Fast sample preparation and liquid chromatography–tandem mass spectrometry method for assaying cell lysate acetylcholine

Available online at www.sciencedirect.com

Journal of Chromatography A, 1183 (2008) 100–107

Fast sample preparation and liquid chromatography–tandem mass spectrometry method for assaying cell lysate acetylcholine Nils Helge Schebb a , Daniel Fischer b , Eva-Maria Hein c , Heiko Hayen c , Josef Krieglstein b , Susanne Klumpp b , Uwe Karst a,∗ a

Westf¨alische Wilhelms-Universit¨at M¨unster, Institut f¨ur Anorganische und Analytische Chemie, Corrensstraße 30, 48149 M¨unster, Germany b Westf¨ alische Wilhelms-Universit¨at M¨unster, Institut f¨ur Pharmazeutische und Medizinische Chemie, Hittorfstraße 58-62, 48149 M¨unster, Germany c ISAS – Institute of Analytical Science, Bunsen-Kirchhoff-Straße 11, 44139 Dortmund, Germany Received 31 August 2007; received in revised form 4 January 2008; accepted 7 January 2008 Available online 24 January 2008

Abstract A fast liquid chromatography–tandem mass spectrometry (LC–MS/MS) method was developed for the analysis of acetylcholine (ACh) in cultured cells. [2 H4 ]Acetylcholine (ACh-d4 ) was used as an internal standard for calibration. ACh was extracted from the cell lysate with acetonitrile (ACN)/water (80/20, v/v) and the crude extract was analyzed without further purification. Isocratic hydrophilic interaction chromatography (HILIC) with (10 mM) ammonium formate/ACN (35/75, v/v) as mobile phase was used for separation. ACh was eluted within 5 min and detected using electrospray-MS/MS in the positive ion mode. The limit of detection (LOD) was found to be 1.5 fmol (0.3 nmol/L) ACh with a S/N ratio of 3:1. The approach was used for the measurement of ACh in undifferentiated SN56 cells and the ACh content was determined to be 1272 ± 109 pmol/mg protein. © 2008 Elsevier B.V. All rights reserved. Keywords: Acetylcholine; SN56 cells; Tandem mass spectrometry; HPLC; HILIC; (3-Carboxypropyl)-trimethylammonium

1. Introduction Acetylcholine (ACh) is an essential messenger, which is involved in neurotransmission in both the peripheral and central nervous system (CNS). It has an effect on alertness, memory and learning. Decreased levels of ACh are found in patients with Alzheimer’s disease [1]. In order to develop drugs against this disease, the ongoing processes in the neuronal cells, which lead to a depletion of ACh have to be understood. Cell cultures are a common and useful system to investigate the molecular biology of tissues. To investigate the role of the signal cascades or the effect of potential drugs on the ACh synthesis in the cells, the ACh concentration in the cells has to be determined. Analyses of ACh in biological samples have been performed mostly by liquid chromatography (LC) connected to a combination of enzymatic reactor and electrochemical detection (ED)

∗

Corresponding author. E-mail address: [email protected] (U. Karst).

0021-9673/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2008.01.033

[2,3] or to electrospray-mass spectrometric detection (ESI-MS) [4–13]. A radioimmunoassay method has also been used for this purpose [14]. In contrast to LC-ED, where ACh must be converted into hydrogen peroxide using a post-column enzyme reactor, LC–MS provides lower limits of detection (LODs), allows a direct detection of ACh and is more robust. As ACh is a small, charged molecule (Fig. 1), it shows no or poor retention in reversed-phase (RP) chromatography. Therefore, suitable LC separation techniques are ion-pair chromatography (IPC) [2,4–7] or cation-exchange chromatography (IEC) [3,12,13]. For these separation techniques, ion-pair reagents or concentrated buffers must be used, thus possibly leading to suppression in the ESI process. Hydrophilic interaction chromatography (HILIC) has also been used for the analysis of ACh [8–11,17]. In HILIC, aqueous mobile phases containing usually more than 50% of organic solvent and a polar stationary phases are used. Alpert [15] suggested that the retention mechanism involves partitioning of the analyte between the mobile phase and a layer of mobile phase enriched with water on the surface of the stationary

