APPLICATION SOLUTIONS FOR STRUCTURAL PROTEOMICS

APPLICATION SOLUTIONS FOR STRUCTURAL PROTEOMICS

TABLE OF CONTENTS The Application of SYNAPT High Defintion Mass Spectrometry for the Conformational Studies of Protein Complexes... 1 High Definition Mass Spectrometry as a Tool for Structural Investigation of High m/z Ion Species... 5 Investigating Gas Composition on Transport and Desolvation of High m/z Species in the First Vacuum Stages of a Mass Spectrometer... 9 High Mass Precursor Ion Selection Utilizing Monoatomic and Polyatomic Collision Gases... 13

THE APPLICATION OF SYNAPT HIGH DEFINTION MASS SPECTROMETRY FO R T H E CONFO RMAT IO NA L S T U DIES O F P ROT EIN COM P L E X E S Iain Campuzano, Kevin Giles, Thérèse McKenna, Christopher J. Hughes, and James I. Langridge Waters Corporation, Manchester, UK INT RODUCTION Electrospray (ESI) is a gentle form of ionization that enables the transfer of large multi-protein structures with little or no fragmentation into the gas phase. The coupling of ESI to mass spectrometry (MS) allows the detection and accurate mass measurement of non-covalently assembled macromolecular protein complexes. This transfer of non-covalently associated proteinprotein complexes from solution to the gas phase generally results in the formation of ions possessing relatively few charges. As such, the m/z values are often above 10,000 and, in some cases MS/MS activation of such complexes can produce ions with m/z values in excess of 20,000. The Waters Synapt TM High Definition Mass Spectrometry TM (HDMS TM ) System combines high-efficiency ion mobility-based (IMS) measurements and separations with a hybrid quadrupole orthogonal acceleration time-of-flight (oa-tof) mass spectrometer. This combination enables both accurate mass measurements of intact biomolecular complexes and the potential to measure their collisional cross sections, including differences in cross section, produced upon activation. The IMS separation allows the detection of subtle conformation differences that are not evident from spectral data alone. 1 In this application note, we describe the analysis of GroEL (Figure 1), an 800 kda non-covalently associated protein-protein complex using the Synapt HDMS System. EXPERIMENTAL Sample The GroEL protein was buffer exchanged into an aqueous solution of 100 mm ammonium acetate, to a final working protein concentration of 3 µm. Figure 2. Schematic of the Synapt HDMS System, which incorporates Triwave technology. Instrumentation Samples were introduced using nano-electrospray ionization. Ions produced were sampled by the ZSpray TM source of the Synapt HDMS System (Figure 2). The ions pass through a quadrupole and are either set to transmit a substantial mass range or to select a particular m/z before entering the Triwave TM device. Triwave is comprised of three T-Wave TM devices. 2 The first device, the Trap T-Wave, accumulates ions. These stored ions are gated (500 µsec) into the second device, the IMS T-Wave, where they are separated according to their mobilities. The final Transfer T-Wave is used to transport the separated ions into the oa-tof for MS analysis. The pressure in the Trap and Transfer T-Wave regions was 7e-2 mbar of Argon and the pressure in the IMS T-Wave was 0.5 mbar of Nitrogen. Figure 1. Space-filling model of the protein complex, GroEL, illustrating the overall ring shaped topology from both arial (left) and side views (right), respectively. 1

TOF-MS analysis was carried out to determine the accurate mass of the intact GroEL tetradecamer complex from the m/z of the different charge-states present. The instrument was calibrated over the m/z range 1000 to 32,000 using a solution of caesium iodide. In a second analysis, the precursor ions from the intact tetradecamer were selected using the quadrupole, which in this instrument enables selection of ions up to an m/z value of 32,000 Da. The selected ions were then fragmented in the gas-filled Trap T-Wave. HDMS mode analysis was also conducted whereby mobility separations in the IMS T-Wave enabled the separation of species by their mobilities and drift time data could be derived. RESULTS AND DISCUSSION The mass spectrum obtained for the intact GroEL tetradecamer protein is shown in Figure 3 (left), where multiply charged ions distributed around m/z 12,000 can be seen. These represent charge states centered around +68, which demonstrate that under these native conditions it is possible to preserve the interactions of the 14, 57 kda subunits. As a result, the intact mass observed is that of the intact tetradecamer (14 mer, 800 kda) with a mass of 800 kda. Activation and subsequent fragmentation of the GroEL complex, Figure 3 (right), occurs in the Trap T-Wave with injection voltages of up to 150 V, which disrupts the large macromolecular assemblies. Operating the Trap T-Wave at elevated pressures allows for efficient fragmentation of the GroEL complex. These high pressures also Figure 3. Schematic representation of the activation of the GroEL complex and the MS and MS/MS spectra obtained. 2

