Use of 3-nitrobenzonitrile as an additive for improved sensitivity in sonic-spray ionization mass spectrometry.

Research Article Received: 19 June 2014

Revised: 23 September 2014

Accepted: 23 September 2014

Published online in Wiley Online Library

Rapid Commun. Mass Spectrom. 2014, 28, 2661–2669 (wileyonlinelibrary.com) DOI: 10.1002/rcm.7064

Use of 3-nitrobenzonitrile as an additive for improved sensitivity in sonic-spray ionization mass spectrometry Katerina Kanaki and Spiros A. Pergantis* Environmental Chemical Processes Laboratory, Department of Chemistry, University of Crete, Voutes Campus, Heraklion 71003, Greece RATIONALE: Sonic-spray ionization (SSI) has been shown to produce gas-phase ions for a wide range of compounds, without the application of voltage or a laser. However, it remains to be shown that it can also provide similar sensitivities to those obtained by electrospray ionization mass spectrometry (ESI-MS). METHODS: Here we report on an attempt to further improve the sensitivity of SSI-MS, more specifically a version of SSI that is referred to as Venturi easy ambient sonic-spray ionization (V-EASI) MS, by adding a signal-enhancing additive to the sample solution. The additive used is 3-nitrobenzonitrile (3-NBN), which has recently been used with success in a new ionization approach named matrix-assisted ionization vacuum. In order to conduct this study we have analyzed a range of compounds, including peptides, metalloproteins, and some organometalloids. During the V-EASI-MS analyses molecular ion and protonated molecule signal intensities as well as their corresponding signal-to-noise (S/N) ratios, obtained in the presence and absence of the 3-NBN, were compared. RESULTS: The 3-NBN-assisted V-EASI-MS approach developed here provides significant improvement in sensitivity relative to conventional V-EASI-MS for almost all compounds tested. More specifically, for peptides a 1.6- to 4-fold enhancement was realized, for proteins the enhancements were from 2- to 5-fold, and for some metalloid species enhancements reached up to 10-fold. However, optimum additive concentration and ion transfer capillary temperature were found to be compound-dependent and thus require optimization in order for maximum enhancements to be achieved. In most cases the 3-NBN-assisted V-EASI-MS approach provides comparable sensitivities and S/N ratios to ESI-MS on the same ion trap mass spectrometer. CONCLUSIONS: The use of 3-NBN with V-EASI-MS gives rise to a novel 3-NBN-assisted MS technique, which has demonstrated considerable signal enhancement for most of the compounds analyzed, thus improving its competitiveness towards the well-established and dominating ESI-MS technique. Copyright © 2014 John Wiley & Sons, Ltd.

Even though sonic-spray ionization (SSI) mass spectrometry (MS) was first introduced in 1994,[1] it remains an underevaluated and under-utilized mass spectrometric technique, not for lack of adequate analytical figures of merit, but most likely because of the dominance of electrospray ionization (ESI)-MS in the years prior to and after that. Recently, however, the technique of SSI-MS has been revived through the work of Eberlin and co-workers[2–6] and others,[7] demonstrating significant potential to become a competitor to ESI-MS. In particular, for the analysis of coordination compounds and oxometals and metalloid anions, some distinct advantages of SSI, over ESI, have recently been documented.[7] Milder ionization conditions affording less fragmentation of labile compounds, lack of redox artefacts as no electrical potentials are applied, and extreme simplicity of use,[8] are some of the advantages of SSI-MS. In addition, in a recent study,[7] the use, for the first time, of commercially available pneumatic nebulizers from the field of atomic

Rapid Commun. Mass Spectrom. 2014, 28, 2661–2669

Copyright © 2014 John Wiley & Sons, Ltd.

2661

* Correspondence to: S. A. Pergantis, Environmental Chemical Processes Laboratory, Department of Chemistry, University of Crete, Voutes Campus, Heraklion 71003, Greece. E-mail: [email protected]

spectrometry, followed by the construction and use of a prototype SSI source, has further improved the potential of SSI-MS to become a more widely accepted and utilized ionization source for a diverse range of analytical needs. This source was based on the report by Santos et al. who described a modified SSI technique, named VL-EASI (Venturi Liquid–Easy Ambient Sonic Spray Ionization),[5] in which the analyte solution is introduced into the mass spectrometer by selfaspiration as a result of the Venturi effect, thus eliminating the need to use electric pumps. In brief, SSI simultaneously generates negative and positive gas-phase ions from their pre-existing solution ions via conventional pneumatic nebulization, a process used for decades to convert liquid samples into fine aerosols and introduce them into a range of analytical atomic techniques. It has been observed that in SSI-MS optimum results are obtained at nebulizer gas flows equal to or higher than sonic velocity. A portion of the aerosol enters the mass analyzer through a vacuum interface, with or without heating, depending on the type of instrument used. Complete ion desolvation possibly occurs in the intermediate pressure region of the interface between atmospheric pressure and the high vacuum inside the analyzer.[9] The mechanism responsible for gas-phase ion formation in SSI is not yet clear.