N.H. Schebb et al. / J. Chromatogr. A 1183 (2008) 100–107

101

Fig. 1. Different fragmentation pathways of (I) acetylcholine (ACh), (II) [2 H4 ]acetylcholine (I.S.) and a suggestion of the fragmentation of (III) (3carboxypropyl)trimethylammonium (CPMA). The empirical formula and the calculated exact mass are shown for all fragments.

phase. Hemstrom and Irgum [16] reviewed the different column materials, surface chemistries and applications of HILIC. Due to the high content of organic solvent, the mobile phase in HILIC is highly volatile and well suited for compound ionisation by ESI-MS, thus leading to excellent sensitivity. The most widely described LC–MS methods determine ACh in brain microdialysates [5,11–13] or demonstrate the sensitivity of the technique only by using standards [8,9]. However, the matrix of cell lysates originating from tissue or cell culture samples is more complex, and a good sample preparation strategy as well the use of an appropriate internal standard are essential. Only a few LC–MS methods are described for the determination of ACh in cell lysates: Acevedo et al. [4] determined the ACh release from rat pheochromocytoma cells (PC12) with IPC–ESIMS. Koc et al. [10] described a HILIC–ESI-MS method for the determination of different choline compounds in tissues includ-

ing lysates from cultured cells. Furthermore, Reubsaet et al. [6] presented a RP-ion-pair chromatography-ESI-MS/MS method for the determination of ACh in bovine cornea. Only two of these methods [4,10] use an isotope-labelled internal standard (I.S.), allowing an accurate determination of ACh even if unknown matrix compounds co-elute with the analyte. However, for both methods, a time-consuming sample preparation is needed and the separation requires more than 20 min. In cell culture experiments, the samples have to be analyzed after a defined time of growing and frequently, many cell culture experiments are carried out simultaneously. Therefore, sample preparation should be less laborious and as fast as possible, to allow the analysis of a large number of samples. For this reason, we developed a rapid and robust LC–MS/MS method for the determination of ACh in cultured cells using a deuterated I.S. A low limit of quantification (LOQ) allows omitting time-consuming preconcentration steps, which are used

102

N.H. Schebb et al. / J. Chromatogr. A 1183 (2008) 100–107

in the established methods [4,10] for determination of ACh in cell lysate. The HILIC separation as online sample clean up allows the direct analysis of the cell lysate. Furthermore, the use of an isotope-labelled internal standard assures the accuracy of the ACh analysis. The development and application of this method is presented within this work. 2. Experimental 2.1. Chemicals [1,1,2,2-2 H4 ]Acetylcholine chloride (Ach-d4 ; 99.1 at.% 2 H) was delivered by CDN Isotopes (Quebec, Canada), neostigmine sulphate 0.5 mg/mL was purchased from DeltaSelect (Dreieich, Germany), and acetonitrile (ACN) in gradient grade quality was delivered by Merck (Darmstadt, Germany). Gentamycin, Dulbecco’s modified eagle’s medium (DMEM) and fetal calf serum (FCS) were obtained from PAA (Marburg, Germany). Readymade trypsin/EDTA solution (0.05%/0.02%, w/v) in phosphate-buffered saline (PBS) was delivered by Biochrom (Berlin, Germany). Acetylcholine chloride (99%), ammonium formate, (3-carboxypropyl)trimethylammonium (technical grade), formic acid and all other chemicals were obtained from Sigma–Aldrich (Steinheim, Germany). Purified water for HPLC analyses and sample dilution was generated by a Milli-Q Gradient A 10 system and filtered through a 0.22-␮m Millipak 40 filter unit (Millipore, Billerica, MA, USA). 2.2. Cell culture SN56.B5.G4 (SN56) is a cholinergic cell line originally generated from murine cholinergic septal neurons fused with N18TG2 neuroblastoma cells [17]. The SN56 cells were maintained in DMEM supplemented with 10% (v/v) FCS and 50 ␮g/mL gentamycin at 37 ◦ C in a humidified 5% CO2 /95% air atmosphere. Media were changed every 2–3 days and cells were splitted every 7 days. In order to determine the cellular ACh content, cells were seeded at a density of 640 000 cells/dish (60 mm × 15 mm, Greiner, Frickenhausen, Germany) and were grown for 4 days. The media were changed every day. 2.3. Preparation of stock solutions A stock solution of ACh and ACh-d4 as I.S. was prepared in 80% ACN and 20% water. This solution was stored at −20 ◦ C and was stable for at least 3 weeks. Fresh dilutions were prepared daily. For calibration, ACh was dissolved and diluted with 80% ACN and 20% water containing 100 nM of I.S. and was diluted at concentrations ranging from 0.01 nM to 10 ␮M. In the solutions of I.S. up to 1000 nM, no signals corresponding to ACh were detected. 2.4. Sample preparation For quantification of ACh, cell culture medium was removed and SN56 cells were washed with PBS (1.05 mM KH2 PO4 ,

3.71 mM Na2 HPO4 , 154 mM NaCl, pH 7.4) at room temperature. Cells were detached using a trypsin/EDTA solution (0.5 mL/dish; 0.05%/0.02% (w/v) in PBS). The reaction was stopped after 5 min by addition of buffered washing solution (BWS; 1 mL/dish; 90 mM NaCl, 30 mM KCl, 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 0.02 mM EDTA, 32 mM sucrose, 0.015 mM neostigmine, pH 7.4) plus 10% FCS. Cells from five dishes were combined and transferred into a 15-mL tube. After centrifugation (7 min, 700 × g, room temperature) cells were washed with 3 mL BWS. After centrifugation (7 min, 700 × g, room temperature) cells were resuspended in 2 mL BWS, splitted into two 1.5 mL reaction tubes and centrifuged (7 min, 700 × g, room temperature). Cells from one tube were used for protein quantification, while the second aliquot was used for quantification of the cells ACh content. For protein quantification, cells were resuspended in 10 mM HEPES, pH 7.4, 2 mM EDTA, homogenized and sonicated. The protein concentration was determined according to the method of Lowry [18] using bovine serum albumin (BSA) as a standard. For quantification of ACh, cells were resuspended in ACN/water (80%/20%, v/v) containing 0.015 mM neostigmine and centrifuged (20 min, 20 000 × g, 4 ◦ C). The supernatant containing ACh was diluted with 80% ACN and 0.015 mM neostigmine to a corresponding cell protein content of 1.0 mg/mL. 100 microliters of this solution was diluted with 900 ␮L of 80% ACN/20% water containing 111.11 nM I.S., leading to a final concentration of 100 nM I.S. in the sample. This solution was used for HPLC–MS/MS analyses directly or stored at −80 ◦ C. 2.5. LC–MS/MS instruments For LC–MS/MS analyses, a Shimadzu (Langenfeld, Germany) HPLC system was used, consisting of two HPLC pumps (LC-10ADVP), a degasser (DGU-14A), an autosampler (SILHT-A), a column oven (CTO-10AVP) and an UV detector (SPD-10AVVP) operating at 254 nm connected to an API 2000 tandem mass spectrometer (Applied Biosystems, Darmstadt, Germany) equipped with a turbo ionspray (pneumatically assisted ESI) source. The software used for controlling LC and MS was Analyst 1.4.1 (Applied Biosystems). For separation, a zwitterionic ZIC-HILIC column (Sequant, Haltern, Germany) with the dimensions 150 mm × 2.1 mm, particle size 3.5 ␮m and pore size of 20 nm was selected. 2.6. LC-separation conditions The injection volume was 5 ␮L and the sample was eluted isocratically in HILIC mode at 0.15 mL/min for 10 min using 65% ACN and 35% aqueous phase containing 10 mM ammonium formate and 0.02% formic acid. After every 10 samples, the following binary gradient was used to rinse those matrix compounds from the column, which may not elute under isocratic conditions: from 0–1 min linear from 75% to 40% ACN, 1–5 min isocratic, 5–7 min linear from 40% to 75% ACN and then isocratic for 10 min to equilibrate the column and re-

N.H. Schebb et al. / J. Chromatogr. A 1183 (2008) 100–107

103

establish the HILIC conditions. To prevent sample carryover, the autosampler was rinsed before every injection with 2 mL of 50% ACN.

3. Results and discussion

2.7. Mass spectrometric conditions

The aim of our studies was to establish both a rapid and robust method for the quantification of ACh in cultured cells. We focused not only on developing an adequate method for separation and detection, but also on a fast and reproducible procedure for sample preparation. In previously reported methods, cells were washed in physiological salt solution for up to 1 h [10,19,20], salt was removed and methanol or methanol plus formic acid were added to detach the cells. After centrifugation the cells were resuspended in chloroform/water to extract ACh [10,17,20]. Alternatively, Pedersen et al. [19] directly used methanol to detach and break the cells, then added formic acid plus acetone. After centrifugation the supernatant was dried and used for quantification of ACh. We applied a fast approach without time consuming extraction and drying steps: After washing the cells, they were detached from the dishes with trypsin solution. Subsequently, trypsin was inhibited by adding FCS in BWS. After two washing steps (FCSfree BWS), the cells were resuspended in 80% ACN/20% water (v/v). This solvent destroys the cell membranes, thereby releasing ACh. Denatured proteins were removed by centrifugation and ACh could be measured directly from the supernatant fraction. The supernatant containing ACh was diluted 1:10, spiked with the I.S. and directly used for HPLC–MS/MS. This way, the method is not only faster, but detaching the cells by trypsin in contrast to scrape them off also enhances the reproducibility of the method. Furthermore, ACh is directly extracted into the mobile phase of the LC, which is beneficial for the chromatographic separation. A similar sample preparation strategy of cell lysates has been already described for other analytes, e.g., homocysteine [22].

The analytes were ionised in the interface with an ionspray voltage of 1500 V, using 40 psi nebulizer gas and 85 psi drying gas at a temperature of 450 ◦ C. The analytes were detected in the positive ion mode using tandem mass spectrometry in the multiple reaction monitoring mode (MRM). The potentials were optimized under LC conditions for ACh, thus leading to the following values: declustering potential (DP) 24 V, entrance potential (EP) 6.5 V and focussing potential (FP) 400 V, collision energy 19 eV, and collision exit potential 4 V. Two MRM transitions were monitored with a dwell time of 20 ms for ACh (146 → 87, 60) and for ACh-d4 (I.S.) (150 → 91, 60). The first transition was used for quantification because of its higher intensity. For MS/MS spectra (40–160 u) and full scan spectra (100–1000 u), the same source parameters were used. 2.8. Fourier transform ion cyclotron resonance-mass spectrometry conditions ESI-Fourier transform ion cyclotron resonance-mass spectrometric detection (FTICR-MS) experiments were carried out using a LTQ-FT Fourier transform ion cyclotron resonance hybrid mass spectrometer (Thermo Fisher Scientific, Bremen, Germany), equipped with a 7.0 T actively shielded superconducting magnet and a nano-ESI source. The instrument was operated in the positive ionisation mode. Ion transmission into the linear trap and signal intensity was automatically optimized for maximum ion signal intensity of the analytes. The parameters were: Source voltage 0.8–0.95 kV, capillary voltage 35 V, capillary temperature 275 ◦ C, and tube-lens voltage 110 V. For MS/MS experiments, carried out using collision-induced dissociation (CID), the resolving power of the FTICR mass analyser was set to 50 000 (full width at half maximum (FWHM) at m/z = 400) and a mass range from m/z 50–180 was scanned. The instrument was calibrated externally with the Thermo-Cal-Mix (MRFA, Ultramark 1621 and caffeine in 50% methanol/50% water (v/v) containing 1% acetic acid). The deviation of measured m/z of the fragments to the theoretical m/z were calculated with the following formula: Relative mass deviation (ppm) = (m/zexperimental − m/zcalculated )/m/zcalculated × 1 × 106 . 2.9. Calculation of ACh content The ACh content of the samples was calculated directly by comparison of the ACh peak area detected with that of the I.S. in MRM mode. For calibration, the ACh to I.S. ratios were fitted in a linear way (weighted by standard deviation), using zero as datapoint, with Origin 6.1 (OriginLab, USA).

3.1. Sample preparation

3.2. Chromatographic separation In the development of the chromatographic separation, the main emphasis was set on the separation of ACh from matrix compounds. The cell lysate was directly injected after denaturation of proteins with 80% ACN. As shown in previous studies [23], this process is insufficient to remove all proteins from the sample, thus leading to irreproducible ionisation in ESIMS. Besides proteins, involatile salts, high concentrations of the added ACh-esterase (AChE) inhibitor neostigmine and endogenous compounds originating from the cytosol of the cells have to be separated from ACh to assure an efficient and reproducible ionisation process. HILIC was selected that are separation technique that allows effective retention of polar or even charged substances like ACh [16] without the addition of ion-pair reagents or high concentrations of buffer salts that are required for RP or IEC separations. Furthermore, the mobile phase in HILIC contains high concentrations of organic solvent, which is favourable for ESI. Different HILIC methods for the separation of ACh have been described so far, using either unmodified silica [10,9], diol [11] or aminopropyl and amide [8] stationary phases. As known

104

N.H. Schebb et al. / J. Chromatogr. A 1183 (2008) 100–107

in Fig. 2) shows exactly the same retention time and peak shape as ACh. If any substance should co-elute with ACh and suppress or enhance its ionisation in the ESI process, the I.S. would be affected to the same extent and the ratio of analyte to internal standard would remain constant, thus allowing an accurate determination of the ACh content. 3.3. Mass spectrometry

Fig. 2. LC–MS/MS chromatograms of a typical separation of a SN-56 cell lysate sample with 100 nM I.S. added. Top: full scan chromatogram with the injection peak (1), the peak of neostigmine (2, m/z 223) added during sample preparation, an unknown substance from the cell lysate (3, m/z 239) and the peak of choline (4, m/z 104) from cell culture media. Middle: MRM chromatogram (146 → 87) showing the peak of ACh (5) and, enlarged, the peak of CMPA produced by SN-56 cells (6). Bottom: MRM chromatogram (150 → 91) containing only the peak of the added I.S. (7).

from NP chromatography, the equilibration of silica columns is technically demanding and time consuming. A zwitterionic ZIC-HILIC column with sulfobetaine groups on the surface was selected as stationary phase. Eluent composition, pH and column temperature were optimized to realize a fast separation of ACh from matrix compounds. Volatile ammonium formate solution (10 mM), adjusted to pH 4.2 with formic acid was selected as aqueous phase and resulted in a good peak shape. The equilibration time in HILIC is comparably long. Therefore, the separation was carried out under isocratic conditions. At a concentration of 75% ACN in the mobile phase, ACh was well retained (peak 5 in Fig. 2). A separation from the dominating peak at m/z 223, being the added AChE inhibitor neostigmine (peak 2 in Fig. 2) and from choline (m/z 104) (peak 4 in Fig. 2) was achieved. Additionally, a small peak appears in the MRM trace of the samples (peak 6 in Fig. 2). By comparison of retention times and mass spectra (Fig. 3) with those of a standard, this substance was identified as (3-carboxpropyl)trimethylammonium (CPMA, Fig. 1). This isomer of ACh has been described before as endogenous substance in brain [5,11] and cornea [6]. CPMA shows the same MRM transitions as ACh (Fig. 3), but is baseline separated from ACh (Fig. 2). Therefore, it does not interfere with the ACh quantification. The TIC of the full scan analysis shows another unknown peak (peak 3 in Fig. 2) with m/z 239 that elutes close to ACh. However, the deuterated I.S. (peak 7

The following MRMs were monitored in LC–MS/MS: m/z 146 → 87 for ACh and m/z 150 → 91 for the I.S. The MS/MS spectra show that both ACh and the I.S. exhibit high intensities for their respective transition (Fig. 3). Even at optimized conditions, both substances show this fragmentation already during ionisation in the ESI source. The structure for the product ion m/z 87 of ACh has been proposed to be a radical cation (pathway a in Fig. 1) [13] or a cation (pathway b in Fig. 1) [7]. In all reactions, a second noncharged fragment with 59 u is formed (Fig. 1), because an ion of m/z 60 is detected in MS/MS (Fig. 3) and is assumed to be the positively charged ion of another part of the molecule occurring in an alternative fragmentation reaction. The mass difference of the first transition of the I.S. is the same as for ACh (Fig. 3). In both reaction pathways, the four deuterium atoms are part of the product ion (II in Fig. 1). Therefore, this transition cannot be used to identify the reaction pathway. To identify the product ion monitored in the MRM, the fragmentation reaction was investigated by a FTICR-MS. The accurate mass for the ACh transition fragment was 87.04404 u and for the I.S. transition 91.06915 u. With a deviation to the calculated mass of 0.18 ppm for ACh and 0.14 ppm for the I.S., it is evident that ACh is fragmented by heterolytic cleavage of the nitrogen–carbon bond. This reaction pathway (pathway b in Fig. 1) is proved by the exact mass of the protonated other fragment (60.08076 u), which could be identified as (CH3 )3 NH+ (deviation 0.26 ppm). These findings are well in line with the findings of Uutela et al. [11], who determined the exact masses with a quadropole time-of-flight mass spectrometer. We found the same exact masses (87.04398 u and 60.0874 u) for the fragments of CPMA. Therefore, we suggest a similar heterolytic fragmentation (III in Fig. 1) of the nitrogen carbon bound leading to the formation of a tetrahydrofuran ring. Even if the fragments of ACh and CPMA have the same m/z, their abundance of fragment ions is different. CPMA is less fragmented under the used conditions and more fragments with m/z 60 are found (Fig. 3). The MS/MS spectrum of Peak 6 (Fig. 2) shows exactly the same pattern, underlining the presence of CPMA in the samples. 3.4. Method calibration and validation The ratio of the peak area of ACh divided by the peak area of 100 nM I.S. was linear over a range from 1 nM to 10 ␮M ACh (Table 1), leading to a calibration function with the regression function y = 0.0099 c + 0.0034 (R2 = 0.997). The slope of the calibration function for ACh alone decreases at high concentrations and is only linear up to 200 nM. It must be assumed that the ion-

N.H. Schebb et al. / J. Chromatogr. A 1183 (2008) 100–107

105

Fig. 3. MS/MS spectra of standards of acetylcholine (ACh), [2 H4 ]acetylcholine (I.S.), (3-carboxypropyl)trimethylammonium (CPMA) and the endogenous substance with m/z 146 from the separation of cell lysate (Peak 6 in Fig. 2). The rhomb indicates the peak of the unfragmentated ion.

isation efficiency is lowered at high concentrations because the peak area of the constant concentration of I.S. decreases with increasing concentration of ACh. However, this has no impact on linearity of ACh/I.S., which is used for quantification. The LOD (S/N = 3) for ACh was determined to be 0.3 nM (1.5 fmol on column) and the limit of quantification (LOQ, S/N = 10) for ACh was 1 nM (5 fmol on column), which is within the range of earlier described HILIC–MS methods [5,6,8,9,11]. To verify that the method yields accurate results in the presence of cell lysate matrix, samples spiked with ACh were measured and the content was determined by standard addition. The deviation of these results to the calculation by ACh/I.S. ratio was less than 8% (n = 4) in all cases. A deuterated I.S.

is most beneficial if its concentration is similar to that of the analyte. The concentration of 100 nM of the internal standard was selected, because the method is two orders of magnitude linear each above and below this concentration. If possible, the samples were diluted to a final ACh concentration of approximately 100 nM. This is achieved by dilution of the supernatant, equivalent to a cell protein concentration of 0.1 mg/mL. The approach of an isocratic separation bears a slightly larger risk that matrix compounds remain on the column or elute in following runs. To exclude this, one sample of cell lysate was injected 13 times. A relative standard deviation (RSD) of 1.5% was obtained demonstrating that there are no adverse effects from late eluting compounds. However, the column was rinsed

Table 1 Precision and accuracy of calibration standards Concentration (nM)

Ratio ACh/I.S.

RSD (%)

Calculated concentration (nM)

Accuracy (%)

2 10 20 100 200 1,000 2,000 10,000

0.022 0.102 0.195 0.969 1.945 10.118 20.453 98.937

7.52 2.03 2.33 1.28 1.24 2.78 1.65 2.82

1.87 10.01 19.48 98.02 197.16 1026.86 2076.12 10044.07

93.7 100.1 97.4 98.0 98.6 102.7 103.8 100.4

The measured ratio of the peak area of the signals in MRM mode of ACh (2–10 000 nM) to the peak area of I.S. (100 nM) of a typical calibration curve is shown. The RSD (%) is based on triple injections. The calculated content by the regression function (R2 = 0.997) is divided by the concentration to determine the accuracy for each standard.

106

N.H. Schebb et al. / J. Chromatogr. A 1183 (2008) 100–107

Table 2 Measured acetylcholine content in SN56 cells Cell lysate

Day 1 acetylcholine (pmol/mg protein)

RSD (%)

Day 2 acetylcholine (pmol/mg protein)

RSD (%)

Day 3 acetylcholine (pmol/mg protein)

RSD (%)

Mean (pmol/mg protein)

Interday RSD (%)

Passage # 32 1 2 3 4 5 6 7 8 9 10

1220 1110 1183 1081 1271 1113 1226 1231 1289 1175

0.53 1.68 0.44 0.68 0.79 1.02 1.21 0.30 0.88 0.83

1222 1143 1180 1099 1251 1122 1170 1223 1283 1232

0.01 1.06 0.24 0.35 1.18 2.41 0.43 0.93 1.69 1.24

1229 1102 1182 1106 1299 1101 1197 1189 1282 1183

0.93 1.36 2.29 0.42 1.27 0.42 1.96 0.54 0.97 0.69

1223 1118 1182 1095 1274 1112 1198 1214 1285 1197

0.38 1.98 0.10 1.21 1.90 0.92 2.33 1.85 0.27 2.57

Average

1190

0.84

1193

0.95

1187

1.08

1190

1.35

Passage # 33 1 2 3 4 5

1281 1290 1439 1525 1362

0.84 1.41 0.58 0.24 1.90

1198 1252 1419 1471 1398

1.19 2.09 0.78 0.91 0.82

1198 1271 1404 1470 1347

0.87 0.84 1.16 0.49 0.52

1225 1271 1421 1488 1369

3.93 1.52 1.21 2.12 1.91

Average

1379

0.99

1347

1.16

1338

0.78

1355

2.14

1272

1.75

Over all average

ACh is determined for two passages in 15 independent cell lysates. The RSD is based on triple injections. The interday deviation was calculated by the measurement of the samples on three different days.

with a gradient after every 10 samples to avoid any possible long-term effects.

be due to the direct sample preparation, leading to less analyte loss.

3.5. Application to SN56 cells

4. Conclusion

The method was applied to determine the ACh content of SN56 cells. Three samples of three passages of the cells were analyzed. A content of 1272 ± 109 pmol/mg protein was determined in cells from passage 32 and 33 (Table 2). The method shows a low RSD of approximately 1% in triple injections. When the sample is measured on three different days, a RSD in the same range is observed, underlining the robustness of the method. The higher deviations (RSD 5.5%) between the measured protein content of three cell samples of the same passage indicate that the determination of the protein content is more prone to errors than the LC–MS measurement. However, the deviation between two passages, which have been independently grown is in the same range (8.5%), thus demonstrating the reproducibility of the approach. The determined ACh content of 1272 pmol/mg protein is similar to previous studies: Blusztajn et al. [17] found a content of about 900 ± 200 pmol/mg protein in undifferentiated SN56 cells applying a LC-ED approach. Szutowicz et al. [21] measured a content of 434 ± 24 pmol/mg protein in undifferentiated SN56 cells and Koc et al. [10] found a content of 280 ± 120 pmol/mg protein in undifferentiated SN56 cells with a LC–MS method. Pedersen et al. [19] described a strongly varying ACh content in SN56 ranging from 600–1000 pmol/mg protein. These contents underline the wide biological variation range of the ACh content in SN56 cells. The fact that the amount of ACh determined with this method are the highest compared with literature data might

A new HILIC–MS/MS method designed for the measurement of ACh in cultured cells has been developed. The sample preparation procedure is much faster than that of the methods described previously, allowing the measurement of ACh in cultured cells within less than 1 h. The separation of ACh from matrix compounds, e.g. ACh-esterase inhibitor and choline, was achieved. The LOD of the method is low (LOD 1.5 fmol; 0.3 nM) and the use of a deuterated internal standard assures the accuracy of the determined content. The identity of the peak assigned to ACh was confirmed by its retention time and high-resolution mass spectrometry. Acknowledgements We would like to thank Carsten Culmsee (Department of Pharmacy, LMU M¨unchen, Germany) for providing the SN56 cells. This work was supported by grants from the “Deutsche Forschungsgemeinschaft” (KR 354/20-1 and KL 601/11-2). We gratefully acknowledge the financial support of the “Studienstiftung des deutschen Volkes” (Bonn, Germany) in the form of a Ph.D. scholarship for N.H.S. References [1] G. Ehrenstein, Z. Galdzicki, G.D. Lange, Biophys. J. 73 (1997) 1276. [2] P.E. Potter, J.L. Meek, N.H. Neff, J. Neurochem. 41 (1983) 188.

N.H. Schebb et al. / J. Chromatogr. A 1183 (2008) 100–107 [3] E. Haen, H. Hagenmaier, J. Remien, J. Chromatogr. 537 (1991) 514. [4] L.D. Acevedo, Y.D. Xu, X. Zhang, R.J. Pearce, A. Yergey, J. Mass. Spectrom. 31 (1996) 1399. [5] Y. Zhu, P.S. Wong, M. Cregor, J.F. Gitzen, L.A. Coury, P.T. Kissinger, Rapid Commun. Mass Spectrom. 14 (2000) 1695. [6] J.L.E. Reubsaet, E. Ahlsen, K.G. Haneborg, A. Ringvold, J. Chromatogr. Sci. 41 (2003) 151. [7] R. Dunphy, D.J. Burinsky, J. Pharm. Biomed. Anal. 31 (2003) 905. [8] Y. Guo, J. Liq. Chromatogr. Rel. Technol. 28 (2005) 497. [9] E.S. Grumbach, D.M. Wagrowski-Diehl, J.R. Mazzeo, B. Alden, P.C. Iraneta, LC–GC N Am. 22 (2004) 1010. [10] H. Koc, M.H. Mar, A. Ranasinghe, J.A. Swenberg, S.H. Zeisel, Anal. Chem. 74 (2002) 4734. [11] P. Uutela, R. Reinila, P. Piepponen, R.A. Ketola, R. Kostiainen, Rapid Commun. Mass Spectrom. 19 (2005) 2950. [12] M.Y. Zhang, Z.A. Hughes, E.H. Kerns, Q. Lin, C.E. Beyer, J. Pharm. Biomed. Anal. 44 (2007) 586.

107

[13] M.E. Hows, A.J. Organ, S. Murray, L.A. Dawson, R. Foxton, C. Heidbreder, Z.A. Hughes, L. Lacroix, A.J. Shah, J. Neurosci. Methods 121 (2002) 33. [14] T. Koyama, Y. Nakajima, T. Fujii, K. Kawashima, Neurosci. Lett. 265 (1999) 33. [15] A.J. Alpert, J. Chromatogr. 499 (1990) 177. [16] P. Hemstrom, K. Irgum, J. Sep. Sci. 29 (2006) 1784. [17] J.K. Blusztajn, A. Venturini, D.A. Jackson, H.J. Lee, B.H. Wainer, J. Neurosci. 12 (1992) 793. [18] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, J. Biol. Chem. 193 (1951) 265. [19] W.A. Pedersen, B. Berse, U. Schuler, B.H. Wainer, J.K. Blusztajn, J. Neurochem. 65 (1995) 50. [20] D. Zhao, M.A. Frohman, J.K. Blusztajn, BMC Neurosci. 2 (2001) 16. [21] A. Szutowicz, M. Tomaszewicz, A. Jankowska, B. Madziar, H. Bielarczyk, Metab. Brain Dis. 15 (2000) 29. [22] Y. Huang, Z.Y. Lu, K.S. Brown, A.S. Whitehead, I.A. Blair, Biomed. Chromatogr. 21 (2007) 107. [23] A. Liesener, U. Karst, J. Sep. Sci. 28 (2005) 1658.