首页   按字顺浏览 期刊浏览 卷期浏览 Chapter 2. Physical methods and techniques. Part (ii) Mass spectrometry
Chapter 2. Physical methods and techniques. Part (ii) Mass spectrometry

 

作者: M. A. Baldwin,  

 

期刊: Annual Reports Section "B" (Organic Chemistry)  (RSC Available online 1989)
卷期: Volume 86, issue 1  

页码: 19-32

 

ISSN:0069-3030

 

年代: 1989

 

DOI:10.1039/OC9898600019

 

出版商: RSC

 

数据来源: RSC

 

摘要:

2 Physical Methods and Techniques Part (ii) Mass Spectrometry By M. A. BALDWIN Department of Pharmaceutical Chemistry School of Pharmacy University of London 29/39 Brunswick Square London WClN 1AX 1 Introduction It seems appropriate that the third and final biennial review of mass spectrometry from this author covers the literature to the end of the 1980s a decade that has seen a transformation in mass spectrometry. Before 1980 despite a limited number of pioneering studies organic mass spectrometry was largely confined to the analysis of volatile compounds. Today the routine analysis of salts polar drug conjugates and biopolymers such as proteins and glycoproteins oligonucleotides and oligo- and polysaccharides is readily achievable. This has opened up mass spectrometry to the biological and medically related sciences to a degree that would have been inconceivable 10 years ago and I make no apology for biasing this Report heavily towards biochemical analysis.The other aspect of mass spectrometry that has changed dramatically is the diversity complexity and sophistication of the commercial mass spectrometers that are now in widespread use. Quadrupole and sector (magnetic) mass spectrometers have been supplemented by the quadrupole ion trap the Fourier transform mass spectrometer (derived from the previously little used ion cyclotron resonance spec- trometer) and the revitalized time-of-flight instrument. As multidimensional tech- niques have greatly broadened the scope of nuclear magnetic resonance so tandem methods have had a similar impact on mass spectrometry.New desorption modes of ionization that supplement electron impact and chemical ionization which include fast atom bombardment (ca. 8 keV bombardment) caesium ion bombardment (20 keV) synchrotron and fission fragment bombardment (MeV) and laser desorp- tion/multiphoton ionization have been further complemented by several liquid-to- vapour phase spray methods. 2 High Mass Analysis At the time of writing the previous review in this series at the end of 1987,' the concept of high mass in mass spectrometry related to an upper limit of -23 000 Da for proteins by fast atom bombardment (FAB) and -38 000 Da by 252Cf plasma desorption (PD). Within a few months there were conference reports of analyses ' M.A. Baldwin Annu. Rep. hog. Chem. Sect. B Org Chem. 1988 84 15. 19 M. A. Baldwin up to several hundred kDa and there are now two totally unrelated techniques that provide this sort of performance. Both methods are derived from techniques that had been in existence for some considerable time namely laser desorption (LD) and electrospray (ESP). Matrix-assisted UV-laser Desorption/Ionization.-The novel feature that has been introduced into LDMS is to disperse the sample at very low concentration (1 lo4-10’) in a matrix compound that absorbs at the laser frequency initially nicotinic acid absorbing strongly at 255 nm.2 Energy absorbed by the matrix is transferred to the analyte and ions are ejected into the gas phase thus giving laser desorption/ioniz- ation (LDI).The initial publication showed mass analysis at 67 000 Da and this was quickly extended to >250 000 Da.374 The study of such massive ions relies on two features the ability to create ions in the gas phase and the use of a time-of-flight (TOF) analyser which has no theoretical upper mass limit and is ideally suited to the pulsed ionization resulting from the laser pulses. Using 3 keV ion kinetic energy the TOF mass analyser gave a resolving power of only 50 but accuracy of mass measurement was 0.1-0.2%. The sensitivity was very high compared with FAB sub-pmol quantities being sufficient for spectra to be obtained successfully. Usually spectra are accumulated over 20-50 laser shots but adequate signals can be obtained from single shots.It is estimated that a single shot consumes only 10 amol of material. An alternative sample preparation method reported was to disperse the sample in a fine suspension of cobalt powder in a liquid matrix ionization being achieved with a 337 nm laser yielding spectra of proteins up to 25 000 Da with cluster ions up to 100000 Da.5 Further developments of UV-LDI have been made by Beavis and Chait who presented an analysis of factors affecting UV ionization and desorption of proteins including the use of alternative matrix compounds.6 The use of 20 keV ion energy and a reflectron increases the mass resolution and apparently increases the mass accuracy by up to one order of magnitude. Several derivatives of cinnamic acid e.g. ferulic acid (3-methoxy-4-hydroxycinnamic acid) have been used as alternative matrices.An advantage over nicotinic acid is a reduced tendency to form adduct ions by photochemical reactions and the greater mass difference between the protonated ion and the matrix adduct ion (+206 for ferulic acid and +123 for nicotinic acid) makes the separation of these peaks more fea~ible.~ 355 nm laser radiation is just as effective as that of 266nm with these different matrices.8 One of the most encouraging aspects of matrix-assisted UV-LDIMS is the very high sensitivity that is maintained throughout the working mass range,’ unlike FAB secondary ion mass spectrometry from liquids (LSIMS) and PD for which the ion M. Karas and F. Hillenkarnp Anal. Chem. 1988 60,2299. M.Karas U. Bahr and F. Hillenkamp lnt. J. Mass Spectrom. fon Processes 1989 92 231. M. Karas U. Bahr A. Ingendoh and F. Hillenkamp Angew. Chem. Int. Ed. Engl. 1989 28 760. K. Tanaka H. Waki Y. Ido S. Akita Y. Yoshida and T. Yoshida Rapid Commun. Mass Spectrom. 