“'E'SI mammary-x \M’YQ“'¢ cog”. ‘13! x 93-~e | . .Q n I, ... u :. A“ »— u.- Kiril'4Il-1og-st I Alxplfi‘y‘yOI‘Q the total number of basic residues (i.e. combined number of Arg, Lys and His), or “partially mobile” when the number of Arg residues 5 the total number of protons S the total number of basic residues [40, 41]. 20 H-I- :1 O 39 O H2N Ava /c\\/J\\ .H. II T T\ 0’ OH ll 0 R' H O R'" L L L. _. X3 Y3 23 x2 Y2 z2 X1 Y1 21 h Scheme 1.1 Nomenclature for peptide fragment ions. Adapted from Reference 38 Under “non-mobile” proton conditions, higher energies are required to mobilize the proton to initiate “charge-remote” fragmentation pathways (i.e. no proton involvement) [40, 42]. Whereas under “mobile” or” partially-mobile” proton conditions, fragmentation of most protonated peptides requires the involvement of a proton at the cleavage site, i.e. the cleavages are “charge~directed” [43, 44]. During a “charge-directed” process, after the proton is transferred to the cleavage site, dissociation of the amide bond is initiated by nucleophilic attack from an adjacent nucleophile, e.g. amide carbonyl, resulting in an ion-molecule complex, followed by further dissociation without or with intermolecular proton transfer giving b- and/or y- type ions. As mentioned above, BID cross-linked neurotensin was observed to yield extensive benzyl cations only in its 5+ charge state [26], indicating that a “mobile- proton” condition was required for CID MS/MS experiments. However, even when a mobile proton was present, other processes, including cleavages along the peptide 21 backbone, as well as cleavage of the amide bond in the cross-linker chain were also observed. The cross-linking reagent (SuDPG) developed by Soderblom et al. [27] is based on the incorporation of an aspartyl-prolyl bond within the linker region for specific gas- cleavage site. As shown in Scheme 1.2, cleavage is initiated by transferring a labile proton from the aspartyl side chain to the basic backbone amine of the adjacent prolyl residue, and then the carboxy anion attacks the carbonyl carbon, resulting in a cyclized anhydride and a free prolyl residue. However the Kapp et al. study indicated that fragmentation of Asp-Pro bonds is enhanced only for peptides under non-mobile conditions [40]. As a result, realizing preference cleavage in both BID and SuDPG cross- linked peptides is limited by their charge states. 6 0 O H O H O H N (the N Lys residue/ N NW \Lys residue Peptide 2 SuDPG ll 0 r 0 0 'fi 0 ii Lys residue / N N 25L N/\n/ N ‘ Lys residue I I H O H O D PG Peptide 1 O Peptide 2 Su Scheme 1.2 Fragmentation mechanism at aspartyl-prolyl bond within SuDPG cross- linked peptides. Adapter from Reference 27 22 1.5.2 -Controlling Peptide Fragmentation Reactions via the use of Fixed Charge Derivatization The mechanistic limitations associated with the fragmentation of protonated cross- linked peptide ions has led to consideration of alternate strategies for controlling or directing the fragmentation of cross-linked peptides toward the selective formation of analytically useful product ions. A number of different methods for controlling the fi'agmentation of non cross- linked peptide ions, involving the introduction of a fixed charge to the peptide, have recently been developed. Recently, the Reid group has described a novel chemical derivatization strategy for selective MS/MS based peptide identification, characterization and quantitative analysis, involving ‘fixed charge’ sulfonium ion derivatization of specific functional groups (e.g., the side chains of methionine (Scheme 1.3) or cysteine containing amino acids) within a peptide of interest [45-48]. From an earlier study on the gas-phase fragmentation behavior of phenacyl sulfonium ion derivatized methionine containing peptides, the exclusive fragmentation pathway observed corresponded to selective cleavage adjacent to the site of the fixed charge within the peptide ion resulting in the neutral loss of phenacyl methyl sulfide. Furthermore, this cleavage was found to occur independently of the precursor ion charge state or amino acid composition (i.e., proton mobility) [46]. Based on molecular orbital calculations on a simple peptide model, together with experimental studies on multistage dissociation of methionine derivatized sulfonium ion containing peptide, a further investigation in the mechanism responsible for this selective fixed charge side-chain fragmentation indicates that SN2 reactions occur, involving the N- and C- terminal amide bonds adjacent to the methionine side chain, 23 resulting in the formation of stable cyclic five- and six-membered ring product ions respectively (Scheme 1.4) [46]. These stable structured product ions were then subjected to further dissociation by MS3 to obtain additional structural information. Extensive studies reveal that though selective fragmentation at the fixed charge containing side chain is independent of amino acid composition and charge state of the peptide ion, loss of the side-chain fragment either as a neutral or charged species was observed highly dependent on the proton mobility of the precursor ion and the identity of sulfonium substitute [48]. S 3 $A9 C 2 O ¢H2 (I) BfCHzCOCeHs t I 0 9H2 .. o 9H2 .CH.C, 4..., (")MS ,JL .CH_C, «a, it 5 u 5 [M] [M+nH+CH2COCgH5]("”)* Scheme 1.3 Fixed charge derivatization of methionine sulfonium ion derivatized peptide ions. Adapted and modified from Reference 47 9 o \SSR Pathway 1 “‘C‘N Cl Pathway 2 Pathway 1 H HivJ o ) CID * “‘63; O -CH3SR 0/1 H H f) ——’ G e “I , N ”N \H‘" Pathway 2 ”J N C VP. Scheme 1.4 Gas-phase fi'agmentation behavior of the sulfonium ion derivative of methionine-containing peptides. Adapted and modified from Reference 47 24 The studies above provide a means to form characteristic product ions in high relative abundance, thereby enabling side chain fixed charge derivative containing peptides to be selectively identified from within complex mixtures, resulting in mixture simplification and improved dynamic range for qualitative and/or quantitative proteome analysis Based on the studies of gas-phase fragmentation reactions of methionine side chain fixed charge sulfonium ion containing peptides, the incorporation of ‘fixed charge’ sulfonium ion derivatives into chemical cross-linking reagents would enable this strategy to be extended toward the development of improved MS/MS based approaches for the identification of specific interaction sites in proteins and multi-protein complexes. In the first stage of MS/MS, ionic cross-linked peptides could be identified by either a product ion scan or a neutral loss scan from their characteristic fragmentation pattern on the cross-linker, then following MS’, cleavages along the peptide backbone will take place giving rise to a series of product ions containing information about the peptide sequence and the modification sites. Thus the cross-linking information could be used to derive a set of distance constraints for structure modelling. 1.6 Aims of this Thesis To develop improved chemical cross-linking and tandem mass spectrometry methodologies for the analysis of protein structures and protein-protein interactions, the aims of this thesis are: 1. Synthesis of a series of novel water soluble ‘fixed charge’ sulfonium ion containing cross-linking reagents. 25 2. Evaluation of the multistage gas-phase fragmentation reactions of cross-linked peptide ions formed by reaction with these reagents, using synthetic peptides and standard proteins and protein complexes. 26 CHAPTER TWO EXPERIMENTAL 2.1 Materials All chemicals were analytical reagent (AR), or of a comparable or higher grade and used without further purification. Thioglycolic acid, 3-Mercaptopropionic acid, N,N’- Dicyclohexylcarbodimid and Phenacyl bromide were purchased from Fluka (Switzerland). S-Bromovaleric acid and Thiourea were from Sigma—Aldrich (St. Louis, MO, USA). N-hydroxysuccinimide was purchased from Pierce (Rockford, IL, USA). Sodium hydroxide, Sodium phosphate dibasic (crystal), Potassium phosphate monobasic (crystal), Dimethyl sulfate and N,N’-Dimethyl formamide (DMF) were purchased from Spectrum Chemicals (Gardena, CA, USA). Sodium chloride, Sodium sulfate anhydrous, Hydrochloric acid and Ethyl acetate were purchased from Columbus Chemical Industries (Columbus, WI, USA). Potassium chloride, Sulfuric acid and Ethyl ether were purchased from Jade Scientific (Canton, MI, USA). Glacial acetic acid, Dichloromethane, Chloroform, Methanol (anhydrous) and Ethanol were purchased from Mallinckrodt Chemicals (Phillipsburg, NJ, USA). Iodomethane was purchased from EMD Chemicals (San Diego, CA, USA), and Acetonitrile was from Fisher Scientific (Hampton, NH, USA). The peptide VTMAHFWNFGK was obtained from Auspep (Parkville, Australia), Neurotensin was purchased from Sigma-Aldrich (St. Louis, MO, USA), and Insulin-like growth factor I (57-70) was purchased fiom American Peptide Company (Sunnyvale, CA). 27 Water was deionized and purified by a Barnstead nanopure diamond purification system (Dubuque, Iowa, USA). DMF was dried over 3 A molecular sieves (Spectrum Chemicals) and filtered. All reactions were performed in oven dried glassware. lH-NMR spectra were obtained on Varian Inova 300 MHz or 500 MHz instruments and are reported in parts per million (ppm) relative to the solvents resonances (8), with coupling constants (J) in Hertz (Hz). 2.2 Mass Spectrometry A Thermo model LCQ Deca quadrupole ion trap mass spectrometer (San Jose, CA), equipped with nanoelectrospray ionization was used for all analysis reported. Samples were introduced into the mass spectrometer at a flow rate of 0.5 uL / min. The capillary voltage was set to 2.5 KV and capillary temperature was in the range of 150 to 200 °C, other parameters were obtained as a result of auto tune function Optimized for each individual compound. Low energy CID MS/MS and MSn experiments were performed using helium as a collision gas, at an activation q value of 0.25 and an activation time of 30 ms. Collision energies were individually optimized for each compound of interest. 2.3 Synthesis of Ionic Cross-Linking Reagent S-Methyl 2-acetoyl,5’- pentanoyldihydroxysuccinimide sulfonium (Compound I in Scheme 2.1 ) 28 o 1)40 % NaOH. 0 0H /\/\/[L 4050 °C' 24 h HO /\/\)J\ its/\n/ + Br OH 4' rs OH 0 0 2) HCI (conc), 53 % 1 O + O O O CHCI3, R.T.. overnight. 31 °/o O O 0 JD 0 O O 2 O O l (CH3)2$O4 _ O ’0 +/\/\/U\ I? CH30N V 3:: If? 0 o O I Scheme 2.1 The synthesis route of ionic cross-linking reagent S-methyl 2-acetoyl-5’- pentanoyldihydroxysuccinimide sulfonium (1) 2.3.1 Carboxymethylmercaptopropionic Acid (1) Following the method by Rabinovich, et al. [49], a freshly prepared solution of 5- bromovaleric acid (4.0 g, 22.1 mmol) in 40 % NaOH (10 ml) was added dropwise to an ice bath cold solution of thioglycolic acid (2.1 g, 22.8 mmol) in 40 % NaOH (10 ml). The resulting reaction mixture was stirred at 40-50 °C for 24 hours. After the heating was terminated, the product mixture was acidified with concentrated hydrochloric acid to pH=1 and repeatedly extracted with dichloromethane (50 ml x 5). Extracts were combined, dried over anhydrous Na2SO4, filtered and evaporated under reduced pressure to give a compound (1) as a white solid in 53 % (2.3 g) yield. 1H NMR (500 MHz, CDC13): 5 1.64-1.77 (m, 4H), 2.38 (t, 2H, J = 7 Hz), 2.67 (t, 2H, J = 7 Hz), 3.23 (s, 2H), 10.18 (s, broad, 2H); 13C NMR (125 MHz, CDC13): 6 23.48, 28.04, 32.14, 33.35, 33.38, 176.61 and 179.73. 29 The product was dissolved in methanol:H2O 50:50 v/v to a concentration of 20 [1M and characterized by negative ion mode nanoelectrospray mass spectrometry and MS/MS as described above. The deprotonated pseudo-molecular ion [M-H] ' at m/z 191.2 (Figure 2.1A) was selected for fragmentation by CID-MS/MS, the fragments obtained were consistent with the structure, and the fragmentation pathways are demonstrated in Scheme 2.2. A l00 ‘ A e: U U S: C3 '0 S D 50 . < d) .2 ‘5 E h [M-H] 191.2 [2M-H]‘ 3812.8 MS IOO 200 300 400 500 600 700 800 900 1000 m/z 100‘ 50‘ MS/MS 91.0 B [M-Hl' 191.1 129.1 99-1 L 147.0 Li 3 l 00 80 100 120 140 I60 180 200 m/z Figure 2.1 Characterization of (1) by nanoelectrospray quadrupole ion trap mass spectrometry and CID MS/MS. (A) MS and (B) CID-MS/MS of m/z 191.1 30 0 {render 0 HO 8 + if” i i O O m/z = 99 6”,» /\/\/u\ _. /\/\/U\ ——->'H O //O 2 C O mlz = 147 111/2 = 129 4. C02 Scheme 2.2 Fragmentation scheme of (l) anions. The spectrum is displayed on Figure 2. l B. Square brackets indicate ion-molecule complexes 2.3.2 Thio 2-acetoyI-5’-pentanoyldihydroxysuccinimide (2) Following methods analogous to other N-hydroxysuccinimide esters [SO-52], compound 1 (0.8 g, 4.2 mmol) and N-hydroxysuccinimide (NHS) (1.2 g, 10.4 mmol) were dissolved in 3:1 v/v mixture of chloroform and dichloromethane (DCM) (40 ml) and stirred for 5 min at room temperature. N,N’-dicyclohexylcarbodiimide (DCC) (2.2 g, 10.6 mmol) was then added and the mixture stirred overnight. After filtration of dicyclohexylurea (DCU) precipitate and solvent removal under reduced pressure, the oily residue was dissolved in a minimum amount of ethyl acetate. The remaining DCU was precipitated and removed by filtration. Following rotary evaporation of the ethyl acetate, the residue was dissolved in DCM, washed with 1M NaOH and H20 then evaporated to near dryness. The residue was then recrystallized with ethyl ether containing a trace amount of acetone to give 0.5 g (31 %) of compound (2). 