RAPID COMMUNICATIONS IN MASS SPECTROMETRY Rapid Commun.Mass Spectrom.2004;18:
49–58
Published online in Wiley InterScience(www.).DOI:10.1002/rcm.1276
A study of ion suppression effects in electrospray ionization from mobile pha additives and solid-pha extracts
Claude R.Mallet*,Ziling Lu and Jeff R.Mazzeo
Waters Corporation,34Maple St.,Milford,MA01757-3696,USA
径字组词
Received25September2003;Revid20October2003;Accepted22October2003
Since the wide adoption of liquid chromatography/tandem mass spectrometry(LC/MS/MS),the ion
suppression/enhancement phenomenon is the latest barrier to high-throughput analysis.This con-
quence of a nonoptimized analytical method can lead to adver effects during
poor accuracy and precision).Previous papers have reported that ion suppression is a direct result
of endogenous material prent in biological samples.However,in the ca of a solid-pha liquid
chromatography/tandem mass spectrometry(SPE/LC/MS/MS)system,the measured result is the
combination of veral operating conditions and parameters.Little has been done to effectively
monitor and/or choo optimized conditions for the complete quence of extraction,clean up,
碳酸钙咀嚼片paration and analysis.This paper describes a simple tup for quantification of ion suppres-
sion/enhancement.Several mobile pha additives,ion-pairing agents and SPE extracts were mea-
sured and compared against a standard reference.The results demonstrated that a clean up of
plasma extracts bad on ion exchange leads to minimal ion suppression/enhancement for the com-
pounds that were investigated.Copyright#2003John Wiley&Sons,Ltd.
Today,liquid chromatography/tandem mass spectrometry (LC/MS/MS)is without question the preferred platform for the analysis of organic molecules from complex matrices. Variousfields such as pharmaceuticals,toxicology,environ-mental and clinical have made the transition from the tradi-tional one-dimensional LC/UV,etc.)to hyphenated mass spectrometric technologies(LC/MS, LC/MS/MS,etc.).This transition was made possible by the pioneering work of Dole1and Fenn2with the development of a novel atmospheric pressure interface for the coupling of a liquid chromatograph and a mass spectrometer.The con-quences of this work can be en in the number of papers published during the last decade and the emergence of new fields of interest(proteomics,metabolomics,etc.).Most appli-cations utilize the electrospray ionization(ESI)interface (about80%of published papers)versus atmospheric pres-sure chemical ionization(APCI).Several reasons can be pre-nted to explain this trend.ESI is an interface that is relatively easy to u,exhibits low solvent consumption and can be ud for large analytes(up to100kDa),has a wide polarity range and can be applied to thermally labile compounds.
A survey of the literature indicates that current LC/MS/MS platforms are far from being problem-free turnkey solutions. With high analytical demands,such as analysis speed,low analysis costs,reduced sample volume,time constraints and limited resources,it is not surprising to read papers or reviews urging for the next generation of integrated LC/MS/MS systems.During recent years,the phenomenon of ion suppression has quickly become a major concern.3,6,8,11–13In fact,the phenomenon of ion enhancement or suppression has become a difficult bottleneck.This matrix effect,if not well understood,can lead to errors in the quantitation of an analyte of interest.Since the introduction of MS/MS instrumentation, many groups have perceived that LC/MS/MS systems have tremendous benefits,especially the capacity to eliminate interferences due to their inherent lectivity.3The tandem mass spectrometer could be ud without sample pre-treatment and/or with little chromatography,and no inter-ference signals were obrved when the mass spectrometer was t up in multiple reaction monitoring(MRM)mode.This practice quickly became the norm for the analysis of biological samples(plasma,tissue,urine,etc.).However,when required detection limits drop to the sub-ng/mL level,matrix effects can produce a total suppression of signal.4It becomes apparent that,even if interference signals are not recorded, their prence can drastically influence the reproducibility and accuracy of any assay.
A quick survey of the literature shows a widespread adoption of ESI-MS for the analysis of biological samples. However,ESI is ud without a thorough understanding of the ion production mechanism.Furthermore,the ESI signal can be affected by veral phenomena,such asflow instability,background noi,mobile pha and sample
Copyright#2003John Wiley&Sons,Ltd.
*Correspondence to:C.R.Mallet,Waters Corporation,34Maple St.,Milford,MA01757-3696,USA.