1988 2 151. R. C. Beavis and B. T. Chait Rapid Commun. Mass Spectrom. 1989 3 233. R. C. Beavis and B. T. Chait Rapid Commun. Mass Spectrom. 1989 3 432. R. C. Beavis and B. T. Chait Rapid Commun. Mass Spectrom. 1989 3 436. M. Karas A. Ingendoh U. Bahr and F. Hillenkamp Biomed. Mass Spectrom. 1989 18 841. Physical Methods and Techniques -Part (ii) Mass Spectrometry 21 currents drop very sharply with increasing mass. For peptides- in the range 500-5000 Da the protonated molecular ions decrease by several orders of magnitude for ionization with MeV particles in PDMS and for 8 keV FAB the drop-off is 1-2 orders of magnitude worse." Apparently there is no corresponding mass- related reduction in ionization efficiency for UV-LDIMS although it had been suggested that there could be a problem with very massive ions travelling so slowly and having insufficient momentum to produce secondary electrons at the electron mu1tipIier.l' An example of a protein studied by this method is catalase of measured molecular mass 236 230 Da which consists of four identical sub-units.The most intense signal was obtained for the single sub-unit measured as 59 060 Da. Doubly charged peaks were also observed for both the single sub-unit and the intact protein but it could be concluded that the peak at m/z 59 060 was not a quadruply charged species as there was no evidence of triply charged ions.4 Electrospray and Ionspray.-The last review in this series gave two references to a technique referred to as ionspray (ISP),'2,'3 in the context of HPLC interfacing rather than high mass analysis.In 1988 it became apparent that ISP and the associated method of ESP offered a new method of high mass analysis when Covey et al. demonstrated its use to analyse peptides proteins and olig~nucleotides.'~ A remark- able aspect of this work as applied to high mass analysis is that this performance was achieved with a 'low mass' quadrupole analyser. ESP and ISP both work by spraying a liquid solution of the analyte through a hypodermic needle held at a high potential into a chamber at atmospheric pressure.At low flow rates of -5 pl min-' the spray is nebulized with the formation of charged droplets. The spray is drawn via a pin-hole orifice through evacuated chambers into the mass spectrometer and as the solvent is stripped from the droplets isolated ions are formed. This technique developed originally by Dole and refined by Fenn," is ESP. ISP differs only in that a current of gas is used to assist the nebulization allowing higher liquid flow rates of 100-200 p1 min.-' Related atmos- pheric pressure ionization (API) methods at 1 ml min-' have been described.16 A quadrupole analyser can be used for the analysis of compounds of molecular mass perhaps fifty times greater than the nominal upper mass limit because mass analysers separate ions according to their mass/charge (m/z) ratios.If a protic solvent such as water/methanol/acetic acid is used for the spray as the solvent is stripped from the liquid droplets protons are transferred to basic sites of the analyte molecules forming multiply charged species. At a critical drop size ions can 'evaporate' from the surface and both ESP and IS? have been referred to as field-assisted ion 10 W. Ens D. E. Main K. G. Standing and B. T. Chait Anal. Chem. 1988 60,1494. A. Hedin P. Hakansson and B. U. R. Sundqvist Int. J. Mass Spectrom. Ion Processes 1987 75 275. A. P. Bruins T. R. Covey and J. D. Henion Anal. Chem. 1987 59 2642. l3 T. R. Covey A. P. Bruins and J. D. Henion Org. Mass Spectrom.1988 23 178. l4 T. R. Covey R. F. Bonner B. I. Shushan and J. D. Henion Rapid Commun. Mass Spectrom. 1988 2 249. l5 J. B. Fenn M. Mann C. K. Meng S. F. Wong and C. M. Whitehouse Mass Spectrom. Rev. 1990,9,37. l6 M. Sakairi and H. Kambara Anal. Chem.. 1989 61 1159. 22 M. A. Baldwin evaporation.” Thus a protein of mass 50 kDa which acquires 50 positive charges will pass through the mass analyser with an apparent mass of 1000 Da. Most proteins contain numerous sites that can be protonated a series of protonated species being formed generally having a maximum intensity at m/z -1000. Multiply charged negative ions can also be formed by multiple proton loss from compounds such as oligonucleotides. From the spacing of the multiply charged ion peaks in the mass spectrum the molecular mass of the protein can be determined.I8 The method is equally applicable to low mass ionic compounds and in the negative ion mode has successfully been used to study a series of anionic species such as ADP and ATP introduced as the trisodium salts.” Positive ion API of small molecules is controlled by gas-phase basicities (GB).The sensitivity for basic compounds (GB > 50 kJ mol-’) is relatively independent of compound class as protonation depends upon the kinetics of fast proton transfer from protonated water clusters whereas less basic compounds give weaker signals owing to the reduced stability of the intermediate species BH+(H20)n.20 These spray methods enjoy an advantage over the UV-LDI technique of extreme simplicity in sample introduction and the mode of ionization and perhaps greater accuracy of mass measurement.The multiple peaks allow a series of mass measure- ments to be made thereby improving the statistics of the final result. Personal experience by this reviewer has shown the mass accuracy