engineering software patch provided by Thermo. Lowering the q eject value is related to two parameters that are derived from q, namely and  To lower q eject, a reduced value and a lower resonance ejection frequency ( during the mass selective instability scan are required. For a q eject of 0.300, the corresponding is approximately 0.2161 an d the resonance ejection frequency is 130.61 kHz. (For
52 comparison, at a qeject of 0.44, is equal to approximately 0.3244 and the resonance ejection frequency is 196.07 kHz.) Since the mass range was extended past the normal mass calibration range, a r ough two point calibration was performed using bradykinin (Sigma Aldrich) and A 1 42. Afterward, the instrument was calibrated using the high mass standard within the calibration kit with A 1 42 added. Results and Discussion Two issues were addressed to extend the m/z range of the instrument beyond m/z 4000 space charging within the trap and mass resolution. Both issues are related, as increased space charge decreases mass resolution. Although the LIT has a larger volume to collect ions when compared to a 3 D Paul trap, there is still a finite number of ions that can be collected before loss of resolution due to space charging occurs  Space charge effects can be overcome by tailoring the number of laser shots per spot and/or the % laser power of each shot. In these studies, the optimum full scan MS conditions to prevent space charging were three laser shots at 12% laser power. Full scan Mass Spectra of H igh MW peptides After the instrument conditions (laser power, number of laser shots) were optimized, the full scan spectrum of A 1 40 was acquired with the instrument (Figure 2 1). The monoisotopic m/z of the [M+H] + A 1 40 ion is 4327.3. The isotope pattern of the recorded A 1 40 peptide spectrum (Figure 2 1A) closely matches the isotope pattern of the simulated spectrum (Figure 2 1B). The simulated spectrum was generated using the QualBrowser software included with the LTQ instrument. The
53 mass resolution for the [M+H] + ion of the peptide was determined to be 6000 FWHM (full width at half max). The detection of amyloid peptide with an AP/MALDI 3 D ion trap has been previously reported  However, only the +2 charge state was detected due to the upper mass limit of the ion trap. With this instrument, the singly charged ion is detected in the full scan mass spectrum with little int erference from the background. MS n Experiments To take full advantage of the capabilities of the LIT, MS 2 experiments were performed on the [M+H] + ion of both the A and A peptides (Figure 2 2). MS Product within ProteinProspector was used to identify several of the major ions for both spectra  It is worth noting that optimal isolation of parent ions for MS/MS analysis required an isolation window of 12 Da centered on the most abundant isotope of the A peptide. The widening of the isolation window is a consequence of increasing the % laser power for the MS n experiment. The % laser power for the MS n experiments was increased (~25% laser power, twice that for full scan MS experiments) to increase the number of A peptide ions available for subsequent isolation and fragmentation. With the increased number of ions due to the higher laser power, space c harge broadened the frequency distribution of the ions during the isolation step. The isolation window was therefore widened to ensure that all the A peptide ions were isolated. After the isolation step, the effects of space charge are greatly reduced, as all ions outside the isolation window have been ejected from the trap. This reduces the ion population within the trap to levels below which space charge is an issue.
54 The dominant ions in the MS 2 spectra of the [M+H] + A 1 40 ion (aside from the loss of water and ammonia) include the y 33 y 37 y 39 b 22 and b 23 ions (Figure 2 2A). For the MS 2 fragmentation of the [M+H] + A 1 42 ion, the major fragments detected were the y 35 y 39 y 41 b 22 and b 23 ions (Figure 2 2B). The dominance of the y ions in th e mass spectra can be explained by preferential fragmentation adjacent to acidic amino acid residues (aspartic acid and glutamic acid)  The MS 2 spectra of the A and A ions provide minimal structural information due to preferential cleavage adjacent to acidic amino acids additional stages of mass spectrometry are necessary. The y 33 ion present in Figure 2 2A was isolated and fragmented; th e MS 3 mass spectrum is displayed in Figure 2 3A. The fragmentation coverage of the y 33 fragment ion is much higher, with a number of y type ions and ions due to internal fragmentation observed. Further coverage of the peptide was achieved by isolating an d fragmenting the ( b 23 y 33 ) 16 ion for an MS 4 experiment (Figure 2 3B). In this fashion, identification of the A peptide was demonstrated with MALDI/MS n Although acidic amino acid residues caused limited fragmentation coverage in MS 2 additional stages o f mass spectrometry overcame this obstacle. This technique could be useful for the identification of unknown peptides, particularly for those not in peptide databases. Isotopically L abeled S tandards Isotopically labeled standards are typically used as in ternal calibrants for quantitative mass spectrometry. Although a single stage of MS can be used for quantitation, analyte specificity can be significantly improved by using tandem MS. The
55 selectivity of MS n is of particular value when sample pretreatment and/or chromatographic separation is not available, as in MALDI imaging studies  However, using alternating MALDI/MS/MS experiments of an analyte and its isotopically labeled standard provides poor reproducibility [8 5] This poor precision was attributed to laser shot to shot variability and non homogenous MALDI matrix crystals. To improve precision, the [M+H] + ions of the analyte and its isotopically labeled standard were isolated for MS/MS in a single scan usin g a wide isolation window. Although it is a simple matter to widen the isolation window to encompass the analyte and its isotopically labeled standard when they differ by only a few AMU, a mass window wide enough to encompass a peptide and its uniformly 15 N labeled analog is more challenging because of the large m/z difference. (The difference between the [M+H] + ions of the A and its uniformly 15 N labeled standard is 53 m/z units). Several experiments were performed to examine the capabilities of the instrument to accommodate a wide isolation window. Equal concentrations of A 1 40 and its 15 N analog were spotted onto a MALDI well plate spot. Figure 2 4A shows the full scan spectrum of this mixture. To isolate and fragment both ions in one even t, an isolation window of m/z 80 was used for the MS/MS experiment centered at m/z 4355 (the midway point between A 1 40 and its 15 N analog); the MS/MS spectrum is shown in Figure 2 4B. A drawback to this method is trapping and fragmenting unwanted ions that fall in the m/z range between the analyte and its isotopically labeled compound; the fragment ions from these interferences may mask the analyte/internal standard response. Nevertheless, in this case the b and y fragment ion peaks were readily identi fied in the MS/MS spectrum. In addition, equal intensities for both the analyte, 15 N
56 analog, and their respective fragments were maintained in both the full scan MS and MS/MS spectra. Conclusions The experiments and data reported here demonstrate the p otential for amyloid peptide analysis on a linear ion trap by intermediate pressure MALDI employing mass range extension. The isotope pattern of the full scan MS matches that of the expected spectrum, and MS n experiments confirm the identity of the pept ides. Wide isolation and fragmentation of an amyloid peptide and its 15 N analog were also demonstrated; with this approach, quantitation of the amyloid peptide would be feasible. With the extended mass range, limitations such as increased space charging and loss of mass resolution were noted. However, careful instrument optimization minimized the effects of these two issues. While all analytes were analyzed on a stainless steel MALDI plate, preliminary experiments suggest that analysis of large molecular weight peptides from intact tissue will be possible. Analysis of large peptides from intact tissue poses several problems, as a tissue section presents an extremely complex sample matrix. Optimizing MALDI extraction efficiency and reduci ng background interferences are among the many issues to address for tissue studies. MALDI extraction efficiency will require careful selection of MALDI matrix compound and solvents, as well as optimization of MALDI matrix application conditions. Althoug h sample background will be increased due to the number of potentially interfering species present in tissue, MS n experiments should prove useful to enhance selectivity and provide sample elucidation and identification.
57 Furthermore, the use of MALDI/MS n o n the LIT offers potential for imaging these peptides in intact tissue.
58 Figure 2 1 Full scan mass spectrum of A 1 40. Inset A is an expansion around the A 1 40 [M+H] + peak. The isotope distribution matches that of the simulated spectrum in inset B. The peptide sequence of A 1 40 is DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVV.
59 Figure 2 2 MS/MS spectra of the [M+H]+ ions of A 1 40 (A) and A 1 42 (B). The m/z values of the major peaks match the expected y and b series ions for their respective parent. Coverage of the A peptide is incomplete due to preferential cleavage adjacent to acidic amino acid residues within the peptide.
60 Figure 2 3 MS 3 (A) and MS 4 (B) experiments for the A 1 40 ion. Further fragmentation of the A 1 40 ion was performed for structural elucidation. The y 33 ion was isolated and fragmented for the MS3 experiment. The ( b 23 y 33 ) 16 ion (marked with asterisk in A) wa s isolated and fragmented to provide additional coverage of the A peptide.
6 1 Figure 2 4 MS (A) and MS 2 (B) mass spectra of a 1:1 mixture of A 1 40 and uniformly 15 N labeled analog. For the MS 2 spectrum, the isolation window was an m/z range of 4315 to 4395 so as to include the [M+H] + ions of both the A 1 40 and the uniformly 15 N labeled analog
62 CHAPTER 3 IN SITU ANALYSIS OF INTACT L ARGE PEPTIDES BY MAL DI LIT MS N Introduction dy that affects over five million people in the United States  This insidious disease leads to memory loss, deterioration of the thought process, and personality changes that may eventually end with the death of the AD patient. The hallmark characteristic of AD is the presence of neuritic plaques in the cortex and hippocampus regions of the brai n  These neuritic plaques consist of amyloid beta (A ) peptide fragments ( lengths of 3 8 to 43 amino acid residues ) that originate from the amyloid precursor protein (APP). These fragments are cleaved from APP by the aspartyl protease b secretase (BACE) and t he presenilin dependent protease secretase At present, there is a need to localize A fragments in brain tissue for AD research. Imaging techni ques (e.g., optical microscopy and PET) offer avenues to determine the dis position of A peptides within a tissue section  This is especially useful for the determination of A within live organisms  However these techniques offer limited chemical selectivity; rather, current imaging te chniques rely on the selectivity of the molecular label (e.g., an antibody, stain, or radiolabel) for analyte detection Whereas a he detection of other molecules that may be medically relevant to determining the progression/treatment of a disease. For example, a common A stain (Thioflavi n S) will stain amyloid plaques blue, yellow, or green (depending upon the filter) in a tissue section  How ever if a lipid or other
63 progression of the disease, these changes would be invisible to the A specific stain. Thus a complementary method that identifies both A peptides a nd other AD relevant molecules is desirable. Analyte i dentification based upon the molecular weight using mass spectrometry (MS) would be more selective than staining especially if paired with tandem mass spectrometry to further increase molecular selecti vity For example, LC MS is used to examine relative concentration of the various A peptides  However, the sampling preparation necessary for such an experiment requires tissue homogenization prior to analysis by LC MS. Unfortunately, this necessary sample preparation prior to LC MS analysis leads to a loss of spatial information. O ne technique that offers the selectivity of MS, while retaining spatial information is mass spectrometric imaging (MSI)  MSI is a micro destructive, label free imaging technique that utilizes a microprobe to collect mass spectra at discrete points across a tissue section. Upon collecti ng the mass spectra, the data are reprocessed using an imaging program and a resultant MS image is generated. The number of MS images that may be generated from a single MSI experiment is limited only by the mass range of the instrument mass spectral resolution, and ionization efficiency of an analyte, as an MS image can be generated at each collected m/z The sp atial resolution of a MSI experiment is dependent upon the spot size of the microprobe and the distance between each collected point (step size of the microprobe). Complete coverage over the tissue section is achieved by matching the step size to the micr oprobe spot size. However, analysis times can be decreased by increasing the micro probe step size (at the cost of degrading the MS image resolution).