1H NMR (500 MHz, CDCl3): 8 1.71-1.77 (m, 2H), 1.81-1.87 (m, 2H), 2.62 (t, 2H, J= 7 Hz), 2.74 (t, 2H, J= 7 Hz), 2.80 31 (s, broad, 8H), 3.45 (s, 2H); 13C NMR (125 MHz, CDC13): 5 23.40, 25.53, 25.54, 27.41, 29.98, 30.24, 31.53, 165.82, 168.27, 168.86 and 169.17. For mass spectrometry, compound 2 was dissolved in acetonitrile to a concentration of 10 11M. In the positive ion mode, the abundance of the pseudo molecular [M+H]+ ion was much lower than the Na+ and K+ adducts due to poor protonation of the diester. The dimer adducts are also observed. The peak of m/z at 272.2 in Figure 2.2A corresponds to a fragment generated in the ionization source from precursor ions of ammonium adduct [M+NH4]+, which is also a product ion present in the MS/MS and MS3 spectra in Figure 2.2C and 2.2D. The dominant fragments of [M+Na]+ and [M+H]+ ions were from a neutral loss of 1 15; the fragmentation pathway is displayed in Scheme 2.3 and 2.4. 32 ,3 409.1 “g [2M+Na]+ é [M+NH 41+ 794.8 3 50 ‘ 40:3 [M+K]+ \[2M+K]+ 50" < [M+H] 425.3 810.8 + t); 387.1 / [M+Na] 3 272.2\\ 409.0 g L ITIAHMLIAJ [A LMAAIA. .IIII AAAIL L 13180 100 200 300 400 500 600 700 800 900 1000 150 200 250 00 350 400 m/z z MS/MS M83 100 - C 3863 100' 1) 271.9 3 271.9 2., <1) 0 1: M '2 3 50 50~ < D .2 + *5 [M+NH4] 2 403.9 386-8 L l 130 200 250 300 350 400 150 200 250 300 350 400 m/z m/z Figure 2.2 Characterization of (2) by nanoelectrospray quadrupole ion trap mass spectrometry and MS/MS. (A) Mass spectrum of (2), due to the poor protonation of diester, the [M+Na]+ adduct is the most intensive peak; (B) CID MS/MS of [M+Na]+ adduct, neutral loss of 115 was observed; (C) CID MS/MS of [M+NH4]+ adduct, yielding [M+H]+ and m/z at 272+ fragment ions; (D) CID MS3 of [M+H]+ coming from [M+NH4]+, giving rise to the fragment at m/z of 272+, by neutral loss of 115 33 //C¢\S O O Na transfero \EENNaO’ OH + CA S/\/\/cl)l\ :3? mlzo=138 Scheme 2.3 Fragmentation of Na+ adduct of (2). The spectrum is displayed on Figure 2.2B. Square brackets indicate ion-molecule complexes (N <9 o o H HO H r /\/\)J\ N’Qfin/ks o’N o o O C+D o O HO _ N-OH N CH gCAS/W/ILO’N /S/V\jI\ON O O O mlz= 272 Scheme 2.4 Fragmentation of protonated (2). The spectrum is displayed on Figure 2.2D. Square brackets indicate ion-molecule complexes. The protonated (2) is supposed to have the similar fragmentation pathway as Na+ adduct of (2) for the same neutral loss of 115 observed but no proton transfer observed to give product ion at m/z of 116 due to NHS has lower affinity to proton than that of Na+ 2.3.2 S-Methyl 2-acetoyI-5’-pentanoyldihydroxysuccinimide Sulfonium (I) 34 A mixture of compound 2 (193 mg, 0.5 mmol) and dimethyl sulfate (2.52 g, 20 mmol) in 2 ml acetonitrile was allowed to react at room temperature for 4 days. Following freeze drying, a dark brown oily residue was obtained. Figure 2.3 shows the MS and MS/MS spectra obtained by analysis of the sulfonium (I) methylsulfate. The fragmentation pathways are displayed in Scheme 2.5, where two competitive routes are observed giving rise to product ions at m/z of 198+ and 286Jr respectively. ,3 100- A w 100“B 9) 401.2 d) O 5 C‘- +- 3 50 50- M < 401.0 0 ..>. 5 286.0 d) 04 inn..illl.1.-2.. r A J 100 200 300 400 500 600 700 800 900 1000 150 200 250 300 350 400 m/z m/z Figure 2.3 Characterization of ionic cross-linking reagent I by nanoelectrospray quadrupole ion trap mass spectrometry. (A) The methyl sulfonium ion is at m/z 401 +; (B) CID-MS/MS product ion spectrum of M+ o o O ,0 / Pathway2 / e o o o O o p H b0 O (C /\/\/U\ N m/z = 198 N’ v %J (9’ O O O I Pathway 1 0 \ O O mlz = 401 Pathway 2 A @/\/\/U\ ,N + N—OH ,c s o O l o o mlz = 286 Scheme 2.5 Fragmentation of sulfonium ion (I). The spectrum is displayed on Figure 2.3B. Fragment of m/z at 198+ is the main process; otherwise the pathway for the neutral loss of 115 is also present 35 2.4 Synthesis of Ionic Cross-Linking Reagent S-Methyl 3-propionoyl,5- pentanoyldihydroxysuccinimide sulfonium (Compound 11 in Scheme 2.6 ) 1) 4o % NaOH 0 o o H H AAA “50°92” _ W SW0 + 3' OH ' HOMS OH 0 2) HCI (cone), 70 % 3 Ei”°“*Q...Ol CHCI3. R.T.. overnight. 97 °/o $9.130.qu O O CH I (CH ) SO 0 0 U2? 3 or 32 4 N /\/\/lk 4 ‘OMS O’N o I O CH3CN Scheme 2.6 Synthesis route of ionic cross-linking reagent S—Methyl 3-propionoyl-5’- pentanoyldihydroxysuccinimide sulfonium (11) 2.4.1 l-Carboxyethylmercaptopropionic Acid (3) This acid was synthesized following the method described for the 1- carboxymethylmercaptopropionic acid (1), from a solution 5-bromovaleric acid (2.0 g, 11.0 mmol) in 40 % NaOH (5 ml) and solution of 3-mercaptopropionic acid (1.2 g, 11.3 mmol) in 40 % NaOH (5 ml) in 70 % (1.6 g) yield. lH NMR (500 MHz, CDC13): 5 1.60- 1.67 (m, 2H), 1.69-1.75 (m, 2H), 2.37 (t, 2H, J = 7 Hz), 2.54 (t, 2H, J = 7 Hz), 2.64 (t, 2H, J= 7 Hz), 2.76 (t, 2H, J: 7 Hz), 11.45 (s, broad, 2H); 13C NMR (125 MHz, CDC13): 5 23.62, 26.45, 28.61, 31.62, 33.46, 34.68, 178.25 and 179.84. The characterization by mass spectrometry and MS/MS is shown in Figure 2.4; the fragments obtained were consistent with the referring structure (Scheme 2.7). 36 MS MS/MS - _ 133.0 A 100 A [M-H]' 100 B 33 205.1 0 E g 50 .4 50 " <1 [M-Hl' «2 133.1 205.0 .5 / [2M-H]' T“; 410.9 M r. I .1 l l L 71L'0 A 150 200 250 300 350 400 450 500 550 600 100 150 200 250 300 m/z m/z Figure 2.4 Characterization of (3) by nanoelectrospray quadrupole ion trap mass spectrometry. (A) The pseudo molecular ion [M-H] ' at m/z 205.1 ' and the dimer at m/z 410.9 '; (B) CID-MS/MS of selected parent ion at m/z 205.1 ' H‘t 8f \9'8 0 U + HSNOH mlz=71 Scheme 2.7 Fragmentation scheme of (3) anions. The spectrum is displayed on Figure 2.4B. Square brackets indicate ion-molecule complexes 2.4.2 Thio 3-propionoyl-5’-pentanoyldihydroxysuccinimide (4) This diNHS-ester was synthesized and purified in the same manner as thio 2- acetoyl-S’-pentanoyldihydroxysuccinimide (2). (3) (1 g, 4.8 mmol), N- hydroxysuccinimide (1.4 g, 12.2 mmol) and DCC (2.5 g, 12.1 mmol) were dissolved in chloroform (30 ml) and stirred overnight. After purification a light yellow solid (4) was 37 obtained in 97 % (1.9 g) yield. ‘H NMR (500 MHz, CDC13): 5 1.67-1.73 (m, 2H), 1.80- 1.86 (m, 2H), 2.55-2.62 (m, 4H), 2.80-2.90 (m, 12H); 13C NMR (125 MHz, CDC13): 8 23.49, 25.50, 26.09, 28.17, 30.37, 31.35, 31.96, 167.13, 168.25, 168.94 and 169.12. Following analysis by ESI-MS, the singly protonated ion was observed at m/z 400.9, however due to poor protonation, the [M+NH4]+ adduct at m/z 418.2 was observed as the most abundant ion. The ion at m/z 286.0 in Figure 2.5A was found to correspond to a source CID fragment of the [M+NH4] + adduct, by analysis of its CID-MS/MS product ion spectrum in Figure 2.5B. The MS3 spectrum (data not shown) of the [M+H] + ion (m/z 400.8) formed by fragmentation of the [M+NH4] + precursor gave rise to the product at m/z 286 as well the neutral loss of 115 Da, similar to that seen for (2) in Scheme 2.4. MS 286.1 MS/MS + 100- @1007 A [M+NH4] B 4008 a, 418.2 Q) U I: CC "3 + g [2M+NH4] , .3 501 [M+Na]+ 3177 +50“ [M+NH4I q, [M+H]” 423.2[M Kl" [2:43:21 417.9 3- 400.9 ’/ + . 33 439.1 / o. 286.0\ a: l. 11-. - 100 200 300 400 500 600 700 800 900 1000 150 200 250 300 350 400 m/z m/z Figure 2.5 Characterization of (4) by nanoelectrospray quadrupole ion trap mass spectrometry. (A) Mass spectrum of (2), [M+NH4]+ adduct is the most abundant peak due to poor protonation of the diester; (B) CID MS/MS of [M+NH4]+ adduct, neutral loss of NH3 and fiirther loss of 115 was observed 2.4.3 S-Methyl 3-propionoyl-5’-pentanoyldihydroxysuccinimide Sulfonium (II) 38 The methylation of (4) can be performed by reacting with dimethyl sulfate as methods mentioned above, from a starting substrate of 200 mg (0.5 mmol). Alternatively, methylation can be achieved by stirring of (4) with iodomethane (2.1 g, 15 mmol) in 2 ml acetonitrile at room temperature in the dark for 24h. A light yellow solid was obtained alter freeze drying. The two procedures gave the methyl sulfonium ion with different counter ions. Both ions exhibited identical fragmentation behavior following low energy CID MS/MS. Figure 2.6 demonstrates positive mode mass spectrometry of molecular ion at m/z 415 and MS/MS spectra obtained by analysis of the sulfonium (II) iodide. A little different from the fragmentation process of (I) as shown in Scheme 2.5, pathway 1 is still the predominant mechanism for (11) resulting in 6-membered ring oxonium at m/z of 198 and complementary ion at m/z of 218, however the neutral loss of 115 was too weak to be observed. / MS , 1979 MS MS ,3 1001 A M. 100 Q 415.3 0 Q C.‘ CU 'U ‘5} . o MI 9 217.8 415.3 (6 T, a 1L L11. .1111 A I ‘-L.Iu‘_ __‘..._..L.‘-.-.J J ‘ 100 200 300 400 500 600 700 800 150 200 250 300 350 400 500 m/z Vm/z Figure 2.6 Characterization of ionic cross-linking reagent II by nanoelectrospray quadrupole ion trap mass spectrometry. (A) Mass spectrum of methyl sulfonium ion at m/z 415 +; (B) CID-MS/MS product ion spectrum of M+ ion Phenacyl bromide is another commonly used alkylating reagent yielding a phenacyl substituted sulfonium ion. A mixture of (4) (200 mg, 0.5 mmol) and phenacyl bromide 39 (1.99 g, 10 mmol) in acetonitrile (5 ml) was allowed to react at room temperature in the dark for 5 days. After freeze drying, the residue was characterized by mass spectrometry and MS/MS without separation of excess reagent. The obtained sulfonium ion has different m/z value from that of methylation products, but still gave the same main fragment at m/z 198 (complementary ion at m/z of 322) as shown in Figure 2.7B, product ions at m/z of 161 and 289 can be derived only from a rearrangement due to the present of the phenacyl group. 197.9 A 100- A M. MS 1001 B MS/MS to, 519.1 8 5 '13 g 50 «t D 50 <5 161.1 .2 206.8 M, a 9- T) 289.1 321.3 51 l ‘3‘ A . r . “I 12 #2 -L l \‘- 1 100 200 300 400 500 600 700 800 900 1000 150 200 250 300 350 400 450 500 m/z m/z Figure 2.7 Characterization of phenacylsulfonium ion derivative of (II) by nanoelectrospray quadrupole ion trap mass spectrometry. (A) Mass spectrum of phenacylsulfonium ion as an alternative for compound II, showing M+ at m/z 519 +; (B) CID-MS/MS yields the same product ion at m/z 198 2.5 Synthesis of Ionic Cross-Linking Reagent S-Methyl 5,5’- dipentanoylhydroxysuccinimide sulfonium (Compound III in Scheme 2.8 ) 40 1 EtOH, reflux, 20 h Br/\/\/U\OH + HZNJLNHZ ) : HS\/\/\n/OH 2) 7.5 M NaOH, 90 °C, 16 h 3) 2 M H2804 5 0 O O O /\/\/U\ /U\/\/\ /\/\/U\ ‘ Br OH HO 8 OH ‘ 6 1) 40 % NaOH, 40-50 °C. 24 h 2) HCI (cone). 78 % O (1 O o O EéN—OH N—C-N N JCJ)\/\/\ WOL Q = ‘0 S O’ 0 0 O CHCI3, R.T., overnight, 84 % 7 O O N M /\/\)CJ)\ 1:? CH3I or (CH3)ZSO4 ‘o s o’ = Q I O CH3CN Scheme 2.8 Synthesis of ionic cross-linking reagent S-Methyl 5,5’- dipentanoylhydroxysuccinimide sulfonium (III) 2.5.1 S-Mercaptopentanoic Acid (5) The procedure was adapted from that of Jessing [53] for 5-bromovaleric acid. 5- bromovaleric acid (2.2 g, 12.1 mmol) and thiourea (1.4 g, 18.4 mmol) were dissolved in ethanol (25 ml) and refluxed for 20 h. The solvent was removed under reduced pressure and 7.5 M NaOH (aq) (25 ml, 187 mmol) was added. The mixture was stirred for an additional 16 h at 90 °C. Then with cooling on an ice bath, 2M H2804 (aq) was added slowly under stirring until pH = 1. The product was extracted with CH2Cl2 (2x100 mL), combined extracts were dried with MgS04 and concentrated by rotary evaporation to give the title acid (5) as a colorless oil in quantitative yield. The product was then used without further purification. ‘H NMR (300 MHz, CDC13): 8 1.32(t, 1H, J: 7.8 Hz), 1.59- 41 1.73 (m, 4H), 2.32 (t, 2H, J = 7.5 Hz), 2.49 (q, 2H, J = 6.9 Hz), 8.95 (s, broad, 1H); l3C NMR (75 MHz, CDC13): 5 23.21, 24.06, 33.07, 33.35 and 179.49. The characterization by mass spectrometry and MS/MS analysis is shown in Figure 2.8, the fragmentation behavior of deprotonated [M-H]' is consistent with the referring structure giving the product ion at m/z of 99 from a neutral loss of H28 as displayed in Scheme 2.9. MS 990 MS/MS A 100- A [MH]. 