for 10-20 kDa proteins to be within 1 Da. Limits of detection may be sub-pmol although typical sample loadings are 10-100pmol. The methods are also directly compatible with HPLC allowing direct introduction of the eluent into the liquid stream and they may offer unique advantages for peptide sequencing by tandem mass spectrometry. Both these aspects will be elaborated below. On the other hand UV-LDI appears to offer extreme sensitivity for both positive and negative ions,21 which could be important for the analysis of compounds such as oligonucleotides oligo- saccharides and phosphorylated sulphated or glycosylated proteins.It will be fascinating to see how these two very different methodologies develop and become established. From the commercial point of view there is clearly a strong bio- chemical market for a simple ‘molecular mass machine’ that can be operated by non-specialists as has been shown by the success of PDMS with a simple TOF analyser. 3 Other Soft Ionization Techniques Although the methods referred to in the previous section are begining to dominate the high mass area the other soft ionization techniques such as chemical ionization (CI) and the various desorption methods (FAB PD and LD) continue to have an essential role to play.” T. R. Covey B. I. Shushan B. A. Thomson and J. D. Henion Proceedings of the 37th ASMS Annual Conference on Mass Spectrometry and Allied Topics Miami Florida 1989 p. 558. 18 M. Mann C. K. Meng and J. B. Fenn Anal. Chem. 1989,61 1702. 19 K. W. M. Siu G. J. Gardner and S. S. Berman Org. Mass Spectrom. 1989 24 931. 20 J. Sunner G. Nicol and P. Kebarle Anal. Chem. 1988 60,1300. M. Salehpour I. Perera J. Kjellberg A. Hedin M. A. Islamain P. Hakansson and B. U. R. Sundqvist Rapid Commun. Mass Spectrom. 1989 3 259. Physical Methods and Techniques -Part (ii) Mass Spectrometry 23 Chemical Ionization.-An advantage of CI is the ability to match the reagent gas to the analyte under study thereby optimizing the analysis.Bartmess has presented a very comprehensive compilation of gas-phase thermochemical data such as basicities acidities and electron hydride and fluoride affinities.22 Electron capture negative ion chemical ionization (NICI) is a very sensitive and selective technique which has been widely used for the analysis of compounds such as biogenic amines in body fluids and tissues. Other compound classes less sensitive to NICI may be derivatized with a group that enhances electron capture e.g. the sensitivity for corticosteroids was increased tenfold by forming pentafluorobenzyloxime-trimethyl-silyl (TMS) derivatives rather than methoxime-TMS derivative^.^^ It is worth noting that derivatization is still an important component in many analyses even though desorption methods have largely removed the need for sample volatility.Anderegg has reviewed derivatization and has shown how it can enhance the molecular ion or fragment ions reduce the role of strongly directing groups be used for isomer differentiation and increase the information content of collision-induced Electron impact (EI) spectra are broadly reproducible and library searching is generally reliable for compound identification provided that an appropriate refer- ence spectrum is available whereas CI is perceived to be more variable. The reproducibility of NICI was tested for twelve compounds by six different laboratories. There was good agreement even between quadrupole and sector instruments differences arising through dissociative reactions and formation of ions such as (M + H -C1)+ from polychlorinated compounds e.g.end~sulphan.~~ Posi-tive ion CI (PICI) is generally best for generating high-intensity quasimolecular ions from a diverse range of samples whereas NICI is much more compound-specific and frequently gives high-intensity fragment ions associated with particular function- alities. Pulsed PICI/ NICI allows the optimum detection of specific compound classes in complex mixtures and has been used for the assay of organophosphorus pesticide residues in food.26 CI lends itself to interfacing with supercritical fluid chromatography (SFC) as the fluid used in the SFC separation can also act as the reagent gas. Ammonia CI has been employed for the SFC-MS analysis of polymers and derivatized oligosa~charides.~~ SFC- MS with helium charge exchange (CE) gives EI-like spectra which is valuable for interpretation as has been shown for a series of drugs and metabolites.28 CI can be viewed as part of the broader area of study of ion-molecule reactions developments having been encouraged by the increasing use of ion trap (IT) Fourier transform (FT) and tandem mass spectrometers.One novel variation of CI is to 22 J. E. Bartmess Mass Spectrom. Rev. 1989 8 297. 23 J. M. Midgley D. G. Watson T. Healey and C. N. J. McGhee Biomed. Environ. Mass Spectrom. 1989 18 651. 24 R. J. Anderegg Mass Spectrom. Rev. 1988 7 395. 25 E. A. Stemmler R. A. Hites B. Arbogast W. L. Budde M. L. Deinzer R. C. Dougherty J. W. Eichelberger R.L. Foltz C. Grimm E. P. Grimsrud S. Sakashita and L. J. Sears Anal. Chem. 1988 60 781. 26 H.-J. Stan and G. Kellner Biomed. Environ. Mass Spectrom. 1989 18 645. 27 X. X. Pinkston G. D. Owens L. J. Burkes T. E. Delaney D. S. Millington and D. A. Maltby Anal. Chem. 1988 60 962. 28 E. D. Lee S.-H. Hsu and J. D. Henion Anal. Chem. 1988 60,1990. 