64 Conversely spatial resolution can be increased by de creasing the microprobe step size in exchange for l engthening the experimental time. Analysis times for an MSI experiment are dep endent upon a number of factors the spot size and step size of the microprobe, the type of MS experiment, the duty cycle of the MS instrument, and the area of the tiss ue surface to be examined that must be balanced to avoid lengthy MSI experiments. Indeed, an MSI experiment can take anywhere from one hour to twelve hours depending upon the factors above However, two hour experiments tend to be the norm for a tissue section wit h dimensions of ~7.5 mm by x 5 mm (i.e., the size of a sagittal mouse brain section) with a spot size and step size of 100 m on an LIT system Detection and localization of A in tissue by MSI offers comple me ntary spatial information to existing imaging techniques. Previous AD MSI work has utilized a time of flight (ToF) mass spectrometer, both for its mass range and mass accuracy [68, 69, 90] However, ToF MS instruments tend to lack structural elucidation capabilities (at best two stages of tandem mass spectrometry is achieved if a ToF ToF or Q ToF is used). An alternative to a ToF MS instrument for MSI experiments is a linear ion trap (LIT) mass spectrometer. LIT experiments offer multiple stages of mass analysis ( MS n ) that enable structural elucidation of unknowns and enhance chemical selectivity through identification of characteristic product ions. However the MAL DI LIT commercial instrument has an upper m/z cutoff of m/z 4000. This limits the instrument to examining compounds with a molecular weight of less than 4000 (assuming a singly charged ion). This poses a problem to the analysis of large peptides such as A 1 40 and A 1 42 that have molecular weights of 4329 and 4512 Da, respectively. A strategy to overcome
65 the upper m/z limitation is to extend the mass range of the LIT instrument  This chapter will describe the methods developed to analyze A peptides on a n LIT. Images collected through the MSI experiments will be compared to a histological stained serial section to determ ine the feasibility and limitations of this technique. Experimental Methods Chemicals and M aterials Trifluoroacetic acid (TFA) and MALDI matrix 2,5 dihydroxybenzoic acid (DHB), were purchased from Acros Organics (Morris Plains, NJ) The other two MALDI matrices, sinapinic acid (SA) and cyano 4 hydroxycinnamic acid (CHCA) along with Thioflavin S (for histology staining) were purchased from Sigma Aldrich (St. Louis, MO). Glacial acetic acid, ammonium bicarbonate, PBS, neutral buff ered formalin, and HPLC grade methanol and water were purchased from Fisher Scientific (Fairlawn, NJ). Distilled water from a Purelab Flex water system was used for histological staining. A standard of A 1 40 was purchased from rPeptide (Athens, GA). DH B MALDI matrix solutions were prepared at a concentration of 40 mg/mL in a 70:29.9:0.1 solution of methanol:water:TFA. Animals A generous donation of mice brains (f emale Tg2576 mice ) were received from Bristol Myers Squibb ( Wallingford CT) These mice exhibit a fivefold increase of A 1 40 and fourteen fold increase of A 1 42 over levels within young, unimpaired mice  These mice were sacrificed at 30 months of age Immediately after euthanization b rains were excised flash frozen and then stored at 8 0 C.
66 Tissue P reparation The left hemisphere of the mouse brain was mounted onto the ta rget plate by surrounding the lower half of the brain with water and allowing the water to freeze the brain into place. OCT should not be used for MSI studies as the polymer may interfere and/or contaminate the MS spectrum. Sagittal sections of the mouse brain were collected using a Microm HM cryostat at a temperature of 25 C. Brain sections were thaw mounted onto glass microscope slide and then stored a t 80 C until analysis. Note that MSI experiments utilizing an LIT do not require the use of condu ctive glass slide s as in MALDI ToF experiments because charge build up will not affect mass analysis  Prior to MALDI analysis, all tissue sections were placed in a vacuum desiccator for 30 minutes to remove excess w ater. Af terward, a series of washes was performed prior to MALDI matrix application to remove endogenous lipids and other endogenous compounds The series of washes consisted of four steps : 1) 30 seconds in 70:30 ethanol:water 2) 30 seconds in 70:30 ethanol:water 3) 30 seconds in chloroform 4) 30 seconds in 90:9:1 ethanol:water:acetic acid. After washing the tissue section was placed in a v acuum desiccator for approximately 15 min to remove any residual moisture. An important aspect to gener ating quality MSI spectra is matrix application  Due to the size of the amyloid plaques, uniform coverage over the entire tissue surface along with matrix crystal lengths of less than 100 m is desirable. Several different options for matrix ap plication are available including pneumatic spraying and inkjet
67 printing [6, 17] For this work, m atrix was applied using an artistic airbrush (Azt ek A470, Testors, Rockford, IL)  Visually, the airbrush offered a more uniform coating of MALDI matrix compared to dried droplet and inkjet printer deposition The matrix solution consisted of 40 mg/mL DHB in 70:29.9:0.1 methanol:water:TFA. Approximately 20 passes were necessary to fully coat the tissue surface. Instrumentation A Thermo LTQ linear ion trap fitted with an intermediate pressure (70 mTorr) MALDI source (San Jose, CA) was used for all experiments. The MALDI source utilizes a N 2 laser (337 nm) with a repetition rate of 60 Hz and maximum energy of ~90 J/pulse and has a laser spot diameter size of approximately 100 m. Experiments were typically performed with automatic gain control ( AGC ) off, using 2 3 laser shots per spot and approximately 6 8 J/pulse. The step size for these experiments was either 100 m or 50 m. For experiments examining in tact A peptides, the mass range of the instrument was extended past m/z 4000 up to m/z 5000, as described in a previous paper [ 45] For MS n experiments of intac t peptides, an isolation window 15 AMU wide was used to capture the maximum number ions for tandem MS experiments. In addition, the laser energy was increased to 12 J/pulse t o increase the number ions available for fragmentation. Although increasing the laser energy may lead to mass shifts during MS scans due to space charging, there was less of a concern for space charging due to the limited number of ions trapped during MS n experiments. Data P rocessing Data were processed using ImageQuest v1.1 software (Thermo Fisher, San Jose, CA). A linear smoothing method was applied to the generated MS image. For MS images, the analyte response was divided by the total ion current (TIC) to generate an
68 m/ z /TIC vs. position image. This was performed to normalize the analyte response across the tissue surface. For MS n images, no normalization step was performed. Typically a window of a pproximately 4 AMU wide was selected when constructing the MS image. Optical M icroscopy For optical microscopy studies, frozen tissue sections were thawed (15 minutes) at room temperature prior to fixation. Formalin fixation was performed for ten minute s in 10% neutral buffered formalin. Following the fixation, slides were rehydrated in distilled water for 5 min. Sections were thrice washed in 40% ethanol for one minute. Afterward, the tissue sections were stained with 0. 0125% Thioflavin /40% ethanol/p hosphate buffered saline (PBS) Slides were then placed in 50% ethanol/PBS and transferred to PBS for thirty seconds each. After dipping the sections gently into distilled water to remove adherent salts the sections were mounted with vecta st ain Sect ions were examined on a Zeiss Pasca l LSM5 (Zeiss; Oberkochen, Germany ) confocal laser scanning m icroscope A 488 nm absorbance wavelength and 505 nm emission wavelength was selected to examine the fluorescence of the Thioflavin S stain. Adobe Illustrator CS5 was use d to merge the resultant images (~ 1.3 by 1.3 mm) to generate a single histological image. Results and Discussion Effects of S ample Preparation on MS S pectra MSI a nalysis of large molecules is quit e different from small molecule analysis For example, MSI analysis of endogenous lipids or exogenous small drug molecules discourages the use of water/alcohol washes. Whereas alcohol/water washes remove endogenous salts, other soluble compounds of interest may undergo analyte migration
69 (or at worst, complete elimination of the analyte from the tissue section ). However for the analysis of A peptides, these endogenous compounds and salts may interfere (e.g., ion suppression, salt adduction, etc.) with MSI experiments. Figure 3 1 illustrates two Tg2576 brain sections that contain A plaques in the cortex and hippocampus regions of the brain Lipids dominate the low mass region of the mass spectrum collected from an unwashed tissue section (Figure 3 1A); in addition, response from A plaques a ppears to be suppressed (inset of Figure 3 1A). However after washing, lipids in the m/z 700 1000 range that are present in the top half of the figure are dramatically reduced (Figure 3 1B). More importantly, ion suppression of the A peptides is decreas ed after the wash procedure allowing detection of the A peptides (inset of Figure 3 1B). The increase in s ignal after the washing step may be explained by the removal of endogenous salts and lipids, whose presence may lead to ion suppression of the targ et analyte. MSI of Intact A P eptides using a M ass Range E xtension Sagittal sections of the anterior portion of a Tg2576 mouse brain were selected for MSI experiments. Several abundant ions were identified in the cortex region of the brain section (Figure 3 2). Three ions abundant in the frontal cortex region of the brain ( m/ z 4132, 4330, and 4515) correspond to the MW of A peptides 1 38, 1 40, and 1 42 respectively Table 3 1 lists the m/z of the [M+H] + ions of the observed A peptides in the MS spectrum of Figure 3 2 along with the expected average MW weight of the peptides The observed m/z of each of the peptides was within 0.1 Da of its corresponding expected m/z The MS image s illustrate a similar distribution in the
70 tis sue section for all three ions. Th e A distribution is consistent with the expected localization of plaques with in the cortex regio n of this anterior brain section Lateral to the tissue section examined in Figure 3 2 two serial whole brain sections wer e examined for A plaques (Figure 3 3). The same t hree abundant A peptides (A 1 38, 1 40, and 1 42) were imaged by MSI (Figure 3 3 A C, respectively) in one sagittal brain section with an adjacent serial section stai ned with Thioflavin S (Figure 3 3E ). Thioflavin S is a common histologica l stain that stains A plaques green under the current experimental settings. MSI l ocalization of the A plaques matched that of the Thioflavin S stain, with localization predominately in the cortex and hippocampus regions of the brain. Although the image contrast for the A 1 42 image is much lower compared to the A 1 38 and A 1 40 images, it is not surprising, given that the response for A 1 42 is at least 20 times less than the A 1 40 response. In addition t o the A peptide distributions, t he distribution of DHB MALDI matrix (Figure 3 3D) was examined Uneven crystallization of MALDI matrix could le ad to hot spots in the MS image. The uniform response of the DHB ion ( M + m/z 154) in the MS image of the ti ssue section indicates an even DHB crystallization across the tissue surface. Earlier studies have demonstrated MSI imaging of A peptides using a ToF instrument [68, 69, 90] Although a ToF instrument is sufficient for performing single stage MSI experiments, a ToF/ToF instrument would be required for any tandem mass spectrometry experiments. And at best, this instrument would be limited to two st ages of mass spectrometry. The lack of tandem mass spectrometry limits structural identification of isobaric species (ions that fall at the same nominal mass ) Multiple
71 stages of mass spectrometry may prove useful in further establishing the identity of t he observed peptides To demonstrate the capabilities of the LIT for MS n peptide imaging experiments, separate MS I experiments isolating and fragmenting the A 1 40 ions were performed. The MS 2 spectra averaged over a region in the cortex region of th e mouse brain section matched that of previous c ollected MS 2 spectrum of an A 1 40 standard ( Figure 3 4 ). The most abundant ion in both fragmentation spectra is the loss of 18 Da water (Figure 3 4A) This ion results from a relatively uninformative neutral loss (many ions exhibit a loss of water upon fragmentation) and thus offers little structural information. The lack of the water neutral loss peak in the standard spectrum could be explained as a lack of isobaric compounds (which readily lose wat er) that are native to tissue. Therefore, other more informative ions were selected to generate MS images After the neutral loss of water, one of the next most abundant ions lies at m/z 3459. This MW corresponds to the y 33 ion of a fragmented A 1 40 ion. The distribution of this product ion (Figure 3 5 A) further supports the ident ity of the parent ion as A 1 40 because the y 33 distribution matches that of the histological stain (Figure 3 5D) Additional evidence supporting the identity of the A 1 40 ion is provided by MS 3 experiments. For this set of MS 3 experiments, a different y type ion (y 39 m/z 4512) was selected for dissociation by CID ( A 1 40 y 39 ) This CID experiment produced two highly abundant product ions, m/z 3459 (Figure 3 5B) and m/z 2645 (Figure 3 5C) The first ion corresponds to y 33 ions, originating from fragmented y 39 ions T he second ion cor responds to a cleavage of the 1 22 residues on the N terminus side of the A 1 40
