Stability comparison between sample preparation procedures for mass spectrometry-based targeted or shotgun peptidomic analysis

Analytical Biochemistry 407 (2010) 290–292

Contents lists available at ScienceDirect

Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio

Notes & Tips

Stability comparison between sample preparation procedures for mass spectrometry-based targeted or shotgun peptidomic analysis Francis Beaudry Groupe de recherche en pharmacologie animal du Québec (GREPAQ), Département de biomédecine vétérinaire, Faculté de médecine vétérinaire, Université de Montréal, Saint-Hyacinthe, Québec, Canada J2S 2M2

a r t i c l e

i n f o

Article history: Received 30 June 2010 Received in revised form 30 July 2010 Accepted 12 August 2010 Available online 18 August 2010

a b s t r a c t The quantification of neuropeptides may play a significant role in future drug development targeting central nervous system functions. Adequate method precision and accuracy is essential, and sample stability is an important factor. This study compares three sample preparation protocols and assesses the stability of targeted neuropeptides under standard laboratory conditions. The results show that the concentrations of substance P, dynorphin A, and calcitonin gene-related peptide (CGRP) change significantly in time when spinal cord tissues are homogenized in phosphate-buffered saline (PBS) buffer or PBS buffer containing a mammalian protease inhibitor cocktail but is stabilized when tissues are homogenized in a 0.25% trifluoroacetic acid solution. Crown Copyright Ó 2010 Published by Elsevier Inc. All rights reserved.

Sample preparation is a very important step in targeted and shotgun peptidomic analysis. The initial steps include peptide extraction and prefractionation [1,2]. Extracting peptides from biological fluids or tissues is now common in biomarker discovery [3,4]. However, this process can be challenging and should be undertaken rigorously [5]. Several reviews and research articles have discussed different trends and challenges in sample preparation in proteomics [6,7], but very few have discussed this issue specifically for peptidomic analysis mainly because many assume that peptidomic is subproteome analysis and sample preparation and treatment should be similar [8]. The connection between peptidomics and the closely related field of proteomics is important especially in that many peptides have significant biological functions and are formed in vivo by proteolysis of precursor proteins [9,10]. Specifically, neuropeptides play a significant role in pain transmission in the central nervous system (CNS),1 where they modulate neuronal communication by acting on pre- and/or postsynaptic cell surface receptors. Moreover, peptidomic and proteomic analysis with an emphasis on neuropeptide and precursor protein regulation may reveal phenomic fingerprints leading to the discovery of specific pain-related biomarkers used in drug discovery. However, the discovery and validation phase depends profoundly on the method precision and accuracy. The sample preparation steps are fundamental to ensure the quality and reliability of the analytical results [11]. Bioanalytical chemists learn that the hard way in small

E-mail address: [email protected] Abbreviations used: CNS, central nervous system; SPE, solid phase extraction; MWCO, molecular weight cutoff; CGRP, calcitonin gene-related peptide; TFA, trifluoroacetic acid; HPLC, high-performance liquid chromatography; ESI, electrospray ionization; MS/MS, tandem mass spectrometry. 1

molecule discovery, but currently many proteomic and peptidomic studies are conducted by molecular biologists who were not necessarily involved in the early phase of bioanalytical mass spectrometry. Peptidomic analysis requires an extraction step and prefractionation. In biological fluids such as serum and plasma, endogenous peptides are extracted by precipitation, solid phase extraction (SPE), or filtration [12,13]. In several studies, serum samples were not treated prior to starting the extraction procedure, and others used protease inhibitor cocktail without any sort of validation [8,12,13]. Neuropeptide analyses in the CNS are performed on the cerebrospinal fluid (CSF) but also in brain and spinal cord homogenates. Tissue disruption requires manual homogenization, vortexing, grinding, and sonication, and sometimes aqueous solution containing a cocktail of protease inhibitors is used [8]. Following tissue homogenization and stabilization, samples are prefractionated using size exclusion chromatography, SPE, filtration with specific molecular weight cutoff (MWCO), or simply protein precipitation. Contrary to protein analysis, peptidomic analysis does not require specific consideration for gel electrophoresis separation [11] or enzymatic fragmentation specific for mass spectrometry analysis (i.e., trypsin) [14]. The objective of the current study was to verify the stability of neuropeptides in tissue homogenate with and without treatments. Substance P, calcitonin gene-related peptide (CGRP), and dynorphin A were purchased from Phoenix Pharmaceutical (Belmont, CA, USA). Neuropeptides were dissolved in 0.1% trifluoroacetic acid (TFA) solution (100 lg/ml). Protease/peptidase inhibitor cocktail optimized for mammalian cell and tissue extracts containing inhibitors with a broad specificity for serine, cysteine, and acid proteases, as well as aminopeptidases, was purchased from Sigma–Aldrich (St. Louis, MO, USA). Other chemicals, including acetonitrile, TFA, formic