24 M. A. Baldwin select the reagent ion and collide it with the analyte in the collison cell of a triple quadrupole. Linked to GC this technique has been shown to give picogram sensi- tivity with a very high degree of specificity. An example is the CE ionization of 2,4-dimethylaniline for which Art' transfers sufficient energy to fragment the analyte ions whereas benzene+' transfers less energy and tends to give molecular ions only allowing the (CE) process to be separated from competing proton transfers and hydride abstraction reactions that occur in the conventional CI ion source.29 The dual cell FTmass spectrometer similarly prevents reactions between the analyte and the neutral reagent gas.30 Also using FTMS it has been shown that Fe+ forms bridged adducts with disubstituted benzenes that allow clear identification of the ortho isomers although the meta and para isomers were harder to distinguish.In a study of several such isomers xylene provided the only example that could not be identified.31 Many groups have investigated low-energy endothermic collisons between mass-selected ions and ammonia or amines. Protonated peptide ions for example form proton-bound adducts with ammonia that either lead to peptide fragment ions or eliminate the ammonium ion with formation of a neutralized peptide-potentially interesting species for study by neutralization/reionization mass spe~trometry.~~ Fast Atom Bombardment.-The mechanism of ionization in FAB and LSIMS is still not clear but Sunner has presented more evidence in support of the phase explosion model based on the concept of CI-like processes in the relatively high-pressure fluid at the interface between the liquid and the vacuum.The ionization can be modelled by classical gas-phase thermodynamic^.^^ Chemical reactions that occur in the liquid phase have been reviewed by Cooks and co-~orkers,~~ who have also shown that desorption of zwitterions induces intermolecular reactions that reduce the overall number of charges desorbed.This effect is reduced in proton-rich solvents.35 Reactions between the matrix and the analyte are a particular problem e.g. rn-nitrobenzyl alcohol can give a covalently bonded adduct of +133 Da with analytes containing amine functions such as peptides. This arises from oxidation of the alcohol to an aldehyde followed by Schiff base formation to give an imine.36 The nature of the matrix is important and ideally should be matched to the physical properties of the analyte in order to reduce discrimination e.g. dodecanol as a matrix for fatty acids reduces surface activity and micelle formation and gives a more linear response with varying c~ncentration.~~ For detection at the trace level 29 M.E. Hail D. W. Berberich and R. A. Yost Anal. Chem. 1989 61 1874. 30 R. B. Cody Anal. Chem. 1989 61 2511. 31 A. Bjarnason J. W. Taylor J. A. Kinsinger R. B. Cody and D. A. Weil Anal. Chem. 1989 61 1889. 32 R. Orlando C. Murphy C. Fenselau G. Hansen and R. J. Cotter Anal. Chem. 1990 62 125. 33 J. Sunner A. Morales and P. Kebarle Int. J. Muss Spectrom. Ion Processes 1988 86 169. 34 L. D. Detter 0. W. Hand R. G. Cooks and R. A. Walton Muss Spectrom. Rev. 1988 7 465. 35 0. W. Hand and R. G. Cooks Int. J. Mass Spectrom. Ion Processes 1989 88 113. 36 M. Barber D. J. Bell M. Morris L. W. Tetler M. D. Woods J. J. Monaghan and W. E. Morden Rapid Commun.Mass Spectrom. 1988 2 181. 37 K. A. Caldwell and M.L. Gross Anal. Chem. 1989 61 494. Physical Methods and Techniques -Part (ii) Mass Spectrometry 25 it may be necessary to increase surface activity at the expense of linearity e.g. formation of the more hydrophobic hexyl esters of peptides is significantly more effective than ethyl ester formation for detection at the pmol level.38 The time limit for an analysis as determined by the rate of evaporation of the liquid can be extended by pulsing the bombarding atom or ion beam which also allows the surface to be replenished between shots and increases the signal It has been reported that the nature of the support surface has a significant effect on the ion currents obtained. Nitrocellulose enhances the signal even using a liquid matrix.41 Organics can be ionized from a dry surface by SIMS but they are generally short-lived unless the flux of bombarding species is kept very low.Fragmentation has been found to decrease with increasing primary ion energy owing to a cubic relationship between the impact energy and the number of energized molecules; thus at higher impact energy the average internal energy per molecule is less.42 Continuous flow fast atom bombardment (CF-FAB) was described in the previous review in this series. A major reason for interest in this technique relates to the ability to interface it with HPLC etc. but it also offers advantages in its own right as a constantly replenished sample is available a wider range of solvents can be employed and there is less selectivity and suppression in the analysis of mixtures.Interference in the low mass region arising from matrix and other cluster ions is reduced allowing the analysis of sub-ng quantities of compounds in the 100-200 Da mass range. Sample recoveries were aided by the addition of a structural analogue or homologue as sample adsorption was reduced.43 The analysis of drugs of molecular mass < 500 Da has been described with cocaine being detected from the direct analysis of The measurement of accurate masses is achieved more readily by CF-FAB than by static FAB as the sample and reference materials can be injected sequentially. Even with a low resolution quadrupole analyser it proved possible to measure precise masses for the naturally occuring anticancer alkaloids vincristine and vinblastine with an accuracy of 6 p.p.m.for the base peak (M + H)+ at mlz 825.45 Plasma Desorption.