provide efficient collisional cooling and focusing of the intact 14 mer and subsequent 13 mer generated from the activation process. The acquired HDMS data was analyzed in Driftscope TM Software, where the drift time of each ion is plotted versus the m/z of the ion. The intensity of an ion is indicated using a color scale from least (blue/red) to most (yellow/white) intense. The multiply-charged series of peaks for the intact GroEL, Figure 4 (left), range in m/z from 11,000 to 12,000 with drift times ranging from 13 msec for the higher charge states to 19 msec for the lower charge states. Activation of the intact GroEL and subsequent ion mobility separation, Figure 4 (right), shows that there are clear differences in drift time of the monomer, the precursor ion, and the remaining 13 mer (744 kda) complex. This 13 mer carries relatively few charges (between +33 and +50). An expanded Driftcope plot of the 13 mer region is shown in Figure 5, where the individual charge states have well defined drift time differences. Figure 5. Expanded drift time vs. m/z region of Figure 4, showing the conformational heterogeneity of the GroEL 13 mer. Figure 4. Left - HDMS analysis (drift-time vs. m/z) of the intact GroEL tetradecamer. Right - HDMS analysis (Drift-time vs. m/z) of GroEL (+68 charge state), activated with 150 V in the TRAP T-Wave. Driftscope Software where each pixel represents one ion with color representing intensity from low (blue) to high (yellow). 3

It is evident that there are two distinct drift time populations for the 13 mer (744 kda), region A and region B. When the mass spectra are extracted from these regions, Figure 6, there are two different charge state envelopes, both of which deconvolute to the mass of the 13 mer (744 kda). Region A is far more intense than region B. The two different populations are postulated to be a result of the monomer being ejected from a different position within the GroEL 14 mer, with one mechanism of ejection being favoured over another, and thus represent different conformations of the GroEL 13 mer. CONCLUSIONS: n The Synapt HDMS System was used to separate and mass analyze different conformations of a large intact protein species in the gas phase. n Ions with high m/z ratios can be isolated for fragmentation (MS/MS) using a Synapt HDMS System equipped with a 32 kda quadrupole. n HDMS analysis provided new evidence on the fragmentation mechanism of the GroEL protein complex. n This additional dimension of specificity obtained in HDMS analysis has provided insight into protein complex conformation (and fragmentation pathways), which would be impossible by MS or MS/MS alone. References 1. Ruotolo, Giles, Campuzano, Sandercock, Bateman & Robinson, Science, 9th December 2005, vol 310, 1569-1724 2. Travelling Wave Ion Propulsion in Collision Cells K. Giles, S. Pringle, K. Worthington and R. Bateman. Presented at the 51st ASMS Conference, Montreal, Canada 2003. 3. The traveling wave device described here is similar to that described by Kirchner in US Patent 5,206,506; 1993. Figure 6. Extracted spectra from Region A and Region B in Figure 5, showing the different charge-state distributions for each potential confirmation of the GroEL 13 mer. Waters is a registered trademark of Waters Corporation. The Science of What s Possible, Synapt, High Definition Mass Spectrometry, HDMS, Triwave, T-Wave, ZSpray and Driftscope are trademarks of Waters Corporation. All other trademarks are property of their respective owners. 2007 Waters Corporation Printed in the USA September 2007 720002343EN AG-PDF Waters Corporation 34 Maple Street Milford, MA 01757 U.S.A. T: 1 508 478 2000 F: 1 508 872 1990 www.waters.com 4