E-mail:claude_
interferences,and competition from bulk ions (bile pha additives)inside the droplets.15Several mechanisms of ion suppression are currently under investigation,generally bad on the Coulomb fission mechanism propod by Kerbale 5and the ion evaporation model of Iribane and Thomson.18Meanwhile,veral papers offer solutions to the problem,including multiple extraction protocols,7ballistic gradients,8and on-line SPE.19Matrix effects are mostly caud by the prence of endogenous materials from biological samples (i.e.salts,amines,fatty acids,etc.).6Thus,an understanding of the nature of the ESI mechanism(s)is crucial in order to derive an effective solution to this suppression/enhancement problem.Many groups are cur-rently working t
o understand aspects of ion suppression/enhancement,especially mechanisms and questions such as the location of suppression/enhancement (liquid or gas pha).In parallel,there is a major effort to develop methods whereby it can be circumvented.
Most papers suggest that suppression occurs due to a reduction of solvent evaporation in the source or to an increa of surface tension in droplets due to a high concentration of matrix interferences.13,14Bad on the evaporation theory,u of hydrophilic interaction chromato-graphy (HILIC)(high organic mobile pha),versus that of traditional reverd-pha chromatography (low organic mobile pha),should be less prone to signal suppression.The scope of this paper is not to elucidate the mechanism of ion suppression/enhancement;rather it is intended to ben-chmark various elements that constitute an SPE/LC/MS/MS system,mainly various mobile pha additives and the importance of an optimized sample clean-up protocol.The latter aspect is certainly not novel,as many current papers prent the same conclusions.15–17Several methodologies for measuring ion suppression are currently available.9,10,14Bad on one of the protocols,14we developed a slightly modified version and ud this method to quantitate the suppression and enhancement effects for various test solu-tions including LC mobile pha additives,LC ion-pairing agents and SPE extracts.
EXPERIMENTAL Chemicals and reagents
All basic drugs (propranolol,trimethoprim,pipenzolate,resperidone,terfenadine,methoxyverapamil,benextramine and rerpine),acidic compounds (fumaric acid,malic acid,canrenoic acid,cholic acid)and sugar (raffino)were pur-chad from Sigma (St.Louis,MO,USA).The biphosphonate
drugs etidronic acid and clodronic acid were purchad from LKT Laboratories Inc.(St.Paul,MN,USA).Acetonitrile and methanol (both HPLC grade)were procured from J.T.Baker (Phillipsburg,NJ,USA).Deionized water purified with a Milli-Q ultrapure water system from Millipore Corporation (Bedford,MA,USA)was ud.Drug-free rat plasma (EDTA)was purchad from Equitek (Kerrville,TX,USA).Drug-free human plasma (heparine)was purchad from Lampire Biological Laboratories (Piperville,PA,USA).Acidic additives (formic acid 98.9%,acetic acid 99%,and tri-fluoroacetic acid (TFA)99%),basic additives (ammonium hydroxide 29.7%and pyrrolidine)and salt additives (tetra-ethylammonium,ammonium formate,ammonium bicarbo-nate and ammonium biphosphate)were purchad from Sigma ).The SPE 30mg 96-well plates (Oasis 1HLB,MCX and MAX)were from Waters Corporation (Milford,MA,USA).
Instrumentation and equipment
The liquid chromatograph ud was an Alliance 2790HT system coupled to a ZQ single quadrupole mass spectro-meter from Waters Corporation.The mass spectrometer was equipped with an ESI interface and controlled by compu-ter software MassLynx version 4.0.The syringe pump,Harvard PHD 2000,was purchad from Harvard Apparatus (Holliston,MA,USA).
Preparation of standards
Stock solutions (10mL)of all drugs (1.0mg/mL)were pre-pared in methanol.Subquently,two 1.0L test solutions were prepared in 50/50methanol/water (no additives),one containing all basic drugs and the other one containing all acidic compounds and raffino.The concentrations of all the drugs are reported in Table 1.Four aliquots of 10mL (50/50methanol/water)were transferred to 20mL vials and spiked with formic acid at 0.05,0.1,0.5and 1.0%.The same procedure was repeated for acetic acid,trifluoroa-cetic acid,ammonium hydroxide and pyrrolidine.For the ion-pairing agents (tetraethylammonium,ammonium for-mate,ammonium bicarbonate and ammonium biphosphate),the aliquots were spiked at 5,10,20and 50mM.