K. Kanaki and S. A. Pergantis However, when comparing it to ESI it is evident that the initial charge separation mechanisms are dissimilar. Subsequently, gas-phase ion formation from charged droplets is expected to also be affected by some differences between the two techniques concerning droplet size distribution, droplet charge densities and the pressure conditions under which gas-phase ions form. Recently, Zilch et al.[10,11] put forward the bag model, which, in combination with the electric double layer at the surface of the water droplet,[12] was used to explain how charge separation occurs under field-free SSI conditions. However, this proposed model has not yet received broader acceptance and thus additional studies are needed. As already mentioned, SSI entered the field of mass spectrometric analysis at a time when ESI was realizing unprecedented success and development in the hands of academics, from a wide range of disciplines, as well as instrument manufacturers. One of the critical aspects for this progress was the improved sensitivity that rapidly became available on commercial instruments as a result of source and analyzer improvements. In contrast, the improvement in sensitivity in SSI-MS has so far received limited attention. With respect to this, we wish to report here on our initial efforts to explore new approaches in order to achieve improved sensitivities with V-EASI-MS, as well as to try to understand the critical parameters that most influence V-EASI-MS sensitivity. Our efforts are based on a recent study by Trimpin and Inutan,[13] which describes a matrix-assisted ionization vacuum (MAIV) technique that does not require high voltages, a laser beam, or applied heat, but uses a matrix, in most cases 3-nitrobenzonitrile (3-NBN), and the vacuum of the mass spectrometer to achieve analyte ionization. In their studies it was mentioned that 3-NBN sublimation[14] and triboluminescence upon crystal fracturing[15,16] may be the critical properties promoting the production of bare gas-phase analyte ions. These reports have provided us with the idea to use 3-NBN as a potential signal-enhancing additive in V-EASI-MS analyses. In order to conduct this study we have analyzed a range of different classes of compounds, including peptides, metalloproteins, and some organometalloids. During the V-EASI-MS analyses molecular ion and protonated molecule signal intensities as well as their corresponding signal-to-noise (S/N) ratios, obtained in the presence and absence of the 3-NBN, were compared. The objective was to achieve higher sensitivities and to investigate if 3-NBN-assisted V-EASI-MS has the potential to compete with ESI-MS in terms of sensitivity.

EXPERIMENTAL Chemicals

2662

3-Nitrobenzonitrile (assay: 98%), used as an additive, and L-glutathione reduced (assay ≥ 97%, HPLC sum of enantiomers) were purchased from Sigma-Aldrich Chemie GmbH (Steinheim, Germany). The peptide MRFA (Met-Arg-Phe-Ala) was purchased from Research Plus, Inc. (Bayonne, NJ, USA). Arsenobetaine (CH3)3As+CH2COOH was prepared from a stock solution used as a Certified Reference Material (BCR No 626, European Commission). Seleno-DL-methionine (SeMet) (≥99.0%) was obtained from Fluka (Biochemika, Sigma–Aldrich,

wileyonlinelibrary.com/journal/rcm

Buchs, Switzerland). The trimethylselenonium ion (TMeSe+) was synthesized[17] and donated by Professor Kevin A. Francesconi (University of Graz, Austria). All protein standards, i.e. cytochrome c from equine heart (≥95%), myoglobin from horse heart (lyophilized powder, minimum 90% phast gel), and lysozyme from chicken egg white (lyophilized powder ≥90%, ≥40000 units/mg protein), were purchased from Sigma-Aldrich Chemie GmbH. The arsenic compounds 4-hydroxyphenylarsonic acid and 3-nitro-4-hydroxyphenylarsonic acid (roxarsone) were obtained from Eastman Kodak and ICN Biochemicals (Cleveland, OH, USA), respectively. Finally, acetonitrile (CHROMASOLV®, gradient grade for HPLC, ≥99.9%) and formic acid (puriss. p.a. ACS; 88-91% (T)) were purchased from Sigma-Aldrich Chemie GmbH. Sample preparation All samples analyzed in this study were prepared in 1:1 acetonitrile/water solutions containing 0.1% v/v formic acid. This solvent composition was adopted because it had previously been used for the preparation of 3-NBN solutions for MAIV analysis.[13] It was not the intention of this preliminary study to optimize the solvent composition used for 3-NBN preparation and V-EASI-MS analysis. However, 100% aqueous solutions were not used because of 3-NBN solubility problems. Future studies will focus on solvent optimization for specific compound groups, e.g. peptides and proteins. All sample solutions were analyzed both with and without 3-NBN. In order to find the amount of 3-NBN that provided the highest sensitivity all samples were prepared in solutions containing 3-NBN concentrations ranging from 0.1 to 1000 μg mL1. Higher concentrations of 3-NBN were not used because this caused frequent clogging of the 100 μm i.d. fused-silica capillary used for introducing and aspirating the liquid sample into the mass spectrometer. Instrumentation For the V-EASI-MS experiments a prototype V-EASI source (Elemental Scientific, Omaha, USA) fitted with a modified glass pneumatic nebulizer (TR-30-C1, Meinhard®, Golden, CO, USA) was installed onto an ion trap mass spectrometer (LCQ Advantage, Thermo Scientific).[7] The pneumatic nebulizer was modified by inserting a 15–20 cm long polyimide-coated fused-silica capillary (100 μm i.d. × 200 μm o.d.) into the nebulizer’s sample uptake channel, until it was aligned with the glass nebulizer spray tip. The fusedsilica capillary was secured in place by tightening a PEEK sleeve around it in the back of the nebulizer. A photograph of the prototype V-EASI source is presented in Fig. 1. This source is a variation of conventional SSI because sample uptake is achieved solely as a result of the Venturi effect (self-aspiration) from the pressure reduction occurring at the nebulizer tip. It should also be noted that no voltage is applied to the nebulizer or to the front of the mass spectrometer and thus no electrospray is occurring during V-EASI-MS. Once the V-EASI source had been mounted onto the mass spectrometer the position of the nebulizer tip in relation to the ion transfer capillary tube entrance was adjusted in order to obtain maximum sensitivity. The optimization was conducted using a sodium trifluoroacetic acid (STFA) solution which is commonly used to tune and calibrate