-There is a perception that PD is essentially a high-mass tech- nique so it will be interesting to monitor its future progress in the face of competition from ESP and UV-LID. In reviewing PD-TOFMS Cotter has referred to its ‘coming of age’ and has shown that by comparison with other methods it offers impressive performance at a modest price. He describes the technique and the theory and chooses examples from peptides proteins glycolipids phospholipids and analyses 3n A. M. Falick and D. A. Maltby Anal. Biochem. 1989 182 165. 39 D. F. Hunt J. Shabanowitz J. R. Yates 111 N. Z. Zhu D. H. Russell and M. E. Castro Proc. Narl. Acad. Sci. USA 1987 84 620.40 R. E. Tecklenberg jun. M. E. Castro and D. H. Russell Anal. Chem. 1989 61 153. 41 R. B. van Breeman and J. C. Le Rapid Commun. Mass Spectrom. 1989 3 20. 42 B. E. Winger 0. W. Hand and R. G. Cooks Int. J. Mass Specrr$m. Ion Processes 1988 84 89. 43 T.-C. L. Wang M.-C. Shih S. P. Markey and M. W. Duncan Anal. Chem. 1989 61 1013. 44 W. E. Seifert jun. A. Ballatore and R. M. Caprioli Rapid Commun. Mass Spectrom. 1989 3 117. 45 J. Roboz J. F. Holland M. A. McDowall and M. J. Hillmer Rapid Commun. Mass Specrrom. 1988 2 64. 26 M. A. Baldwin from whole cells to demonstrate its power and ~ersatility.~~ PD-TOFMS is still being used effectively for peptide and protein analysis:’ but perhaps more impor- tantly in terms of competing with ESP it has also been shown to be successful for the analysis of lipo-oligosaccharides glycopeptidolipids and phenolic glycolipids from mycobacterial sources.48 Laser Desorption.-Lasers are used both for desorption and ionization sometimes simultaneously as in the matrix-assisted LDI experiments described above but not necessarily with a matrix and sometimes in two isolated steps with lasers operating at different wavelengths.Desorption without ionization is generally achieved with an IR laser although this will also sputter ions from appropriate samples. A UV laser is required for ionization in the gas phase. There have been two independent reviews of gas-phase multiphoton ionization (MUPI) by Grotemeyer and S~hlagg~~ and by L~bman.’~,’’ The number of commercial laser instruments in use is still relatively low but some very impressive results suggest that they will take an increasing role in routine mass spectrometric analysis.LDI from thin layer chromatography (TLC) plates has proved to be sensitive giving good spatial resolution. Although TLC-FABMS equipment is available com- mercially the focusing ability of the laser offers advantages over FAB particularly for incompletely resolved components on the TLC plate e.g. adenine and g~anine.’~ IR-LDI-FTMS was compared with FAB for the analysis of mixtures of amino acids and small peptides. LDI showed less selectivity all fifteen amino acids in a mixture being ionized to give (M + 2Na -H)+ ions but only ten being detected by conven- tional static FAB,53 although CF-FAB might have been more successful in this respect.There has also been a comparison of LDI with PD and FAB for the analysis of carbohydrates LDI offering significant advantages in giving fragment ions and thereby structural information which could only be obtained by derivatization or tandem methods using the other desorption methods.54 LDI can be carried out directly within an IT or FT mass spectrometer as has been shown for peptides and disaccharides in ITMS.” LDI from dry solids without a liquid matrix enjoys the advantage of being compatible with high-vacuum operation which is particularly important for ITMS. UV-LDI-ITMS of polymers was found to be superior to FAB or SIMS giving more regular distributions less fragmentation and less discrimina- tion for a series of alkoxylated pyrazoles and hydrazine~.~~ A second laser can be used for photodissociation of the ions formed by LDI e.g.IR-LDI-FTMS of 46 R. J. Cotter Anal. Chem. 1988 60,781A. 47 I. Jardine G. F. Scanlan A. Tsarbopoulos and D. J. Liberato Anal. Chem. 1988 60 1086. 48 I. Jardine G. Scanlan M. McNeil and P. J. Brennan Anal. Chem. 1989 61 416. 49 J. Grotemeyer and E. W. Schlagg Angew. Chem. Int. Ed. Engl. 1988 27 447. 50 D. M. Lubman Mass Spectrom. Rev. 1988 7 535. 51 D. M. Lubman Mass Spectrom. Rev. 1988 7 559. 52 A. J. Kubis K. V. Somayajula A. G. Sharkey and D. M. Hercules Anal. Chem. 1989 61 2516. 53 M. P. Chiarelli and M. L. Gross Anal. Chem. 1989 61 1895. 54 W. B. Martin L.Silly C. M. Murphy T. J. Raley jun. R. J. Cotter and M. F. Bean Int. J. Mass Spectrom. Ion Processes 1989 92 243. 55 D. N. Heller I. Lys R. J. Cotter and 0. M. Uy Anal. Chem. 1989 61 1083. 56 L. M. Nuwaysir and C. L. Wilkins Anal. Chem. 1988 60 279. Physical Methods and Techniques -Part (ii) Mass Spectrometry 27 tetrakis-(4-fluorophenyl)porphyrininvolved attachment of iron and UV-laser dis- sociation gave losses of the fluorphenyl groups not observed in the conventional mass spectrum.57 4 Liquid Chromatography and Electrophoresis Mass spectrometry is generally most successful when applied to the analysis of pure compounds rather than complex mixtures hence the emphasis on interfacing with chromatographic methods. There is continuing interest in SFC-MS for compounds of low volatility a system having been described for converting a GC-MS instrument into dual GC-SFC operation with simple modification^.^' The thermospray (TSP) interface has become the standard for conventional bore reverse-phase HPLC at flow rates of -1 ml min.-' 4study of drugs antibiotics and peptides was carried out in deuterated solvents showing characteristic mass shifts which revealed the number of acidic hydrogens on various f~nctionalities.