ESI conditions
The mass spectrometer was operated in positive mode for the analysis of basic drugs and in negativ
e mode for the acidic drugs.The source ttings for positive and negative modes were the same (with appropriate polarity reversals of applied potentials)and were as follows:source temperature 1508C,
Table 1.Concentrations of acidic and basic drugs ud as test analytes for suppression/enhancement evaluation
Compound [M–H]À[](ng/m L)
Compound [M þH]þ[](ng/m L)
Fumaric acid 115.0 1.0Propranolol 260.2 1.0Malic acid 133.1 1.0Trimethoprim 291.3 1.0Etidronic acid 205.28.0Pipenzolate 354.4 1.0Clodronic acid 243.2 1.0Resperidone 411.40.5Niflumic acid 281.4 6.0Terfenadine
472.6 1.0Canrenoic acid 357.6 6.0Methoxyverapamil 485.6 1.0Cholic acid 407.710.0Benextracmin 591.67.0Raffino
503.4
10.0
Rerpine
609.6
5.0
Copyright #2003John Wiley &Sons,Ltd.Rapid Commun.Mass Spectrom.2004;18:49–58
50 C.R.Mallet,Z.Lu and J.R.
Mazzeo
desolvation temperature 3508C,capillary voltage 3.5kV,cone voltage 30V,extractor 3.0V,RF lens 0.3V,
desolvation gas 550L/h,cone gas flow 50L/h,ion energy 0.5and multiplier 650V.The quadrupole was t for maximum resolution (high mass and low mass parameters were t at 15.0,the unit-less values reprent a ratio of RF/DC for unit mass resolution).
System tup
福州市花
As shown in Fig.1,the Alliance 2795HT is directly connected to the ESI interface via a stainless steel tee.Line A was ud to deliver,at constant flow (0.2mL/min),the basic or acidic compound test solutions.The infusion pump line,t at 0.2mL/min,was ud to introduce various LC additives and SPE extracts for quantitation of suppression or en-hancement.Data acquisition was done using the multiple channel acquisition (MCA)tting for a period of 1.0min
with m/z from 100to 1000with 1.0s scan cycle time.The ana-lytical quence involved the infusion of a blank (50/50water/methanol),followed by the test sample and then back to the blank solvent.Each acquisition was done in tripli-cate;the signal intensity for each drug was measured as the mean value.The blank solvent solution was ud as a refer-ence zero.Subquently,any suppression effect obrved was reported as a negative percentage and an enhancement effect as a positive value.The next t of analys was per-formed only after a thorough infusion line flush with solvent u
ntil background levels returned to the values obrved for the solvent blank in the previous data t.
Protein precipitation and SPE extracts
The protein precipitation extract was prepared as follows:0.5mL of raw plasma was placed in a 1mL Eppendorf tube,0.5mL of methanol was added,and the tube was spun in a microcentrifuge at 5000rpm for 5min;the supernatant was collected and transferred to a clean Eppendorf tube.The SPE extracts for the 1D,2D and mixed-mode methods were prepared according to the protocol described in Table 2.
RESULTS AND DISCUSSION丁卤肉
The effect of ion suppression was measured utilizing a mod-ification of the protocol describe by King et al .14This approach 14gives a qualitative view of suppression effects during the cour of a chromatographic paration of a blank sample extract (zero analyte)using an added (constant)infu-sion of the target analyte solution (fixed concentration)via a post-column tee.In the abnce of any interferences eluting from the column,the post-column infusion of analyte will give a constant signal in the MRM channel.The resulting chromatogram will show a reduction of this constant baline signal due to suppression of the analyte signal by elution of endogenous material from the sample (e.
g.,extract of urine,plasma,etc.).With this method,14valuable information is obtained on any potentially high suppression regions in the chromatogram,so that the chromatographic conditions can be modified to reduce or increa the retention time
of
Figure 1.Instrumentation tup for suppression/
enhancement quantitation.
Table 2.SPE protocols for the 1D,2D and mixed mode plasma extracts
Reverd pha Acidic &basic (1wash)Acidic analyte (3washes)
Basic analyte (3washes)Condition 1mL water 1mL water 1mL water Condition 1mL MeOH 1mL MeOH 1mL MeOH Load 1mL plasma 1mL plasma
1mL plasma
Wash 1mL 5%MeOH 1mL 5%MeOH þ2%FA
1mL 5%MeOH þ2%FA
Wash none 1mL 5%MeOH þ2%NH4OH
1mL 5%MeOH þ2%NH4OH Wash none
1mL 50/50MeOH/water þ2%NH4OH 1mL 50/50MeOH/water þ2%FA Elute 0.5mL MeOH 0.5mL MeOH þ2%NH4OH 0.5mL MeOH þ2%FA Dilute 0.5mL water
0.5mL water 0.5mL water Mixed mode Basic analyte
Acidic analyte
Condition 1mL water 1mL water Condition 1mL MeOH 1mL MeOH Load 1mL plasma
1mL plasma
Wash 1mL water þ0.1N HCl 1mL water þ2%NH4OH Wash 1mL MeOH 1mL MeOH Wash none
none
那年花好月正圆Elute 0.5mL MeOH þ2%NH4OH 0.5mL MeOH þ2%FA Dilute
0.5mL water
0.5mL
water
ESI suppression by additives and SPE extracts 51
Copyright #2003John Wiley &Sons,Ltd.