3-Nitrobenzonitrile for improved sensitivity in SSI-MS where I is the absolute intensity of the analyte ion peak; N stands for the average background levels calculated using MS intensity data spanning 3 m/z units (approximately 45 data points) preceding the analyte peaks; and σ N is the standard deviation of the background data points. In the case of the protein mass spectra, S/N ratios of the deconvoluted protein mass were calculated using MagTran 1.02 mass spectral deconvolution software.[20]

RESULTS AND DISCUSSION Peptide analysis using 3-NBN-assisted V-EASI-MS

Figure 1. Prototype V-EASI source (Elemental Scientific, Omaha, USA) used in this study. ESI-MS systems within the mass range of 100–4000 Da.[18] In general, optimum signals were obtained with the nebulizer tip located 1–3 mm from the ion transfer capillary, but slightly off-axis to it. V-EASI-MS operating conditions included a nitrogen gas flow of 60 psi, along with ion transfer capillary temperatures ranging from 65 to 300°C. Such gas flow rates in combination with the used solvent resulted in a liquid sample uptake flow rate, via the Venturi effect (self-aspiration), between 10 and 15 μL min1. The original ESI source of the instrument was operated at sample flow rates of 10–12.5 μL min1, delivered via a syringe pump. Data analysis S/N ratios have been calculated according to Marginean et al.[19] using the following equation: S=N ¼ I N =3σ N

100

The SSI-MS analysis of a solution containing the peptide MRFA with 3-NBN showed a substantially higher signal intensity for protonated MRFA at m/z 525 (3–4-fold higher; Fig. 2(a)) compared with the signal observed for the same peptide ion in the absence of 3-NBN (Fig. 2(b)). Similarly, the S/N ratio determined in the presence of 3-NBN was 2.8 times greater than the S/N ratio determined in the absence of 3-NBN. However, optimum 3-NBN concentration and ion transfer capillary temperature had to be used in order to realize such an enhancement effect, i.e., 2.5 μg mL1 and 200°C, respectively. No 3-NBN ions were observed during this analysis. It is of interest to note that when using ESI-MS to analyze the same MRFA solutions, with and without 3-NBN, a 2-fold signal enhancement was observed for the protonated peptide molecule in the presence of 200 μg mL1 of 3-NBN. In contrast, the S/N ratio deteriorated from 865 to 428. Overall, comparable intensities were observed for this peptide using both additive-assisted V-EASI-MS and additive-assisted ESI-MS. For more detailed comparisons, S/N data for all the compounds analyzed and techniques used in this study are given in Table 1. V-EASI-MS analysis of a solution containing reduced glutathione (GSH) with 3-NBN resulted in a 1.6-fold enhancement in signal intensity, along with a 1.8-fold

[M+H]

a

80

Relative Abundance

60 40 20 0 100

b

80 60 40 20 0 200

300

400

500

600

700

800

900

1000 m/z




2663

Figure 2. V-EASI mass spectra of MRFA peptide with 2.5 μg mL1 of 3-NBN (a) and without 3-NBN (b), acquired with the ion transfer capillary at 200°C. Signal intensities were normalized to the most intense protonated molecule.

0.1 0.9 13.0 n.d. n.d. n.d. 1.1 n.d. 0.9 0.9 2.0 0.4 0.6 n.d. n.d. n.d. 0.4 n.d. 1.0 0.9 200 125 250 n.a. n.a. n.a. 200 n.a. 250 100 428 280 14 n.a. n.a. n.a. 451 n.a. 208 256 865 119 8.6 n.a. n.a. n.a. 175 n.a. 207 222 2.8 1.8 1.5 3.0 1.5 7.4 7.4 1.0 2.0 0.9


n.a. not analysed; n.d. not determined

3-NBN (μg mL1)

2.5 1.25 0.25 0.1 1 500 2 1.2 250 100 33 139 123 16 58 13 67 295 89 226 93 245 182 48 87 96 495 307 178 210 MRFA GSH TMSe+ 4-hydroxyphenylarsonic acid roxarsone SeGalNAc selenomethionine cytochrome c myoglobin lysozyme