High Definition Mass Spectrometry as a Tool for Structural Investigation of High m/z Ion Species Iain Campuzano and Kevin Giles Waters Corporation, Manchester, UK INTRODUCTION Over the past 10 years, interest in high mass non-covalent analysis has increased due to the ability of the current mass spectrometers and electrospray sources to preserve the non-covalent interactions, allowing one to analyze compounds in their native conformation and stoichiometry. 1 The transfer of non-covalently associated complexes from solution to the gas phase using electrospray ionization generally results in the formation of ions possessing relatively few charges. As a result, these species appear high on the m/z scale, making time-of-flight (TOF) mass spectrometry ideal for their mass analysis. The utility of ion mobility spectrometry (IMS) in probing the structures of relatively large complexes has been highlighted previously. 1 Here we present the use of high-efficiency IMS (Triwave TM Technology) on a SYNAPT TM High Definition Mass Spectrometry TM (HDMS TM ) System for analysis of high m/z caesium iodide clusters over the m/z range 1,000-20,000. This demonstrates the utility of the SYNAPT HDMS System for the mass measurement of high m/z species, such as in the analysis of non-covalent protein complexes. EXPERIMENTAL The instrument used in this study was a SYNAPT HDMS System, which combines high-efficiency ion mobility based measurements and separations with a hybrid quadrupole orthogonal acceleration Figure 1. Schematic of the SYNAPT HDMS System. 5

time-of-flight (oa-tof) mass spectrometer, as shown in Figure 1 2. Samples were introduced with a borosilcate-glass nano electrosprayspray tip and sampled into the vacuum system. The ions pass through a quadrupole mass filter to the enabling Triwave device. Triwave consists of three travelling wave (T-Wave TM ) ion guides. The TRAP T-Wave accumulates ions (with high efficiency), after which these ions are released as discrete packets into the IMS T-Wave, where the ion mobility separation of ions is performed. The TRANSFER T-Wave is used to deliver the ion mobility-separated ions into the oa-tof analyzer. Each IMS separation was 51 ms long and the ions were released from the TRAP T-wave in 500 µs wide packets. The gas pressure in the TRAP and TRANSFER T-Wave regions was 0.07 mbar (Argon) and the pressure in the IMS T-Wave was 0.5 mbar (Nitrogen). The traveling wave used in the IMS T-Wave for ion mobility separation was operated at a velocity of 250 m/sec. The wave amplitude was ramped from 0 to 30 V over the period of the mobility separation for optimum performance over the large m/z range used (m/z 1,000 to 32,000). Figure 2. Mass spectrum of CsI (m/z 1,000 to 20,000). RESULTS Upon MS acquisition of a concentrated caesium iodide solution, intense ion clusters can be observed as high as m/z 20,000, with each cluster s composition based on the formula Cs(n+1)In. From the mass spectrum generated it is evident that a number of overlapping series, differing in charge state and intensity profile, are present over the m/z scale (Figure 2). However, the HDMS (IMS/MS) analysis clearly illustrates discrete distributions, which are related by their m/z and drift-time, as shown in Figure 3. Figure 3. HDMS analysis: Drift-time versus m/z plot of CsI (m/z 1,000 to 32,000). MS conditions MS system: SYNAPT HDMS System Ionization mode: nanoesi positive Capillary voltage: 1000 V Cone voltage: 150 V Source temp: 40 C Acquisition range: 1,000 to 32,000 m/z IMS T-wave ramp: 0 to 3 0V over IMS experiment IMS T-wave speed: 250 m/sec IMS pressure: 0.5 mbar (nitrogen) 6

Figure 4. Charge state distributions of CsI. The drift-times of the cluster ions seem to increase monotonically with increasing m/z values for the different charge state species, although with increasing charge state series up to +4, distinct m/z stability regions become clear, as shown in Figure 4. 7