Rapid Commun.Mass Spectrom.2004;18:49–58
the target analyte such that the retention time is outside of the suppression zones.However,while the analyte is shifted in the chromatogram,the suppression region may also shift and testing must be repeated.
This method14provides a miquantitative view of suppression/enhancement phenomena in the LC/MS/MS method.However,the question remains as to how the ion suppression can be fully quantitated,and which elements of an SPE/LC/MS/MS system are the major contributors to the suppression or enhancement.The prent work us a variant of the experimental method;14the LC column was omitted(Fig.1)to eliminate any potential suppression/ enhancement contributions from the support or ligand material),so as to concentrate on the effects of various mobile pha additives and SPE extracts.The test solution (e.g.a solution of mobile pha additives or SPE extracts)is mixed into a constant stream of a target solution(either acidic or basic analytes in50/50methanol/water solution with no additives).There is no chromatographic paration orflow injection;rather the two streams are continuously mixed. When the syringe isfilled with a reference pure mobile pha instead,the analyte signals provide a reference point against which each test bile pha additives or SPE extracts)will be quantitated for suppression or enhancement. Figure2shows three partial mass spectra acquired with two common mobile pha additives to demonstrate the ion suppression and enhancement effect.The middle spectrum was acquired with50/50methanol/water.The ion of interest in the spectra is at m/z472,the[MþH]þion of terfenadine. When the analyte stream is mixed with an equivalentflow rate of0.5%TFA from the test solution stream(bottom spectrum),the m/z472ion shows a reduction of75%in signal intensity.Howeve
r,when the analyte stream was mixed with 0.5%ammonium hydroxide(top spectrum),the signal shows an enhancement effect of41%.
The following ctions will prent results measured for various mobile pha additives and SPE extracts.An important consideration concerns the experiments with mo-bile pha additives.It is common practice in some labo-ratories to u an unbuffered mobile pha for LC/MS/MS applications.U of unbuffered mobile pha can lead to retention time variation(high relative standard deviations (RSDs)for retention time),and therefore u of a buffered mobile pha is recommended.The usual rule-of-thumb is to u a mobile pha buffered at2pH units higher or lower than the pK a of the target analyte.In the prent experimental arrangement,since the chromatographic column was omitted,the mobile pha additives were prepared without buffering capacity for the following reasons.Since the main objective of this study was to quantify the effect of ion suppression/enhancement,each additive was measured parately at typical percentages or mM levels.Thus,the results will reflect the contribution of that particular additive alone.If buffered solutions were ud,results could be difficult to explain since it would be unclear which component caud the suppression or enhancement effect. For example,it is difficult to predict if a buffered mobile pha shows an additive effect or a more complicated situation;this question concerning effec
ts of multiple additive components was not addresd in this study and will be investigated in future work.Furthermore,since some of the additives are also ud in various SPE protocols
for Figure2.Examples of ion suppression and enhancement on terfenadine.(A)Terfenadine mixed with
50/50MeOH/H2Oþ0.5%NH4OH;(B)terfenadine mixed with50/50MeOH/H2Oþno additive;and(C)
terfenadine mixed with50/50MeOH/H2Oþ0.5%TFA.
52 C.R.Mallet,Z.Lu and J.R.
Mazzeo
Copyright#2003John Wiley&Sons,Ltd.Rapid Commun.Mass Spectrom.2004;18:49–58
rat and human plasma samples,and can be prent in the final extract,this reprents an additional argument for the prent unbuffered approach.
The 16compounds chon as test analytes are shown in Table 1;the choice of the mixtures was bad on veral criteria.ESI signal intensity is compound-dependent.So the choice of test compounds was bad on a diversity of mass range,polarity and structure.For example,the drug pipenzolate is a quaternary amine (pK a >14.0)and is not expected to be influenced by the solution chemistry of droplets during the nebulizing step.Also,etidronic and clodronic acid,both biphosphonate drugs,have pK a values below 1.0,and thus have the same expectations with respect to additive effects as pipenzolate.