3-NBN S/N (with 3-NBN) (μg mL1) S/N (without 3-NBN) Without 3-NBN With 3-NBN S/N (with 3-NBN) S/N (without 3-NBN)

S/N

ESI-MS V-EASI-MS S/N

With Without 3-NBN 3-NBN Compound

2664

Table 1. S/N ratios and their comparison for V-EASI-MS and ESI-MS analyses with and without addition of 3-NBN

S/N (V-EASI-MS) S/N (ESI-MS)

K. Kanaki and S. A. Pergantis enhancement in S/N ratio, for protonated GSH (m/z 308), compared to its intensity and S/N ratio in the absence of 3-NBN. Figure 3 shows the V-EASI mass spectra obtained for 1 μg mL1 GSH in 1:1 H2O/acetonitrile solutions containing 0.1% v/v formic acid with 1.25 μg mL1 of 3-NBN (Fig. 3(a)) and without 3-NBN (Fig. 3(b)). In the absence of 3-NBN, the optimum intensity for protonated GSH was obtained by setting the ion transfer capillary temperature to 150°C, whilst, in the presence of 1.25 μg mL1 3-NBN, the optimum temperature was 200°C. Also observed were some low intensity K+ and Na+ adducts of the GSH dimers and trimers. We also considered it relevant to conduct a more systematic comparison between 3-NBN-assisted V-EASI-MS and ESI-MS sensitivities, the latter being the predominant mass spectrometric technique used for peptide analysis. However, it must be clarified that such a comparison of sensitivity is a complicated task that can easily lead to erroneous general conclusions. This is because the sensitivities observed in both ionization techniques are compound-dependent, i.e., some compounds give higher intensities with one of the ionization techniques, whereas other compounds give higher intensities with the other.[7,21] Therefore, depending on compound selection one may arrive at false conclusions regarding the overall sensitivity of each of these ionization techniques. In the present study we randomly selected compounds from those we have been analyzing in our laboratory by ESI-MS and V-EASI-MS during the last couple of years. However, this does not necessarily remove all bias from the sensitivity comparison since only a limited number of compounds were analyzed. Another factor that complicates the comparison is the fact that numerous ESI source designs are currently commercially available with even more types in use. In our case the LCQ Advantage ion trap mass spectrometer (Thermo Scientific) is a decade old instrument that has remained equipped with its original ESI source. This potentially means that this older source may not provide comparable sensitivity to that achieved by some of the latest-generation ESI sources used on newer MS instruments. However, we do stress that the aforementioned ESI source is still in excellent working order and provides similar sensitivities to those achieved upon its original installation that fulfilled manufacturer specifications a decade ago. Also confounding the comparison is the fact that, for a given compound, optimum solvent conditions may differ between V-EASI and ESI. Such differences were not investigated in the present study as the focus was to work under the 3-NBN solution conditions that were described by Trimpin and Inutan.[13] Because of all these limitations we have refrained from arriving at any general conclusions regarding the sensitivity and S/N comparisons. We can at best provide a preliminary compound-specific sensitivity comparison between 3-NBN-assisted V-EASI and ESI for our particular instrument setup. For this comparison we used the same 1 μg mL1 GSH solutions with and without 3-NBN. In summary, under optimum ESI conditions which include an ion transfer capillary temperature of 100°C, the protonated GSH molecule (m/z 308) gave intensities of 1.50e5 (S/N 119) and 1.12e5 (S/N 280) with and without 3-NBN, respectively. The optimum 3-NBN concentration in this case was 125 μg mL1. In contrast to



3-Nitrobenzonitrile for improved sensitivity in SSI-MS [M+H] 100

a

80 60

Relative Abundance

40 20 0 100 80

b

60 40 20 0 200

300

400

500

600

700

800

900

Figure 3. V-EASI mass spectra for 1 μg mL1 GSH with 1.25 μg mL1 of 3-NBN (a) and without 3-NBN (b), acquired with the ion transfer capillary at 200°C and 150°C, respectively. Ion intensities in both mass spectra were normalized to the same protonated molecule. V-EASI-MS the presence of the 3-NBN in ESI-MS showed minor signal enhancement and deterioration in S/N ratio. Also, both of these intensity values obtained for ESI are substantially lower than those observed using additive-assisted V-EASI, which gave 4.0e5 (S/N 245) with 3-NBN, and 2.6e5 (S/N 139) without 3-NBN. Overall, the S/N ratios obtained for additiveassisted V-EASI-MS and conventional ESI-MS were similar, i.e. 245 vs 280. However, it should be mentioned that the sample uptake rates were slightly different for each ionization technique, i.e. 15 μL min1 for V-EASI and 10–12.5 μL min1 for ESI. Protein analysis using 3-NBN-assisted V-EASI-MS