0003-2697/$ - see front matter Crown Copyright Ó 2010 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2010.08.017

Notes & Tips / Anal. Biochem. 407 (2010) 290–292

acid, sodium chloride, and potassium phosphate monobasic and dibasic, were purchased from Fisher Scientific (Pittsburgh, PA, USA). The analytical method used to identify, characterize, and quantify targeted neuropeptides was based on a previously published article [15]. Briefly, the instrumentation consisted of a Thermo Surveyor high-performance liquid chromatography (HPLC) system and a Thermo LCQ Advantage MAX quadrupole ion trap mass spectrometer (San Jose, CA, USA). The chromatographic condition used gradient mobile phase and a microbore column (Thermo Biobasic C8, 100  1 mm, particle size 5 lm). The initial mobile phase conditions consisted of acetonitrile and 0.2% formic acid in water at a ratio of 0:100. From 0 to 1 min, the ratio was maintained at 0:100. From 1 to 20 min, a linear gradient was applied up to a ratio of 40:60 and was maintained for 2 min. The mobile phase composition ratio was reverted to the initial condition, and the column was allowed to reequilibrate for 10 min for a total run time of 32 min. The flow rate was fixed at 75 ll/min, and substance P eluted at 16.3 min, CGRP at 18.2 min, and dynorphin A at 14.1 min. Samples (2 ll) were injected, and the total run time was set at 32 min. The mass spectrometer was interfaced with the HPLC system using a pneumatic-assisted electrospray ion source. The sheath gas was set to 5 U, and the electrospray ionization (ESI) electrode was set to 4000 V. The capillary temperature was set at 300 °C, the capillary voltage was set to 6 V, and the collision energy was set at 30% for neuropeptides. The mass spectrometer operated in full-scan tandem mass spectrometry (MS/MS) mode and used selected mass transition with 674 ? [254 + 600] for substance P, 952 ? [963 + 1215] for CGRP, and 716 ? [630 + 944] for dynorphin A, an isolation width of 3 Da, and a maximum injection time set to 200 ms. Spinal cord homogenate stability experiments were performed on specific neuropeptides. The lumbar region of the spinal cord of male Sprague–Dawley rats (n = 9) was rapidly collected following deep anesthesia with isoflurane. Spinal cord tissues were snapfrozen in cold hexane (kept in a dry ice bath), and samples were kept frozen at 80 °C prior to analysis. The tissues were weighed frozen and homogenized. One set of tissues (n = 3) was homogenized (1:5 [w/v] ratio) in phosphate-buffered saline (PBS) buffer (pH 7.4) without any treatment (control group). Another set of tissues (n = 3) was homogenized in PBS buffer (pH 7.4, 1:5 [w/v] ratio) and treated with a mammalian protease inhibitor cocktail as described by the supplier (Sigma–Aldrich, product code P8340), and the last set of tissues (n = 3) was homogenized in a 0.25% TFA solution (1:5 [w/v] ratio). The samples were incubated at 25 °C (i.e., representative laboratory temperature). Aliquots were taken at time 0 and after 5, 15, 30, 60, 90, 120, and 240 min, and an equal volume of ice-cold acetonitrile was added. The samples were immediately vortexed and centrifuged at 12,000g for 10 min, and the supernatants were filtered with Microcon centrifugal filters (10 kDa MWCO). The filtrate was analyzed in triplicate, and relative quantification of targeted peptides was based on peak areas. Statistical analyses were performed using PRISM 5.0c GraphPad software (La Jolla, CA, USA). Student’s paired t test was used with the level of significance fixed to 0.05. Bioanalytical chemistry results are profoundly dependent on sample integrity. The core of biomarker discovery and validation relies on the intrinsic quality of preclinical and clinical samples analyzed. This is particularly true for protein and peptide biomarker studies performed using biological fluids and tissues because many enzymes play an important role in protein and peptide metabolism [9,10]. Several studies have used protease/ peptidase cocktail inhibitors for protein and peptide analysis, assuming that the biodegradation reaction will be adequately inhibited without proper validation [8,16,17]. The current study tested this hypothesis using three specific neuropeptides found in rat spinal cord [15]: substance P, CGRP, and dynorphin A. These