100- B ' é 133.1 0 U 5 [M-H]' 'Cg’ [2M-H]' 132.9 .9 50‘ 266.8 50* <: O .2. E D m I In; LL ll 2.; 100 150 200 250 300 350 400 450 500 60 70 80 90 100 110 120 130 140 150 m/z m/z Figure 2.8 Characterization of (5) by nanoelectrospray quadrupole ion trap mass spectrometry. (A) The pseudomolecular ion [M-H] ’ at m/z 133.1 ’ and the dimer at m/z 266.8 '; (B) CID-MS/MS of selected parent ion at m/z 132.9 ‘ b0 f‘ H) O 8 Wk)- —> H: ‘J ——> \ + H S HS) 0 0 O 2 mlz = 133 mlz = 99 Scheme 2.9 Fragmentation mechanism of (5) anion. The spectrum is displayed on Figure 2.8B. Square brackets indicate ion-molecule complexes 2.5.2 5,5'-Thiodipentanoic Acid (6) 42 The acid (6) was obtained by the method described for the 1- carboxyrnethylmercaptopropionic acid (1), from 5-bromovaleric acid (3.4 g, 18.8 mmol) dissolved in 40 % NaOH (8 ml) and freshly prepared (5) (2.5 g, 18.7 mmol) dissolved in 40 % NaOH (8 ml) in 78 % (3.4 g) yield. 1H NMR (300 MHz, CDCl;;): 5 1.49-1.67 (m, 8H), 2.26 (t, 4H, J = 7.2 Hz), 2.44 (t, 4H, J = 7.2 Hz); 13c NMR (75 MHz, coch): 6 23.84, 28.80, 31.42, 33.39 and 176.79. The characterization by mass spectrometry and MS/MS is shown in Figure 2.9. The deprotonated pseudo-molecular ion [M-H] at m/z 233 (Figure 2.9A) was selected for fragmentation by CID-MS/MS, the fragments obtained were consistent with the referring structure, and the fragmentation pathways are demonstrated in Scheme 2.10. MS MS/MS 171-1 214.9 .3. 100-A [M-H]' 100 ‘B Q, 233.2 0) U c: m "C 5 .2 50‘ 50‘ 133.1 [M-H]' 1; 233.1 .5 [2M-H]' T“: 466.9 99-0 a: IL L- 1-.. I 1. I I1 I. 100 200 300 400 500 600 700 80 100 120 140 160 I80 200 220 240 m/z m/z Figure 2.9 Characterization of compound (6) by nanoelectrospray quadrupole ion trap mass spectrometry. (A) The pseudo molecular ion [M-H] ' at m/z 233.2 ' and the dimer at m/z 466.9 '; (B) CID-MS/MS of selected parent ion at m/z 233.1 ' 43 0 O O (SD/MOH + O 0 (it: O/v\/IL {I f mlz=133 S OH \rjms are 0 O Pathway 2 o \ N + HS OH Pathway 2 O O m/z = 99 9 gm N“ m /z = 233 Pathway 1 Pathway 11 -H20 0 0 co 9 “W l’ ' l’ 66 k4 SWC ____2.. (DA/\ch mlz = 215 mlz = 171 Scheme 2.10 Fragmentation mechanism of (6) anion. The spectrum is displayed on Figure 2.98. Square brackets indicate ion-molecule complexes 2.5.3 5,5'-Thiodipentanoylhydroxysuccinimide (7) This diNHS-ester was synthesized and purified in the same manner as thio 2- acetoyl-5’-pentanoyldihydroxysuccinimide 2. Thiodiacid (6) (1.17 g, 5.0 mmol), N- hydroxysuccinimide (1.44 g, 12.5 mmol) and DCC (2.58 g, 12.5 mmol) were dissolved in a 10:1 v/v mixture of chloroform and dichloromethane (55 ml) and allowed to react at room temperature overnight. The final crystallized product (7) was obtained as a light yellow solid in 84 % (1.8 g) yield. 1H NMR (300 MHz, CDC13): 5 1.63-1.71 (m, 4H), 1.76-1.84 (m, 4H), 2.50 (t, 4H, J = 7.5 Hz), 2.60 (t, 4H, J = 7.5 Hz), 2.78 (s, 8H); 13C NMR (75 MHz, CDC13): 5 23.60, 25.50, 28.35, 30.42, 31.03, 168.32 and 169.14. Upon characterization by mass spectrometry and MS/MS, the singly charged [M+H]+ ion at m/z 429.1 was the most abundant ion, the NHa+ and Na+ adducts were also 44 present. The peak of m/z at 314.2 corresponds to an in source fragment of the [M+H]+ ion, via the neutral loss of 115, a similar fragmentation pathway as that of (2) in Scheme 2.3 but no proton transfer observed. [M+Hl’ MS MS/MS 314.0 100 - A 429" 100 - B 314.2 [M+ NHJI Relative Abundance (%) ‘6 446.2 50 - [M+H]+ [M+Nal‘ 428.9 451.3 In 1 L1. 1L ..1 I .I 1 ..ng 100 200 300 400 500 600 700 800 900 1000 150 200 250 $10 350 400 450 m/z m/z Figure 2.10 Characterization of (7) by nanoelectrospray quadrupole ion trap mass spectrometry. (A) The pseudo molecular ion [M+H] + at m/z 429.1 +, the NH4+ and Na+ adducts are present; (B) The CID-MS/MS of selected parent ion at m/z 428.9 +, m/z at 314.0 is the main fragment, which corresponds to 314.2 fragment in source in panel A 2.5.4 S-methyl 5,5'-Thiodipentanoylhydroxysuccinimide (III) The methylation of (7) was performed by reacting with either iodomethane or dimethyl sulfate as methods mentioned above, from a starting substrate of 0.5 mmol (214 mg) scale. Both methods gave the same sulfonium ion with different counter ions; identical fragmentation behavior was also observed following low energy CID MS/MS. Figure 2.11 demonstrates positive mode mass spectrometry of molecular ion at m/z 433 and MS/MS spectra obtained by analysis of the sulfonium (III) methylsulfate. The sulfonium ion (III) was observed giving the same main product ion, 6-membered ring oxonium at m/z of 198, as shown in Figure 2.1 18. Fragment at m/z 131 could be derived from a further fragmentation of complementary ion of 198. 45 MS 1979 MS/MS A 100 - A M. 1001 B § 443.3 d) 0 C.‘ (U '0 C. .8 50 502 < .‘2’ 130.9 . E M g / 443.1 .. m . 1 1 I 200 400 600 800 1000 150 200 250 300 350 400 450 m/z m/z Figure 2.11 Characterization of ionic cross-linking reagent III by nanoelectrospray quadrupole ion trap mass spectrometry. (A) The methyl sulfonium ion at m/z 443 +; (B) CID-MS/MS product ion spectrum of M4r The sulfonium III iodide was purified by precipitation from dichloromethane/ hexanes. 1H NMR (500 MHz, CDCI3): 5 1.89-1.91 (m, 8H), 2.68 (t, 4H, J = 6.5 Hz), 2.78 (s, 8H), 3.02 (s, 3H), 3.53 (t, 4H, J= 7.5 Hz); l3c NMR (125 MHz, CDC13): 6 22.51, 22.66, 23.00, 25.53, 29.99, 40.34, 168.21 and 169.73. 2.6 Peptide Cross-Linking Reactions Peptides were dissolved in 40 11L of either DMF or phosphate buffered saline (PBS, 800 mg NaCl, 217 mg Na2HP04°7H20, 20 mg KCl and 20 mg KH2PO4 per 100 ml, pH 7.5) to a concentration of 0.6 mM. The cross-linker solutions were prepared in DMF immediately before use. 4 uL of cross-linker solution was added to the 40 11L peptide solution and the reaction allowed to proceed at room temperature for 30 min-24 h. Cross- linking reactions were optimized by varying the cross-linker, cross-linker concentration and reaction time. The product solutions were directly diluted after reaction with a solution of acetonitrilezwaterzacetic acid 60:40:1 v/v to a concentration of 10 uM for 46 nanoelectrospray when DMF was used as the reaction buffer. When PBS buffer was used, the products were desalted by sep-pak purification with elution performed by applying 2 ml of 40 %, 60 % and 80 % acetonitrile aqueous solution containing 0.05 % formic acid successively, then the product was subjected to mass spectrometry analysis without further dilution [20, 26]. 47 CHAPTER THREE ALKYL CHAIN LENGTH EFFECTS ON THE STABILITY AND FRAGMENTATION BEHAVIOR OF IONIC CROSS-LINKING REAGENTS 3.1 Introduction An ionic cross-linking reagent that could have preference for cleavage under low energy CID MS/MS conditions has to satisfy a number of prerequisites: (1) the reagent must be stable when subjected to cross-linking reaction conditions, as minimum requirements, it must exhibit certain stability in aqueous solutions at mild pH values; (2) it must contain an appropriate linker length for detecting specific interaction sites within protein complexes, usually a short (4-8 A) spacer arm is used in intra-molecular cross- linking studies, while intermolecular cross-linking is favored with a cross-linker containing a long spacer arm [7]; (3) the energy required for cleavage within the linker region should be lower than that for peptide backbone cleavage. 0n the basis of previous experience, introducing a fixed charge could realize selective cleavage at a site adjacent to the fixed charge within a peptide ion [45-48]. Back et al. proposed to quatemarize the amine of BID in order to have a reaction channel independent of the charge states of cross-linked peptide for yielding marker ions [26]. From a previous study in our laboratory of the low energy CID-MS/MS spectra obtained from sulfonium and ammonium fixed-charge containing peptides [45], selective loss of dialkylsulfide from sulfonium derivatives is an energetically favored process; however the loss of trimethylarnine from ammonium ion derivatives appears to be highly dependent on proton mobility. In a further investigation into the electronic effects of 48 sulfonium substitute [54], the positive charge on sulfur produces a strong dipole with resultant weakening C-S bond, and the ability of the sulfur atom to accept the electron pair of C-S bond enabled the sulfonium to be a better leaving group than an ammonium. These results suggest the incorporation of fixed charge sulfonium ions into chemical cross-linking reagents for approaches aimed at selective gas phase fragmentation. The reactive groups of cross-linking reagents generally target primary amines or sulfliydryl groups in proteins. NHS esters are highly reactive amine-specific cross-linking reagents [55-57], targeted to the free N-termini and s-amino groups in lysine side chains of proteins to create stable amide bonds. Due to the large number of lysine residues present on the surface of most proteins [2], it is likely that two lysine side chains will be close to each other at a protein-protein interface. Because of their low abundance, cysteines are less common on the surface of proteins, and because they are easily subject to oxidation reactions forming disulfide bonds, the sulfhydryl group is not as an attractive target functional group for the novel ionic cross-linking reagents, though it could also be employed. The application of NHS esters is usually limited by their poor water solubility for large scale protein complex studies, resulting in the use of the more expensive sulfo- NHS esters. However, simple members of ionic sulfonium salts typically exhibit high solubilities in water [54], thus sulfonium containing cross-linkers are expected to have an advantage of substantial solubility in both organic and aqueous solutions. Cross-linking reactions often employ the use of a series of cross-linking reagents with different linker region lengths to map the comprehensive topology of protein and protein complex. Our initial design of cross-linking reagents has a general structure with NHS pentanoate at one side and an NHS acetate or propanoate at the other side of the 49 sulfonium sulfur. The third sulfur substituent could be either a methyl or phenacyl group, dependent on which alkylating agent was to be employed. NHS pentanoate forms a pentanamide bond following cross-linking reaction with primary amine in protein. When subject to CID MS/MS, nucleophilic attack from the carbonyl group on the linker would result in the formation of a protonated six-membered iminotetrahydropyran product ion upon cleavage on the C-S bond. This process is expected to have a low-energy transition state compared to the energy required for amide bond cleavage based on the previous theoretical calculations and experimental results described by Amunugama et al. [47]. In this chapter, the reaction characteristics as well as the application of these first two versions of the ionic cross-linking reagents (I and II in Scheme 2.1, 2.6) are discussed in detail. Experiments based on the multistage MS/MS of cross-linked neurotensin were carried out to provide evidence to support the formation of a protonated six-membered iminotetrahydropyran product ion as the dominant mechanism. 3.2 Cross-Linking Reagent S-Methyl 2-acetoyl,5’- pentanoyldihydroxysuccinimide sulfonium (Compound I in Scheme 2.1 ) 3.2.1 Stability of I following Reaction with Iodomethane Iodomethane is a commonly used methylation agent, and an example of moderately soft electrophilic center [58], thus the methylation with iodomethane tends to occur at the soft nucleophile center such as a dialkylsulfide. Together with its volatility, iodomethane is expected to be a suitable alkylating agent to prepare sulfonium from di-NHS ester 50 intermediates. However, a reaction from 2 and iodomethane was not able to give sulfonium I but a dimethyl substitute sulfonium as shown in Scheme 3.1. Figure 3.1 demonstrates the positive mode mass spectrometry of the molecular ion at m/z 260 and the MS/MS spectra obtained by analysis of the sulfonium (8) iodide. The proposed 6-membered ring oxonium product ion observed at m/z of 198 from the neutral loss of dimethyl sulfide from (8) was consistent with the referring structure. O O o 0 o ,O N ,N CHal \*/\/\/u\ ,N é: \ll/\S 0 CchN 1? O O O 2 8 mlz=260 Scheme 3.1 Dimethyl sulfonium 8 obtained from methylation of 2 with iodomethane S D 0 100.0 5 E ..o 50 ~ 5()- 99.0 < E 272.1 E 4040 g L. 83.0 260.0 All I1- .1 L. L 1 11 1 200 400 600 800 1000 1200 80 100 120 140 160 180 200 220 240 260 280 m/z m/z Figure 3.1 Characterization of dimethyl sulfonium 8 by nanoelectrospray quadrupole ion trap mass spectrometry. (A) The dimethyl sulfonium ion at m/z 260 +, peak at m/z 404 corresponds to ammonium adduct of reactant 2 and m/z 272 is a fragment generated in source from it; (B) CID-MS/MS product ion spectrum of M+ Iodide is generally a rather poor nucleophile at carbon centers to initiate a nucleophilic addition reaction at the carbonyl group, however when the carbonyl carbon is only one methylene group apart from a sulfonium positive charge center, it becomes 51 more electron deficient and hence becomes more easily subject to iodide nucleophilic attack, leading to reactions giving rise to 8 shown in Scheme 3.2. ')a MJtZTQ {1: “DAN o o 0 O ’0 If 6’ WZQ e\®/\/\)Ol\p + lion {:1 WSW S O o 0 o I I o 0 Q @WL \s o’N | O 8 mlz = 260 Scheme 3.2 Mechanism of dimethyl sulfonium formation by reaction of iodomethane with I Cross-linking reagent I was not successfully prepared by methylation with iodomethane due to the electron deficient carbonyl carbon. Presumably due to the steric hindrance resulting from the bulky NHS ester group, phenacyl bromide could also not be employed to alkylate 2. The close arrangement of the NHS ester group and sulfur center in space not only brings difficulty to the preparation of sulfonium from 2 but also has great impact on the stability of sulfonium NHS ester functional group which has tendency to undergo hydrolysis and transesterification reactions [7]. 