2665

In this study three proteins (horse heart cytochrome c, horse heart myoglobin and chicken egg lysozyme) were analyzed using 3-NBN-assisted V-EASI-MS, and significant signal enhancement was observed for two of them (cytochrome c and myoglobin). More specifically, a 4–5-fold signal enhancement, along with a 2-fold enhancement in S/N ratio, were observed for myoglobin in the presence of 250 μg mL1 of 3-NBN (Fig. 4). In the absence of 3-NBN a slight bimodal charge distribution around the +13 (m/z 1305) and +10 (m/z 1697) charge states was observed (Fig. 4(b)), whereas, in the presence of 3-NBN, a single distribution appeared around the +11 charge state (Fig. 4(a)). Even though it is not clear why this occurs it signifies that protein ionization mechanisms must differ between bare V-EASI and 3-NBN-assisted SSI. In comparison to the work of Trimpin and Inutan for the MAIV analysis of myoglobin,[13] some differences are evident. These include the much lower source temperature of 50°C used in their study, and the resulting higher charge states centred mainly on +13 to +16, going up to a maximum of about +23 that they observed. The most obvious difference between the two techniques (MAIV and 3-NBN-assisted V-EASI) is the

way the samples are introduced into the mass analyzer. In the case of Trimpin and Inutan it was a 3-NBN-precoated paper strip (dry) to which the myoglobin solution was added, and this was held against a modified capillary inlet tube at a source temperature of 50°C. Because analyte and matrix were in dry form, and thus solvent evaporation was not required, we consider this to be the reason for which significantly lower source temperatures can be used in MAIV, whereas, in the case of 3-NBN-assisted V-EASI-MS, elevated temperatures of 150–300°C are required to assist in solvent evaporation from the aerosol droplets in order for bare gas-phase ions to form. However, in all studies so far the elevated ion transfer capillary temperatures used in V-EASI and 3-NBN-assisted V-EASI do not seem to contribute significantly to analyte degradation. In fact, we have observed that for a given ion transfer capillary temperature less thermal degradation is observed in V-EASI than in ESI.[7] ESI-MS analysis of the same myoglobin solutions revealed the absence of signal-enhancing effects when adding 3-NBN; in fact a slight suppression was observed in the presence of 250 μg mL1 of 3-NBN. Upon comparing the sensitivities observed in ESI and V-EASI, it was clear that ESI offered higher sensitivity compared to conventional V-EASI-MS, but only approximately half of the signal intensity when compared to 3-NBN-assisted V-EASI-MS. This indicates a significant improvement in sensitivity for the analysis of myoglobin when using 3-NBN-assisted V-EASI-MS compared to ESI-MS. However, S/N ratios are similar for both, i.e., 178 for additive-assisted V-EASI-MS and 208 for conventional ESI-MS. Similar observations were made during the 3-NBN-assisted V-EASI-MS analysis of 10 μg mL1 of cytochrome c in which case a 2-fold signal enhancement was recorded in the presence of 1.2 μg mL1 of 3-NBN, whereas the S/N ratio remained almost the same (Supplementary Fig. S2, Supporting Information). Once again a high ion transfer capillary temperature (300°C) was found to be optimum.

K. Kanaki and S. A. Pergantis 100

1542.5

9.73E5

a

1696.7

1414.1

16957.1

1305.5 1885.0

50

Relative Abundance

1212.3 1131.5 0 100

b

16958.3

50

+13 +14 +15 1305.5 +16 1131.5 1212.3 1061.1 0 1000

1100

1200

1300

+12 1414.3

1400

+11 1542.8

1500

1600

+10 1696.8

1700

+9 1885.1 1800

1900

Figure 4. V-EASI mass spectra for myoglobin (20 μg mL1) with 250 μg mL1 of 3-NBN (a) and without 3-NBN (b), acquired with the ion transfer capillary at 300°C. Insets show the molecular mass of the protein as it was calculated using MagTran deconvolution software.[20] Ion signals in both mass spectra were normalized to the same most abundant ion. Finally, in the case of 10 μg mL1 of lysozyme no enhancements in S/N ration or signal were observed for 3-NBN-assisted V-EASI-MS. However, higher 3-NBN concentrations (>100 μg mL1) resulted in the peak distribution shifting to higher charge states and a broader charge distribution (Supplementary Fig. S3, Supporting Information), without any significant signal intensity changes. Similar results were also obtained using 3-NBN-assisted ESI-MS. Selenium and arsenic species analysis using 3-NBN-assisted V-EASI-MS

2666

V-EASI-MS analysis of 0.37 μg mL1 of the cationic trimethylselenonium ion (TMSe+), a Se species of biological relevance,[22] revealed signal enhancement for its molecular ion (m/z 125) ranging from 2- to 3-fold in the presence of 0.25 μg mL1 of 3-NBN and an ion transfer capillary temperature of 200°C. Under such conditions, the S/N ratio was also enhanced by a factor of 1.5. Figure 5 shows the effect of 3-NBN concentration and ion transfer capillary temperature on the molecular ion signal intensity when analyzed using V-EASI-MS. It is observed that for any given ion, at transfer capillary temperature from 65–300°C, the addition of 3-NBN to the TMeSe+-containing solution resulted in increased molecular ion intensity. Similar graphs for some of the other analyzed compounds (GSH, 4-hydroxyphenylarsonic acid and selenomethionine) are presented in Supplementary Fig. S1 (see Supporting Information). The selenosugar 2-acetamido-2-deoxy-1-seleno-β-D-galactopyranoside (SeGalNAc), which is the main selenium metabolite found in human urine,[23] showed substantial signal and S/N enhancement in 3-NBN-assisted V-EASI-MS. Almost an order of magnitude higher intensity, accompanied by a 7.4 higher S/N ratio, was observed for its protonated molecule (m/z 300) in the presence of 500 μg mL1 of 3-NBN using an ion transfer capillary temperature of 200°C (Fig. 6).