Conclusion n HDMS provides greater definition of even relatively simple samples, such as inorganic CsI clusters. n High mass analysis of CsI clusters up to m/z 20,000 has been shown. n CsI clusters have been successfully separated into five charge states series, demonstrating the ability of HDMS (highefficiency IMS combined with oa-tof mass spectrometry) for differentiating high mw species based on their mobility. n CsI can be used to tune, optimize, and calibrate the SYNAPT HDMS System over a wide m/z range prior to high mass acquisitions being carried out. References: 1. Ruotolo, Giles, Campuzano, Sandercock, Bateman & Robinson, Science, 310 (2005) 1658-1661. 2. Pringle, Giles, Wildgoose, Williams, Slade, Thalassinos, Bateman, Bowers, Scrivens, Int. J. Mass Spectrom., 261 (2007) 1-12. Waters is a registered trademark of Waters Corporation. The Science of What s Possible, Triwave, SYNAPT, HDMS, T-Wave, and High Definition Mass Spectrometry are trademarks of Waters Corporation. All other trademarks are property of their respective owners. 2008 Waters Corporation. Produced in the USA May 2008 720002632EN AG-PDF Waters Corporation 34 Maple Street Milford, MA 01757 U.S.A. T: 1 508 478 2000 F: 1 508 872 1990 www.waters.com 8

Investigating Gas Composition on Transport and Desolvation of High m/z Species in the First Vacuum Stages of a Mass Spectrometer Iain Campuzano and Kevin Giles Waters Corporation, Manchester, UK INTRODUCTION Here, we present a method that provides efficient desolvation and effective transport of large biological ions from atmospheric pressure into the vacuum system of a mass spectrometer, through modification of the atmospheric pressure gas (cone gas) composition. No other source modifications were carried out in this work. Significant enhancements of the transmission and desolvation of large multiply-charged protein ions were achieved by changing the cone gas from nitrogen to a significantly heavier polyatomic gas. Nanoelectrospray is an ionization technique that efficiently generates large biological gas-phase ions. Transfer of non-covalently associated protein-protein complexes from solution to the gas phase generally results in the formation of ions possessing relatively few charges, and consequently m/z values are often above 10,000. This depends on the size of the protein complex under investigation. Biological samples analyzed under non-denaturing aqueous conditions are heavily solvated, resulting in non-gaussian mass spectral peak shape. An efficient desolvation process results in an accurately measured intact mass of the complex, due to the stripping of non-specific solution adducts. EXPERIMENTAL The instrument used in these studies was a SYNAPT HDMS System, which has a hybrid quadrupole/ion Mobility-based (IMS) orthogonal acceleration time-of-flight (oa-tof) geometry. Briefly, samples were introduced by a borosilcate glass nanoelectrospray spray tip and sampled into the vacuum system through a Z-Spray source. The ions passed through a quadrupole mass filter to the IMS section of the instrument. This section is comprised of three travelling wave (T-Wave ) ion guides. The trap T-Wave accumulates ions, while the previous mobility separation occurs. These ions were released in a packet into the IMS T-Wave in which the mobility separation was performed. The transfer T-Wave was used to deliver the mobility separated ions into the oa-tof analyzer. SAMPLES and GASES Alcohol dehydrogenase (147.5 kda), GroEL-SR (400 kda, single ring), Proteasome-S (700 kda), and GroEL (800 kda) were all buffer exchanged into an aqueous solution of 100 mm ammonium acetate, to a final working protein concentration of 1.5 µm. Sulphur hexafluoride (SF 6 ) was obtained from BOC Gases LTD. Octafluoropropane was obtained from F2-Chemicals, UK. SF 6 and Octafluoropropane were introduced as cone gas through the sheath cone assembly, as shown in Figure 1. The flow rate was controlled by means of a software controlled Bronkhorst gas flow controller (standard). The flow rate was varied from 0 L/hour to 100 L/hour to investigate optimal conditions for GroEL (800 kda) detection and transmission. Figure 1. Schematic of the Z-Spray Source Block and point of introduction of cone gas. 9

