Real Estate

A matrix-assisted laser desorption/ionization mass spectrometry approach to the lipid A fromMesorhizobium loti

Description
A matrix-assisted laser desorption/ionization mass spectrometry approach to the lipid A fromMesorhizobium loti
Categories
Published
of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  RAPID COMMUNICATIONS IN MASS SPECTROMETRY Rapid Commun. Mass Spectrom.  2005;  19 : 1725–1729Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/rcm.1982 Matrix-assisted laser desorption/ionization massspectrometry of collected bioaerosol particles Jae-Kuk Kim, Shelley N. Jackson and Kermit K. Murray* Department of Chemistry, Louisiana State University, Baton Rouge, LA 70803, USA Received 12 November 2004; Revised 26 April 2005; Accepted 26 April 2005 A method was developed for collection and analysis of bioaerosols by matrix-assisted laser desorption/ionization (MALDI) time-of-flight mass spectrometry using a modified Andersen N6bioaerosol collector. The overall goal of the study was to develop methods for obtaining mass spec-tra with minimal reagents and treatment steps for potential use in remote collection and analysissystems. Test bioaerosol particles were generated from a nebulized  E. coli  bacterial suspension andcollected on MALDI targets placed in an Andersen N6 single-stage aerosol impactor. The bioaer-osols were mixed with matrix either by deposition on a bare target with the matrix solution addedlater, or by deposition on a target pre-coated with matrix. The matrix compounds  a -cyano-4-hydroxycinnamic acid (CHCA) and sinapic acid (SA) were tested and the SA matrix was foundto give the best results in number of peaks, resolution, and signal-to-noise ratio. Deposition ofbioaerosol particles onto the matrix pre-coated target did not produce signal in the  m/z  region above1000, but the signal could be recovered with the addition of a 1:1 (v/v) acetonitrile/water solvent.Addition of solvent by pipette to the pre-coated targets after particle deposition recovered signalcomparable to the dried-droplet sample preparations, whereas solvent sprayed into the impactor recovered fewer peaks. Deposition on pre-coated targets with post-collection solvent additionwas superior to deposition on bare target followed by post-collection addition of matrix solution.Copyright # 2005 John Wiley & Sons, Ltd. Biologicalagentsarelivingorganismsortoxinsderivedfromliving organisms that can be used as weapons of combat orterror. 1,2 Theseagentsaresimpleandinexpensivetoproduceand can be manufactured in small laboratories that can beconcealed, transported, or disguised as pharmaceutical orfood production facilities. Dispersal of biological agents istypically through an aerosol spray, but sabotage of food,water, or public spaces by dispersal of material is also possi- ble. Because biological agents are not detectable with thenaked eye or by smell and because their effects are notobservedforhoursordaysafterexposure,theyarepsycholo-gically threatening and thus particularly effective weaponsof terror.The development of techniques for the detection of  biologicalagentsischallenging.Thesensitivityofabiologicalagent detector must be high because of their high toxicityrelative to their mass. In addition, the detector must beselective and able to discriminate against an ambient back-ground that includes a high level of naturally occurringparticulate. Development of analytical instrumentation for biological agent detection has followed two general paths: 3 the adaptation of existing analytical instruments, and thedevelopmentofaffinitysensorsthatarebasedonbiochemicalselectivity. 