291

neuropeptides and associated propeptides play an important role in pain transmission and hold great potential to be used as clinical biomarkers in pain-related studies. However, the biomarker validation process is dependent on the accuracy of the results that strongly depend on the sample integrity. To evaluate peptide stability, a series of aliquots was taken for up to 4 h and the relative concentrations were determined by comparing each aliquot with the sample at time 0. Fig. 1 shows interesting results following the analysis in time of targeted neuropeptides. In spinal cords homogenized in PBS buffer, the substance P and dynorphin A concentrations increased significantly by 51% (P = 0.0033) and 41% (P = 0.0089), respectively, whereas the CGRP concentration decreased 18% (P = 0.0361) after 4 h at 25 °C. Interestingly, after 15 min of incubation, we observed gains of 7% and 6% for substance P and dynorphin A, respectively, and a loss of 10% for CGRP. These results suggest that the degradation of protachykinin [18] and prodynorphin [19] in situ significantly increased the concentration

Fig.1. (A) Substance P stability experiments indicate an increase of concentration up to 51% (P = 0.0033) in positive control and 24% (P = 0.0185) with protease inhibitors. Substance P was stable in TFA homogenates (P = 0.8108). (B) Dynorphin A stability experiments indicate an increase of concentration up to 41% (P = 0.0089) in positive control and 17% (P = 0.0349) with protease inhibitors. Dynorphin A was stable in TFA homogenates (P = 0.1035). (C) CGRP stability experiments indicate a decrease of concentration of 18% (P = 0.0361) in positive control and 14% (P = 0.0437) with protease inhibitors. CGRP was stable in TFA homogenates (P = 0.6184).

292

Notes & Tips / Anal. Biochem. 407 (2010) 290–292

of substance P and dynorphin A in time and that the enzymatic degradation of CGRP is significant. The cocktail used contains inhibitors with a broad specificity for serine, cysteine, and acid proteases, as well as aminopeptidases, and is widely used in proteomics. The results shown in Fig. 1 suggest that the protease inhibitor cocktail is not efficient in stopping the degradation of protachykinin, prodynorphin, and CGRP. We observed significant increases in the concentrations of substance P and dynorphin A and a decrease in the concentration of CGRP in rat spinal cord homogenates. However, when the tissues were simply homogenized in a TFA solution, the results indicate that all three neuropeptides were stable in rat spinal cord homogenates for up to 4 h at 25 °C. The pH of the homogenate is below 3.0, and several (but not all) proteases are inactive at low pH. However, incompatibility with the resulting homogenates and sample preparation procedures may hamper application in proteomics. For peptidomic studies using simple depletion techniques (e.g., filtration, SPE), this method appears to be efficient and less risky than protein precipitation using acidified methanolic solution due to potential recovery issues, an important concern for in vivo quantitative peptidomic studies. These preliminary results suggest that special attention is needed when preparing biological fluid and tissues for protein and peptide analysis. More validation is required to ascertain sample integrity in biomarker discovery despite the availability of several inhibitor cocktails, recipes, and extraction protocols used blindly. Consequently, the validation process needs to consider the stability of the targeted proteins and peptides within the biological preparations to avoid artificial study outcomes. Acknowledgment This study was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). References [1] L.D. Fricker, J. Lim, H. Pan, F.Y. Che, Peptidomics: identification and quantification of endogenous peptides in neuroendocrine tissues, Mass Spectrom. Rev. 25 (2006) 327–344. [2] O.N. Yatskin, O.V. Sazonova, D.P. Khachin, E.Y. Blishchenko, A.A. Karelin, V.T. Ivanov, Isolation of peptides from rat tissues: peptidomics vs. degradomics, Adv. Exp. Med. Biol. 611 (2009) 399–400.

[3] I. Vandenbroek, R.W. Sparidans, J.H. Schellens, J.H. Beijnen, Quantitative assay for six potential breast cancer biomarker peptides in human serum by liquid chromatography coupled to tandem mass spectrometry, J. Chromatogr. B 878 (2010) 590–602. [4] J.A. Morón, L.A. Devi, Use of proteomics for the identification of novel drug targets in brain diseases, J. Neurochem. 102 (2007) 306–315. [5] L. Paulson, R. Persson, G. Karlsson, J. Silberring, A. Bierczynska-Krzysik, R. Ekman, A. Westman-Brinkmalm, Proteomics and peptidomics in neuroscience: experience of capabilities and limitations in a neurochemical laboratory, J. Mass Spectrom. 40 (2005) 202–213. [6] J.R. Yates, C.I. Ruse, A. Nakorchevsky, Proteomics by mass spectrometry: approaches, advances, and applications, Annu. Rev. Biomed. Eng. 11 (2009) 49– 79. [7] M.M. Shaw, B.M. Riederer, Sample preparation for two-dimensional gel electrophoresis, Proteomics 3 (2003) 1408–1417. [8] R. Chen, M. Ma, L. Hui, J. Zhang, L. Li, Measurement of neuropeptides in Crustacean hemolymph via MALDI mass spectrometry, J. Am. Soc. Mass Spectrom. 20 (2009) 708–718. [9] V. Hook, L. Funkelstein, D. Lu, S. Bark, J. Wegrzyn, S.R. Hwang, Proteases for processing proneuropeptides into peptide neurotransmitters and hormones, Annu. Rev. Pharmacol. Toxicol. 48 (2008) 393–423. [10] N.G. Seidah, M. Chrétien, Proprotein and prohormone convertases: a family of subtilases generating diverse bioactive polypeptides, Brain Res. 848 (1999) 45–62. [11] Y. Jmeian, Z. El Rassi, Liquid-phase-based separation systems for depletion, prefractionation, and enrichment of proteins in biological fluids for in-depth proteomics analysis, Electrophoresis 30 (2009) 249–261. [12] M. Ma, J. Wang, R. Chen, L. Li, Expanding the Crustacean neuropeptidome using a multifaceted mass spectrometric approach, J. Proteome Res. 8 (2009) 2426– 2437. [13] Y. Kawashima, T. Fukutomi, T. Tomonaga, H. Takahashi, F. Nomura, T. Maeda, Y. Kodera, High-yield peptide-extraction method for the discovery of subnanomolar biomarkers from small serum samples, J. Proteome Res. 9 (2010) 1694–1705. [14] B. Cañas, C. Piñeiro, E. Calvo, D. López-Ferrer, J.M. Gallardo, Trends in sample preparation for classical and second generation proteomics, J. Chromatogr. A 1153 (2007) 235–258. [15] F. Beaudry, C.E. Ferland, P. Vachon, Identification, characterization, and quantification of specific neuropeptides in rat spinal cord by liquid chromatography electrospray quadrupole ion trap mass spectrometry, Biomed. Chromatogr. 23 (2009) 940–950. [16] J. Tenorio-Laranga, M.L. Valero, P.T. Männistö, M. Sánchez del Pino, J.A. GarcíaHorsman, Combination of snap freezing, differential pH two-dimensional reverse-phase high-performance liquid chromatography, and iTRAQ technology for the peptidomic analysis of the effect of prolyl oligopeptidase inhibition in the rat brain, Anal. Biochem. 393 (2009) 80–87. [17] X. Zheng, S.L. Wu, M. Hincapie, W.S. Hancock, Study of the human plasma proteome of rheumatoid arthritis, J. Chromatogr. A 1216 (2009) 3538– 3545. [18] A. Ernst, J. Suhr, J. Köhrle, A. Bergmann, Detection of stable N-terminal protachykinin A immunoreactivity in human plasma and cerebrospinal fluid, Peptides 29 (2008) 1201–1206. [19] F. Merg, D. Filliol, I. Usynin, I. Bazov, N. Bark, Y.L. Hurd, T. Yakovleva, B.L. Kieffer, G. Bakalkin, Big dynorphin as a putative endogenous ligand for the jopioid receptor, J. Neurochem. 97 (2006) 292–301.