~~ A similar study applied to nucleosides in D20 showed 97% exchange of acidic hydrogens.60 The optimum conditions for the chromatographic separation may not coincide with optimum TSP performance but this may be compensated for by post-column addition of solvent.Glutathione conjugates were shown to need a 0-20% acetonitrile gradient for separation but TSP required 30-60% acetonitrile.100% acetonitrile at 0.4ml min-' was added to the column eluent flowipg at 0.8 ml min,-' thereby optimizing both elements in the analysk6' TSP can also be used for monitoring reaction mixtures without chromatographic separation. The electrochemical oxidation pathways for uric acid were revealed by direct monitoring of the electrochemical and the kinetics of a Michael condensation between glutathione and the neurotoxin acry- lamide have been in~estigated.~~ A major demand in biochemical analysis is the separation detection and charac- terization of ever smaller quantities of compounds in complex mixtures. To this end there has been a move towards microbore HPLC techniques and the adoption of capillary zone electrophoresis (CZE),64 which offers very high resolution on very small quantities of charged species.CF-FAB and ESP are two alternative ionization modes for the continuous introduction and ionization of polar compounds in liquid solutions at flow rates in the pl min-' range. Microbore HPLC can be coupled with either technique with relative ease as the column outlet can be fed via a capillary directly to the end of the FAB target or directly to the hypodermic needle of the ESP. However neither ionization technique works effectively with all HPLC solvents and as with TSP it can be necessary to make up the flow with another liquid. One approach is to use coaxial tubes the column eluent flowing through the 57 L.M. Nuwaysir and C. L. Wilkins Anal. Chem. 1989 61 689. 58 H. T. Kalinoski and L. 0. Hargiss J. Chromatogr. 1989 474 69. 59 M. M. Siegel Anal. Chem. 1988 60 2090. 60 C. G. Edrnonds S. C. Pornerantz F. F. Hsu and J. A. McCloskey Anal. Chem. 1988 60 2314. 6' M. F. Bean S. L. Pallente-Morell D. M. Dulik and C. Fenselau Anal. Chem. 1990 62 121. 62 K. J. Volk R. A. Yost and A. Brajter-Toth Anal. Chem. 1989 61 1709. 63 M. E. Harrison and M. A. Baldwin Org. Mass Specirom. 1989 24 689. 64 A. G. Ewing R. A. Wallingford and T. M. Olefirowicz Anal. Chem. 1989 61 292A. 28 M. A. Baldwin inner tube and the additional liquid through the outer tube the two mixing on the FAB target. Using both open tubular and packed microcolumns at flow rates of <lo0 nl-min-' with the addition of 25% glycerol/75% water at 200 nl-min,-' single scan spectra can be obtained at the low fmol level and signal averaging brings this down to 500 amol.Examples quoted include peptides phospho- lipids oligosaccharides and unsaturated fatty acids6' A quantitative study of peptide neurotransmitters assayed in spinal cord by HPLC-CF-FAB gave results consistent with radioimmunoassay.66 In addition to reverse-phase chromato-graphy CF-FAB can be coupled with ion exchange ion pair and ion exclusion ~hromatography.~~ For interfacing CZE with mass spectrometry an additional problem is the preven- tion of flow of the separating medium through the capillary which occurs with direct coupling to the vacuum system with the inherent danger of laminar flow degrading the resolution (although this is minimized with the very low flow rates reported for the CF-FAB co-axial interface).68 In an alternative co-axial interface described by Smith et al.for ESP operation the inner tube which is the electrophoresis capillary does not enter the vacuum. Its low voltage end terminates close to a restriction at the end of the outer tube from which the spray is formed. This interface has a dead volume of only 10nl corresponding to <1 s at the flow rates employed for The same experimental arrangement was also shown to be compatible with isotachophoresis for which lo-'' M solutions were analysed with amol limits of dete~tion.~' This sensitivity for CZE is very similar to that obtained with 32P radiochemical detection72 and electrochemical detection.73 An alternative is a liquid- liquid junction interface in which the outlet of the electrophoresis column is held very close to the end of another capillary that takes the liquid flow to the mass spectrometer ionization region.The junction is immersed in the make-up liquid which is drawn through by the vacuum. Excellent electrophoretic separation has been obtained on a mixture of peptides from a tryptic digest of bovine somatotrophin fragment ion information on individual components being obtained from the triple quadrupole mass analy~er.~~ The resolution reported was greatly superior to that shown in the HPLC-CF-FAB separation of a digest from ribonuclease B although the use of single ion monitoring in the latter effectively enhanced the resolution ~onsiderably.~~ 65 L.J. Deterding M. A. Mosely K. B. Tomer and J. W. Jorgenson Anal. Chem. 1989 61 2504. 66 C. A. Lisek J. E. Bailey L. M. Benson T. L. Yaksh and I. Jardine Rapid Cornrnun. Mass Spectrom. 1989 3 43. 67 A. S. Al-Omair and J. F. J. Todd Rapid Commun. Mass Spectrom. 1989 3 405. 68 M. A. Moseby L. J. Deterding K. B. Tomer and J. W. Jorgenson Rapid Commun. Mass Spectrom. 1989 3 87. 69 R. D. Smith C. J. Baringa and H. R. Udseth Anal. Chem. 1988 60,1948. 70 C. G. Edrnonds J. A. Loo C. J. Baringa H. R. Udseth and R. D. Smith J. Chrornatogr. 1989,474,21. 71 H. R. Udseth J. A. Loo and R. D. Smith Anal. Chem. 1989 61 228. 12 S. L. Pentony jun.R. N. Zare and J. F. Quint Anal. Chem. 1989 61 1642. 73 R. A. Wallingford and A. G. Ewing Anal. Chem. 1989 61 98. 74 E. D. Lee W. Muck J. D. Henion and T. R. Covey Biorned. Environ. Mass Specrrom. 1989 18 844. 75 K. Mock J. Firth and J. S. Cottrell Org. Mass Spectrom. 