72 peptide resulting in a (b 23 y 39 ) 22 ion. The doubl e backbone cleavage is preferred over continued fragmentation along the C terminal side of the peptide due to t he presence of internal acidic amino acid residue s (in this case glutamic acid) in the peptide structure  (Fragmentation adjacent to ac idic amino acid residues is preferred over other sites on the peptide backbone ). The localization of both MS 3 product ions further supports the ion as A 1 40 peptide as the loc alization is similar to that of the histological stain (Figure 3 6) If the M S images offered a different localization, there would be some doubt to the identification of the peptides as isobaric species could have a different distribution within the tissue. Although this technique lacks a separation step such as liquid chromatog raphy or immunoprecipitation tandem MSI offers a technique for peptide identification and localization complementary to traditional histological staining methods. H istological versus MSI Image Resolution of Individual Plaques A plaques range in size from 50 m to 200 m (see Figure 1 11 for an example) Whereas larger plaques should not be difficult to image with current instrument ation localization of smaller aggregations may pro ve more problematic. Figure 3 7A illustrates the localization of A 1 40 taken with a raster step size of 100 m. Two regions (indicated by the white boxes) have been expanded to examine individual amyloid plaque clusters (figure 3 7C and D). The histological image superimposed over the Figure s 3 7 C and D illustrate s the difference in spatial resolution between a confocal microscope and an MSI instrument. The spatial resolution of the MSI instrument could be defined as being equal to the spot size of the laser (assuming the raster step size is equal to the spot
73 size and t he size of the MALDI crystals are smaller than the spot size of the laser). Although the localization of the A peptide fragments are generally similar to the histological image, close examination of individual plaque formations indicate a blurring of the MS image. This is due in part to the spot size of the laser being greater than the diameter of individual plaques. Moreover the spot size of the laser is greater than the distances between individual plaque in Figures 3 7 C and D. As a result the s moothing function of the imaging software blurs the regions between the plaques such that it appears that one single plaque is present in each of the expanded areas. To improve the spatial resolution, a n MSI experiment with a reduced raster step size (50 m rather than 100 m) was performed Figure 3 7B illustrates the localization of A 1 40 taken with the reduced raster step size. The image quality and resolution of the 50 m raster step size image are sharper than in the image with the 100 m raster s tep size. Closer examination of the boxed areas supports the sharpened image conclusion (Figures 3 7 E and F). However, this improvement was at the cost of quadrupling of analysis time. For experiments that examine a whole body tissue section, this incr ease in analysis time may prove impractical ; however for this application, oversampling may be useful since we are interested in a single organ A lthough the spatial resolution of a confocal microscope (~200 nm)  is vastly superior to a n MSI instrument, an MSI experiment can detect a wider range of compounds with a single M SI experiment while generating compound specific spatial information for multiple analytes
74 Conclusion s Mass spectrometric imaging offers many advantages over conventional histological stains. Primarily, m ore chemical information is available per experiment than in any single histological experiment. Using tandem mass spectrometry, analyte selectivity can be evaluated by matching fragment patterns of an ion to that of a standard. However MSI is not withou t limitations. The washing and fixation step used for these experiments may wash away endogenous lipid s and /or wash soluble xenobotic analytes within the tissue This could be problematic if, for example, the distribution s of an AD drug and A deposits were to be compared Thu s two experiments on serial sections may be necessary if one wished to determine the disposition of the small drug molecule and the larger peptide Even if analyte migration is not an issue the MSI technique does not offer the spatial resolution that an optical technique will achieve. This is especially true in these AD studies, as MSI does not have the resolution to distinguish between pl aques and neurofibrillary tangles. Nevertheless MSI of large peptides (such as A peptid es) on a linear ion trap provides valuable and complementary information to traditional imaging techniques.
75 Figure 3 1. Eff ect of tissue washing on a Tg2576 mouse brain section Lipids dominate the mass spectrum in A. There is a noticeable lack of signal in the region where the amyloid peptides should reside. After washing however B most of the lipid signal has been eliminated. In addition, the ions corresponding to the amyloid peptide s are readily detected
76 Figure 3 2 MSI images from the anterior portion of a sagittal Tg2576 mouse brain section. The three images illustrate the localization of a few prominent peptides that originate from APP, A 1 38 (A, m/z 4130 4133), A 1 40 (B, m/z 4329 4332), and A 1 42 (C, m/z 4214 4217). The mass spectrum in this figure is the average over the area indicated by the white box of inset B.
77 Table 3 1. Observed amyloid peptides in situ during MSI experiments Predicted average [M+H] + Observed [M+H] + Peptide (m/z) (m/z) 1 37 4075.5 4075.4 1 38 4132.6 4132.5 1 40 4330.8 4330.7 1 40 Oxid 4346.8 4346.8 1 42 4515.1 4515.1
78 Figure 3 3. MSI images a sagittal Tg2576 mouse brain section. The first three images illustrate the localization of a three prominent peptides that originate from APP, A 1 38 (A, m/z 4130 4133), A 1 40 (B, m/z 4329 4332), and A 1 42 (C, m/z 4214 4217). Image D illustrates the distribution of DHB ( m/z 154) across the tissue surf ace. Image E is an image of the histological stain of the serial section adjacent to the MS imaged tissue section. Cx indicates the cortex region of the brain. Hy indicates the hippocampus region of the brain.
79 Figure 3 4. Comparison of MS 2 fr agmentation of A 1 40 from tissue (A) and an A 1 40 standard (B). Although the neutral loss of water is more abundant in A, the gener al fragmentation pattern is similar between both spectra.
80 Figure 3 5 MS n image maps of the fragmentation of endoge nous A 1 40 The localization of the y 33 ion, A, illustrates a distribution similar to the histological image, D. The fragmentation of the y 39 ion was performed to demonstrate the ability to construct images from ions found in the MS 3 spectra (y 33 ,B, and (b 23 y 39 ) 22 C) For further information about the MS n fragmentation of A 1 40 see Chapter 2.
81 Figure 3 6 MS image of A 1 40 ion, A, with three regions, indicated by the white boxes, expanded (B D). T he Thioflavin S histological image was superimposed atop the expanded MS images. The white scale bars in B D i ndicate a length of approximately 100 m. The raster step size employed for this MS image was 100 m. Although the general MS image indicates regions of amyloid plaques analogous to the histological image, the MS image does not differentiate plaques that are in close proximity to each other. A B C D
82 Figure 3 7. Effect of step size on an MSI experiment. Figures A and B illustrates the localization of A 1 40 (red) across two different MSI experiments utilizing a 100 m and 50 m raster step size, respectively. Fi gures C through F are expansions of the indicated white boxes in A and B. The histological stain image (green) is superimposed over the MS images to demonstrate the differences in spatial resolution between the two techniques. The scale bars present in C through F represents a distance of approximately 100 m. F D MS image of A 1 40 100 m SS A MS image of A 1 40 50 m SS B C E
83 CHAPTER 4 DIGESTION OF LARGE P EPTIDES IN SITU PRIO R TO MSI STUDIES Introduction The a nalysis of peptides and proteins offer s scientists valuable information about the biological processes related to disease states within the body. Protein discovery and identification are typically performed by mass spectrometry due to the high sensitivity and structural elucidation capabilities. Various MS platforms are avail able for peptide identific ation, e.g., LIT s, ToFs, FT ICRs, and orbitraps are MS instruments commonly found in MS proteomic cores However, LITs have a distinct advantage over ToFs and or bitrap instruments because multiple stages of mass spe ctrometry (MS n ) enables identification o f larger sections of the peptide sequence FT ICR instruments also offer MS n capabilities; however the cost of an FT ICRs is much greater than that of an LIT. MS experiments typically examine proteins and peptides as intact molecules ( or as smaller fra gments if an enzymatic digestion step is performed prior to MS analysis ) The previous chapter demonstrated the abilities of a n LIT for the analysis of intact large peptides. However, no commercial LIT MS instruments have the capability to analyze peptide ions with an m/z greater than 4000. Thus an alternative is to enzymatically digest the sample s prior to MS analysis. Peptide/protein digestion is a technique that utilizes an enzyme to digest proteins into smaller peptides. One advantage of enz ymatic digestions is that s maller peptides are easier to sequence versus larger compounds due to potentially poor ionization efficiencies and s tructural modifications on a protein  Moreover, digestion of the proteins increases the likelihood of detect ion by MS This is because a large protein is
84 likely to have an m/z greater than the upper m/z limit of the MS instrument and thus will not be discovered in an MS scan There are ma ny different enzymes for protein digestion ; h owever, trypsin is by far the most common protease for enzymatic digestion experiments  Both its low cost and its specificity make it an attractive protease Trypsin selectively cleaves proteins at the carboxy terminal side of any ar ginine or lysine residues present in a protein molecule. Tryptic digestions are performed on numerous sample types, e.g., cell cultures, tissue extracts, etc. For trypt ic digestion of tissue samples, protein extraction precedes the digestion step. Howe ver, extraction from a tissue section will lead to a loss of spatial localization of the protein. Although protein identification is important, localization of the protein within a tissue section is equally important. Mass spectrometric imaging of protei ns and peptides has been demonstrated to illustrate the localization of large proteins within tissue sections [3, 5] But what if the protein of interest has an m/z greater than the upper m/z limit of the instrument and digestion is necessary ? In situ digestion of proteins atop intact tissue is feasible, albeit difficult [46, 94, 95] Current equipment for in situ digestion of tissue sections utilizes automated spotters to dispense nL volumes of trypsin solution (diameter ~55 m) to the tissue surface  Unfortunately equipment to dispense tr ypsin solution at these low volumes is expensive (> $50,000) and may pose a funding barrier to emerging (or even well established) research laboratories. Moreove r, trypsin is not applied across the entire tissue surface. Rather, trypsin is applied at dis crete spots such that each spot is separate from its
85 neighbor. This deposition method is analogous to an undersampling experiment (see Figure 1 4B). Thus, t his spotting method generates a series of trypsin digestion experiments that reduces the likeliho od of analyte migration. However, this method is not useful for A experiments as demonstrated by the results in chapter 3. An oversampling experiment was necessary to achieve high levels of spatial resolution to identify closely spaced amyloid plaques. The goal of this chapter is to demonstrate a cost effective enzymatic digestion alternative that will perform in situ tryptic digestion of A peptides across the entire tissue surface. Experimental Methods Chemicals and M aterials Trifluoroacetic acid ( TFA) and MALDI matrix 2,5 dihydroxybenzoic acid (DHB), were purchased from Acros Organics (Morris Plains, NJ) The other two MALDI matrices, sinapinic acid (SA) and cyano 4 hydroxycinnamic acid (CHCA) along with trypsin and octyl / glucoside (OcGlc), were purchased from Sigma Aldrich (St. Louis, MO). Glacial acetic acid, ammonium bicarbonate, ammonium hydroxide (NH 4 OH) and HPLC grade methanol and water were purchased from Fisher Scientific (Fairlawn, NJ). A standard of A 1 40 was purchased from rPep tide (Athens, GA). A 1 40 standards were resuspended at a concentration of 1 mg/mL in 1% NH 4 OH DHB MALDI matrix solutions were prepared at a concentration of 40 mg/mL in a 70:29.9:0.1 solution of methanol:water:TFA. Trypsin was prepared at a concentrat ion of 125 ng/L in either 10 mM ammonium bicarbonate 0.1% OcGlc or 70:30 ethanol:10mM ammonium bicarbonate solution.