Figure 5. Effect of 3-NBN concentration and ion transfer capillary temperature on the signal intensity observed for the TMeSe+ ion analyzed using V-EASI-MS and 3-NBN-assisted V-EASI-MS. Another selenium compound of biological interest, analyzed in the present study, is selenomethionine (SeMet). The V-EASI-MS analysis of SeMet in 1:1 aqueous acetonitrile solution containing 0.1% v/v formic acid, with and without 3-NBN, resulted in the mass spectra presented in Fig. 7. In these mass spectra it is observed that the protonated molecule of SeMet at m/z 198 exhibits an intensity of 6.5e4 in the absence of the 3-NBN matrix; whereas in the presence of 2 μg mL1 of 3-NBN its intensity undergoes a 3-fold increase, i.e. 2.04e5. S/N enhancement was also substantial, i.e. increased 7.4 times. The optimum ion transfer capillary temperature during this experiment was 150°C. Subsequent analysis of the same compound using ESI-MS resulted in an almost 2-fold enhancement in the protonated molecule signal in the presence of 200 μg mL1 of 3-NBN, i.e., resulting in a protonated molecule with an intensity of 1.6e5 using an ion transfer capillary temperature of 200°C. However, the S/N ratio in ESI-MS decreased in the presence of 3-NBN by 2.5



3-Nitrobenzonitrile for improved sensitivity in SSI-MS 100

300.3 2.54E5

a

c 298.2

Relative Abundance

50

296.4

0 100

302.3

300.3

3.10E4

b 298.3 50

296.2 302.1

0 290

292

294

296

298

300

302

304

306

308

310

Figure 6. V-EASI mass spectra for 1 μg mL1 SeGalNAc with 500 μg mL1 3-NBN (a) and without 3-NBN (b), acquired with the ion transfer capillary at 200°C. The shown mass spectral region corresponds to the protonated molecule isotopic pattern. In (c) the theoretical isotopic pattern is provided.

a

b

Figure 7. V-EASI mass spectra for 1.6 μg mL1 SeMet with 2 μg mL1 3-NBN (a) and without 3-NBN (b), acquired with the ion transfer capillary at 150°C. In both mass spectra intensities were normalized to the same most intense ion (2.04e5).


In contrast to the 3-NBN-assisted V-EASI-MS analysis of most of the previously described compounds no signal enhancement was observed when analyzing the zwitter ion arsenobetaine. This finding is in line with the observation that analyte signal enhancement caused by 3-NBN is compoundspecific. Other arsenic compounds, including arsenic animal-feed additives, were also analyzed using 3-NBN-



2667

times. The S/N comparison between 3-NBN-assisted V-EASI-, ESI- and 3-NBN-assisted ESI-MS revealed that the highest SeMet protonated molecule S/N was obtained using 3-NBNassisted V-EASI-MS (Table 1). Also, optimum results for V-EASI and ESI were obtained at different ion transfer capillary temperatures, additive concentrations, capillary voltage and tube lens voltage.

K. Kanaki and S. A. Pergantis assisted V-EASI-MS. More specifically, for the analysis of 0.1 μg mL1 4-hydroxyphenylarsonic acid, a significant 5–6-fold enhancement in signal was observed in the presence of 0.1 μg mL1 3-NBN, using an ion transfer capillary temperature of 300°C (Supplementary Fig. S4, Supporting Information), corresponding to a 3-fold improvement in S/N ratio. Finally, when analyzing 0.5 μg mL1 of 3-nitro4-hydroxyphenylarsonic acid (roxarsone) with 1 μg mL1 3-NBN-assisted V-EASI-MS, a 1.5-fold enhancement in signal and S/N ratio was observed for its protonated molecule when using an ion transfer capillary temperature of 250°C and 1 μg mL1 of 3-NBN.

unravel the mechanism of function of 3-NBN as well as its effect on mixtures and interfering substances, especially in combination with high-performance liquid chromatography.

Acknowledgements The authors wish to acknowledge co-funding of this research by the European Union-European Regional Development Fund and Greek Ministry of Εducation/EYDE-ETAK through program ESPA 2007-2013/EPAN II/Action “SYNERGASIA” (09ΣΥΝ-13-832). The authors also thank Elemental Scientific and Meinhard for providing nebulizers and a prototype sonic-spray ionization source for this study.