3.2.2 Transesterification and Hydrolysis Reactions of I 52 I was easily subject to transesterification and hydrolysis because of the electron deficient carbonyl carbon on the short chain length side of the sulfonium ion crosslinking reagent. When 50 % MeOH + 50 % H2O + 1 % AcOH was employed for analysis of I by MS, no molecular ion was observed. Instead, dominant methyl transesterification product (m/z 318) and significant hydrolysis product (m/z 304) were observed as shown in Figure 3.2 and Scheme 3.3. '00“ 318.3 a: U U C. (B '0 E 504 :‘r’ g 304.3 ': 260.3 8 \- M I I 198.1 L II. 1 I L11 II AL AJAA11L AI mg I I I I I T I I I 1 100 200 300 400 500 600 700 800 900 1000 m/z Figure 3.2 Nanoelectrospray quadrupole ion trap mass spectrum of sulfonium I methyl sulfate in 50 % MeOH + 50 % H2O + 1 % AcOH MS buffer 0 O o W o W O +/\/\/U\ N OH +/\/\/lk N / I I it”? 0 0 it”? 0 O O O 9 10 m/z = 318 m/z = 304 Scheme 3.3 Structures of methyl transesterification and hydrolysis products of I Transesterification and hydrolysis were also observed when I was stored in dry DMF solutions or when diluted in 100 % AcCN. A 3 mM stock solution of I in dry DMF was diluted to 20 uM by AcCN at different time and immediately injected to the mass spectrometer. A notable increase in the relative abundance of 9 and 10 and a decrease of I 53 was observed during a 4-hour period storage in DMF (Figure 3.3), giving evidence for the instability of I even when potential nucleophiles were minimized. Trace amounts of water could initiate substantial hydrolysis reactions in a relative short time following the proposed mechanism in Scheme 3.4, while a similar mechanism for methanol could result in methyl esterification. 100- A M. '00“ B M+ 401.2 40” 9 . 10 318.2 50 9 50 304.2 10 318.3 \ 304.3 A 2 JII 1 II , ,. ,1 I11. 1.11 1 III II1x - -. - -.1 100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000 m/z m/z A 100‘ C M+ [00-1 D MT 9:, 10 9 401.2 10 9 401,2 0.) § 304.3 318.2 30K“ '2 "a \ j .8 50~ I 50; < O .2 ‘5 T.) l L M.L11__.....I,~ II I 1 1 41 A L 11 .I_11 1II I r L .1 1 1 L 100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000 m/z m/z |001 E Mi. 1001 F [0 10 401.1 304.2 304.2 9 9 318.2 // 318.2 M+ 50- ‘ 50« /401.2 1.1 .IIL i ILI -1 L 1 lunniLA1. 1LI II1 I 11 100 200 300 400 500 600 700 800 900 1000 m/z 100 200 300 400 500 600 700 800 900 1000 m/z Figure 3.3 Nanoelectrospray quadrupole ion trap mass spectra of sulfonium I methyl sulfate stored in dry DMF, 100 % CH3CN was employed as MS solvent; (A) 0 min; (B) 45 min; (C) 75 min; (D) 2 hr; (E) 3 hr; (F) 4 hr 54 o o) e CON HOH {:IKIWOn—fi’ &:p®s/1\/\JMO:§Q O O O O O O N k WAS o’ *— n’0 s o’N | O O 0 H8 OQH O O OH +/\/\)k ’N + ’N S O OH m/z = 304 Scheme 3.4 Mechanism of hydrolysis of I, methanol has the similar process as a nucleophile for methyl transesterification 3.2.3 Supplementary Evidence from 1-Carboxymethylmercaptopropionic Acid (Compound 1 in Scheme 2.1) Additional evidence for a highly reactive carbonyl carbon one methylene group from the sulfonium positive charge center was obtained from the reaction of 1 with iodomethane under the same conditions as those employed for 2 (Figure 3.4 and Scheme 3.5). 55 MS MS/MS A 100- A 163.0 100 ‘ B 101.0 95: D 33’ g 50‘ 50‘ < 0 .2. g 1010 82.9 - 221.0 a: 163.0 -l - ‘L—ALWfl—L‘L— \ I A 100 200 300 400 500 600 700 800 900 1000 60 80 100 120 I40 160 180 m/z mlz Figure 3.4 Nanoelectrospray quadrupole ion trap mass spectra from methylation of 1 with iodomethane. Similar mechanism as indicated in Scheme 3.2 to form dimethyl sulfonium. (A) The dimethyl sulfonium ion at m/z 163 +, peak at m/z 101 corresponds to a fragment in source; (B) CID-MS/MS product ion spectrum of ion at m/z 163 + O CH I O OH /\/\/u\ 3 : \+/\/\/1L r3 01" R.T., 20 hr, CchN ? 0H 0 1 11 mlz = 163 '2‘ /\ t to?) C9 0 H r‘ / \+/\/\/II\’2 O 'H20 O S‘) OH mlz =163 mlz =101 mlz = 83 + /S\ Scheme 3.5 Methylation reaction of 1 with iodomethane and potential mechanisms for gas phase fragmentation of dimethyl sulfonium product Based on the results described above, cross-linking reagent I would be expected to undergo fragmentation via a charge-directed mechanism resulting in formation of a dominant six-membered ring product ion. However, the highly reactive carbonyl carbon limits the application of this reagent under aqueous buffer conditions. 56 3.3 Cross-Linking Reagent S-Methyl 3-propionoyl,5- pentanoyldihydroxysuccinimide sulfonium (Compound 11 in Scheme 2.6 ) 3.3.1 Stability in Iodomethane The only difference in the structure of II and I is the inclusion of an additional methylene group at the short NHS ester end. Because of this methylene group, the electron density of the carbonyl carbon is not so strongly withdrawn by the positive charge on the sulfur, making it much less reactive than that of I. As a result, II is quite stable when present together with iodomethane and no significant dimethyl sulfonium could be observed. 3.3.2 Stablity under Aqueous Conditions The addition of one methylene group improves the stability of II not only in the presence of excess iodomethane, but also under aqueous conditions. The mass spectra of freshly prepared 11 in 100 % CH3CN or in 50 % CH3CN + 50 °/o H2O MS solvent are quite similar (Figure 3.5). 57 100 ~ A M. 100 - B M, 5: 415.1 415.2 Q) U C. (U '0 I: .8 50 ~ 50 « < 4) .2 35. 198.2 [2M+|1‘ 318.3 g 2603‘ 956.7 1 .I IL 1 a 1 I ALI 1 1 ILI 200 400 600 800 1000 1200 200 400 600 800 1000 1200 m/z m/z Figure 3.5 Nanoelectrospray quadrupole ion trap mass spectra of sulfonium II iodide. (A) In 100 % CH3CN buffer, small peak of dimethyl sulfonium (m/z = 260, compound 8 in Scheme 3.2) observed; (B) in 50 % CH3CN + 50 % H2O buffer, a small peak resulting from methyl transesterification (m/z = 318, compound 9 in Scheme 3.3) was observed 3.3.3 Supplementary Evidence from l-Carboxyethylmercaptopropionic Acid (Compound 3 in Scheme 2.6) To further demonstrate the improvement of stability brought by adding one methylene group between the carbonyl carbon and the sulfur atom, similar methylation experiment with iodomethane was carried out for 3. On the basis of our previous experience, we could expect methylated 3 (compound 12, m/z =221) as the main product rather than the dimethyl sulfonium (compound 11, m/z =l63). As shown in Figure 3.6, m/z 163 was observed at only low abundance compared to m/z 221. 58 MS MS/MS A 100] A 221.1 100‘ B 100.9 s: Q.) Q I: «3 'U C 3 50- 50+ < 6 220.9 .2 iii or; 163.0 82.9 11.1 1 131.- . - .- . '29-8 100 200 300 400 500 600 700 800 900 1000 80 100 120 140 160 180 200 220 240 m/z m/z Figure 3.6 Nanoelectrospray quadrupole ion trap mass spectra from methylation of 3 with iodomethane. (A) Expected product 12 at m/z 221 is dominant peak, small amount of dimethyl sulfonium (m/z = 163) (Compound 11 in Scheme 3.5) by-product observed. (B) CID-MS/MS product ion spectrum of ion at m/z 221 + O O C JOKA O ’u\/\ /\/\/lk H3| - +/\/\/II\ 0“ 3 0“ R.T., 20 hr. CH3CN 0“ ‘T’ O” 3 12 mlz=221 OH’Lvs/MTSH (299 I») Hn J /00 mlz=221 0 ._1-12_o_> C I mlz=101 mlz=83 + O o (DH/Ivy H+ Transfer 0 0 0H 0 OHMS/ , mlz = 121 Scheme 3.6 Methylation reaction of 3 with iodomethane and potential mechanisms for gas phase fragmentation of methylated product 12 59 The improved stability of cross-linking reagent II was experimentally demonstrated above. The addition of one methylene group also alleviates steric hindrance on sulfur brought by the large NHS ester group. As a result, phenacyl bromide could be employed to alkylate 4 (Scheme 2.6) to yield a phenacyl substituted sulfonium ion (Figure 2.7). 3.3.4 Suspected E2 Elimination Reactions Might Lead to Unintended Cleavage on Cross-Linker The additional methylene group makes it possible, however, for E2 eliminaiton reactions to occur. During this process, a proton from the methylene group adjacent to the carbonyl on the short chain is transferred to a nucleophile, which could come from the same or other molecules. As illustrated in Scheme 3.7, for the case of an intermolecular cross-linked peptide, pathway 2 would result in intramolecular proton transfer and bond cleavage to yield alternative product ions. 0 o 0 Pathway1/U\/\S/ + GIL—3‘: :K km 0”. NLIJ Pathway 2 Mi /u\/ + \ /\/\/u\ 1'1 Pathway 2 Scheme 3.7 Potential fragmentation mechanisms of II intermolecular cross-linked peptides Based on the previous study by Amunugama et a1. [47], the formation of protonated six-membered tetrahydropyran product ion via pathway 1 is predicted to be the energy favored fragmentation process. However, the possibility of cleavage on the other C-S 60 bond due to the E2 mechanism is not able to be excluded. The coexistence of two fragmentation pathways of II cross-linked peptides would complicate the resultant MS/MS spectra. To simplify the identification of cross-linked peptides, a symmetrical ionic cross-linking reagent III was developed. 3.4 Cross-Linking Reagent S-Methyl 5,5’-dipentanoylhydroxysuccinimide sulfonium (Compound III in Scheme 2.8 ) III is analogous to I and II, but contains 4 methylene groups on both chains. On the basis of our previous experiments, we could expect III to be most stable among three reagents. Under ClD-MS/MS conditions the symmetrical structure results in formation of the same products upon cleavage of either C-S bond in III. 3.4.1 Stability in Aqueous Conditions and Methanol The hydrolysis reaction of NHS-ester is a major competing reaction of the NHS- ester acylation reaction when carrying out cross-linking reactions. Consequently, stability of cross-linking reagents in aqueous buffers must be considered. Similar experiments as described above were performed using freshly prepared III in water or/and methanol containing MS solvents. No obvious hydrolysis or transesterification was observed from 111 in water or methanol, even under acidic conditions (Figure 3.7). Because of the solubility and stability of this reagent under both aqueous and non-aqueous conditions, 111 has the potential to cross-link a large range of proteins under varying conditions. 61 100 I A M+ 100 1 B M+ 443.3 443-3 50 'l 50 _( [M+CH3SO4+H]+ 3 555.5 a, 360.2 / [2M+CH3SO4]+ 8 I. L 997.0 g - ._“L I L 1 11 1 1 _I g 200 400 600 800 1000 200 400 600 800 1000 '2'" m/z m/z .3 100 1 C M+ 100 ' D M+ 2 443.3 443.2 0 o: 50 4 50 - 360.3 [2M+CH33041+ [2M+CH3SO4]+ 996.9 M J 996.9 JAIJIIAALI 1 III- I1I 1 1 1I jJJLJl ...4 1 I II 11A. A -I 200 400 600 800 1000 200 400 600 800 1000 m/z m/z Figure 3.7 Nanoelectrospray quadrupole ion trap mass spectra of sulfonium III' methylsulfate. (A) In 100 % CH3CN buffer, small peak of methyl ester sulfonium (m/z = 360, analogous to compound 9 in Scheme 3.3) observed; (B) in 50 % CH3CN + 50 % H2O buffer; (C) in 50 % AcCN + 50 % H20 + 1 % AcOH buffer; (D) in 50 % MeOH + 50 % H2O + 1 % AcOH buffer 3.4.2 Comparison among Three Ionic Cross-Linking Reagents and Their Intermediates The structure of the three cross-linking reagents and their synthetic intermediates only differ in the length of the carbon chain on one side of the sulfur atom. For the sulfonium compounds, the methylene group in the shorter chain is more reactive due to electron withdrawing effects resulting from the vicinal carbonyl group and positive charge centers. One consequence is that this reagent is more vulnerable to nucleophilic attack. Although III was found to satisfy all the principal requirements for a gas phase 62 cleavable cross-linking reagent, I and 11 could also be useful for varying the arm chain length in some specific applications and reaction conditions. 