RESULTS 100 11452 To enable accurate comparisons of GroEL transmission when assessing SF 6 and Octafluoropropane, the deconvolution algorithm Maximum Entropy 1 was used to generate an intensity value for the entire GroEL multiply-charged envelope. This figure was then plotted against the cone gas flow rate, as shown in Figure 2. SF 6, when used as a cone gas, cools the large ions more efficiently than Octafluoropropane. The multiply-charged GroEL ions were not detectable when Nitrogen, Argon, or Xenon (data not shown) was used as a cone gas. % 0 100 % 3843 11290 11620 11793 11133 11972 10978 11782 11959 11611 12142 11444 12333 11281 12526 8910 Octoflouropropane 8.5 L/hour Sulphurhexafluoride 50 L/hour Large multiply-charged ions possess a large energy spread within the mass spectrometer. The kinetic energy spread reduced, so the ions could be focused and ultimately detected. Polyatomic gases, such as SF 6 and Octaflouropropane, cool or thermalize large ions, as shown in Figures 3 and 4, which resulted in enhanced detection. Polyatomic gases possess far more degrees-of-freedom than monoatomic gasses, such as Argon or Xenon. For example, SF 6 (Mw 146) and Octafluorpropane (Mr 188) possess 21 and 33 degrees-of-freedom (3N) respectively, as opposed to the 3 degrees-of-freedom of Xenon (Mr 131). 0 2500 5000 7500 10000 12500 15000 17500 20000 22500 25000 27500 30000 Figure 3. Multiply charged GroEL spectra obtained at optimal flow rates of Octafluorpropane (85 L/hour) and SF 6 (60 L/hour). 100 % 0 100 % 1806 0 5678 5467 5905 3081 2843 3822 5273 6152 8520 8903 8343 8923 9109 8174 9335 Yeast ADH Tetramer 147.5 kda GroEL-Single Ring 400 kda m/z 6 5 SF 6 100 % 11980 12042 11590 12203 12255 11412 12405 Rabbit Proteasome-S 720 kda 11217 4 Octafluoropropane 0 2500 5000 7500 10000 12500 15000 17500 20000 22500 25000 27500 30000 m/z 3 2 Figure 4. TOF-MS data acquisition of Yeast ADH, GroEL-SR, and Rabbit Proteasome-S, using SF 6 as a cone gas. 1 Argon 0 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 S1 Figure 2. Investigating the effect of cone gas SF 6, Octafluoropropane, and Argon indicated flow rates (L/hour) on the transmission of intact GroEL (800 kda). 10

CONCLUSIONS Polyatomic gases used as a cone gas dramatically improved transmission and desolvation of large multiply charged, high m/z ions without the need for any source modification. Improved transmission and efficient desolvation of large multiply-charged ions allowed for accurate mass measurement of large protein/protein noncovalent complexes. Waters is a registered trademark of Waters Corporation. The Science of What s Possible, ZSpray, SYNAPT, HDMS, and T-Wave are trademarks of Waters Corporation. All other trademarks are property of their respective owners. 2008 Waters Corporation Produced in the U.S.A. July 2008 720002744EN AG-PDF Waters Corporation 34 Maple Street Milford, MA 01757 U.S.A. T: 1 508 478 2000 F: 1 508 872 1990 www.waters.com 11