1989 24 591. Physical Methods and Techniques -Part (ii) Mass Spectrometry 29 5 Tandem Mass Spectrometry By the end of the 1980s tandem mass spectrometers had made the transition from home-made research instruments in the laboratories of a few pioneers in the 1970s to become an established part of the analytical scene. This has been spurred on by soft ionization methods which generally give protonated molecular ions that are relatively stable and give little structural information unless fragmentation can be induced by the deposition of more internal energy i.e.by photoexcitation or more commonly by collisions with gas atoms or molecules. (Cooks and co-workers have also carried out a series of studies on collisions with a solid surface e.g. for more than seventy odd-electron organic ions.76) There is a hierarchy of instrument types based on cost starting with triple quadrupoles (QQQ) in which the initial and final quadrupoles are used for mass analysis with collisions taking place in the radio- frequency-only central quadrupole through the hybrids in which a double focusing two-sector instrument is followed by two quadrupoles (SS/QQ) finally to the double focusing/ double focusing four-sector (SS/ SS) high performance instruments.Inevitably the low-cost QQQ instruments are by far the most widely used whereas SS/SS instruments at about US $1000 000 each are relatively few and far between. Tandem experiments are also possible by trapping in instruments such as the IT and FT mass spectrometers. This diversity of instruments and their applications have been re~iewed.~' A quadrupole is a low resolution mass analyser compared with a double focusing spectrometer and is unable to separate isobaric ions of the same nominal mass whereas SS/SS combinations can provide high-resolution analysis of both parent and daughter ions. This is also true for FTMS which in a dual cell instrument with a pulsed gas valve with selective daughter ion excitation gave a mass resolution of 500 000 for daughter ions and a mass measurement error of <5 ~.p.m.~~ The ion kinetic energy in a quadrupole is 10-100 eV whereas sectors routinely operate at -lo4eV so the kinetic energy transferred into internal energy is substantially less for QQQ and SS/QQ (-5 eV) than for SS/SS (-30 eV).79 For peptide sequencing the backbone fragmentations can be monitored by any of these instrument types but the analytically useful high-energy fragmentations of the amino acid side-chains which for example allow differentiation of the isomers leucine and isoleucine are mostly only observed in SS/ SS experiments." High-energy collisons yield tandem mass spectra of peptides that are less dependent on amino acid type and although QQQ and SS/QQ combinations are successful for most small peptides and have been used in the 2000 Da region," many peptides of masses above -600 Da give little useful data compared with SS/ SS spectrometer^.^^.^^ One advantage of the 76 T.Ast M. A. Mabud and R. G. Cooks Int J. Mass Specfrom. Ion Processes 1988 82 131. 77 K. L. Busch G. L. Glish and S. A. McLuckey 'Mass Spectrometry/Mass Spectrometry. Techniques and Applications of Tandem Mass Spectrometry' VCH New York 1988. 78 R. B. Cody Anal. Chem. 1988 60,917. 19 L. Poulter and L. C. E. Taylor Znf. J. Mass Specfrom. Ion Processes 1989 91 183. 80 R. S. Johnson S. A. Martin and K. Biemann Inf. J. Mass Spectrom. Zon Processes 1988 86 137. *' D. F. Hunt J. R. Yates 111 J. Shabanowitz S. Winston and C. R. Hauer Roc. Nafl. Acad. Sci.USA 1986 83 6233. 82 A. J. Alexander and R. K. Boyd Int. J. Mass Spectrom. Ion Processes 1989 90 211. 30 M.A. Baldwin quadrupole mass analyser is its much greater acceptance angle giving higher sensitivity for QQQ or SS/QQ as ions are less likely to be scattered out of the beam.83 The sensitivity of high-performance SS/SS instruments can be enhanced by several orders of magnitude (but at considerable expense and with lowered resolving power) by an array detector that allows anything up to 50% of the mass range to be monitored sim~ltaneously.~~ For the sequencing of peptides tandem mass spectrometry is particularly useful for identifying modified amino acids either from a post-translational event or perhaps from xenobiotic modification.The addition of styrene oxide to haemoglobin was identified by the presence of a tryptic peptide 120 Da too high the site of modification being located at histidine by the identification of a so-called immonium ion which was shifted from 110 to 230 Da and the corresponding 120 Da shifts in the C-terminal ion series8’ Methods are being developed for computer interpreta- tion of peptide tandem mass spectra which offer considerable promise for fast and reliable sequencing.86 There are reported advantages in obtaining tandem spectra of alkali metal ion adducts rather than protonated species. FAB shows sodium binding to the site of highest alkali ion affinit~,~’ and an elimination from the C-terminal residue allows identification of the C-terminal amino acid.88 A very promising development is the use of ESP/QQQ studies on doubly charged ions.Tryptic peptides tend to be protonated on both N-and C-termini cleavage into two singly charged ions giving two overlaid fragment series. The high sensitivity and efficiency of ionization by ESP gives strong signals for quite large peptides such as a 19-residue example of mass 2314Da giving a doubly protonated ion at m/z 1158.89 The fragmentation into singly charged species causes half of the fragment ions to have higher m/z values than the doubly charged molecular species putting more demands on the mass range of the final quadrupole analyser. Nevertheless the fragmentation of melittin could be rationalized up to m/z 2800.90 Although tandem mass spectrometry is particularly successful for peptide sequenc- ing many other classes of biomolecules have been studied as has been reviewed by T~mer.