CONCLUSIONS

2668

This study demonstrates an approach for enhancing the sensitivity of V-EASI-MS using 3-NBN as a solution additive for a wide range of compounds, including low molecular weight organometalloids, peptides, as well as high molecular weight biomolecules (e.g. proteins). In order to explain the observed signal enhancement for 3-NBN-assisted V-EASI-MS compared to bare V-EASI-MS, as well as the observed minor or insignificant signal changes in 3-NBN-assisted ESI-MS and bare ESI-MS, we have considered the properties of 3-NBN to sublime and to exhibit triboluminescence upon crystal fracturing. In the case of Trimpin and Inutan’s study on the development of MAIV, these properties of 3-NBN were highlighted as reasons for ion formation and ion detection directly from solids and liquids pressed against the inlet of the mass spectrometer. These properties along with the vacuum of the mass spectrometer are possible reasons for the formation of gas-phase analyte ions directly from solids and solutions without the need for inlet heating. Based on the explanation provided by Trimpin and Inutan for ion formation in MAIV with 3-NBN,[13,24] we attempt to discuss the varying effects of the additive in ESI- and V-EASI-MS. In the case of ESI, generation of gas-phase ions takes place mainly exterior to the mass analyzer, in the atmospheric pressure region of the ion source, where the high voltage applied to the sprayed analyte solution causes the formation of gas-phase analyte ions from charged droplets. As a result of the applied high voltage the charged droplets may also be depleted in 3-NBN. Thus analyte and 3-NBN do not remain together long enough in the droplets in order for 3-NBN to have any substantial effect on analyte ion formation. On the other hand, although the mechanism for ion formation in SSI is not yet well understood, it has been suggested that bare ions are mainly formed in the intermediate pressure region of the mass spectrometer.[9] As a result the SSI technique makes more efficient use of the sublimation and triboluminescent properties of 3-NBN. This is because 3-NBN remains for a longer period of time together with the analyte in the sonicspray droplets, and well into their travel through the heated ion transfer capillary and into the vacuum region of the mass spectrometer. In conclusion, the use of 3-NBN with V-EASI-MS gives rise to a novel V-EASI-MS technique, which has demonstrated considerable signal enhancement for most of the compounds analyzed, thus improving its competitiveness towards the well-established and dominating ESI-MS technique. In order to further pursue its development we will be attempting to


REFERENCES [1] A. Hirabayashi, M. Sakairi, H. Koizumi. Sonic spray ionization method for atmospheric pressure ionization mass spectrometry. Anal. Chem. 1994, 66, 4557. [2] R. Haddad, R. Sparrapan, T. Kotiaho, M. N. Eberlin. Easy ambient sonic-spray ionization-membrane interface mass spectrometry for direct analysis of solution constituents. Anal. Chem. 2008, 80, 898. [3] R. Haddad, R. Sparrapan, M. N. Eberlin. Desorption sonic spray ionization for (high) voltage-free ambient mass spectrometry. Rapid Commun. Mass Spectrom. 2006, 20, 2901. [4] A. C. H. F. Sawaya, P. V. Abdelnur, M. N. Eberlin, S. Kumazawa, M.-R. Ahn, K.-S. Bang, N. Nagaraja, V. S. Bankova, H. Afrouzan. Fingerprinting of propolis by easy ambient sonic-spray ionization mass spectrometry. Talanta 2010, 81, 100. [5] V. G. Santos, T. Regiani, F. F. G. Dias, W. Romão, J. L. P. Jara, C. F. Klitzke, F. Coelho, M. N. Eberlin. Venturi easy ambient sonic-spray ionization. Anal. Chem. 2011, 83, 1375. [6] N. V Schwab, A. M. Porcari, M. B. Coelho, E. M. Schmidt, J. L. Jara, J. V Visentainer, M. N. Eberlin. Easy dual-mode ambient mass spectrometry with Venturi self-pumping, canned air, disposable parts and voltage-free sonic-spray ionization. Analyst 2012, 137, 2537. [7] M. M. Antonakis, A. Tsirigotaki, K. Kanaki, C. J. Milios, S. A. Pergantis. Bipolar mass spectrometry of labile coordination complexes, redox active inorganic compounds, and proteins using a glass nebulizer for sonic-spray ionization. J. Am. Soc. Mass Spectrom. 2013, 24, 1250. [8] N. V Schwab, M. N. Eberlin. Mass spectrometry made easy: the quest for simplicity. Drug Test. Anal. 2013, 5, 137. [9] A. Stindt, C. Warschat, A. Bierstedt, U. Panne, J. Riedel. Characterization of an inexpensive sonic spray ionization source using laser induced fluorescence imaging and mass spectrometry. Eur. J. Mass Spectrom. 2014, 20, 21. [10] L. W. Zilch, J. T. Maze, J. W. Smith, G. E. Ewing, M. F. Jarrold. Charge separation in the aerodynamic breakup of micrometer-sized water droplets. J. Phys. Chem. A 2008, 112, 13352. [11] L. W. Zilch, J. T. Maze, J. W. Smith, M. F. Jarrold. Freezing, fragmentation, and charge separation in sonic sprayed water droplets. Int. J. Mass Spectrom. 2009, 283, 191. [12] V. I. Parfenyuk. Surface potential at the gas–aqueous solution interface. Colloid J. 2002, 64, 588. [13] S. Trimpin, E. D. Inutan. New ionization method for analysis on atmospheric pressure ionization mass spectrometers requiring only vacuum and matrix assistance. Anal. Chem. 2013, 85, 2005.