3.5 Cross-Linked Neurotensin Neurotensin (pELYENKPRRPYIL, MW 1672.9) contains only one lysine residue at position six, with a cyclized N-terminal glutamic acid. Under cross-linking reaction conditions, the primary amine at this lysine residue is the only available reactive group toward NHS-esters. Consequently, only intermolecular cross-link and/or mono-link products could be observed from a cross-linking reaction. The use of a gas phase cleavable cross-linking reagent to aid the identification of cross-linked neurotensin has previously been described by Back et al. using their BID-NHS reagent (Scheme 3.8), which could yield a low mass specific marker ion under low energy CID conditions [26]. As a comparison, cross-linking reactions with I, II and BID-NHS were performed following a method modified from that of Back et. al., using a 1:1 molar ratio of cross- linking reagent to neurotensin in DMF, a reaction time of 30 min, and a 20-fold excess of water added after reaction for 15 min to hydrolyze unreacted NHS-ester groups. iiz’c’fjwofi CH °° Scheme 3.8 Structure of the cross—linker N-benzyliminodiacetoylhydroxysuccinimid (BID-NHS). Adapted from Reference 26 3.5.1 Cross-Linking Reaction with BID-NHS 63 The proposed fragmentation reaction from BID-NHS intermolecular cross-linked peptides is illustrated in Scheme 3.8. W2 =91 Scheme 3.9 Proposed mechanism for the formation of benzyl cations from BID-NHS cross-linked peptide. Adapted from Reference 26 Similar to that described by Back, this reaction was only found to occur during the fragmentation of cross-linked neurotensin in its 5+ charge state (Figure 3.8A). At other charge states, the dominant processes were peptide backbone cleavage or side chain neutral losses that were are not able to give diagnostic information for identification (Figure 3.8B-D). '00 “ A [M+5H-CéH5CH2]4+ '00 ‘ B [M+411-1.14+ 861.7 856.2 [M+411]4+ 50 4 50 ~ 884.8 [M+5H])+ A 708.1 5 \l l D U E ‘ - 4.11 JL [I J A 1 Au 5, 400 600 800 1000 1200 1400 1600 1800 2000 400 600 800 1000 1200 1400 1600 1800 2000 1% m/z m/z O - E '00 C [M+3H'M‘313+ ml D [M+2H-NH3]2+ (U 3 ”72-9 1758.8 Cd 50 ~ [M+31113+ 50 . 1179.2 2, [M+2H] 1 | / 1767.2 1 A .111 L ‘ 400 600 800 10001200140016001800 2000 600 800 1000 1200 1400 1600 1800 2000 m/z m/z Figure 3.8 Low energy CID MS/MS of BID—NHS cross-linked neurotensin. (A) Precursor ion in the 5+ charge state, product ion at m/z 861 corresponds to the loss of benzyl cations; (B) precursor ion in the 4+ charge state, most abundant product ion at m/z 856 corresponds to peptide backbone cleavage between I and L residues; (C) precursor ion in the 3+ charge state, dominant product ion resulted from neutral loss of NH; from N or R side chains; (D) precursor ion in the 2+ charge state, dominant product ion resulted from neutral loss of NH3 from N or R side chains A “mobile proton” condition is required to realize the formation of expected marker ions from BID-NHS cross-linked peptides. In the case of neurotensin, 4 ionizing protons are expected to be ‘sequestered’ on the side chains of arginine, lysine and histidine residues in the peptide sequence. Thus, to initiate cleavage between the nitrogen and the benzyl moiety, a fifth proton would be required. In order to realize the intended cleavage independent of the charge state of the cross-linked peptides, Back proposed the possibility of quatemarizing the amine in the linker chain. However, presumably due to severe steric hindrance of the NHS-ester and benzyl functional groups, fixing the charge 65 on the amine next to the desired leaving group could not be achieved here directly from BID-NHS. 3.5.2 Cross-Linking Reaction with Ionic Cross-Linking Reagent I and II The proposed fragmentation reactions of I and II intermolecular cross-linked peptides are illustrated in Scheme 3.9. The labile C-S bond at the long chain length side of reagents as indicated by dashed lines allows for preferential cleavage to produce two unique peptides, each containing an additional mass modification corresponding to the remaining portion of the cross-linker fragments. Because I and II only differ by one methylene group, fragmentation of I and II intermolecular cross—linked peptides are expected to generate one peptide with an additional mass of 83 u for the ITP (iminotetrahydropyran) modification, and another peptide with either a 88 u or 102 u sulfide modification (S). Tandem mass spectrometry of individual peptides at higher energy CID then allows peptide identification and determination of the sites of modifications via interpretation of the y- and b-type ions generated. Note that the feature of ITP which distinguishes it from other neutral modifications is that it results in the addition of one charge to the product ion. 66 H l H H l O N Cs N ~Cs\/u\/§HL «Wavy We 8 Iw 1w AWN? “fig“? 511 5' Amass=+102u + Amass=+88u H 0 HQ 6) ITP A mass = + 83 u Scheme 3.10 Proposed fragmentation mechanism of l and II intermolecular cross-linked peptides. The resulting mass additions (relative to the single peptide) following cleavage on the C-S bond are'indicated, where S denotes a sulfide modification and ITP denotes protonated iminotetrahydropyran modification; I and 11 indicate cross-linking reagent The cross-linked neurotensin in its 5+ charge state, formed by reaction with either I or II, were analyzed by MS/MS (Figure 3.9). As expected, the fragmentation occurred exclusively at the desired C-S bonds and the peptide chains remained intact to generate simple MS2 spectra. Two peaks emerged: one corresponding to the triply charged neurotensin for the protonated ITP modification ([MH2+ITP]3+), which is the same product ion from I or II, and the other corresponding to doubly charged neurotensin for the s modification ([MH2+S]2+). 67 MS/MS MS/MS [2M+4H+ITP-S“]5+ .00 , [2M+4H+1T1>-s,]5+ 100 . A A [M+211+17‘P]3+ B [M+2H+ITP]3+ <3 586.0 586.1 4) U C: ('3 "U S ..D 50 'l 2+ 50 .. < [h 1+2H+S“] [M+2H+sl]2+ Q, 888.5 _>_ 881.6 E Q) °‘ 1 l 200 400 600 800 1000 1200 1400 I600 1800 2000 200 400 600 800 1000 1200 1400 1600 1800 2000 m/z m/z Figure 3.9 Low energy CID MS/MS spectra of ionic cross-linked neurotensin. (A) From cross-linker II; (B) from cross-linker I. The arrow indicates the precursor ions: (A) m/z 706.9, 2 = 5; (B) m/z 704.2, 2 = 5 To obtain additional sequence information, the two peaks resulting from the MS/MS fragmentation reactions were selected and subjected to M83. The sequence of neurotensin and the sites of modification were then identified from the product ion spectra. As shown in Figure 3.10A and 3.108, the observation of a series of b- and y-type ions provided the sequencing information for the S modified neurotensin. In particular, the existence of fragment ions at y7 which do not contain the modification and yg-NH3 which does confirmed that the modification occurred at Lys 6 of the peptide. The MS/MS spectrum of unmodified neurotensin in its 2+ charge state is shown in Figure 3. It is noticeable that the S modifications did not significantly affect the fragmentation pattern. 68 MS’ _ ‘ 2+ 100 A 1b.2 +H201 2+ y m . [M+2H+Sl] 9 ‘24- o b - 2+ 9 bl2 2+ le Y7'NH3 ‘2... 50 -‘ .21,“ 8'2 / ‘-NH3 89 ,2,\ 3’8 b,‘ N,H .2+ bll . y9 -NH3 b -NH Y 11 3 7 .1 2 -JL -..ull .11 400 600 800 l 000 l200 1400 I 600 1800 2000 m/z M33 — " 2+ M+2H+S 2“ A 100 B [biz +H20] [ n] i\: b C. E b9..2+ 2 ..2+ it .0 '8 Y9 2+ biz 2+ '3 )7' 'NH3 5 so - .o < 0.2+ '2 \ l "-NH3N n 0 ‘.2+ '2 bll u Y9“ % Yakx Y5 \ I] b8\ M I] y b "-NH 1J1 I 17 - I - zéli// H 1 3 - 400 600 800 1000 I400 I600 l800 2000 m/zl2 MS/MS “’0 ‘ C thump?" [M+2H]2+ 2+ 2+ biz Yo a 2+ b 2+ 50 o [y ~NH 12 '2 ‘3 9 3 2+ ya 9 1. y NH, b _ y, Y1 3 bQ-ll blO—ll 2 5 y41 / ‘I y9-NH3 “’34 \ 5’3 / t J ys-NH, l Y7 100:3; \ b9 l l 1 .13. I - u -311 - . \l. 21.11 Lilli ..UI 1L 1 . “Lu 400 600 800 1200 m/z Figure 3.10 Low energy CID Ms3 or MS/MS of neurotensin. (A) Ms3 of 8 modified neurotensin from cross-linker I, * indicates SI modification; (B) MS3 of S modified neurotensin from cross-linker II, ** indicates S" modification (C) MS/MS of unmodified neurotensin in 2+ charge state Similarly, the triply charged ITP modified precursor ions at m/z 586 were isolated and fragmented (Figure 3.1 1A). The fragmentation pattern of ITP modified neurotensin was observed to be quite similar to that of unmodified neurotensin in 2+ charge state instead of its 3+ charge state (Figure 3.1 18), presumably due to the extra charge being 69 fixed on the ITP group, and thus does not change the proton mobility conditions of the peptide. M83 [M+2H+ITP]3+ IOO y9#3+ # 2+ [bl2 +H,0] 50 ~ y ,3 i \ o: 1 1 -11L “.111 A § 200 300 400 500 600 700 800 900 1000 E m/z .D < MS/MS 3 ' [M+3H13+ E 100 ‘7 y 2+ odd) [3’10 "NH3]2+ 9 2+ Ylo _ 2+ [Y9 NH3] [bu-NH3]2+ 2+ 3’11 50 _4 3+ 2+ y 3+ bl} [3’11 ‘NH3] 2:} 2+ a'22+ b1224- y., 3!; b 2* / [1’124'Hzol2+ 85 b b3-4 a, b L ym [9 \ l/ 1 " 3'1l 31 1 -LlLL 11 J J11. 1 200 400 600 800 1000 1200 m/z Figure 3.11 Low energy CID MS3 or MS/MS of neurotensin. (A) MS3 of ITP modified neurotensin, both I and 11 give same product ions; (B) MS/MS of unmodified neurotensin in 3+ charge state The capability of our ionic cross-linking reagents and methodology was experimentally demonstrated from the preliminary work of I and 11. Exclusive cleavage on desired C-S bonds under low energy CID makes identification of cross-linked peptides straightforward. However, the instability of these reagents under aqueous conditions limited their utility. 70 CHAPTER FOUR IDENTIFICATION OF CROSS-LINKED PEPTIDES FOR PROTEIN INTERACTION STUDIES USING A SYMMETRICAL IONIC CROSS-LINKING REAGENT AND MASS SPECTROMETRY 4.1 Introduction Due to its improved stability under aqueous conditions, cross-linking reactions with III could be optimized in PBS buffer; one of most commonly used reaction buffers for such purpose [11, 14, 23]. Generally, there are three main cross-link types generated from a single cross-linking reaction as described earlier, mono-link (type 0), intra-peptide cross-link (type 1), and inter-peptide cross-link (type 2). Usually mono-link is the modification of peptide with one reactive group of cross-linking reagent and the other one hydrolyzed. However from our previous experience there is still some unhydrolyzed mono-link observed under typical conditions. Thus two types of mono-link are described in this work: hydrolyzed and unhydrolyzed. Potential mechanisms for fragmentation of the three cross-link types generated by reaction with III are depicted in Scheme 4.1. For an intra-peptide cross-link (Figure 4.1A), the desired C-S bond cleavage is expected to be the energetically favored process. However, no change of m/z will be observed during the event allowing further fragmentation to directly occur to yield b- and y-type ions. Due to the symmetrical structure of 111, two C-S bonds are expected to be cleaved with similar efficiencies, though only one is displayed in Scheme 4.1A. Therefore it should be noted that a given b or y product ion may be observed containing both ITP and S modifications. 71 l Hi0 {is N“; MSWS. NHQ, /s NH {1 ii 11? “Vi 3‘6 W71 "1’ 3111 Amass=+83u Amass=+130u Scheme 4.1 Proposed fragmentation mechanism of III cross-linked peptides. The resulting mass additions (relative to the single peptide) following cleavage on the C-S bond are indicated, where S denotes a sulfide modification and ITP denotes protonated iminotetrahydropyran modification. (A)Intramolecular cross-link; (B) unhydrolyzed and hydrolyzed mono-link; (C) intermolecular cross-link For the unhydrolyzed and hydrolyzed mono-link (Scheme 4.1B), it is predicted that the major cleavage takes place in the C-S bond closer to the peptide chain due to the amide nitrogen is greater nucleophile compared to either the ester or acid oxygen. Similar to I and II described earlier, two separated peptide chains would be generated upon low energy CID from an inter-molecular cross-linked peptide of III (Scheme 4.1 C): one corresponding to ITP modification with an additional mass of 83 u 72 and the other corresponding to S modification with an additional mass of 130 u. If this is a homo-dimer cross-linked peptide, i.e. the two peptides connected together are the same; single ITP and S modified product ions will emerge in the MS/MS scan due to symmetrical structure of III. Otherwise for hetero-dimer cross-linked peptides, a pair of ITP and S modified product ions will be generated. Further tandem mass spectrometry (M83) of the individual product ions allows the identification of the peptide sequence and modification sites. 4.2 Cross-Linking Reaction with Neurotensin 4.2.1 Cross-Linking Reactions in DMF and PBS Neurotensin was dissolved in solvents to a concentration of 0.6 mM for all the cross-linking reactions as mentioned before. The cross-linking of neurotensin in DMF, carried out at 1:2 molar ratio of cross-linker to peptide and a reaction time at 75 min, without hydrolysis, was found to give the highest relative yield of inter-molecular cross- link (Figure 4.1A). However, prominent unmodified neurotensin peaks were observed indicating an incomplete reaction. For cross-linking in PBS, 3 first attempt was made to optimize reaction conditions by varying molar ratios of cross-linker to peptide at 1:2, 2.5:1 and 5:1 and by using reaction times of 45, 90 and 120 min for each reaction ratio. The best relative yield of mono-link products (Figure 4.18) was obtained at a 2.511 molar ratio of cross-linker to peptide for 90 min. Hydrolyzed mono-link product appeared and was observed the most abundant cross—linking products when the reaction was conducted in PBS buffer at the referring conditions but the inter-molecular cross-link still could not 73 be improved by this strategy. Though unmodified neurotensin peaks predominated in the mass spectra of product mixtures from cross-linking reactions, signals of cross-linked products were strong enough for subsequent fragmentation and unambiguous identification. 