Mobile pha additives
The u of additives and buffers in HPLC mobile phas rves two purpos.First,they are primarily ud to add buffering capacity in order to achieve reproducible retention times during a typical chromatographic run (isocratic or gra-dient).Second,depending on the nature of the additive,they can also be ud as an ion-pairing agent to create a pudo-
neutral species with the target analyte and thus produce a sharper peak shape and longer retention t
腐乳肉的家常做法
imes (increa in paration power and theoretical plates).The are tried and tested practices since the early days of LC/UV systems.The u of additives with LC/MS/MS systems can rve a third purpo,which is to protonate basic molecules when operating the ESI source in positive mode,and vice versa for acidic molecules.However,veral papers 20–22have reported that protonated analytes can be obrved with high pH mobile phas and deprotonated analytes under low pH conditions.U of an acidic mobile pha for the ana-lysis of basic compounds creates a difficult situation for reverd-pha chromatography since the protonated ana-lyte will show little retention (low k 0)under acidic conditions.This situation could have a huge impact on assays that require the analysis of multiple components.The con-quence of low k 0will lead to multiple peaks co-eluting into the ESI source,with potential cross suppression/enhance-ment effects.Thus,it is often necessary to u baline chro-matographic resolution to eliminate this potential cross-talk.Table 3summarizes the results obtained for the test solutions of both basic and acidic compounds in the prence
Table 3.Suppression and enhancement effects for the pH additives
0.05%
0.10%
0.50%
1.00%
0.05%
0.10%
0.50%
1.00%
Formic acid
Ammonium hydroxide Positive test solution Propranolol 36.528.8 4.5À8.3À2.2 2.0210.211.4Trimethoprim 41.730.1À5.3À17.5À5.4À5.4 4.28.9Pipenzolate À0.1À0.2À5.5À9.50.020.020.020.01Resperidone À27.5À37.1À54.2À59.4 6.19.616.116.8Terfenadine
17.311.6À7.9À16.510.821.357.966.6Methoxyverapamil 22.817.1À1.8À10.738.841.146.649.1Benextr
amine À39.77À44.1À52.7À52.822.130.737.938.3Rerpine
21.421.417.28.9À12.1À11.9À6.2À3.2Negative test solution Fumaric acid À11.9À29.5À64.7À68.1À38.4À41.1À45.8À57.8Malic acid À11.2À27.9À62.2À63.9À35.5À38.8À42.4À53.4Etidronic acid 29.817.8À17.2À30.9À61.9À63.5À75.9À70.3Clodronic acid 5.7À15.7À58.3À66.60.3À1.3À5.3À27.7Niflumic acid À0.28À21.4À60.9À64.514.111.1 5.3À11.6Canrenoic acid 13.8À11.1À51.6À57.6196.1202.5201.9127.3Cholic acid 31.9 3.7À40.8À44.7420.5454.9403.1352.8Raffino
À4.6
À26.3
À39.4
À43.7
60.9
61.9
66.6
32.1
Trifluoroacetic acid
Acetic acid Propranolol À54.8À62.8À74.7À77.125.525.817.3À0.2Trimethoprim À40.1À58.1À73.9À76.618.310.4À0.4À7.1Pipenzolate À27.5À37.4À43.9À43.70.01À0.01À0.4À1.7Resperidone À53.7À62.3À68.2À69.3À2.1À16.8À37.7À44.2Terfenadine
À24.4À44.6À61.5À64.815.911.97.5À2.8Methoxyverapamil À59.9À57.3À70.2À72.619.516.68.9À4.8Benextramine À29.4À41.8À42.7À38.7À21.9À28.9À29.9À27.8Rerpine
À32.5À52.8À71.7À75.719.315.612.411.1Negative test solution Fumaric acid À87.4À89.7À91.1À91.2À15.1À29.1À51.3À59.5Malic acid À84.1À86.9À88.4À88.1À14.5À27.6À48.3À58.04Etidronic acid À71.9À73.1À71.6À65.929.226.810.1À17.8Clodronic acid À95.6À97.4À98.8À98.8 4.9À4.9À36.5À49.
6Niflumic acid À91.7À94.8À98.2À98.138.520.6À22.1À34.7Canrenoic acid À93.8À96.1À96.8À96.1À16.7À33.8À67.4À59.4Cholic acid À95.2À97.5À99.5À99.6À18.9À33.1À48.5À63.9Raffino
À84.1
À91.2
À96.5
À97.6
À5.7
炸辣椒油的做法À19.4
À26.3
À
63.6
ESI suppression by additives and SPE extracts 53九峰山景区
Copyright #2003John Wiley &Sons,Ltd.Rapid Commun.Mass Spectrom.2004;18:49–58