86 Animals A generous donation of aged female mice brains (Tg2576 mice are modified to express human A peptides ) and mature rat brains (ma le Sprague Dawley rats) were received from Bristol Myers Sq uibb ( Wallingford CT). The Tg2576 mice exhibit a fivefold increase of A 1 40 and fourteen fold increase of A 1 42 over levels within young, unimpaired mice  Immediately after euthanization brains were excised, flash frozen, and stored at 80 C. Tissue P reparation The left hemisphe re of the mouse brain was mounted onto the target plate by surrounding the brain with water and allowing the water to freeze the brain into place. OCT should not be used for MSI studies as the polymer may interfere and/or contaminate MS spectra Sagittal sections of the mouse brain were collected using a Microm HM cryostat at a tempe rature of 25 C. For rat brains, coronal sections of the brain were collected. Brain sections were thaw mounted onto glass microscope slide and then stored at 80 C until analysis. Prior to MALDI MS analysis, all tissue sections were placed in a vacuum desiccator for 30 minutes to remove excess water. A series of washing step s was then performed prior to the digestion step to remove interfering lipids and salts The was hing step used for these experiments involved a 4 step wash: 1) 30 seconds in 70:30 ethanol:water 2) 30 seconds in 70:30 ethanol:water 3) 30 seconds in chloroform 4) 30 seconds in 90:9:1 ethanol:water:acetic acid
87 The washing steps involved repeatedly dipping the glass slide into and out of the wash solution over a period of thirty seconds for each solution. After the tissue washing step, the tissue section was placed in a vacuum desiccator for approximately 15 mi n utes to remove any residual moisture. Enzymatic D igestion T rypsin was applied to dried tissue sections using a micropi pettor prior to MALDI matrix application. Approximately 15 L of trypsin solution was required to completely cover the tissue section. After the trypsin application, the glass slide was placed into a container (five slide holder) and then the container sealed with parafilm. After sealing the container, the incubation device was then p laced in a sealed bag similar to the setup described by Setout et al.  (Figure 4 1) The container held a wetted kimwipe (1 mL of DI water was added to the kimwipe) so as to maintain a humid environment within the sealed container. This apparatus was placed in a n oven and i ncubated at 37 C for approximately 2.5 h. After removal from the oven, the tissue section was placed in the vacuum desiccator to remove excess moisture. For digestions of A 1 40 standard s equal volumes (1 L each) of the standard and the trypsin solut ion were placed in 0.5 L centrifuge tube s and allowed to digest for one to two hours. MALDI Matrix Application MALDI m atrix was applied using a modified Epson R260 inkjet printer  The inkjet cartridges were filled with 40 mg/mL DHB in 70:29.9:0.1 methan ol:water:TFA. Approximately 40 passes were required to fully coat the tissue surface. For digested standards, a dried droplet method was employed to examine the peptide fragments.
88 Instrumentation A Thermo LTQ linear ion trap fitted with an intermediate pressure (70 mTorr) MALDI source (San Jose, CA) was used for all experiments. The MALDI source utilizes a N 2 laser (337 nm) with a repetition rate of 60 Hz and maximum energy of ~90 J/pulse. Experiments were typically performed using 2 3 laser shots p er spot and approximately 6 8 J/pulse. The raster step size for these experiments was 100 m. For enzymatic digestion experiments, the instrument was scanned from m/z 150 to 2000. For MS n experiments involving digested peptides, an isolation window of 2 AMU and laser energy of 12 J/pulse was used The increased laser energy was less of a concern for space charging effects due to the limited number of ions trapped during MS n experiments. Data P rocessing MSI images wer e generated using ImageQuest v1. 1 software (Thermo Fisher, San Jose, CA). A linear smoothing method was applied to the generated MS image. Typically, a window of approximately 4 AMU wide was selected when constructing the MS image. Results and Discussion Effects of MALDI Matrix Choice for MS Analysis of Enzymatically digested T i ssue S ections Try psin digestion experiments are a powerful tool for AD studies. Trypsin cleaves peptides and proteins at the carboxy terminal side any arginine (R) or lysine (L) residues present within a peptide. This specific cleavage permits in silico determination of peptide fragments resulting f rom a tryptic digestion. A web based program, ProteinProspector  was utilized to predict fragments resulting from tryptically
89 digested A 1 40 Figure 4 2 Although several tryptic peptides are generated during the enzymatic digestion, few of these fragments are useful for A studies. Digestion and instrument paramete rs were optimized to achieve optimal signal of the fragments containing the C terminus of A 1 40. These fragments are shaded in Table 4 1 (t he reasoning for the optimization based upon these select f ragments is explained below ) P eptide experiments by M ALDI MS tend to use one of three MALDI matrices, CHCA, DHB, or SA. Although any of these matri ces are use able for the analysis of intact A peptides, tryptic digestion experiments of A peptides uncovered a drawback to usi ng certain matrices. Figure 4 3 illustrates the mass spectrum of tryptically digested A 1 40 using CHCA, SA and DHB. There are many similar peaks within each spectrum; however several differences were noted. The experiment using CHCA offered the least amount of background noise (Fig ure 4 3 A) especially when compared to SA (Figure 4 3 B). The MS spectra utilizing DHB (Figure 4 3 C) demonstrated A fragments common to the CHCA and SA experiment. However DHB produce d more sodiated (i.e., [M+Na] + ) ions than experiments utilizing CHCA or SA complicating the MS spectrum However, the CHCA related spectrum is less informative than the DHB related spectrum. This is due to the absence of A 6 40, A 17 40, and A 29 40 fragments in the CHCA related spectrum Clinical AD samples will contai n a number of different A isof o rms, differing by their C terminal residues. For analysis of enzymatically digested A to be effective, the C term inus of the peptide must retain the charge ; o therwise, it is difficult to distinguish the origin of a digested peptide fragment. For example if trypsin is added to a mixture of A 1 38, 1 39, 1 40, 1 41, or 1 42 peptides among the tryptic fragments, an A 1 5 fragment is generated (cleavage
90 occurs after the fifth residue (arginine)). However, it would be impossible, without prior labeling, to determine the origin of the A 1 5 fragment. Thus, fragments retaining the terminal 40 th residue (valine) or 42 nd residue (alanine) are more informative for MSI experiments than any of the other fragment ion s that dominate all three spectra. Although the SA related spectrum does contain the A 29 40 fragment, the wide abundance of response throughout the mass spectrum potentially complicates the identification of unknown ions after a digestion. The lack of [M+H] + ions containing the C terminus of the A peptide could be attributed to the lack of basic residues in the remaining fragment ion as the presence of basic residues promote s protonation. Figure 4 4 illustrates a mass spectrum of a trypsin digestion of the A 1 40 standard when more complete digestion is performed Although the incubation time for digestion could be extended, the digestion period was not extended for fear of producing tryptic autocleavage peptides The ion of interest was observed at m/z 1108, [A 29 40+Na] + Sample preparation parameters will focus on optimizing the response of the [A 29 40+Na] + ion because the response of the two fragments with missed cleavages, [A 17 40+Na] + and [A 6 40+Na] + unpredictably varied between experiments In situ Tryptic Digestion of A Peptide Standards Digestion of A peptides offers the advantage of examining both A peptides and other biologically relevant peptides in the process. However, after performing a tryptic digestion atop a tissue surface, the mass spectrum was complex Figure 4 5 A & B illustrates a tryp t ic digestion of an A 1 40 standard deposited atop a Tg2576 mouse tissue surface and allowed to dry prior to tryptic digestion T he response for the [A 29
91 40+Na] + ion in Figure 4 5 A is l ow compared to other peaks within the spectrum Others have demonstrated the use of surfactants to enhance peptide signal for MSI studies [95, 97] Use of a surfactant ( OcGlc ) appear ed to decrease the lipid background response, by a factor of three; and as a result, A fragments of interest were readily identified. Although the use of OcGlc appeared to aid detection of the tryptic A peptides, further use was discarded due to the amount of OcGlc necessary for an experiment and a desire to generate results in a cost effe ctive manner The previous chapter demonstrated the benefits of removing lipids though a washing step in MSI for intact A peptides Figure 4 6 illustrates this poi nt for tryptic peptides The wash protocol set forth in the experimental section was appl ied to a rat brain tissue section prior to tryptic digestion. After the washes, the lipid signal observed in the m/z 700 900 region (Figure 4 6 A) is dramatically reduced compared to that of the unw ashed tissue section (Figure 4 6 B); note that the scale in A is 20X expanded over that in B. H owever even after washing, the response of [A 29 40+Na] + is in di stinguishable from the noise in these tissue sections that underwent tryptic digestion (a 1 L droplet of A 1 40 standard was applied to both brain sections prior to tissue washing) Despite this setback, other tools are available to tease out the [A 29 40+Na] + ion from the noise. Rather than performing an MS experiment as has been previously performed, what if the [A 29 40+Na] + ion was isolated and fragmented by CID (Figure 4 7)? When a A 1 40 standard was tryptically digested and the resulting solution co crystallized with DHB the following MS n spectra were collected. The most abundant ion in the MS 2 spectrum of [A 29 40+Na] + ( m/z 1107.6) was m/z 1008.7, (Figure 4 7A).