100‘ M+3H 3+ MS - 3+ MS A [ l '00 3 [mm] [M+2H+lTP-SmOH]3+ 635.3 558.7 558.9 / [M+2H+rr1>-s,,,N]3+ 3+ + + - 667.6 [2M+4H+1TP-s,,,]5+ [M 2H ITP smN] 712.4 . ' / 50 [Zli'fi’MHH'P'S111]5+ 2+ 712.4 7 W" 2”] M+2H ~+ 111+ / 837.0 111* /[ 837.0] 443.3 [M+H]+ 443.3 1--- 1673.1 11 ii .i ”.1“ ——_T—_—T—I'—'_ l I I T 1 I l I I I a I I T T 1 200 400 600 800 1000 1200 I400 1600 1800 2000 200 400 600 800 1000 I200 I400 I600 l800 2000 m/z m/z Relative Abundance (%) ‘6 Figure 4.1 Nano-electrospray quadrupole ion trap mass spectra of III cross-linked neurotensin. (A) Cross-linking reaction in DMF at 75 min with 1:2 molar ratio of cross- linker to peptide; (B) cross-linking reaction in PBS buffer at 90 min with 2.5:1 molar ratio of cross-linker to peptide. M: unmodified nurotensin, [2M+4H+ITP-Sm]5+ (m/z 712.4): inter-molecular cross-link in 5+ charge [M+2H+ITP-SmN]3+ (m/z 667.6): unhydrolyzed mono-link in 3+ charge state, [M+2H+I'I‘P-SmOH]3+ (m/z 635.3): hydrolyzed mono-link in 3+ charge state 4.2.2 Inter-Molecular Cross-Linked Neurotensin The majority of inter-molecular cross-linked neurotensin were present in its 5+ charge state due to two basic residues contained in each peptide chain and one fixed charge on the cross-linker (Figure 4.1). Some +4 charge state was also observed. The dissociation of inter-molecular cross-links was characterized by low energy CID MS/MS. As indicated in Figure 4.2, MS/MS gave simple fragmentation patterns for both charge states due to specific cleavage of the cross—linker. Thus, we can conclude that the 74 fragmentation of inter-molecular ionic cross-linked neurotensin is not affected by the charge state. 100 A 586.1 50 ~ 2 [M+2H+Sm] * 902.5 Relative Abundance (%) l MS/MS 100 '1 [M+2H+1Tm3+ [21Vl+4H+ITP—Sm]5+ 50‘ B [M+H+1TP]2+ 878.5 I MS/MS [2M+3H+1rP-sm]4+ [M+2H+sm]2+ 902.4 ./ 200 400 600 800 1000 1200 I400 1600 1800 2000 m/z 400 600 800 1000 1200 1400 1600 1800 2000 m/z Figure 4.2 Low energy CID MS/MS of inter-molecular cross-linked neurotensin by III prepared in DMF. (A) Precursor ion in 5+ charge state (m/z 712.5); (B) precursor ion in 4+ charge state (m/z 890.5). The arrow indicates peak of precursor ions The two product ions from the +5 charge state MS/MS spectrum were then selected and subjected to MS}. As shown in Figure 4.3, the observation of a series of b- and y—type ions provided information on the sequence and modification site for the peptides. In particular, the presence of yg ions containing ITP or S tags and the presence of y7 ions without any modification provided evidence that the modification had occurred at e- amine of Lys 6. 75 100 — A y 113+ MS" 9 [6,,”+H,013* [M+2H+1TP]3+ # 2+ [ylo 'NH3] [Y] ifi'NH3]2+ 50 — #3+ 9 #2+ #3+ Yam+ 24?,” \ b13#3+ l ngm y” /# b y \ b ’ l ’ i ‘ / i I ° J. L Li L ..I -J-L“Jl1lh lAll JlJl. 1:11.; .1.- ..l- -1. X A 200 300 400 500 60? 700 800 900 [000 m 2 Ms3 2+ — + + A 100 B .. [blz‘+H,O]2+ [M 2H Sm] \° 2+ “ 3., O U C N -o C 3 .D < °>’ . ‘3 59 'NH; 3 'l . ‘3‘ b” 'NH3 -1 - 1400 1600 1800 2000 MS3 '00 _ [b[2#+HZO]2+ [M+H+ITP]2+ #2+ b13 2+ b # NH b13 9 - 3 # 50 + / V8 'NH3 Y7'NH3 # / b g 3’9 'NHJ Y . 8 / I H “I \ + 400 600 800 1000 1200 1400 I600 I 800 2000 Figure 4.3 CID MS3 product ion spectra of modified neurotensin in Figure 4.2 formed from inter-molecular cross-links. (A) [M+2H+ITP]3+ ion; (B) [M+2H+Sn1]2+ ion; (C) [M+H+ITP]2+ ion. ITP modification is labeled as #, Sm modification is labeled as * 4.2.3 Mono-Linked Neurotensin The majority of cross-linked products existed in mono-link types when cross- linking was performed in PBS buffer (Figure 4.1B). Consistent with the prediction shown in Scheme 4.1B, a product ion containing an ITP tag is the dominant fragment from both unhydrolyzed and hydrolyzed mono-links in 3+ charge state (Figure 4.4). It 76 appears that the carbonyl group of cross-linker-lysine amide bond on the mono-link is a much stronger nucleophile than NHS-ester carbonyl or free carboxylic acid under the experimental conditions, thus much more reactive to initiate the fragmentation reactions. However another pathway involving nucleophilic attack from the carboxylic acid carbonyl was observed as a minor pathway, as evidenced by a minor S modified product ion formed by fragmentation of the hydrolyzed mono-link (Figure 4.4B). MS/MS MS/MS 100- 3+ 100 - 3+ 83 535.9 5860 Q) Q C CU 'U E 4 . J, 50 50 < .52) [M+2H+5111]2+ 3 902.4 9.) a: l l / . ,. .I__.,A* , ,., 2-7, _ ._ _ _. _,7,,,._.v7, . . 1 , - 2 1 .2, ~_,fi 200 300 400 500 600 700 800 900 100011001200 200 300 400 500 600 700 800 900 100011001200 m/z m/z Figure 4.4 Low energy CID MS/MS spectra of mono—linked neurotensin by III prepared in PBS buffer. (A) Unhydrolyzed mono-link, m/z 667.7, 2 = 3; (B) hydrolyzed mono- link, m/z 635.3, 2 = 3. The unassigned arrow indicates peaks of precursor ions Distinct fragmentation behavior was detected in CID MS/MS of the doubly charged hydrolyzed mono-link (Figure 4.5) where an S tagged neurotensin product ion dominated the spectrum. Thus, it appears that charge states can have some influence in the fragmentation behavior of mono-linked neurotensin. However, no matter which C-S bond cleaves, it is still the energetically favorable mechanism and takes place prior to peptide backbone cleavage. 77 MS/MS ‘00“ [M+2H+Sm” [M+H+ITP-SmOH]2+ cf, 902.4 0) U C CU “O ..D < 2 [M+2H+1TP]2+ 3:5 878.5 15:) \ l 400 600 800 1000 1200 1400 1600 m/z Figure 4.5 Low energy CID MS/MS spectra of hydrolyzed mono-link in 2+ charge state prepared by reaction of neurotensin with 111 in PBS buffer. The unassigned arrow indicates peak of precursor ions, m/z 952.3, 2 = 2 The product ion spectra obtained by MS3 dissociation of the doubly and triply charged ITP modified ions, and the doubly charged S modified ions in Figure 4.4 and 4.5 were found to yield the similar product ion patterns as those observed in Figure 4.3 from the inter-molecular cross-link (data not shown). The cross-linker III has been successfully applied to cross-linking reactions with neurotensin in both DMF and PBS medium. The experimental capability and methodology was demonstrated when inter-molecular cross-link and mono-linked peptides were unambiguously identified from low energy CID MS/MS and through MS3. Simple MS/MS patterns make the identification of cross-link types straightforward and the modification attached on the peptide product ions do not affect the ability to generate b- and y-type ions for fiirther characterization of the peptide sequence and modification sites. The capability of this novel cross-linking reagent will be discussed more in following sections. 4.3 Cross-Linking Reaction with Insulin-Like Growth Factor I (57-70) 78 Insulin-like growth factor I (57-70) (IGF, ALLETYCATPAKSE, MW 1495.7), contains two primary amine functional groups: one is on the lysine side chain at position 12 and the other is on the N terminus. With two reactive groups available toward NHS- esters, three cross-linked products might be observed from a cross-linking reaction. Cross-linking of IGF was performed in PBS buffer at 2.5:] molar ratios of cross-linker (15 mM, 4 uL in DMF) and peptide (0.6 mM, 40 11L in PBS buffer) for 90 min at room temperature. The majority of cross-linking led to intra—molecular cross-link type, although a minor hydrolyzed mono-link was observed and no significant inter-molecular product (Figure 4.6). When two reactive amine groups are present in the same peptide chain, there are some questions should be considered before further investigation of MS/MS and MS3 behavior of cross-linked products. One question is whether two amine groups have same possibility to form a mono-link or involved in an inter-molecular cross-link? A related problem might be that for an intra-molecular cross-link, the cleavage of two C-S bonds is random or preferential due to the amide bond formed toward different residues or at different places. 79 100 - MS 2+ [M+H+ITP-Sm] 854.9 [M+2H]2+ 50 * 748.9 [M+H+ITP-SmOH]2+ \ / 8637 11. .111. 200 400 600 800 1000 1200 1400 1600 1800 2000 m/z Relative Abundance (%) Figure 4.6 Nanoelectrospray quadrupole ion trap mass spectrum of III cross-linked IGF. Reaction performed in PBS buffer. M: unmodified IGF; [M+H+ITP-Sm]2+: intra- molecular cross-link in 2+ charge state; [M+H+ITP-SmOH]2+: hydrolyzed mono-link in 2+ charge state To address these questions, a first attempt was made to characterize the fragmentation of intra-molecular cross-linked IGF. Consistent with the prediction shown in Scheme 4.1A, direct b- and y-type product ions were generated upon CID MS/MS (Figure 4.7). For an intra-molecular cross-linking occurring at the N-terminus and Lys 12, we could expect that all the resultant b-type ions are modified with an ITP or S tag while y-type ions starting from y:, should be modified as well. And besides precursor ion, only the b-type ions starting from biz could contain both tags. However, a discrepancy was observed between the theoretical prediction and the experimental results. In particular, y3, y5 and y6 ions were observed without modification. Otherwise b7, b3, b9, b“ and y3 ions were identified containing both ITP and S tags. Based on those results, it appears that some cross-linking had taken place at residue 7, i.e., cysteine. The reactivity of NHS-esters toward sulphydryl functional group was seldom reported before; further evidence is required to open up this new channel for the structural analysis of proteins. 80 MS/MS [b,2#.-H20]2+ 2+ IOO N 12 11 2 -H20 b #t {3 [b13'H20] 9 8.x ys ”$2.4. Q) bl3 #- g / b9' H20 (U '0 #9 c 5: -— b10.12 y, b3 y I2 .0 50 t < yo 6,, “-st \b, . g b2“ b # \ b7 5'. t y“ g y3b11 b#5 11* V; y' #1 ._ 3 4 \ ”\y '0 bll nit) \ l # I “Ii 1 I 11 "4 .1l,..- fink-1i ...m J 111 --.. 21. -1 400 600 800 [200 I400 I 600 I 800 2000 000m/z Figure 4.7 CID MS/MS spectrum of doubly charged intra-molecular cross-linked IGF. ITP modification is labeled as #, Sm modification is labeled as * Although C-S bond cleavage is not able to be detected directly from MS/MS spectrum; the generated b— and y-type ions provided us hints regarding the fragmentation mechanism within the linker. Based on the observations that singly tagged b-type ions tended to contain ITP, and y—type ions with single S modification tended to be present, it appears that cleavage didn’t occur randomly at each C-S bond but was instead preferred at the N-terminal amide side. The controlling factors responsible for this bias are not clear. For the IGF hydrolyzed mono-link, similar results to those described above were also obtained by CID MS/MS and M83. As shown in Figure 4.8A, ITP modified peptide ions were observed as dominant fragment ions fiom the doubly charged precursor. Based on the experience of mono-linked neurotensin, a minor peak corresponding to S modified peptide ions was also observed. However the appearance of a ys ion implies that some peptide backbone cleavage had also occurred during MS/MS. This is probably due to enhanced cleavage at the N-terminal side of proline [40, 59]. Thus we could have confidence by the facts that even when compared to preferential N-terminal proline 81 fragmentation, desired C-S bond cleavage within the linker region is still the dominant pathway. 100 2+ MS/MS 7 A [M+H+ITP] + + _ 2+ 789.9 [M H ITP SmOH] 2+ [M+2H+Sm] 50‘ 814.3 e: y. l O 1 AL A 11‘ .1 .1 Q - - ,5 400 600 800 1000 1200 1400 1600 1800 2000 g m/z .13 < 1; [b,,#-H20]:+ 2+ MS3 5 1°07 3 “"3 '"201 b9” [M+H+ITP]2+ Q.) 9 ' 2 #2 l3 l \ yIZ 50"1 5# b8# b,” 3’5“ # b10.12 #b‘ b,‘\)’6 b-,# y” b”# Iy/b’IH ”I’°"°\/ - -11. . . 1 .....1.11 l ---11 -1 . 400 600 800 1000 1200 1400 1600 1800 2000 m/z Figure 4.8 Low energy CID MS/MS and MS3 spectra of hydrolyzed mono-linked IGF. (A) CID MS/MS of hydrolyzed mono-linked IGF, unassigned arrow indicates peak of precursor ions (mlz 863.8, 2 = 2); (B) CID M83 product ion spectrum of [M+H+ITP]2+ ion from hydrolyzed mono-linked IGF. ITP modification is labeled as # MS3 spectrum obtained by dissociation of doubly charged ITP modified peptide (Figure 4.8 B) was observed to be very similar to that from fragmentation of intra- molecular cross-linked IGF due to the fact that they are at the same proton conditions and the attached ITP or S tag doesn’t affect the peptide backbone cleavage. This observation might be utilized to assist in identifying peptide sequence and modification sites by 82 comparing m/z difference between ITP and S tags, especially when a data base searching program is employed. Since there are three reactive functional groups capable of cross-linking reactions, one related problem is their relative reactivity toward NHS-esters. The presence of a modified y5 ion could only come from mono-link at Lys 12; otherwise the absence of modified yg t0 Y13 ions implies no significant mono-link occurred at Cys 7. On the basis of this observation, a mono-link at N-terminal is responsible for formation of the majority of modified b-ions and unmodified y-ions. A much smaller modified y5 peak indicates that the N-terminal mono-link was the main modification site. IGF is an ideal model to characterize the dissociation of intra-molecular cross-link. However the factors responsible for formation of the intra—molecular cross-link as a major product instead of other cross-link types is not clear. A better control of cross- linking reactions needs further investigation into the reaction conditions as well as peptide compositions. 4.4 Cross—Linking Reaction with VTMAHFWNFGK (MWK) Synthetic peptide VTMAHFWNFGK (MWK, MW 1337) contains two primary amine functional groups: one is on the N terminus and the other on the C terminus lysine side chain. Thus modifications on the N-terminal will be observed from all b-type ions while modifications of the lysine side chain would be detected on all y—type ions. MWK was cross-linked in PBS buffer at room temperature, following the conditions employed for cross-linking of IGF. From a single cross-linking reaction, three types of cross-linked products: intra-molecular, inter-molecular cross-link and mono-link were all identified as 83 shown in Figure 4.9. Multiple cross-linking reactions were also observed, where two cross-linker molecules had reacted with two primary amine groups on the peptide chain by one of their NHS-esters followed by hydrolysis of the other NHS-ester, resulting in double hydrolyzed mono-link. [2M+4H+rr1>-s,,,]5+ MS A 100 — [M+2H+ITP-SmOH]3+ [M+2H+lTP-SmN]3+ 578.5 °\° 555.9 [M+H+2rr1>-smon]3+ 4+ 7; 600.3 [2M+3H+lTP-Sm] 0 722.8 1: 3+ _ 8 [M+3H] [M+H+ITP-S 12+ 1: 447 l 111 3 50 — - , g I" 784. -.—.- 443.3 (6 a \ 300 400 500 600 700 800 900 1000 1 100 1200 Figure 4.9 Nanoelectrospray quadrupole ion trap mass spectrum of III cross-linked MWK. Reaction performed in PBS buffer. M: unmodified MWK; [M+2H+ITP-SmOH]3+ (m/z 523.3): hydrolyzed mono-link in 3+ charge state; [M+2H+ITP-SmN]3+ (m/z 555.9): unhydrolyzed mono-link in 3+ charge state; [2M+4H+ITP-Sm]5+ (m/z 578.5): inter- molecular cross-link in 5+ charge; [M+H+21TP-SmOH]3+ (m/z 600.3): double hydrolyzed mono-link in 3+ charge state; [2M+3H+ITP-Sm]4+ (m/z 722.8): inter- molecular cross-link in 4+ charge; [M+H+ITP-Sm]2+ (775.9): intra-molecular cross-link in 2+ charge state; [M+H+ITP-SmOH]2+ (m/z 784.3): hydrolyzed mono-link in 2+ charge state 4.4.1 Intra—Molecular Cross-Linked MWK Intra-molecular cross-linked MWK is observed dominantly in a 2+ charge state. The majority of mono-links and inter-molecular cross-links were observed in 3+ and 5+ charge states. CID-MS/MS of the intra-molecular cross-linked MWK peptide, shown in Figure 4.10, indicated that ITP and S tags could be attached on the same b- or y-ion, providing evidence that cleavage had occurred at both C—S bonds. The chance of cleavage of either bond was not equal, however, as ITP tags were mainly found on y-ions and S 84 tags were mainly on b-ions. This implies that the favorable cleavage pathway is associated with nucleophilic attack from the carbonyl group of the lysine-linked amide bond. MS/MS 2+ A ys Y6 °\° V #2 8 Yio + r: b” “’ 112+ 2 2+ . ”mo -H,o b .. . g 50 ~ y7:\4y# 6 Y9 < b,“ H ,bo y\ob b ‘ y # . 11 8 b 2 \zbz I: #2 K b8” b9 I0 "' t g y\ Y3 y:\ * b7# \1 b8. b9# 1" M \ a6 yio # - 21 I I I _.11 1 y” 400 600 800 l 000 l 200 I 400 l 600 m/z Figure 4.10 CID MS/MS spectrum of intra-molecular cross-linked MWK. ITP modification is labeled as #, Sm modification is labeled as * 4.4.2 Mono-Linked MWK As cross-linking was performed in PBS buffer, the majority of unreacted NHS-ester functional group underwent hydrolysis to yield a hydrolyzed mono-link product. However, the hydrolysis did not go to completion as a small amount of unhydrolyzed mono-link was still observed. CID MS/MS of the triply (Figure 4.11 A and B) or doubly (Figure 4.11 C and D) charged NHS-ester or hydrolyzed mono-links all resulted in neutral losses of a dialkyl sulfide (either as the methyl(pentanoylhydroxysuccinimide) sulfide or as the methyl(pentanoic acid) sulfide) as the dominant fragmentation pathway, resulting in ITP modified peptide ions. 85 MS/ MS MS/ MS “WA [M+2H+1Tp]3+[M+2H+ITP-SmN]3+ '00‘13 [M+2H+ITP-smom3+ 474.0 [M+2H+1TP]3+ 474.0 50- 50 A 7 ‘3‘: [M+2H+Sm13" [M+2H+sm]~* o 734.2 7347 U i C. .8 I 4 1‘ f A T I I I j F—T A I ‘JIL II I I I I g 200 400 600 800 1000 1200 1400 1600 200 400 600 800 1000 1200 1400 1600 g m/z m/z a,» IO()-C[M+H+1TP]’* MS/MS 2 100.], 2+ MS/MS 7 -.: : _ + [M+H+ITP] _ -+ ,3 7,0,, [M+H+ITP smN] m4 [M+H+ITP smon] CZ 50~ 50< , [M+211+s,,,]~* 1 734.3 1L] 1. III. I. 1. 400 600 800 1000 1200 1400 1600 400 600 800 1000 12100 1400 16,00 m/z m/z Figure 4.11 Low energy CID MS/MS spectra of mono-linked MWK by III prepared in PBS buffer. (A) Unhydrolyzed mono-link in 3+ charge state; (B) hydrolyzed mono-link in 3+ charge state; (C) unhydrolyzed mono-link in 2+ charge state; (B) hydrolyzed mono- link in 2+ charge state The spectra obtained by MS3 dissociation of the triply and doubly charged ITP modified peptides are shown in Figure 4.12 A and B, respectively. Distinct fragmentation behaviors were observed from these two charge states. It can be seen from Figure 4.12 A that b-ions are dominantly present in their ITP modified forms, while most y-ions are unmodified, indicating that the monolink for the 3+ charge state was primarily from reaction at the N-terminus of the peptide. The data in Figure 4.12 B, however, indicated no significant difference in the relative intensity of the b- and y-type product ions with or without the modification, indicating no preference for formation of the mono-link at either the N-terminal or C-tenninal lysine. The results are consistent with a decreased 86 proton affinity for the lysine residue when the cross-linking reaction had occurred at its amine group. M83 [M+2H+1TP]3+ o + I g I '7 7' —~- '1’ I '7” l I I g 200 400 600 800 1000 1200 1400 1600 1: W2 :1 5:” 2 Ms3 23 100 B [M+H+ITP]2+ 8 50 b8 b,“ hm 1 I I. 200 1200 I400 I600 Figure 4.12 CID MS3 product ion spectra of ITP modified MWK in Figure 4.11 formed from hydrolyzed mono-link. (A) [M+2H+ITP]3+ ion; (B) [M+H+ITP]2+ ion. ITP modification is labeled as #; the unassigned arrow indicates the precursor ion, m/z 710.4, 2 = 2 4.4.3 Double Hydrolyzed Mono-Link Although the cross-linking reaction was carried out at a low cross-linker to peptide ratio, a double hydrolyzed mono-link was observed, as shown in Figure 4.9. A peptide which is multiply mono-linked does not provide any information on spatial relationship 87 between two reactive functional groups, but it could yield important information on the surface accessibility of target functional groups in a protein cross-linking experiment. A particular feature of the double mono-link, which distinguishes it from single modifications, is its multistage MS/MS behavior. Similar to that described earlier, the neutral loss of dialkylsulphide (5-(methylthio)pentanoic acid) was the dominant MS/MS fragmentation pathway from the 3+ doubly mono-linked peptide precursor ion of MWK. However, further dissociation of the ITP modified product from this reaction resulted in a second loss of 5-(methylthio)pentanoic acid, indicative of the presence of the second mono-link. Thus, the identification of the peptide sequence and modification sites in this case required the use of an MS4 experiment (Figure 4.14). MS/MS MS" A 100- A [M+H+2np-smon]3+ 100']; [M+H+ITP+ITP-S,”OH]3+ é [M+H+ITP+lTP-SI"OH]3+ [M+H+2lTP]3+ 3 550.7 501.5 C N '3 .8 50. [M+H+21TP13+ 50- < 50l.3 o / .3. E / ' O o: l l l 200 400 600 800 1600 1200 who 1600 200 400 600 300 1000 1200 1400 1600 m/z m/z Figure 4.13 Low energy CID MS/MS and MS3 product ion spectra of double hydrolyzed mono-link of MWK. (A) CID MS/MS product ion spectrum of double hydrolyzed mono- link of MWK; (B) CID M33 product ion spectrum of [M+H+ITP+I'rP-smom3+ ion formed from double hydrolyzed mono-link in Figure 4.13A. The unassigned arrow indicates the precursor ion: (A) m/z 600.2, 2 = 3; (B) m/z 550.8, 2 = 3 88 M34 3+ 100? b #2, y # [M+H+21TP] A 7 4 E: b #2+ a7#2+ b5# 8 2+ -H,o c: - '8 b ” bum+ r: 4 y5 g 50 a y ii V < 3 #2+ 0.) \ I b9 2 _ #24” as» / _c_u b # # a - ‘ | y. ..L .. 7 u i J . . L 400 600 800 1000 l 200 I400 I 600 m/z Figure 4.14 CID MS4 product ion spectrum of double ITP modified MWK ion formed from double hydrolyzed mono-link in Figure 4.13. ITP modification is labeled as #; the unassigned arrow indicates the precursor ion, m/z 501.5, 2 = 3 4.4.4 Inter-Molecular Cross-Link Characterization of the dissociation of inter-molecular cross-linked peptides not only provides us information enabling peptide identification and determination of modification sites, but also allows for investigation into the factors influencing bond cleavage adjacent to the ‘fixed charge’ within the linker. With two reactive amine groups available in MWK for cross-linking, there could be three possible structures of the inter- molecular cross-linked product. Only two product ions will appear upon low energy CID MS/MS of inter-molecular cross-links, corresponding to ITP and S modifications. However, each peak could be representative of peptides modified at two different sites, N-terminal or C-terminal lysine residue, which could be differentiated by two series of b- and y-type ions observed upon MS3 . As described earlier, the cleavage of C-S bond adjacent to the ‘fixed charge’ was found to be independent of the charge state of the cross-linked peptide (Figure 4.15 A and B). Two product ions were generated in each spectrum, containing ITP and S 89 modifications. Interestingly, MS3 fragmentation of the ITP modified triply charged product ion (Figure 4.16A) resulted in the formation of a different product ion spectrum to that obtained from the mono-linked peptide (Figure 4.12A). For this peptide, more y- type ions were observed containing the ITP tag compared to those from mono-links, indicating a greater extent of lysine side chain modification. MS/MS MS/MS 100‘ A [wmmply [2M+4H+ITP-Sm]5+ 100713 [Mmmply [2M+3H+ITP-sm]4+ 474.1 710.5 [M+2H+Sm]2+ 2, 734.4 [M+2H+Sm] 50, 734.3 50, / Anti _ A ______. _.__._4L_l‘i| — Y 1 1 r r I 1 fl 1 I r r r I I 200 400 600 800 1000 l200 1400 1600 200 400 600 800 IOOO 1200 1400 I600 m/z m/z Relative Abundance (%) Figure 4.15 Low energy CID MS/MS spectra of inter—molecular cross-linked MWK by 111. (A) Precursor ion in 5+ charge state, m/z 578.2; (B) precursor ion in 4+ charge state, m/z 722.6. The arrow indicates peak of precursor ions The product ion spectrum obtained by MS3 dissociation of the complementary S modified peptide ion formed from the inter-molecular cross-link is shown in Figure 4.16B. Large numbers of product ions were generated providing extensive sequence information. Thus the fragment ions generated by MS3 of both ITP and S modified peptides supported the assignment of the peptide sequence. The inter-molecular cross-link described here involved two identical peptides cross- linked together; whereas proteolysis of cross-linked proteins or protein complexes would usually produce heterogeneous cross-linked peptide dimers. However, MWK still provides us an ideal model to examine the fragmentation behavior of different cross- 90 linked products and different charge states for the identification of peptide sequence and modification sites. M33 100 A [M+2H+1TP]3+ b” #3+ [If 50 #2+ A ab: Y3\ '36 112+ y3\\ \8\b Ya Ya” 3 {151th l I 8 .l1001\\i. 11 \I‘lll 1 5E 600 800 1000 1200 I400 I600 g m/z i 2 Ms3 '5 100 1 a, 2+ 2 B lyio‘ [M+2H+8111] o -H,O 0’7 94 * ' b9 Y4 w+ y; + b“ 2+ Y5 y6 b o 50 b8 Jo ‘ b6 :- " ' , b a / 'ys b o y° y: 8\ blO y”; . 1310 y“ o , l1 1 \ \ \ 1 ii 1/ i it u 1 .J1 I I ii L L 1 . .1 ‘1‘ 600 800 l000 I200 1400 I 600 m/z Figure 4.16 CID MS3 product ion spectra of modified MWK in Figure 4.15A formed from inter-molecular cross- -links. (A) [M+2H+ITP]3+ ion, (B) [M+2H+S111]2+ ion. ITP modification 15 labeled as #, Sm modification is labeled as * 4.5 Summary The multistage tandem mass spectrometry of cross-linked peptides presented here illustrate that peptides containing ionic cross-linking reagent 111 could be applied to effectively distinguish between mono-linked, intra-, and inter-molecular cross-links by their distinct fragmentation patterns. Upon low energy CID MS/MS, the mono-link 91 experiences a neutral loss giving rise to an intact peptide chain modified with an ITP tag. MS/MS of inter-molecular cross-linked peptides gives rise to two separate peptide chains attached with ITP or S tags. From fiirther fragmentation events, the peptide sequence and modification site(s) could be unambiguously identified. Sequence ions are directly generated from CID MS/MS of an intra-molecular cross-link, by searching for specific mass additions corresponding to an ITP or S tag on the b— and y-type product ions. The dominant cleavage of the OS bond adjacent to the ‘fixed charge’ within the linker region is experimentally demonstrated to be independent of the charge state of the cross-linked peptide. The solubility and improved stability under aqueous conditions make this symmetrical ionic cross-linking reagent feasible to be applied to large scale proteins and protein complexes, Opening up new horizons for the structural analysis of large biological assemblies. Further work is underway to investigate the protein-protein interactions with these cross-linking and fragmentation methods. 92 REFERENCES 1. Horton H.R.; Moran L.A.; Ochs R.S.; Rawn J.D.; Scrimgeour K.G. Principles of Biochemistry, 3rd edition, Pearson Education, Inc., Prentice Hall, 2002, Page 51. 2. Trakselis, M.A.; Alley, S.C.; Ishmael, F.T. Identification and mapping of protein- protein interactions by a combination of cross-linking, cleavage, and proteomics. Bioconj. Chem. 2005, 16, 741-750. 3. Bader, G.D.; Hogue, C.W.V. Analyzing yeast protein-protein interaction data obtained from different sources. Nature Biotechnol. 2002, 20, 991-997. 4. Sinz, A. Chemical cross-linking and mass spectrometry to map three-dimensional protein structures and protein-protein interactions. Mass Spectrom. Rev. 2006, 25, 663- 682. 5. Young, M.M.; Tang, N.; Hempel, J.C.; Oshiro, C.M.; Taylor, E.W.; Kuntz, I.D.; Gibson, E.W.; Dollinger, D. High throughput protein fold identification by using experimental constraints derived from intramolecular cross-links and mass spectrometry. Proc. Natl. Acad. Sci. USA. 2000, 97, 5802-5806. 6. Chakravarti, B.; Lewis, S.J.; Chakravarti, D.N.; Raval, A. Three dimensional structures of proteins and protein complexes from chemical cross-linking and mass spectrometry: a biochemical and computational overview. Current Proteomics 2006, 3, 1-21. 7. Cross-Linking Reagents Technical Handbook, Pierce, 2005, Page 3. 8. Alley, S.C.; Ishmael, F.T.; Jones, A.D.; Benkovic, S.J. Mapping protein-protein interactions in the bacteriophage T4 DNA polymerase holoenzyme using a novel trifunctional photo-cross-linking and affinity reagent. J. Am. Chem. Soc. 2000, 122, 6126- 6127. 9. Trester-Zedlitz, M.; Kamada, K.; Burley, S.K.; Fenyo, D.; Chait, B.T.; Muir, T.W. A modular cross-linking approach for exploring protein interactions. J. Am. Chem. Soc. 2003, 125, 2416-2425. 93 10. Sinz, A. Chemical cross-linking and mass spectrometry for mapping three- dimensional structures of proteins and protein complexes. J. Mass Spectrom. 2003, 38, 1225-1237. 11. Schilling, B.; Row, R.H.; Gibson, E.W.; Guo, X.; Young, M.M. MS2Assign, automated assignment and nomenclature of tandem mass spectra of chemically crosslinked peptides. J. Am. Soc. Mass Spectrom. 2003, 14, 834-850. 12. Sinz, A.; Kalkhof, S.; Ihling, C. Mapping protein interfaces by a trifunctional cross- linker combined with MALDI-TOF and ESI-FTICR mass spectrometry. J. Am. Soc. Mass Spectrom. 2005, 16, 1921-1931. 13. Ahrends, R.; Kosinski, J .; Kirsch, D.; Manelyte, L.; Giron-Monzon, L.; Hummerich, L.; Schulz, 0.; Spengler, B.; Friedhoff, P. Identifying an interaction site between MutH and the C-terrninal domain of MutL by crosslinking, affinity purification, chemical coding and mass spectrometry. Nucl. Acids Res. 2006, 34, 3169-3180. 14. Chowdhury, S.M.; Munske, G.R.; Tang, X.; Bruce, J.E. Collisionally activated dissociation and electron capture dissociation of several mass spectrometry-identifiable chemical cross-linkers. Anal. Chem. 2006, 78, 8183-8193. 15. Reynolds, K.J.; Yao, X.; Fenselau, C. Proteolytic l8O labelling for comparative proteomics: evaluation of endoprotease Glu-C as the catalytic agent. .1. Proteome Res. 2002, 1, 27-33. 16. Back, J.W.; Notenboom, V.; de Koning, L.J.; Muijsers, A.O.; Sixma, T.K.; de Koster, C.G.; de Jong, L. Identification of cross-linked peptides for protein interaction studies using mass spectrometry and 18O labeling. Anal. Chem. 2002, 74, 4417-4422. 17. Chen, X.; Chen, Y.H.; Anderson, V.E. Protein cross-links: universal isolation and characterization by isotopic derivatization and electrospray ionization mass spectrometry. Anal. Biochem. 1999, 273, 192-203. 18. Tavemer, T.; Hall, N.E.; OHair, R.A.J.; Simpson, R.J. Characterization of an antagonist interleukin-6 dimer by stable isotope labeling, cross-linking, and mass spectrometry. Biol. Chem. 2002, 277, 46487-46492. 94 19. Muller, D.R.; Schindler, P.; Towbin, H.; Wirth, U.; Voshol, H.; Hoving, S.; Steinmetz, M.O. Isotope-tagged cross-linking reagents. A new tool in mass spectrometric protein interaction analysis. Anal. Chem. 2001, 73, 1927-1934. 20. Pearson, K.M.; Pannell, L.K.; Fales, H.M. Intramolecular cross-linking experiments on cytochrome c and ribonuclease A using an isotope multiplet method. Rapid Commun. Mass Spectrom. 2002, 16, 149-159. 21. Seebacher, J .; Mallick, P.; Zhang, N.; Eddes, J .S.; Aebersold, R.; Gelb, M.H. Protein cross-linking analysis using mass spectrometry, isotope-coded cross-linkers, and integrated computational data processing. J. Proteome Res. 2006, 5, 2270-2282. 22. Petrotchenko, E.V.; Olkhovik, V.K.; Borchers, C.H. Coded cleavable cross-linker for studying protein-protein interaction and protein complexes. Mol. Cell Proteomics 2005, 4, 1167-1179. 23. Bennett, K.L.; Kussmann, M.; Bjéirk, P.; Godzwon, M.; Mikkelsen, M.; Sorensen, P.; Roepstorff, P. Chemical cross-linking with thiol-cleavable reagents combined with differential mass spectrometric peptide mapping—A novel approach to assess intermolecular protein contacts. Protein Sci. 2000, 9, 1503-1518. 24. Back, J .W.; Sanz, M.A.; de Jong, L.; de Koning, L.J.; Nijtmans, L.G.J.; de Koster, C.G.; Grivell, L.A.; van der Speck, H.; Muijsers, A0. A structure for the yeast prohibitin complex: Structure prediction and evidence fiom chemical crosslinking and mass spectrometry. Prot. Sci. 2002, 11, 2471-2478. 25. Peterson, J.J.; Young, M.M.; Takemoto, L.J. Probing a-crystallin structure using chemical cross-linkers and mass spectrometry. Molecul. Vision 2004, 10, 857-866. 26. Back, J.W.; Hartog, A.F.; Dekker, H.L.; Muijsers, Anton 0.; de Koning, L.J.; de Jong, Luitzen A new cross-linker for mass spectrometric analysis of the quaternary structure of protein complexes. J. Am. Soc. Mass Spectrom. 2001, 12, 222-227. 27. Soderblom, E.J.; Goshe, M.B. Collision-induced dissociative chemical cross-linking reagents and methodology: Applications to protein structural characterization using tandem mass spectrometry analysis. Anal. Chem. 2006, 78, 8059-8068. 95 28. Tang, X.; Munske, G.R.; Siems, W.F.; Bruce, J .E. Mass spectrometry identifiable cross-linking strategy for studying protein-protein interactions. Anal. Chem. 2005, 77, 311-318. 29. Hoffrnann, E.D.; Stroobant, V. Mass Spectrometry Principles and Applications, 2nd edition, John Wiley and Sons, New York, Page 136. 30. Henzel, W.J.; Billeci, T.M.; Stults, J.T.; Wong, S.C.; Grimley, C.; Watanabe, C. Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases. Proc. Natl. Acad. Sci. U. S. A. 1993, 195, 5011- 5015. 31. Eng, J.K.; McCormack, A.L.; Yates, J.R. An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J. Am. Soc. Mass Spectrom. 1994, 5, 976-989. 32. Mann, M.; Wilm, M. Error-Tolertnt identification of peptides in sequence databases by peptide sequence tags. Anal. Chem. 1994, 66, 4390-4399. 33. Johnson, R.S.; Martin, S.A.; Biemann, K.; Stults, J.T.; Watson, J .T. Novel fragmentation process of peptides by collision-induced decomposition in a tandem mass spectrometer: differentiation of leucine and isoleucine. Anal. Chem. 1987, 59, 2621-2625. 34. Hunt, D.F.; Yates, J .R.; Shabanowitz, J .; Winston, S.; Hauer, C.R. Protein sequencing by tandem mass spectrometry. Proc. Natl. Acad. Sci.U. S. A. 1986, 83, 6233- 6237. 35. Simpson, R.J.; Connolly, L.M.; Eddes, J.S.; Pereira, J .J .; Moritz, R.L.; Reid, G.E. Proteomic analysis of the human colon carcinoma cell line (LIM 1215): Development of a membrane protein database. Electrophoresis 2000, 21, 1707-1732. 36. Chait, B.T.; Kent, S.B.H. Weighing naked proteins: Practical, high-accuracy mass measurement of peptides and proteins. Science 1992, 25 7, 1885-1894. 37. Lee, S.—W.; Berger, S.J.; Martinovic, S.; Pasa-Tolic, L.; Anderson, G.A.; Shen, Y.; Zhao, R.; Smith, R.D. Direct mass spectrometric analysis of intact proteins of the yeast 96 large ribosomal subunit using capillary LC/FTICR. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 5942-5947. 38. Biemann, K.; Papayannopoulos, I.A. Amino acid sequencing of proteins. Acc. Chem. Res. 1994, 27, 370-378. 39. Dongre, R.A.; Jones, J.L.; Somogyi, A.; Wysocki, V.H. Influence of peptide composition, gas-phase basicity, and chemical modification on fragmentation efficiency: evidence for the mobile proton model. J. Am. Chem. Soc. 1996, 118, 8365-8374. 40. Kapp E.A.; Schiitz, F .; Reid, G.E.; Eddes, J .S.; Moritz, R.L.; O’Hair, R.A.J.; Speed, T.P.; Simpson, R.J. Mining a tandem mass spectrometry database to determine the trends and global factors influencing peptide fragmentation. Anal. Chem. 2003, 75, 6251-6264. 41. Reid, G. E.; Roberts, K. D.; Kapp, E. A.; Simpson, R. J. Statistical and mechanistic approaches to understanding the gas-phase fragmentation behavior of methionine sulfoxide containing peptides. J. Proteome Res. 2004, 3, 751-759. 42. Lioe, H.; O’Hair, R.A.J. A novel salt bridge mechanism highlights the need for nonmobile proton conditions to promote disulfide bond cleavage in protonated peptides under low-energy collisional activation. J. Am. Soc. Mass Spectrom. 2007, I8, 1109- 1123. 43. O’Hair, R.A.J. The role of nucleophile-electrophile interactions in the unimolecular and bimolecular gas-phase ion chemistry of peptides and related systems. J. Mass Spectrom. 2000, 35, 1377-1381. 44. Schlosser, A.; Lehmann, W.D. Five-membered ring formation in unimolecular reactions of peptides: a key structural element controlling low-energy collision-induced dissociation of peptides. J. Mass Spectrom. 2000, 35, 1382-1390 45. Roberts, K.D.; Reid, G.E. Leaving group effects on the selectivity of the gas-phase fragmentation reactions of side chain fixed-charge-containing peptide ions. J. Mass Spectrom. 2007, 42, 187-198. 46. Reid, G.E.; Roberts, K.D.; Simpson, R.J., OHair, R.A.J. Selective identification and quantitative analysis of methionine containing peptides by charge derivatization and tandem mass spectrometry. J. Am. Soc. Mass Spectrom. 2005, 16, 1131-1150. 97 47. Amunugama, M.; Roberts, K.D.; Reid, G.E. Mechanisms for the selective gas-phase fragmentation reactions of methionine side chain fixed charge sulfonium ion containing peptides. J. Am. Soc. Mass Spectrom. 2006, I 7, 1631-1642. 48. Sierakowski, J .; Amunugama, M.; Roberts, K.D.; Reid, G.E. Substituent effects on the gas-phase fragmentation reactions of sulfonium ion containing peptides. Rapid Commun. Mass Spectrom. 2007, 21, 1230-1238. 49. Rabinovich, M.S.; Levimov, M.M.; Kulikova, G.N.; Yakushina, L.M.; Verkhovtseva, T.P.; Meller, F.M. An investigation in the field of the synthesis of precursors and fragments of antibiotics VII. Carboxy derivatives of mercaptoacetic acid. Zhurnal Obshchei Khimii 1962, 32, 1167-1172. 50. Anderson, G.W.; Zimmerman, J.B.; Callahan, FM. The use of esters of N- hydroxysuccinimide in peptide synthesis. Metab. Clin. Exp. 1964, 13, 1026-1031. 51. Zhang, J.; Jing, B.; Tokutake, N.; Regen, S.L. Transbilayer complementarity of phospholipids. A look beyond the fluid mosaic model. J. Am. Chem. Soc. 2004 126, 10856-10857. 52. Numata, M.; Koumoto, K.; Mizu, M.; Sakurai, K.; Shinkai, S. Parallel vs. anti- parallel orientation in a curdlan/oligo(dA) complex as estimated by a FRET technique. Org. Biomol. Chem. 2005, 3, 2255-2261. 53. Jessing, M.; Brandt, M.; Jensen, K.J.; Christensen, J .B.; Boas, U. Thiophene backbone amide linkers, a new class of easily prepared and highly acid-labile linkers for solid-phase synthesis. J. Org. Chem. 2006, 71, 6734-6741. 54. an, S., Organic Chemistry of Sulfur. Plenum Press, New York, 1977, page 485. 55. Anderson, G.W.; Zimmerman, J .E.; Callahan, F.M., The use of activated esters in peptides synthesis. Metabol. Clin. Exp. 1964, 13, 1026-1031. 56. Bragg, P.D.; Hou, C., Subunit composition, fimction, and spatial arrangement in the (calcium-magnesium ion)-activated adenosinetiiphosphatases of Escherichia coli and Salmonella typhimurium. Arch. Biochem. Biophys. 1975, 1 6 7, 311-321. 98 “luv-fl 57. Lomant, A.J.; Fairbanks, G., Chemical probes of extended biological structures: synthesis and properties of the cleavable protein cross-linking reagent [35S]dithiobis(succinimidyl propionate). J. Mol. Biol. 1976, 104, 243-261. 58. Pearson, R.G.; Sobel, H.; Songstad, J., Nucleophilic reactivity constants toward methyl iodide and trans-[Pt(py)2C12]. J. Am. Chem. Soc. 1968, 90, 319-326. 59. Tang, X.J.; Thibault, P.; Boyd, R.K. Fragmentation reactions of multiply-protonated peptides and implications for sequencing by tandem mass spectrometry with low-energy collision-induced dissociation. Anal. Chem. 1993, 65, 2824-2834. 99