4,5 In both cases, particles are typically collectedprior to analysis. Affinity-based sensors are small, fast,inexpensive and highly specific, but this specificity limitstheir applicability to systems for which affinity targets have been developed. Mass spectrometers are particularly pro-mising analytical instruments for biological agent detection because they are sensitive and specific as with affinitymethods, but at the same time they are generally applicable.The main challenge for biological agent detection by massspectrometry is in ionizing the relatively large moleculesthat are indicative of biological agents without excessivefragmentation.Matrix-assisted laser desorption/ionization (MALDI) isa soft ionization technique that can be used to ionize biomolecule constituents of bacteria 6,7 and viruses 8 withminimal fragmentation. Identification of microbes based onprotein database searching 9 and MALDI spectrum finger-printing 10,11 is being actively developed for homelandand national security applications. With the fingerprintingapproach, a library of specific biomarker peaks is obtainedfrommassspectraoftestsamples.Massspectraofunknownsare compared to the fingerprint library using a comparisonalgorithm. A drawback to the mass spectral fingerprintingapproach is that it often requires bacterial incubation undercontrolled conditions in order to generate reproduciblemass spectra that are not affected by growth stage or other Copyright # 2005 John Wiley & Sons, Ltd. * Correspondence to : K. K. Murray, Department of Chemistry,Louisiana State University, Baton Rouge, LA 70803, USA.E-mail: kkmurray@lsu.eduContract/grant sponsor: National Institutes of Health; contract/grant number: R01ES10497.  conditions. In the proteomics approach, experimentallydetermined protein molecular weights or sequence tags arecompared directly to known protein sequences. 7 Althoughtheproteomicsapproachdoesnotrequirethedevelopmentof fingerprintlibraries,theidentificationofproteinsfromintactmoleculemassspectrarequiresmassspectrawithgoodmassresolution and signal-to-noise.One of the chief issues in MALDI analysis of bioaerosolsis the mixing of the matrix and the analyte. 12–15 This can be done in several ways: collected particles (or extractedmaterial) can be mixed with matrix and deposited onthe target, 6,7 particles can be collected on a matrix-coatedtarget, 12 or matrix can be added to the particles by con-densation. 13–15 Analysis of collected particulate affordsthe most freedom in sample treatment prior to analysis, butit is complicated and requires multiple sample preparationsteps. Addition of matrix by condensation has been demon-strated previously both with off-line 12 and on-line particleionization. 13–15 The condensation matrix addition approachallows the ionization of particulate without deposition, but,as currently implemented, is not sufficiently sensitive forpracticalapplications.Depositiononamatrix-coatedtargetisattractive because it has the potential for rapid analysis andthe process requires few preparatory steps. Initial resultsindicate that the mixing of matrix and analyte after deposi-tion on the target is critical to obtaining good signal andsuppressing background interferences. 16,17 In this study, test bioaerosols containing aerosolized bacteria were generated using a Collison nebulizer andcollected on a MALDI target mounted in a modifiedAndersen N6 bioaerosol impactor. The target was removedfromtheimpactorandanalyzedusingMALDItime-of-flightmass spectrometry (TOFMS). Three different configurationswereused:(1)particlescollectedonabaretargetwithmatrixadded later by a pipette; (2) particles collected on a matrixpre-coated target with post-collection pipette addition of solvent;and(3)particlescollectedonapre-coatedtargetwithpost-collection spray deposition of solvent. The bioaerosoldeposition methods were compared to the standard dried-droplet MALDI sample preparation. The relative merits of the different deposition methods are discussed in terms of  bioaerosol detection with minimal sample treatment andmanipulation for use in remote and unattended MALDI-MS based bioaerosol detection instruments. EXPERIMENTAL Mass spectra were obtained using a Bruker OmniFlex TOF mass spectrometer (Bruker Daltonics, Billerica, MA, USA).The mass spectrometer can be operated in linear and reflec-tron mode and uses a 337nm nitrogen laser at a repetitionrate of 2Hz for ionization. An accelerating voltage of 19kVwas employed with delayed ion extraction and mass spectrawere obtained in linear mode of operation. Signal suppres-sion in a selected  m/z  range was accomplished using deflec-tionplates.Eachmassspectrumshownbelowwasanaveragefrom 50 laser shots.A schematic diagram of the particle collection system isshown in Fig. 1. Test bioaerosol particles were generatedusing 10mL of bacteria suspension that was placed in aCollison nebulizer(BGI, Waltham, MA, USA) and nebulizedusing filtered house air at a 30psi pressure as the carrier gas.Particle size distributions for particles generated in thismanner have been published previously and are ap-proximately 1.5 m m in diameter at a concentration of 0.5 to1  10 4 /cm 3 exiting the nebulizer. 