12

High Mass Precursor Ion Selection UtiliZing Monoatomic and Polyatomic Collision Gases Iain Campuzano and Kevin Giles Waters Corporation, Manchester, UK INTRODUCTION The current Waters SYNAPT HDMS Mass Spectrometer and nanoelectrospray sources lend themselves particularly well to preserving non-covalent interactions, allowing one to analyze compounds in their native conformation and stoichiometry 1. The transfer of non-covalently associated complexes from solution to the gas phase using electrospray ionization generally results in the formation of ions possessing relatively few charges. Therefore, these species appear high on the m/z scale, which makes time-offlight (TOF) mass spectrometry ideal for their mass analysis. Utilizing a 32k amu quadrupole allows for selection and fragmentation of these multiply charged species that appear high on the m/z scale. The benefit of this method is twofold: fragmentation of large macromolecular complexes and determination of individual subunit mass and stiochiometry. Fragmentation of large, highly-charged species enables product ions to possess a wide kinetic energy spread. This energy spread needs to be thermalized. This is achieved by the presence of a collision gas in the Trap T-Wave region of the SYNAPT HDMS System. Here we investigate and present the use of monoatomic (Xenon and Argon) and polyatomic (Sulphurhexafluoride and Octafluoropropane) gases as collision gases when performing MS/MS fragmentation of large multiply-charged precursor ions. In previous studies Argon has either been mixed with, or replaced with, larger inert monoatomic gases 2,3 to improve fragmentation efficiency or transmission of high m/z ions. Polyatomic gases are particularly efficient at thermalizing large ions because they have many degrees of freedom. For example, SF6 (Mw 146) and Octafluorpropane (Mr 188) possess 21 and 33 degrees of freedom (3N) respectively, as opposed to the 3 degrees of freedom of Argon (Mr 40) Xenon (Mr 131). EXPERIMENTAL The instrument used in this study was a SYNAPT HDMS System, which combines high-efficiency ion mobility based measurements and separations with a hybrid quadrupole orthogonal acceleration time-of-flight (oa-tof) mass spectrometer, as shown in Figure 1. Samples were introduced with a borosilcate glass nanoelectrospray-spray tip and sampled into the vacuum system. The ions pass through a quadrupole mass filter to the enabling Triwave device consisting of three travelling wave (T-Wave) ion guides 4. The gas pressure in the TRAP and TRANSFER T-Wave regions varied from 2.4 e -2 to 6.7 e -2 mbar, which was dependant on which collision gas was used. The IMS T-Wave was operated at 0.5 mbar (Nitrogen), but not in ion mobility mode. Figure 1. Schematic of the SYNAPT HDMS System. Extended quadrupole mass selection range up to m/z 32,000. ToF m/z range up to 100,000. 13

MS conditions MS System: SYNAPT HDMS System Ionization Mode: nanoesi Positive Capillary Voltage: 1,000 V Cone Voltage: 150 V Collision energy: 50 180 V Acquisition Range: 1,000 50,000 m/z T-wave Trap/Trans: Argon, Xenon, SF6, and gases Octafluoropropane T-wave Trap/Trans: 2.4 e -2 to 6.7 e -2 mbar pressures The multiply charged ion m/z 11,440 was then selected by the quadrupole and subjected to different collision energies ranging from 50 V to 180 V, in the presence of four different collision gases (Argon, Xenon, SF6, and Octafluoropropane). A typical MS/MS fragmentation profile is observed in Figure 3. At low collision energies (50-80 V), the precursor ion GroEL tetradecamer (14 mer) remained intact. At a collision energy of 90 V, the highly charged monomer appeared with the accompanying tridecameric (13 mer) species. At collision energies above 120 V, the appearance of the dodecamer (12 mer) was observed. RESULTS Upon infusion of GroEL in aqueous solution into the mass spectrometer, a narrow, well-defined multiply-charged envelope was observed at m/z 12,000, corresponding to intact GroEL mass 801 kda, as shown in Figure 2. Figure 3. GroEL MS/MS of precursor ion m/z 11,440 (+71) in SF6 over the collision energy range 50 180 V. Figure 2. GroEL ToF-MS spectrum over the m/z range 1,000 50,000. 14