~~.~~ Stereochemical effects are often seen only in tandem experiments in which ions of low internal energy can be sampled.Zaretskii has published a series of papers on steroid stereochemistry. A recent study involved the selection of the ions formed by the unimolecular or metastable loss of water and subjecting these to collisional dissociation (MS/MS/MS) to distinguish between epimers of the 83 S. J. Gaskell and M. H. Reilly Rapid Commun. Mass Spectrorn. 1988 2 139. 84 K. Biemann and J. A. Hill Proceedings of the 37th ASMS Conference on Mass Spectrometry and Allied Topics Miami Florida 1989 p. 27. 85 S. Kaur D.Hollander R. Haas and A. L. Burlingame J. Biol. Chem. 1989 264 16981. 86 R. S. Johnson and K. Biemann Biomed. Environ. Mass Spectrom. 1989 18 945. 87 D. H. Russell E. S. McGlohon and L. M. Mallis Anal Chem. 1988 60,1818. 88 W. Kulik W. Heerma and J. K. Terlouw Rapid Cornmun. Mass Spectrom. 1989 3 276. 89 D. F. Hunt N.-Z. Zhu ,and J. Shabanowitz Rapid Commun. Mass Spectrom. 1989 3 122. 90 C. J. Baringa C. G. Edmonds H. R. Udseth and R. D. Smith Rapid Commun. Mass Spectrom. 1989 3 160. 9’ K. B. Tomer Mass Spectrom. Rev. 1989 8 445. 92 K. B. Tomer Mass Spectrom. Rev. 1989 8 483 Physical Methods and Techniques -Part (ii) Mass Spectrometry 31 3-hydroxy sterols.93 It has been shown that cis/trans ring junctions at the B/C rings in morphinans can be distinguished by characteristic ions in the tandem mass spectra.94 Determining stereochemistry linkage position and branching in sac-charides and oligosaccharides has proved difficult or impossible by conventional mass spectrometry but the various stereoisomers of per-acetylated glucose and galactose have been identified by tandem mass ~pectrometry,'~ and the stereoiso- meric 2-deoxy-2-aminohexoses have been di~tinguished.~~ On the basis of peak intensity ratios it was possible to characterize 1,2- 1,3- 1,4- and 1,6-linked disaccharides using negative ion FAB/MS/ MS; peaks which were strong in some cases being completely absent in others9' Laine et al.combined theoretical calculations with tandem methods to study linkage position in trisa~charides.~~ Tandem mass spectrometry was used to correct an earlier structure for high- mannose sugars isolated from Saccharomyces cerevisiae in which the branching had been wrongly interpreted by chemical and NMR methods.It is noteworthy that the revised structure would not be distinguishable by 1H NMR from that proposed rigi in ally.^^ Unusual Chemical Species.-Mass spectrometry can often yield information on unusual chemical species. In recent years there have been many examples identified of 'distonic' ions ie. odd-electron ions in which the charge and radical sites are separated corresponding formally to the removal of an electron from a zwitterion. In reviewing these species Hammerum has pointed out that they are both 'common and stable' existing in condensed phase both solid and solution often as intermedi- ates in organic synthesis as well as the isolated gas phase prevailing in the mass spectrometer."' As explained in the previous review in this series the 'existence and stability of some of the corresponding hypervalent or non-classical neutrals can be investigated by neutralization/reionization mass spectrometry (NRMS) in which the ions are collisonally neutralized and then reionized some 4 x lop7 s later.'" NRMS has also been used to demonstrate the stability of some unusual radical species.Protonation of some aromatics showed species such as the benzenium radical C6H7',an intermediate in the Birch reduction to be stable.'02 Ligand-deficient electronically unsaturated half-sandwich complexes such as FeC,H were also shown to exist as stable neutrals within the lifetime of the experiment.The abundance of 93 Z. V. I. Zaretskii J. M. Curtis D. Ghosh and A. G. Brenton Znt. J. Muss Spectrom. Ion Processes 1988 86 121. 94 B. Charles and J.-C. Tabet Rapid Commun.Mass Spectrom. 1988 2 86. 95 R. Guevremont and J. L. C. Wright Rapid Commun.Muss Spectrom. 1988 2 47. 96 R. Guevremont and J. L. C. Wright Rapid Commun. Muss Spectrom. 1988 2 50. 97 A. Ballistreri G. Montaudo D. Garozzo M. Giuffrida and G. Impallomeni Rapid Commun.Mass Spectrom. 1989 3 302. 98 R. A. Laine K. M. Panichi A. D. French R. W. Hall S. A. Abbas R. K. Jain and K. L. Matta J. Am. Chem. SOC.,1988 110 6931. 99 L. M. Hernandez L. Ballou E.Alvarado B. L. Gillece-Castro A. L. Burlingame and C. E. Ballou J. Biol. Chem. 1989 264 11 849. 100 S. Hammerum Muss Specfrom. Rev. 1988 7 123. 101 J. L. Holmes Muss Spectrom. Rev. 1989 8 513. 102 A. W. McMahon F. Chadikun A. G. Harrison and R. E. March Int. J. Muss Spectrom. Ion Processes 1989 87 275. M. A. Baldwin survivor ions was in the order FeCp < CoCp < NiCp the reverse of the ther- mochemical stability. The breakdown products identified in the reionization spec- trum included species such as FeC3H3+ as well as abundant hydrocarbon ions.lo3 The interpretation of such experiments is not always trivial and earlier claims concerning the stability of some hypervalent species have now been dismissed for H30',lo4 and also for CH5' which can only be formed in high Rydberg states followed by rapid disso~iation.'~~ 103 T.Drewello and H. Schwarz Int. J. Muss Spectrom. Ion Processes 1989 93 177. 104 R.E. March and A. 9. Young Int. J. Muss Spectrom. Ion Processes 1988 85 237. 105 J. Bordas-Nagy J. L. Holmes and C. E. C. A. Hop Inr,J. MQSSSpectrom. Ion Processes 1988,85 241.

 



返 回