15 The bioaerosol particleswerethen directed through 10mm i.d.  30cm length Tygontube into an Andersen N6 single-stage bioaerosol impactor(ThermoAndersen,Smyrna,GA,USA).Airisdrawnintotheimpactorataflowrateof10L/minbyavacuumpump.Agapof approximately 2mm between the Tygon tube and theimpactorinletcompensatesforapressuredifferencebetweenoutgoing flow from the nebulizer and incoming flow to theimpactor. The impactor was modified to accept the massspectrometer sample target, which can be rotated by a clockmotorduringparticlecollectiontodispersethesampleonthetarget.Inthiswork,thetargetwasheldstationarytoproducea single ‘spot’ for each of the 400 inlet holes in the impactor.The target protrudes approximately 3mm above the floor of the impactor so that the MALDI target is the appropriatedistance from the impactor inlet holes.Bioaerosol particles were collected on bare or matrix pre-coated targets. Matrix-coated targets were prepared bydepositing400 m Lofsaturatedmatrixsolutionovertheentiresurface of the target and allowing it to dry. After deposition,the target was removed from the impactor and inserted intothe mass spectrometer for analysis. In some cases, matrix orsolvent was added to the target prior to analysis. Aerosolparticle collection times were approximately 5min. For thedried-droplet method, the bacteria suspension and matrixsolution were mixed to 1:1 and 1 m L of the mixture wasdeposited on the target and allowed to dry.Alyophilizedbacteriumof  Escherichiacoli Wstrain(ATCC-9637) was purchased from Sigma-Aldrich (St. Louis, MO,USA). Matrices used in this research were 3, 5-dimethoxy-4-hydroxycinnamicacid(sinapicacid,SA,Sigma-Aldrich)and a -cyano-4-hydroxycinnamic acid (CHCA, Sigma), whichwereusedwithoutfurtherpurification.Bacteriasuspensionswerepreparedbydissolving5mgof  E.coli in1mLof4:1(v/v)acetonitrile (ACN) and 0.1% trifluoroacetic acid (TFA) indeionized water. It was found that the TFA reduced particleclumping in the suspension and produced higher particlecounts from the nebulized suspension. Saturated matrixsolutionswerepreparedbydissolving20mgofthematrixin1mL of 1:1 (v/v) ACN and deionized water. Figure 1.  Schematic diagram of the test particle generatorand modified Andersen N6 impactor particle collector. Copyright # 2005 John Wiley & Sons, Ltd.  Rapid Commun. Mass Spectrom.  2005;  19 : 1725–1729 1726 J.-K. Kim, S. N. Jackson and K. K. Murray  RESULTS AND DISCUSSION MALDI mass spectra of   E. coli  deposited using the dried-droplet method are shown in Fig. 2 and will be used forcomparison with those obtained from the collected bacteria bioaerosol particles shown below. These mass spectra aresimilar to those reported previously in that several dozenpeaksareresolvedinthe m/z rangeabove1000.Thelocationsand relative intensities of the peaks depend on a number of factors including bacterial strain, 18 growth stage, 9,19 and thematrix used. 9,10,18–20 In Fig. 2, between 30 and 40 peakswereobserved in the  m/z  range 1000–13000 with a signal-to-noise(S/N) ratio up to 100 and an average mass resolution of 600and 1100 for CHCA and SA, respectively. Mass resolutionwas calculated from m/ D m, where  D m is the full widthat half maximum (FWHM) of the peak. The CHCA matrixgenerally produced more peaks in the  m/z  range below 6000as compared to SA. Over 95% peaks that have a S/N ratiohigher than 3 were reproduced in mass spectra observed inthis region.In the initial bioaerosol MALDI experiment, test bioaer-osolsweregeneratedfromasuspensionoflyophilized E. coli and collected on a MALDI target for 5min in the modifiedAndersen impactor. After particle collection, the target wasremoved from the impactor and 1 m L of saturated matrixsolution was added to each collected spot on the target.Figure 3(a) shows the mass spectrum resulting from theaddition of CHCA and Fig. 3(b) shows the correspondingspectrum obtained from SA addition. For MALDI analysisof collected bioaerosol particles, approximately one-thirdhigher laser