It is observed from Figures 4 A-D that the larger collision gases Xenon, SF6, and Octafluoropropane provided improved collisional cooling over smaller gases, such as Argon. SF6 provided improved ion thermalizing properties over Xenon, observed by the increased ion intensities of the Dodecamer (12 mer). The polyatomic gas SF6, with its greater number of degrees of freedom (21), thermalized large ions more efficiently than Xenon or Argon. MaxEnt 1 Log Intensity MaxEnt 1 Log Intensity 7 6 5 4 3 2 1 0 7 6 5 4 3 2 1 0 50 70 90 110 130 150 170 50 70 90 110 130 150 170 Trap Collision Energy (V) A 12mer precursor 13mer monomer C 12mer precursor 13mer monomer Figure 4. Decay profiles of GroEL precursor ion m/z 11,400 in gases Ar (A), Xe (B), SF6 (C), and Octafluoropropane (D). Applied collision energy plotted against log summed ion intensities. 7 6 5 4 3 2 1 0 7 6 5 4 3 2 1 0 50 70 90 110 130 150 170 50 70 90 110 130 150 170 Trap Collision Energy (V) B 12mer precursor 13mer D monomer 12mer precursor 13mer monomer CONCLUSIONS: High m/z precursor ions were selected in the high mass quadrupole, fragmented in the in T-wave trap region, and subsequently mass-measured in the ToF analyzer, which provided subunit composition information on biological macromolecules whose subunit stoichiometry may not be known. Polyatomic gases possessed a greater number of degrees of freedom. Therefore, they provided improved collisional cooling of large m/z ions produced by an MS/MS experiment. This resulted in increased sensitivity on high m/z and highly-charged species. References 1. Ruotolo, Giles, Campuzano, Sandercock, Bateman, and Robinson. Evidence for Macromolecular Protein Rings in the Absence of Bulk Water. Science 2005 December 9;310,1651-1661. 2. Schey KL, Kenttamaa HI, Wysocki VH, and Cooks RG. Low-energy Collisional Activation of Polyatomic Ions with Different Target Gasses. Int. J. Mass Spectrom. Ion Proc. 1989 ; 90 71-83. 3. Lorenzen K, Versluis C, van Duijn E, van den Heuvel RHH, Heck AJR. Optimizing Macromolecular Tandem Mass Spectrometry of Large Non-covalent Complexes Using Heavy Collision Gases. Int. J. Mass Spectrom. 2007; 268 198-206. 4. Giles K, Pringle S, Worthington K, and Bateman R. Travelling Wave Ion Propulsion in Collision Cells. Presented at the 51st ASMS Conference, Montreal, Canada 2003. The travelling wave device described here is similar to that described by Kirchner in US Patent 5, 206, 506 (1993). Waters is a registered trademark of Waters Corporation. The Science of What s Possible, SYNAPT, HDMS, T-Wave, and Triwave are trademarks of Waters Corporation. All other trademarks are property of their respective owners. 2008 Waters Corporation. Produced in the U.S.A. July 2008 720002743EN AG-PDF Waters Corporation 34 Maple Street Milford, MA 01757 U.S.A. T: 1 508 478 2000 F: 1 508 872 1990 www.waters.com 15

Sales Offices: Austria 43 1 877 18 07 Australia 61 2 9933 1777 Belgium and Luxembourg 32 2 726 1000 Brazil 55 11 5094 3788 Canada 1 800 252 4752 China 86 10 8586 8899 Czech Republic 420 2 617 11384 Denmark 45 46 59 8080 Finland 358 9 5659 6288 France 33 1 30 48 72 00 Germany 49 6196 400 600 Hong Kong 852 2964 1800 Hungary 36 1 350 5086 India 91 80 2837 1900 Ireland 353 1 448 1500 Italy 39 02 265 0983 Japan 81 3 3471 7191 Korea 82 2 820 2700 Mexico 52 55 52 00 1860 The Netherlands 31 76 508 7200 Norway 47 6 384 6050 Poland 48 22 833 4400 Puerto Rico 1 787 747 8445 Russia/CIS 7 495 727 4490 / 290 9737 Singapore 65 6273 1221 Spain 34 93 600 9300 Sweden 46 8 555 115 00 Switzerland 41 56 676 70 00 Taiwan 886 2 2543 1898 UK 44 208 238 6100 US 1 800 252 4752 Waters Corporation 34 Maple Street Milford, MA 01757 U.S.A. T: 508 478 2000 F: 508 872 1990 www.waters.com Waters is a registered trademark of Waters Corporation. The Science of What s Possible, SYNAPT, High Definition Mass Spectrometry, HDMS, T-Wave, ZSpray, DriftScope, and Triwave are trademarks of Waters Corporation. All other trademarks are the property of their respective owners. 2008 Waters Corporation Printed in the U.S.A. August 2008 720002776EN PC-UM