3-Nitrobenzonitrile for improved sensitivity in SSI-MS [14] J. J. Mitchell, H. D. Deveraux. Atmosphere: Role of detectors in gas chromatography. Anal. Chim. Acta 1978, 100, 45. [15] L. M. Sweeting. Triboluminescence with and without air. Chem. Mater. 2001, 13, 854. [16] L. M. Sweeting, M. L. Cashel, M. M. Rosenblatt. Triboluminescence spectra of organic crystals are sensitive to conditions of acquisition. J. Lumin. 1992, 52, 281. [17] P. Traar, F. Belaj, K. A. Francesconi. Synthesis of methyl 2acetamido-2-deoxy-1-seleno-β-D-gluco- and galactopyranoside: Selenium metabolites in human urine. Aust. J. Chem. 2004, 57, 1051. [18] M. Moini, B. L. Jones, R. M. Rogers, L. Jiang. Sodium trifluoroacetate as a tune/calibration compound for positive- and negative-ion electrospray ionization mass spectrometry in the mass range of 100–4000 Da. J. Am. Soc. Mass Spectrom. 1998, 9, 977. [19] I. Marginean, K. Tang, R. D. Smith, R. T. Kelly. Picoelectrospray ionization mass spectrometry using narrow-bore chemically etched emitters. J. Am. Soc. Mass Spectrom. 2014, 25, 30. [20] Z. Zhang, A. G. Marshall. A universal algorithm for fast and automated charge state deconvolution of electrospray massto-charge ratio spectra. J. Am. Soc. Mass Spectrom. 1998, 9, 225.

[21] H. T. Björkman, P.-O. Edlund, S. P. Jacobsson. Sonic spray ionization interface for liquid chromatography–mass spectrometry. Anal. Chim. Acta 2002, 468, 263. [22] D. Kuehnelt, D. Juresa, N. Kienzl, K. A. Francesconi. Marked individual variability in the levels of trimethylselenonium ion in human urine determined by HPLC/ICPMS and HPLC/vapor generation/ICPMS. Anal. Bioanal. Chem. 2006, 386, 2207. [23] S. Letsiou, V. Nischwitz, P. Traar, K. A. Francesconi, S. A. Pergantis. Determination of selenosugars in crude human urine using high-performance liquid chromatography / atmospheric pressure chemical ionization tandem mass spectrometry. Rapid Commun. Mass Spectrom. 2007, 21, 343. [24] S. Trimpin, E. D. Inutan. Matrix assisted ionization in vacuum, a sensitive and widely applicable ionization method for mass spectrometry. J. Am. Soc. Mass Spectrom. 2013, 24, 722.

SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher’s website.

2669




ionization time-of-flight mass spectrometry of peptides.

Ionization methods in mass spectrometry.

Electrospray ionization mass spectrometry.

Improving detection sensitivity of amino acids in thyroid tissues by using phthalic acid as a mobile phase additive in hydrophilic interaction chromatography-electrospray ionization-tandem mass spectrometry.

Preparation of plutonium-242 for use as an alpha spectrometry and mass spectrometry tracer.

DRILL: An Electrospray Ionization-Mass Spectrometry Interface for Improved Sensitivity via Inertial Droplet Sorting and Electrohydrodynamic Focusing in a Swirling Flow.

Solid-phase extraction and nanoflow liquid chromatography-nanoelectrospray ionization mass spectrometry for improved global urine metabolomics.

Comparison of the sensitivity of mass spectrometry atmospheric pressure ionization techniques in the analysis of porphyrinoids.

Charge-state reduction with improved signal intensity of oligonucleotides in electrospray ionization mass spectrometry.

Applications of chemical ionization mass spectrometry.

An improved scheme of leucine derivative fragmentation in mass spectrometry.

Improved Sensitivity for the Qualitative and Quantitative Analysis of Active Ricin by MALDI-TOF Mass Spectrometry.

Detection mass bias in atmospheric pressure ionization mass spectrometry.

Improving the Sensitivity of Matrix-Assisted Ionization (MAI) Mass Spectrometry Using Ammonium Salts.

Electrostatic spray ionization mass spectrometry imaging.

Active hydrogen by chemical ionization mass spectrometry.

Microfabricated ultrarapid desalting device for nanoelectrospray ionization mass spectrometry.

Extractive electrospray ionization mass spectrometry for uranium chemistry studies.

atmospheric-pressure-ionization mass spectrometry.

Pulse-heating ionization for protein on-chip mass spectrometry.

Characterization of mouse switch variant antibodies by matrix-assisted laser desorption ionization mass spectrometry and electrospray ionization mass spectrometry.

Characterization of anthraquinone-2-carbonyl chloride as an alcohol derivatization reagent for negative ion chemical ionization mass spectrometry.

Rapid high mass resolution mass spectrometry using matrix-assisted ionization.

Soft supercharging of biomolecular ions in electrospray ionization mass spectrometry.