fluence is required to obtain adequate signal inthe higher  m/z  region (above 6000) compared to the dried-droplet sample preparation. Ion suppression was employed below  m/z  3000 to reduce the complicated signal and theraisedbaselinefromfragmentpeaks,matrixpeaksandotherinterferences caused by the increased laser fluence. The S/Nratios are comparable between Figs. 2 and 3, but, in Fig. 3,there are approximately half as many resolved peaks andtherearefewappearingabove m/z 10000.Theobservedmassresolution is a factor of two lower for the particles collectedon SA matrix in Fig. 3(b) than those collected on CHCAmatrixinFig.3(a).TheresolutioninFig.3(a)isapproximately800 for CHCA, which is slightly higher than that observedwith the dried-droplet method (Fig. 2(a)). Overall thereare fewer peaks in Fig. 3 as compared to Fig. 2, however; allof peaks observed in Figs. 3(a) and 3(b) are matchedwith corresponding peaks in those in Figs. 2(a) and 2(b),respectively.Mass spectra obtained by depositing bioaerosols onmatrix-coated targets are shown in Fig. 4. The matrix-coated targets were prepared by depositing 400 m L of matrix solution on the target and allowing it to dry beforethe target was placed in the impactor. A target was removedfrom the impactor after bioaerosol deposition and inserteddirectly into the mass spectrometer for analysis with nofurther treatment; the mass spectrum shown in Fig. 4(a)was obtained as a result. The jump in signal at  m/z  3000reflects the ion suppression pulse and resulting shift in baseline. No signal was observed above 3000, and only anunresolved low  m/z  signal was recorded. However, signalcouldberecoveredifsolventwasaddedtothetargetafterthe bioaerosol was collected on the matrix pre-coated target.Figures 4(b) and 4(c) show the result of adding 200 m L of a1:1 (v/v) mixture of ACN and water on the spot of collectedparticulate deposited on CHCA and SA matrix-coatedtargets, respectively. The SA target mass spectrum inFig. 4(c) is similar in terms of number of peaks, S/N ratioand mass resolution (greater than 800) compared withFig. 2(b), the SA dried-droplet mass spectrum. The CHCAtarget mass spectrum in Fig. 4(b) is somewhat lower in Figure 2.  MALDI mass spectra of  E. coli   bacteria using (a) a -cyano-4-hydroxycinnamic acid (CHCA) and (b) sinapic acid(SA) matrix. Figure 3.  Mass spectra of  E. coli   bacteria obtained bydepositing 1 m L of (a) CHCA and (b) SA matrix on thecollected test bioaerosol particles. MALDI-MS of collected bioaerosol particles 1727 Copyright # 2005 John Wiley & Sons, Ltd.  Rapid Commun. Mass Spectrom.  2005;  19 : 1725–1729  quality with fewer peaks with worse mass resolution(approximately 200) at one-third higher laser fluencecompared to the dried-droplet mass spectrum shown inFig. 2(a).As a final test of the impaction on matrix approach, test bioaerosols were deposited on matrix-coated targets fol-lowed by spray deposition of solvent. These results areshown in Fig. 5. Matrix-coated targets were inserted into theimpactor and test bioaerosol particles were deposited onthe target. Solvent was deposited on the target in the formof aerosol particles delivered to the throat of the impactor.The solution of 1:1 (v/v) ACN/water was nebulized usingthe Collison nebulizer and collected on the target for ca.5min.Itispossiblethatsomeofthesolventwasdepositedbyvapor condensation due to evaporation from the particles;however, sufficient solvent was added to the sample depositto partially recover the high-mass signal. Several peaks areobserved in the mass spectra, but they are relatively few innumber and broad with a mass resolution of 75 and 100 forCHCA and SA, respectively. CONCLUSIONS Using a modified Andersen N6 impactor and test bioaerosolparticles formed by nebulizing a suspension of lyophilized E. coli  bacteria, several methods for particle collection andsubsequent matrix addition for MALDI-MS analysis weretested. The matrixes used were  a -cyano-4-hydroxycinnamicacid (CHCA) and sinapic acid (SA). Of these methods, parti-cles collected on a matrix pre-coated target with post-collec-tion solvent addition by pipette gave the best results, whichwere nearly identical to those obtained using the standarddried-droplet approach. The mass spectra obtained