92 This corresponds to a neutral loss of the terminal valine on the C terminus side of the peptide. Note that the loss of a terminal valine on the C terminus side of the peptide would result in a NL of 117 (not NL 99). Because the parent species is sodiated, the peptide fragment could experience a neutral loss of a valine residue and a gain water forming a [b 11 +Na+H 2 O] + Lin et al. observed a similar phenomenon when performing CID experiments on alkali cationized peptides on a 3D ion trap  ; however, they observed [b n 1 +Na+OH] + ions after CID rather than [b n 1 +Na+H 2 O] + ions for the fragmentation of the A 29 40 ion (a difference of one dalton). Unfortunately further experiments did not explain the discrepancy. Unlike the MS 2 spectrum, the MS 3 spectrum is more straightforward. The spectrum illustrates a second neutral loss of valine on the C terminal side of the peptide (Figure 4 7B). Note that an addition water is not gained during CID. In s itu Tryptic Digestion of Tg2576 Brain Tissue An additional tactic to improve detection of tryptic peptides is to remove non relevant analytes from the tissue section (i.e., remove the lipids from the tissue section). One would anticipate that washing the tissue prior to tryptic digestion is better tha n after digestion because analyte migration is more likely if washing is performed after digestion (lower MW trypic peptides will likely be more soluble than intact A peptides) The comparison between washing before and after digestion is demonstrated in Figure 4 8 An A 29 40 standard (1 L droplet) was applied atop two blank rat brain sections. The washing steps described in the experimental section were applied to one tissue section prior to tryptic digestion and to another section after tryptic digestion Both the droplets of the amyloid standards had similar diameters prior to washing and digestion.
93 However, washing appears to move the analyte in the tissue Figures 4 8 A and B illustrate an MS 3 image of an ion originating from A 29 40 ( m/z 909.6 ). Despite similar starting conditions it appears that the A 29 40 fragment has migrated in the y direction (Figure 4 8 B) if one compares response of the two spots in the y direction (2.6 mm, Figure 4 8A, versus 3.1 mm, Figure 4 8B) Th e glass sli des were raised and lowered into the wash solution along the y axis, perhaps explaining why migration occurred along the y axis After a tryptic digestion was performed on a Tg2576 mouse sample, CID was applied to the ion at [A 29 40+Na] + Figure 4 9 i llustrates the MS 2 and MS 3 images of the dominant ions that are present in each spectrum, m/ z 1008.7 ([b 11 +Na+H 2 O ] + ) and m/z 909.6 ([b 10 +Na+H 2 O] + ) respectively The MS 2 image of m/z 1008.7 appea rs to localize across the anterior (left side of the image) and posterior (right) regions of the brain, with little localization in the top portion of the brain (top of the image) The MS 3 image of m/ z 909.6 offers a weaker signal (poorer S/N) that perhaps suggests similar distribution as in the MS 3 image. The d istribution of the A 29 40 ion in the MS 2 and MS 3 images should not be the case as the amyloid plaques should be confined to the cortex and hippocampus regions of the brain as demonstrated in the previous chapter. Due to the length of time the sections are immersed in the trypsin solution (2.5 h) it is possible that soluble peptides migrat e throughout the tissue section which is feasible because the trypsin solution does not readily evaporate but rather forms a atop the tissue section that st ays wet for the duration of the experiment (If the tissue section remains in a humid environment, the tissue can remain moist for over 12 hours). Indeed, it appears that the peptides have migrated from the cortex regions (see
94 Figure 3 3E of the brain to the anterior and posterior regions (left and right sides of the image) Other in situ digestion methods perform thei r digestions using a robotic spotter to deposit trypsin spots that are isolated from each other [46, 94, 95] This could solve the analyte migration problem; however, the spots are typically 200 m in diameter with 200 m separation between the spot s The size of the spots and the dist ance between the spots would be too large to determine the location of amyloid beta plaques as even the largest plaques are usually 200 m in diameter. Thus a different method must be employed. Reduction of Analyte Migration within Tissue during in situ Peptide Digestion Preliminary experiments have demonstrated that no digestion occur s if the tissue section is not damp through out the digestion period. However, this does not mean a trypsin solution must remain atop the tissue section. An alternative, is to apply a smaller volume trypsin solution atop the tissue surface then allow the solution to evaporate, leaving trypsin atop the tissue section. Then rehydration of the tissue can be effected by placing the tissue section in a humid environment. Initial experiments indi cate that tryptic digestion occurs using this d ry digestion method. Figure 4 10 illustrates the results of the experiment. Trypsin in an ethanol solution was applied to tissue section previously spotted with 1L of 1 mg/mL A 1 40 standard The right side of the figure illustrates a wet digestion where the trypsin solution is not allowed to dry and the right side of the figure illustrates a dry digestion where the trypsin solution is allowed to dry prior to placement into the humid incubation chamber (a kimwipe wetted with 1 mL of DI water wa s placed in the slide holder) To maximize the opportunity for analyte migration, both samples were incubated for a longer period of time (16 hours).
95 The distribution of the A 29 40 tryptic peptide ion in appears uniform acr oss the entire image in the we t digestion case (Figure 4 10 A), whereas the distri bution appears to be confined to the circle where the A 1 40 standard was applied in the dry digestion (Figure 4 10 B). If the MS 3 images (green) are superimposed over the MS images of the A 1 40 ion (red), the overlay images supports the assertion that little analyte migration has occurred in the dry digestio n (Figure 4 10 digestion experiment (Figure 4 10 C). Conclusions Digestion in situ of neuropeptides followed by MSI analysis offers a method to simultaneously examine high mass peptides and proteins while retain ing spatial integrity. A standards and endogenous peptides have been identified using MS 3 ; however initial experiments have indicated that analyte migration c an limit the spatial resolution of the MS image. In situ digestion experiments found extensive analyte migration if the trypsin solution is puddle atop the tissue section. However, preliminary results have indicated that and if trypsin solution is applie d then allowed to dry atop the tissue section and if the tissue section remains moist d igestion still occurs (with less analyte migration occurring) Unfortunately it appears that the trypsin digestion was incomplete as much of the intact A standard rem ained. Further experiments are necessary to viable alternative to expensive automated spotters for in situ tryptic digestion of proteins and peptides.
96 Figure 4 1. Image of trypsin dig estion apparatus. The tissue section bearing microscope slide was placed in a five slot microscope slide holder along with a wetted k imwipe. The slide h older was sealed with parafilm then placed in a sealed plastic bag. The kimwipe ensures a humid environment within the container; whereas the parafilm and plastic bag ensures that moisture will not escape from the container, thus drying out the tissue. The bag was then incubated within an oven to allow in situ tryptic digestion.
97 Figure 4 2 In silico digestion of A 1 40 using ProteinProspector  Although there are many fragments, only a few fragments are diagnostically relevant. Any fragments that do not terminate at the 40 th residue (V) are not usable as they could have a contribution from a different amyloid isoform.
98 Table 4 1 MWs of t ryptic fragments originating from A 1 40 Peak Predicted m/z of [M+H] + Predicted m/z of [M+Na] + Start End # of Missed Cleavages Sequence 1 1085.6 29 40 0 (K) GAIIGLMVGGVV( ) 1* 1107.6 29 40 0 (K) GAIIGLMVGGVV( ) 2 1325.7 17 28 0 (K) LVFFAEDVGSNK(G) 3 1336.6 6 16 0 (R) HDSGYEVHHQK(L) 4 1954.9 1 16 1 ( ) DAEFRHDSGYEVHHQK(L) 5 2392.3 17 40 1 (K) LVFFAEDVGSNKGAIIGLMVGGVV( ) 5* 2414.3 17 40 1 (K) LVFFAEDVGSNKGAIIGLMVGGVV( )
99 Figure 4 3 Comparison of MALDI matrices for analysis of A tryptic fragments. An A 1 40 standard was trypti cally digested and the resulting solution was spotted and analyzed. [M+H] + ions derived from A 1 40 are noted in the figure and the [M+Na] + ions are denoted with an asterisk ( ) CHCA (A) offers the le ast amount of background ions especially when compared to SA, ( B ). However, ions 2 and 3 are not informative as these fragments are common among amyloid beta fragments (e.g., 1 37, 1 38, 1 42, etc.) that have been tryptically digested Peak 1 and its sodiated analog (1*) are peaks unique to 40, as it contains the terminal C termin us. The molecular weights of the numbered peaks are found in Table 4 1.
100 Figure 4 4 MALDI MS spectrum of a t ryptic digestion of A 1 40 atop tissue using DHB as a MALDI matrix. Note the ions of interest (fragments that retain the terminal C terminus) a re the sodiated rather than proto nated.
101 Figure 4 5 Mass spectrum of a series of t ryptic digestion atop tissue Digested fragments (circled in red) are difficult to distinguish from the noise, A. However, use of a deter ge nt (OcG lc) may help improve analyte signal, B. The improvement of the signal could be due to the reduction of lipid response after treatment with OcGlc. The most abundant peak in A has 1.48e 5 counts versus 3.65e 4 for B The response of A 29 40 for both spectr a remained nearly identical (2.24e 4 counts for A versus 2.04e 4 counts for B).
102 Figure 4 6 Comparison of mass spectra from two in situ digestion experiments. A, illustrates the MS spectrum after a series of washes (see experimental section for details) were performed. B, illustrates the MS if no washes are performed. Note the high abundance of lipid in the 700 900 region of the spectrum. Although the lipid response was dramatically reduced, response of the [A 29 40 +Na ] + ion ( m/ z 1108) in A is not visible from the background
103 Figure 4 7 MS 2 and MS 3 spectra of the [ A 29 40 +Na] + ion m/z 1108.0 The top figure A, illustrates the MS 2 fragmentation of [ A 29 40 +Na] + ( m/z 1108 .0 ) The bottom figure B, illustrates the MS 3 fragmentat ion of 1108 .0 1008.7 major ion in each spectrum is due to a neutral loss of valine The sequence of the A 29 40 tryptic peptide is GAIIGLMVGGV V.
104 Figure 4 8 Comparison of w ashing a tissue section before and after tryptic digestion. A standard (a 1 L droplet) was applied atop two tissue sections. Although the A droplet had similar diameters between the two sections, washing after digestion appears to promote analyte migration in the y direction (the response of the tryptic peptide i on is 2.6 mm along the y axis for A versus 3.1 mm for B).
105 Figure 4 9 MS 2 and MS 3 i mages of A 29 40 after digestion of Tg2576 mice brain sections. The MS 2 image of m/z 1008.7 exhibits a localization of signal in the anterior (left in the image) and posterior (right) sections of the image, A; whereas, more even distribution of signal is noted in the MS 3 image of m/z 1108.0 1008.7 909.6
106 Figure 4 10 Comparison of wet and dry digestion techniques. The white circle in the figures indicates the area where the A 1 40 standard was deposited. The images on the left hand side demonstrate a digestion where the trypsin is applied atop the tissue surface and promptly placed in the sealed, humid incubation chamber. The images on th e righ t hand side illustrate a dry digestion where the trypsin solution is allowed to evaporate prior to placement into the sealed, humid incubation chamber. The top set of images (A and B) illustrate the localization of m/z 909.6 after performing an MS 3 expe riment isolating A 29 40 tryptic peptide ion then monitoring the second successiv e loss of valine (see Figure 4 7 for more details). The bottom set of images (C and D) illustrate the localization of the intact A 1 40 ion
107 CHAPTER 5 CONCLUSIONS AND FUTURE WORK Mass spectral imaging offers scientists a unique tool to determine the identity and distribution of a wide range of ions across a tiss ue section. Identification and l ocalization of peptides within normal and diseased tissue can help determine their biological role. Analyses of large peptides are commonly performed on a ToF MS instrument. However, peptide structural elucidation and identification is more difficult without multiple stages of mass spectrometry. The goal of this work was to demo nstrate the advantages and limitations of performing large peptide analysis with a MALDI LIT MS for MSI studies. The m/ z range of a commercial LIT was extended past the normal upper limit of m/ z 4000 to m/ z 5500. MS spectra were collected for A 1 40 an d 1 42 peptides. Additionally, tandem mass spectrometry was performed on these large peptides. An experiment examining the feasibility of the instrument to perform tandem mass spectrometry over a wide window (>80 Da) proved successful. Wide window tande m mass spectrometry experiments may prove useful for future quantitation studies of endogenous A peptides. MSI studies o ffer ed greater potential to localize intact peptides. It was found that i f no washing/fixation was performed during the sample preparation, lipid ions domi nate d the MS spectra and suppress ed the ion signal of the A peptides. A washing and fixation step prior to matrix application enhanced the response of the amyloid peptide fragments. Intact A fragments 1 38 1 40, and 1 42 ions were imaged in the tissue section and these MS i mage plots generally correlated with the Thioflavin S histological
108 images. MS n f ragmentation of A 1 40 ions was performed and these product ions also correlated well with the histologi cal image. Although MS image plots offered general localization of the amyloid plaques, closely grouped individual amyloid plaque formations were indistinguishable from one another in the MS image. The poor MS image resolution was due to the large spot size of the laser (~100 m) The MS image resolution was improved when the raster step size was decreased from 100 m to 50 m. After this change, individual plaque s 100 m apart were identified in the MS image. Experiments of peptides digested in situ proved more difficult than those of the intact species. MALDI matrix selection proved to be important for the study of digested A peptides. DHB appeared to promote the ionization of A fragments contain ing the C terminus of the peptide. CHCA and S A lacked these diagnostic peaks and thus were discarded in favor for DHB for the remaining experiments. Detection of tryptic peptides proved unviable unless a washing step and MS n were used. The washing step should be per formed prior to tryptic digestion ; otherwise peptides could migrate across the tissue section during the wash procedure. Unfortunately, t his was not the only sourc e of analyte migration. In situ digestion s of Tg2576 mice brain sections were performed. The distribution of A tryptic pept ides either indicated localization of amyloid plaques in the anterior and posterior regions of the brain (MS 2 image) or even distribution of the peptide across the tissue section (MS 3 image). This is in disagreement with known localization of the plaques in the cortex and hippocampus regions of the brain. These results suggest significant analyte migration occurred.
109 lyin g atop the tissue surface over the course of the incubation period (2.5 h) The necessary because no digestion was observed if the trypsin solution was allowed to evaporate and tissue was not kept damp. An alternative to this to dry; a fterward, the tissue section was p laced into a humid environment to ensure the tissue section remained damp. Although full digestion of the A 1 40 peptide was not achieved, analyte migration was minimized Further work is needed to determine if this method is viable for in situ digestion of tissue sections. This work would be valuable to pharmaceuti cal companies if a drug compound that targets a peptide biomarker and the biomarker itself could be localized However, the washing and fixation steps would likely eliminate any traces of the drug molecule from the tissue surface. Future studies could involve examining methods to fix both the drug compound and the p eptide in the tissue section to permit simultan eous analysis of both compounds; or, seria l sections could be examined (one section washed and the adjacent section unwashed). Another area of interest is the qua ntitation of peptide biomarkers The best way to perform a quantitative study would be to apply a deuterated or a uniformly N 15 labele d standard to the tissue surface and ratio the response of the peptide to the standard to determine the analyte concentration. However, for this experiment wide isolation MS n window s would be necessary to isolate both the peptide and its isotopically l abeled standard. And alternative to this would be to employ SWIFT waveforms to isolate the analyte and its standard and ignore any response in between the two peptides 
110 In addition to LTQ work, studies of large peptides on a hybrid LIT instrument (LTQ ToF or LTQ Orbitrap) could provide valuable complementary information. A n LTQ ToF instrument could offer analy sis of A peptides and tau proteins simultaneously. Tau proteins have been linked to AD, however, their molecular weight (~45 65 kDa)  is significantly greater than the expanded mass range of the LIT. Whereas an LTQ ToF would be used for its high MW range, the high resolution of an LTQ Orbitr ap instrument could prove helpful for distinguishing peptides from noise. Currently the orbitrap instrument will s can up to m/z 4000. However, a ccording to private communications with the instrument engineers at Thermo, the orbitrap can detect ion up to m/ z 5000 An orbitrap coupled to a n LIT with an extended mass range could prove useful for the analysis of large peptides.
111 LIST OF REFERENCES (1) Nelson, D.; Cox, M. Principles of Biochemistry 3rd ed.; Worth Publishers: New York, 2000. (2) Wouters, F. S.; Verveer, P. J.; Bastiaens, P. I. H. Imaging biochemistry inside cells. Trends Cell Biol. 2001 11 203 211. (3) Caprioli, R. M.; Farmer, T. B.; Gile, J. Molecular imaging of biological samples: localization of peptides and proteins usin g MALDI TOF MS. Anal. Chem. 1997 69 4751 4760. (4) Troendle, F. J.; Reddick, C. D.; Yost, R. A. Detection of pharmaceutical compounds in tissue by matrix assisted laser desorption/ionization and laser desorption/chemical ionization tandem mass spectromet ry with a quadrupole ion trap. J. Am. Soc. Mass Spectrom. 1999 10 1315 1321. (5) Seeley, E. H.; Caprioli, R. M. Molecular imaging of proteins in tissues by mass spectrometry. Proc. Natl. Acad. Sci. U. S. A. 2008 105 18126 18131. (6) Garrett, T. J.; Pri eto Conaway, M. C.; Kovtoun, V.; Bui, H.; Izgarian, N.; Stafford, G.; Yost, R. A. Imaging of small molecules in tissue sections with a new intermediate pressure MALDI linear ion trap mass spectrometer. Int. J. Mass Spectrom. 2007 260 166 176. (7) Schwartz, S. A.; Reyzer, M. L.; Caprioli, R. M. Direct tissue analysis using matrix assisted laser desorption/ionization mass spectrometry: Practical aspects of sample preparation. J. Mass Spectrom. 2003 38 699 708. (8) Skoeld, K.; Svensson, M.; Nilsson, A.; Falth, M.; Svenningsson, P.; Andren, P. The importance of sample handling in neuropeptidomics, 2008; John Wiley & Sons, Inc.; 177 189. (9) Landgraf, R. R.; Conaway, M. C. P.; Garrett, T. J.; Stacpoole, P. W.; Yost, R. A. Imaging of Lipids in Spinal C ord Using Intermediate Pressure Matrix Assisted Laser Desorption Linear Ion Trap/Orbitrap MS. Anal. Chem. 2009 81 8488 8495. (10) Crossman, L.; McHugh, N. A.; Hsieh, Y.; Korfmacher, W. A.; Chen, J. Investigation of the profiling depth in matrix assisted laser desorption/ionization imaging mass spectrometry. Rapid Commun. Mass Spectrom. 2006 20 284 290. (11) Chaurand, P.; Schwartz, S. A.; Billheimer, D.; Xu, B. J.; Crecelius, A.; Caprioli, R. M. Integrating histology and imaging mass spectrometry. Anal. Chem. 2004 76 1145 1155.
112 (12) Scherl, A.; Zimmermann Ivol, C. G.; Di, D. J.; Vaezzadeh, A. R.; Binz, P. A.; Amez Droz, M.; Cochard, R.; Sanchez, J. C.; Glueckmann, M.; Hochstrasser, D. F. Gold coating of non conductive membranes before matrix assisted la ser desorption/ionization tandem mass spectrometric analysis prevents charging effect. Rapid Commun. Mass Spectrom. 2005 19 605 610. (13) Ayorinde, F. O.; Hambright, P.; Porter, T. N.; Keith, Q. L., Jr. Use of meso Tetrakis(pentafluorophenyl)porphyrin as a Matrix for Low Molecular Weight Alkylphenol Ethoxylates in Laser Desorption/Ionization Time of Flight Mass Spectrometry. Rapid Commun. Mass Spectrom. 1999 13 2474 2479. (14) Cohen, L. H.; Gusev, A. I. Small molecule analysis by MALDI mass spectrometry Anal. Bioanal. Chem. 2002 373 571 586. (15) Lemaire, R.; Tabet, J. C.; Ducoroy, P.; Hendra, J. B.; Salzet, M.; Fournier, I. Solid ionic matrixes for direct tissue analysis and MALDI imaging. Anal. Chem. 2006 78 809 819. (16) Meriaux, C.; Franck, J.; Wisztorski, M.; Salzet, M.; Fournier, I. Liquid ionic matrixes for MALDI mass spectrometry imaging of lipids. J. Proteomics 2010 73 1204 1218. (17) Baluya, D. L.; Garrett, T. J.; Yost, R. A. Automated MALDI Matrix Deposition Method with Inkjet Printing f or Imaging Mass Spectrometry. Anal. Chem. 2007 79 6862 6867. (18) Aerni, H. R.; Cornett, D. S.; Caprioli, R. M. Automated Acoustic Matrix Deposition for MALDI Sample Preparation. Anal. Chem. 2006 78 827 834. (19) Hankin, J. A.; Barkley, R. M.; Murphy, R. C. Sublimation as a Method of Matrix Application for Mass Spectrometric Imaging. J. Am. Soc. Mass Spectrom. 2007 18 1646 1652. (20) Puolitaival, S. M.; Burnum, K. E.; Cornett, D. S.; Caprioli, R. M. Solvent Free Matrix Dry Coating for MALDI Imaging of Phospholipids. J. Am. Soc. Mass Spectrom. 2008 19 882 886. (21) Kaletas, B. K.; van der Wiel, I. M.; Stauber, J.; Dekker, L. J.; Guzel, C.; Kros, J. M.; Luider, T. M.; Heeren, R. M. A. Sample preparation issues for tissue imaging by imaging MS. Proteomics 2009 9 2622 2633. (22) Khatib Shahidi, S.; Andersson, M.; Herman, J. L.; Gillespie, T. A.; Caprioli, R. M. Direct Molecular Analysis of Whole Body Animal Tissue Sections by Imaging MALDI Mass Spectrometry. Anal. Chem. 2006 78 6448 6456. (23) Solon, E. G.; Schweitzer, A.; Stoeckli, M.; Prideaux, B. Autoradiography, MALDI MS, and SIMS MS Imaging in Pharmaceutical Discovery and Development. AAPS J. 2010 12 11 26.
113 (24) Karas, M.; Hillenkamp, F. Laser desorption io nization of proteins with molecular masses exceeding 10,000 daltons. Anal. Chem. 1988 60 2299 2301. (25) Hillenkamp, F.; Karas, M.; Beavis, R. C.; Chait, B. T. Matrix assisted laser desorption/ionization mass spectrometry of biopolymers. Anal. Chem. 1991 63 1193A 1203A. (26) Simpson, R. J. Proteins and Proteomics: A Laboratory Manual ; Cold Spring Harbor Laboratory Press: New York, NY, 2003. (27) Cooks, R. G.; Ouyang, Z.; Takats, Z.; Wiseman, J. M. Ambient Mass Spectrometry. Science 2006 311 1566 1570. (28) Takats, Z.; Wiseman, J. M.; Gologan, B.; Cooks, R. G. Mass Spectrometry Sampling Under Ambient Conditions with Desorption Electrospray Ionization. Science 2004 306 471 473. (29) Kertesz, V.; Van Berkel, G. J. Scanning and Surface Alignment Considerations in Chemical Imaging with Desorption Electrospray Mass Spectrometry. Anal. Chem. 2008 80 1027 1032. (30) Takats, Z.; Wiseman, J. M.; Cooks, R. G. Ambient mass spectrometry using desorption electrospray ionization (DESI): Instrumentation, mechanisms and applications in forensics, chemistry, and biology. J. Mass Spectrom. 2005 40 1261 1275. (31) Ifa, D. R.; Wiseman, J. M.; Song, Q.; Cooks, R. G. Development of capabilities for ima ging mass spectrometry under ambient conditions with desorption electrospray ionization (DESI). Int. J. Mass Spectrom. 2007 259 8 15. (32) Pasilis, S. P.; Kertesz, V.; Van Berkel, G. J. Surface Scanning Analysis of Planar Arrays of Analytes with Desorpti on Electrospray Ionization Mass Spectrometry. Anal. Chem. 2007 79 5956 5962. (33) Van Berkel, G. J.; Kertesz, V.; Koeplinger, K. A.; Vavrek, M.; Kong, A. N. T. Liquid microjunction surface sampling probe electrospray mass spectrometry for detection of dr ugs and metabolites in thin tissue sections. J. Mass Spectrom. 2008 43 500 508. (34) Kertesz, V.; Van Berkel, G. J. Fully automated liquid extraction based surface sampling and ionization using a chip based robotic nanoelectrospray platform. J. Mass Spectrom. 2010 45 252 260. (35) Northen, T. R.; Yanes, O.; Northen, M. T.; Marrinucci, D.; Uritboonthai, W.; Apon, J.; Golledge, S. L.; Nordstrom, A.; Siuzdak, G. Clathrate nanostructures for mass spectrometry. Nature 2007 449 1033 1036.
114 (36) Patti, G. J.; Woo, H. K.; Yanes, O.; Shriver, L.; Thomas, D.; Uritboonthai, W.; Apon, J. V.; Steenwyk, R.; Manchester, M.; Siuzdak, G. Detection of Carbohydrates and Steroids by Cation Enhanced Nanostructure Initiator Mass Spectrometry (NIMS) for Biofluid Analysis and Tissue Imaging. Anal. Chem. 2010 82 121 128. (37) Reyzer, M. L.; Hsieh, Y.; Ng, K.; Korfmacher, W. A.; Caprioli, R. M. Direct analysis of drug candidates in tissue by matrix assisted laser desorption/ionization mass spectrometry. J. Mass Spectrom. 20 03 38 1081 1092. (38) Cornett, D. S.; Frappier, S. L.; Caprioli, R. M. MALDI FTICR Imaging Mass Spectrometry of Drugs and Metabolites in Tissue. Anal. Chem. 2008 80 5648 5653. (39) Stoeckli, M.; Staab, D.; Schweitzer, A. Compound and metabolite distribution measured by MALDI mass spectrometric imaging in whole body tissue sections. Int. J. Mass Spectrom. 2007 260 195 202. (40) Bouslimani, A.; Bec, N.; Glueckmann, M.; Hirtz, C.; Lar roque, C. Matrix assisted laser desorption/ionization imaging mass spectrometry of oxaliplatin derivatives in heated intraoperative chemotherapy (HIPEC) like treated rat kidney. Rapid Commun. Mass Spectrom. 2010 24 415 421. (41) Chen, J.; Hsieh, Y.; Knem eyer, I.; Crossman, L.; Korfmacher, W. A. Visualization of first pass drug metabolism of terfenadine by MALDI imaging mass spectrometry. Drug Metab. Lett. 2008 2 1 4. (42) Li, F.; Hsieh, Y.; Kang, L.; Sondey, C.; Lachowicz, J.; Korfmacher, W. A. MALDI ta ndem mass spectrometry imaging of astemizole and its primary metabolite in rat brain sections. Bioanalysis 2009 1 299 307. (43) Atkinson, S. J.; Loadman, P. M.; Sutton, C.; Patterson, L. H.; Clench, M. R. Examination of the distribution of the bioreducti ve drug AQ4N and its active metabolite AQ4 in solid tumours by imaging matrix assisted laser desorption/ionisation mass spectrometry. Rapid Commun. Mass Spectrom. 2007 21 1271 1276. (44) Trim, P. J.; Henson, C. M.; Avery, J. L.; McEwen, A.; Snel, M. F.; Claude, E.; Marshall, P. S.; West, A.; Princivalle, A. P.; Clench, M. R. Matrix Assisted Laser Desorption/Ionization Ion Mobility Separation Mass Spectrometry Imaging of Vinblastine in Whole Body Tissue Sections. Anal. Chem. 2008 80 8628 8634. (45) Magpa rangalan, D. P.; Garrett, T. J.; Drexler, D. M.; Yost, R. A. Analysis of Large Peptides by MALDI Using a Linear Quadrupole Ion Trap with Mass Range Extension. Anal. Chem. 2010 82 930 934.
115 (46) Groseclose, M. R.; Andersson, M.; Hardesty, W. M.; Caprioli, R. M. Identification of proteins directly from tissue: in situ tryptic digestions coupled with imaging mass spectrometry. J. Mass Spectrom. 2007 42 254 262. (47) Schwartz, J. C.; Senko, M. W.; Syka, J. E. P. A two dimensional quadrupole ion trap mass spe ctrometer. J. Am. Soc. Mass Spectrom. 2002 13 659 669. (48) Perchalski, R. Characteristics and Application of a Laser Ionization/Evaporation Source for Tandem Mass Spectrometry University of Florida, Gainesville, FL, 1985. (49) Sleno, L.; Volmer, D. A. Some fundamental and technical aspects of the quantitative analysis of pharmaceutical drugs by matrix assisted laser desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom. 2005 19 1928 1936. (50) Gobey, J.; Cole, M.; Janiszewski, J.; Covey, T.; Chau, T.; Kovarik, P.; Corr, J. Characterization and performance of MALDI on a triple quadrupole mass spectrometer for analysis and quantification of small molecules. Anal. Chem. 2005 77 5643 5654. (51) Ho pfgartner, G.; Varesio, E.; Stoeckli, M. Matrix assisted laser desorption/ionization mass spectrometric imaging of complete rat sections using a triple quadrupole linear ion trap. Rapid Commun. Mass Spectrom. 2009 23 733 736. (52) Kertesz, V.; Van Berkel G. J.; Vavrek, M.; Koeplinger, K. A.; Schneider, B. B.; Covey, T. R. Comparison of Drug Distribution Images from Whole Body Thin Tissue Sections Obtained Using Desorption Electrospray Ionization Tandem Mass Spectrometry and Autoradiography. Anal. Chem. 2 008 80 5168 5177. (53) Strupat, K.; Kovtoun, V.; Bui, H.; Viner, R.; Stafford, G.; Horning, S. MALDI Produced Ions Inspected with a Linear Ion Trap Orbitrap Hybrid Mass Analyzer. J. Am. Soc. Mass Spectrom. 2009 20 1451 1463. (54) Watson, J. T.; Sparkma n, O. D. Introduction to Mass Spectrometry: Instrumentation, Applications, and Strategies for Data Interpretation 4th ed.; Wiley: Hoboken, NJ, 2007. (55) Wong, P. S. H.; Cooks, R. G. Ion trap mass spectrometry. Curr. Separations 1997 16 85 92. (56) Marc h, R. E.; Todd, J. F. J. Quadrupole Ion Trap Mass Spectrometry 2nd ed.; Wiley: Hoboken, NJ, 2005. (57) Todd, J. F. J. Introduction to practical aspects of ion trap mass spectrometry. In Pract. Aspects Ion Trap Mass Spectrom. ; CRC, 1995; Vol. 1, pp 3 24.
116 ( 58) Douglas, D. J.; Frank, A. J.; Mao, D. M. Linear ion traps in mass spectrometry. Mass Spectrom. Rev. 2005 24 1 29. (59) Stafford, G. C., Jr.; Kelley, P. E.; Syka, J. E. P.; Reynolds, W. E.; Todd, J. F. J. Recent improvements in and analytical applicat ions of advanced ion trap technology. Int. J. Mass Spectrom. Ion Processes 1984 60 85 98. (60) Mikesh, L. M.; Ueberheide, B.; Chi, A.; Coon, J. J.; Syka, J. E. P.; Shabanowitz, J.; Hunt, D. F. The utility of ETD mass spectrometry in proteomic analysis. B iochim. Biophys. Acta 2006 1764 1811 1822. (61) Zubarev, R. A. Electron capture dissociation tandem mass spectrometry. Curr. Opin. Biotechnol. 2004 15 12 16. (62) Selkoe, D. J. Degenerative disorders: Alzheimer disease. In Molecular and Genetic Basis of Neurologic and Psychiatric Disease (4th Edition) ; Rosenberg, R. N., Dimauro, S., Paulson, H. L., Ptacek, L., Nestler, E. J., Eds.; Lippincott Williams & Wilkins: Philadelphia, PA, 2008, pp 317 329. (63) National Institute on Aging ( http://www.nia.nih.gov/Alzheimers/ ) (Accessed June 20, 2010). (64) Ford, M. J.; Cantone, J. L.; Polson, C.; Toyn, J. H.; Meredith, J. E.; Drexler, D. M. Qualitative and quantitative characterization of the amyloid beta peptide (Abeta ) population in biological matrices using an immunoprecipitation LC/MS assay. J. Neurosci. Methods 2008 168 465 474. (65) Geula, C. Pathological Diagnosis of Alzheimer's Disease. In Early Diagnosis of Alzheimer's Disease ; Scinto, L. F. M., Daffne r, K. R., Eds.; Humana Press, Inc.: Totowa, NJ, 2000, pp 65 82. (66) Deshpande, A.; Mina, E.; Glabe, C.; Busciglio, J. Different Conformations of Amyloid beta Induce Neurotoxicity by Distinct Mechanisms in Human Cortical Neurons. J. Neurosci. 2006 26 601 1 6018. (67) Stoeckli, M.; Staab, D.; Staufenbiel, M.; Wiederhold, K. H.; Signor, L. Molecular imaging of amyloid beta peptides in mouse brain sections using mass spectrometry. Anal. Biochem. 2002 311 33 39. (68) Rohner, T. C.; Staab, D.; Stoeckli, M. MA LDI mass spectrometric imaging of biological tissue sections. Mech. Ageing Dev. 2005 126 177 185. (69) Stoeckli, M.; Knochenmuss, R.; McCombie, G.; Mueller, D.; Rohner, T.; Staab, D.; Wiederhold, K. H. MALDI MS imaging of amyloid. In Amyloid, Prions, and Other Protein Aggregates, Pt B ; Elsevier Academic Press Inc: San Diego, 2006; Vol. 412, pp 94 106.
117 (70) Petricoin, E.; Wulfkuhle, J.; Espina, V.; Liotta, L. A. Clinical proteomics: Revolutionizing disease detection and patient tailoring therapy. J. Proteo me Res. 2004 3 209 217. (71) Li, W.; Nemirovskiy, O.; Fountain, S.; Mathews, W. R.; Szekely Klepser, G. Clinical validation of an immunoaffinity LC MS/MS assay for the quantification of a collagen type II neoepitope peptide: A biomarker of matrix metallo proteinase activity and osteoarthritis in human urine. Anal. Biochem. 2007 369 41 53. (72) Pisitkun, T.; Johnstone, R.; Knepper, M. A. Discovery of urinary biomarkers. Mol. Cell. Proteomics 2006 5 1760 1771. (73) Wu, S. L.; Amato, H.; Biringer, R.; Cho udhary, G.; Shieh, P.; Hancock, W. S. Targeted proteomics of low level proteins in human plasma by LC/MSn: Using human growth hormone as a model system. J. Proteome Res. 2002 1 459 465. (74) Taylor, S. W.; Andon, N. L.; Bilakovics, J. M.; Lowe, C.; Hanley, M. R.; Pittner, R.; Ghosh, S. S. Efficient High Throughput Discovery of Large Peptidic Hormones and Biomarkers. J. Proteome Res. 2006 5 1776 1784. (75) Kaiser, R. E., Jr.; Cooks, R. G.; St afford, G. C., Jr.; Syka, J. E. P.; Hemberger, P. H. Operation of a quadrupole ion trap mass spectrometer to achieve high mass/charge ratios. Int. J. Mass Spectrom. Ion Processes 1991 106 79 115. (76) Masters, C. L.; Simms, G.; Weinman, N. A.; Multhaup, G.; McDonald, B. L.; Beyreuther, K. Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc. Natl. Acad. Sci. U. S. A. 1985 82 4245 4249. (77) Hardy, J.; Selkoe, D. J. The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeutics. Science 2002 297 353 356. (78) Selkoe, D. J.; Schenk, D. Alzheimer's disease: Molecular understanding predicts amyloid based therapeutics. Annu. Rev. Pharmacol. Toxicol. 2003 43 545 584. (79) Schupf, N.; Tang, M. X.; Fukuya ma, H.; Manly, J.; Andrews, H.; Mehta, P.; Ravetch, J.; Mayeux, R. Peripheral Abeta subspecies as risk biomarkers of Alzheimer's disease. Proc. Natl. Acad. Sci. U. S. A. 2008 105 14052 14057. (80) Graff Radford Neill, R.; Crook Julia, E.; Lucas, J.; Boev e Bradley, F.; Knopman David, S.; Ivnik Robert, J.; Smith Glenn, E.; Younkin Linda, H.; Petersen Ronald, C.; Younkin Steven, G. Association of low plasma Abeta42/Abeta40 ratios with increased imminent risk for mild cognitive impairment and Alzheimer diseas e. Arch. Neurol. 2007 64 354 362. (81) ProteinProspector ( http://prospector.ucsf.edu/ ) (Accessed July 25, 2009).
118 (82) Grasso, G.; Mineo, P.; Rizzarelli, E.; Spoto, G. MALDI, AP/MALDI and ESI techniques for the MS detection of amyloid beta peptides. Int. J. Mass Spectrom. 2009 282 50 55. (83) Qin, J.; Chait, B. T. Preferential Fragmentation of Protonated Gas Phase Peptide Ions Adjacent to Acidic Amino Acid Residues. J. Am. Chem. Soc. 1995 117 5411 5412. (84) Garrett, T. J.; Yost, R. A. Analysis of Intact Tissue by Intermediate Pressure MALDI on a Linear Ion Trap Mass Spectrometer. Anal. Chem. 2006 78 2465 2469. (85) Reich, R. F.; Cudzilo, K.; Yost, R. A. Quantitative Imaging of Cocaine and Its Metabolites i n Postmortem Brain Tissue by Intermediate Pressure MALDI/Linear Ion Trap Tandem Mass Spectrometry, Proceedings of the 56th ASMS Conference on Mass Spectrometry Denver, CO, June 1 5 2008. (86) Centers for Disease Control and Prevention ( http://www.cdc.gov ) (Accessed June 20, 2010). (87) Klunk, W. E.; Engler, H.; Nordberg, A.; Wang, Y. M.; Blomqvist, G.; Holt, D. P.; Bergstrom, M.; Savitcheva, I.; Huang, G. F.; Estrada, S.; Ausen, B.; Debnath, M. L.; Barletta, J.; Price, J. C.; Sandell, J.; Lopresti, B. J.; Wall, A.; Koivisto, P.; Antoni, G.; Mathis, C. A.; Langstrom, B. Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound B. Ann. Neurol. 2004 55 306 319. (88) Westermark, G. T.; Johnson, K. H.; Westermark, P. Staining methods for identification of amyloid in tissue. In Methods Enzymol. ; Academic Press, 1999; Vol. 309, pp 3 25. (89) Pype, S.; Moechars, D.; Dillen, L.; Mercken, M. Characterization of amyloid pe ptides from brain extracts of transgenic mice overexpressing the London mutant of human amyloid precursor protein. J. Neurochem. 2003 84 602 609. (90) Stoeckli, M.; Farmer, T. B. MALDI MS imaging in biomedical research. In Biomed. Appl. Proteomics ; Sanch ez, J. C., Corthals, G. L., Hochstrasser, D. F., Eds.; Wiley VCH: Weinheim, 2004, pp 373 388. (91) Hsiao, K.; Chapman, P.; Nilsen, S.; Eckman, C.; Harigaya, Y.; Younkin, S.; Yang, F.; Cole, G. Correlative Memory Deficits, Abeta Elevation, and Amyloid Plaqu es in Transgenic Mice. Science 1996 274 99 103. (92) Jakobs, S. High resolution imaging of live mitochondria. Biochim. Biophys. Acta, Mol. Cell. Res 2006 1763 561 575. (93) Steen, H.; Mann, M. The abc's (and xyz's) of peptide sequencing. Nat. Rev. Mol. Cell Biol. 2004 5 699 711.
119 (94) Lemaire, R.; Desmons, A.; Tabet, J. C.; Day, R.; Salzet, M.; Fournier, I. Direct Analysis and MALDI Imaging of Formalin Fixed, Paraffin Embedded Tissue Sections. J. Proteome Res. 2007 6 1295 1305. (95) Setou, M.; Hayasaka, T.; Shimma, S.; Sugiura, Y.; Matsumoto, M. Protein denaturation improves enzymatic digestion efficiency for direct tissue analysis using mass spectrometry. Appl. Surf. Sci. 2008 255 1555 1559. (96) CHIP 1000 ( http://www.ssi.shimadzu.com/products/literature/Biotech/CHIP 1000.pdf ) (Accessed December 1, 2010). (97) Djidja, M. C.; Francese, S.; Loadman, P. M.; Sutton, C. W.; Scriven, P.; Claude, E.; Snel M. F.; Franck, J.; Salzet, M.; Clench, M. R. Detergent addition to tryptic digests and ion mobility separation prior to MS/MS improves peptide yield and protein identification for in situ proteomic investigation of frozen and formalin fixed paraffin embe dded adenocarcinoma tissue sections. Proteomics 2009 9 2750 2763. (98) Lin, T.; Payne, A. H.; Glish, G. L. Dissociation pathways of alkali cationized peptides: opportunities for C terminal peptide sequencing. J. Am. Soc. Mass Spectrom. 2001 12 497 504. (99) Reich, R. F.; Cromwell, K. N.; Yost, R. A. MALDI MS n Quantitation by Selective Isolation of Analyte and Internal Standard Ions Using a Multi Notch SWIFT Waveform, Proceedings of the 57th ASMS Conference on Mass Spectrometry Philadelphia, PA, May 31 June 4 2009. (100) Bue, L.; Bussire, T.; Bue Scherrer, V.; Delacourte, A.; Hof, P. R. Tau protein isoforms, phosphorylation and role in neurodegenerative disorders. Brain Research Reviews 2000 33 95 130.
BIOGRAPHICAL SKETCH Daniel Patrick Magparangalan was born in 1977 to Danilo and Anita Magparangalan, the first of three children. Shortly after his birth, his family moved to St. Louis where Daniel spent the majority of his years. Daniel attended Saint Louis University for his undergraduate studies and majored in Chemistry. After graduation, Daniel was employed at Mallinckrodt Inc. in various positions. Initially, he worked as a peptide chemist per forming peptide synthesis to support development effo rts. Later, he moved to the quality control portion of the group. Later, he moved from analyzing large biomolecules to HPLC method development of small drug molecules. It was here that he was first introduced to mass spectrometry Wanting to learn more about MS he eventually left Mallinckrodt left to study at the University of Florida. Daniel joined the chemistry program in 2005 under the advisement of Dr. Richard A. Yost.