frommatrix addition after bioaerosol deposition on a bare targethad nearly as good signal-to-noise ratios as those obtainedwith the dried-droplet approach, but the number of peaksand the mass resolution were lower. Deposition on a matrixpre-coatedtargetwithoutlatersolventadditiondidnotresultinmassspectra:sometypeofpost-collectionreagentaddition(either solvent or matrix solution) was necessary to obtainmass spectra.The fact that solvent or matrix addition is necessary toobtain MALDI mass spectra of collected bioaerosol particleshas important implications for portable and transportableinstrumentation for bioaerosol detection. A remote auto-mated data collection system based on a MALDI-MSapproach must have a solvent reservoir and a mechanismfor the addition of the solvent to the matrix target prior toanalysis. Simple deposition onto matrix pre-coated targetswillmostlikelynotbesufficient.Thiswilladdtothesizeandcomplexity of such an instrument and require periodicreagent replenishment. The mechanical complexity of sucha system might be reduced if the liquid reagents are sprayedinto the impactor after (or potentially during) particledeposition, but operation in this mode will result in lowerperformance. Acknowledgement This research is supported by the National Institutes of Health under grant no. R01ES10497. Figure 4.  MALDI mass spectra of test bioaerosols depos-ited on a target pre-coated with (a) SA matrix and noadditionaltreatment,(b)CHCA plus200 m Lofsolvent,and(c)SA plus 200 m L of 1:1 (v/v) acetonitrile/water solvent. Ionsuppression was used below  m/z   3000, resulting in somebaseline discontinuity. Figure 5.  MALDI mass spectra of test bioaerosols depos-ited on a target pre-coated with (a) CHCA and (b) SA matrixwith 1:1 (v/v) acetonitrile/water solvent added by spraydeposition. 1728 J.-K. Kim, S. N. Jackson and K. K. Murray Copyright # 2005 John Wiley & Sons, Ltd.  Rapid Commun. Mass Spectrom.  2005;  19 : 1725–1729  REFERENCES 1. Fatah AA, Barrett JA, Arcilesi RD, Ewing KJ, Lattin CH,Moshier TF.  An Introduction to Biological Agent DetectionEquipment for Emergency First Responders , National Instituteof Justice Guide 101-00, US Department of Justice,December 2001.2.  Chemical and Biological Defense Primer , Department of Defense Chemical and Biological Defense Program,Washington, DC, October, 2001.3. Branscomb LM, Klausner RD. (co-Chairs, NationalResearch Council Committee on Science and TechnologyforCounteringTerrorism). The Role of Science and Technologyin Countering Terrorism . National Academies Press:Washington, 2002.4. Spurny KR. Chemical analysis of bioaerosols. In  Bioaerosols Handbook , Cox CS, Wathies CM (eds). CRC Press: BocaRaton, 1995.5. Hensel A, Penzoldt K. Biological and biochemical analysisof bacteria and viruses. In  Bioaerosols Handbook , Cox CS,Wathies CM (eds). CRC Press: Boca Raton, 1995.6. Lay JO Jr.  Mass Spectrom. Rev . 2001;  20 : 172.7. Fenselau C, Demirev PA.  Mass Spectrom. Rev . 2002;  20 : 157.8. Siuzdak G.  J. Mass Spectrom . 1998;  33 : 203.9. Demirev PA, Ho YP, Ryzhov V, Fenselau C.  Anal. Chem .1999;  71 : 2732.10. Jarman KH, Cebula ST, Saenz AJ, Petersen CE, ValentineNB, Kingsley MT, Wahl KL.  Anal. Chem . 2000;  72 : 1217.11. WahlKL,WunschelSC,JarmanKH,ValentineNB,PetersenCE, Kingsley MT, Zartolas KA, Saenz AJ.  Anal. Chem . 2002; 74 : 6191.12. Jackson SN, Murray KK.  Anal. Chem . 2002;  74 : 4841.13. Stowers MA, van Wuijckhuijse AL, Marijnissen JCM,Scarlett B, van Baar BLM, Kientz CE.  Rapid Commun. MassSpectrom . 2000;  14 : 829.14. van Baar BL.  FEMS Microbiol. Rev . 2000;  24 : 193.15. Jackson SN, Mishra S, Murray KK.  Rapid Commun. MassSpectrom . 2004;  18 : 2041.16. Scholl PF, Leonardo MA, Rule AM, Carlson MA, AntoineMD,BuckleyTJ.  Johns Hopkins APL Technical Digest 1999; 20 :343.17. Bogan MJ, Agnes GR.  Anal. Chem . 2002;  74 : 489.18. ArnoldRJ,ReillyJP. RapidCommun.MassSpectrom .1998; 12 :630.19. Arnold RJ, Karty JA, Ellington AD, Reilly JP.  Anal. Chem .1999;  71 : 1990.20. Jones JJ, Stump MJ, Fleming RC, Lay JO, Wilkins CL.  Anal.Chem . 2003;  75 : 1340.MALDI-MS of collected bioaerosol particles 1729 Copyright # 2005 John Wiley & Sons, Ltd.  Rapid Commun. Mass Spectrom.  2005;  19 : 1725–1729
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks