Improved Identification of Endogenous Peptides from Murine Nervous

Jan 16, 2009 - Ac- Lipotropin gamma (1-25). KR.ELEGERPLGLEQVLESDAEKDDGPY.R****R. X. GNAS3. Neuroendocrine secretory protein 55 (61--67). RR...
1 downloads 0 Views 758KB Size
Download PDF
Improved Identification of Endogenous Peptides from Murine Nervous Tissue by Multiplexed Peptide Extraction Methods and Multiplexed Mass Spectrometric Analysis A. F. Maarten Altelaar,†,§ Shabaz Mohammed,†,§ Maike A. D. Brans,‡ Roger A. H. Adan,‡ and Albert J. R. Heck*,†,§ Biomolecular Mass Spectrometry and Proteomics Group, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Sorbonnelaan 16, 3584 CA Utrecht, The Netherlands, Rudolf Magnus Institute of Neuroscience, Department of Neuroscience and Pharmacology, University Medical Center Utrecht, Universiteitsweg 100, 3584 CG, Utrecht, The Netherlands, and Netherlands Proteomics Centre, Sorbonnelaan 16, 3584 CA Utrecht, The Netherlands Received June 18, 2008

In recent years, mass spectrometry (MS) based techniques have made their entrance in the analysis of endogenous peptides extracted from nervous tissue. In this study, we introduce a novel peptide extraction procedure using 8 M urea, next to the more established extraction method that uses acetic acid. The extracted peptide mixtures are analyzed by both high-resolution nanoLC MS/MS using collision induced dissociation (CID) on an LTQ-Orbitrap and nanoLC electron transfer induced dissociation (ETD) on a linear ion trap. The combined use of the two extraction methods significantly increased the yield of identified endogenous neuropeptides. The multiplexed use of high mass accuracy mass spectrometry and the ETD fragmentation technique further increased the yield and confidence of peptide identifications. Furthermore, reduction of disulfide bridges during sample preparation was essential in the identification of several endogenous peptides containing cysteine disulfide bonds. Through this study, we identified in total 142 peptides in extracts of the mouse pituitary tissue, whereby 43 uniquely in the urea extract and 11 uniquely in the acetic acid extract. A large number of detected endogenous peptides were reported previously, but we confidently identified 22 unreported peptides that possess characteristics of endogenous peptides and are thus interesting targets to be explored further. Keywords: Endogenous Peptides • Peptide Extraction • ETD • CID • Multiplex Mass Spectrometry

Introduction Many endogenous peptides function as messenger molecules within cell-cell communication in the brain (neuropeptides) and in the neuroendocrine system (peptide hormones). These peptides are the largest and most complex group of messenger molecules, playing a crucial role in animal physiology and behavior. They are involved in almost every neuronal process in the brain and, in combination with their functions as peptide hormones, are involved in a wide variety of behavioral mechanisms including reproduction, reward mechanisms, pain, memory, food intake and circadian rhythms.1-5 Endogenous peptides are derived from larger precursor proteins, which are stored in secretory granules together with processing enzymes (convertases). Convertase cleavage of the peptide from the precursor protein mainly occurs in these secretory granules, often at motifs containing double basic amino acid residues. Tight control over this convertase activity allows for an additional mechanism to regulate signaling pathways. Following * To whom correspondence should be addressed. E-mail: [email protected]. † Utrecht University. ‡ University Medical Center Utrecht. § Netherlands Proteomics Centre.

870 Journal of Proteome Research 2009, 8, 870–876 Published on Web 01/16/2009

convertase cleavage, the C-terminal basic residue of the peptide is removed by a carboxypeptidase, delivering the final endogenous peptide sequence. Precursor proteins may contain multiple copies of the same peptide in their sequence as well as multiple ‘different’ peptides all potentially, with distinct biological activity. These peptides may then be subsequently subjected to a multitude of post-translational modifications (PTMs) before they become bioactive. Currently, known neuropeptide modifications include acetylation, C-terminal amidation, phosphorylation and sulfation, which are essential modifications affecting stability and affinity for the receptors they act upon. This extensive post-translational peptide processing causes endogenous peptides to be inherently difficult to predict from a genetic sequence or even a precursor protein.3,5-7 To establish the sequence of endogenous peptides, several studies have made use of mass spectrometry (MS).2-5,8-10 The strength of MS for peptide sequencing is based on its attributes as a fast, accurate and sensitive method. However, the study of endogenous peptides using tandem MS is challenging since these peptides can be relatively large (compared to tryptic peptides) and possibly contain multiple basic residues and labile PTMs. The presence of multiple basic residues in the 10.1021/pr800449n CCC: $40.75

 2009 American Chemical Society

Identification of Endogenous Peptides from Murine Nervous Tissue peptide will result in a higher charge state, which in turn results in more complicated tandem MS spectra. Furthermore, in conventional tandem MS based (proteomic) studies, collision induced dissociation (CID) is most often applied to fragment the peptide backbone to attain sequence information. In CID, the fragmentation is most often channelled through the lowest possible energy pathways, for example, bond cleavages involving basic residues or PTMs such as phosphorylation and glycosylation, which may result in a limited number of diagnostic sequence ions being produced hampering confident identification. This is especially true for CID fragmentation in ion-trap instruments due to the relatively slow heating of the ions compared to ‘tandem in space’ fragmentation in, for example, Quadrupole Time-of-Flight (Q-ToF) instruments or the recently introduced so-called Higher Energy Collisional Dissociation (HCD) for the LTQ-Orbitrap.11 To increase the peptide assignment confidence, alternative dissociation techniques can be used, such as electron transfer dissociation (ETD).12 In ETD, the interaction between (near) thermal electrons with the positively charged peptide results in peptide dissociation of the N-CR bond. The exact mechanism leading to ETD fragmentation is still under debate and a recent study by Xia et al.13 showed evidence for contributions from competing ETD mechanisms. However, since ETD is relatively indifferent to peptide sequence, length, or PTMs, it might represent a better technique for analysis of endogenous peptides. Another difficulty in endogenous peptide analysis is efficient extraction from tissue while preventing postmortem degradation. It has been shown that endogenous peptides can be degraded within minutes of postmortem delay, by active peptidases. Consequently, in the following analysis, the peptides of interest are masked by the much more abundant protein degradation products.2,10,14 Furthermore, PTM levels can also be modified during postmortem processes, for example, the phosphorylation level of the endogenous peptide corticotropin-like intermediate lobe peptide (CLIP) is reduced by 85% within 3 min postmortem.10 Prevention of degradation and artificial modifications (like dephosphorylation) has been achieved by denaturation of the active enzymes, provoked by rapid heating of the tissue, by either boiling4,15 or microwave irradiation.2,10,16 Peptides are then, commonly, extracted using an acidic solution (e.g., 0.25% acetic acid) by either boiling in the acidic solution or by sonication. Since acidic solvents might favor peptides with relatively higher pI values, we explored an alternative extraction procedure using 8 M urea. Since the solubility of the endogenous peptides will differ in urea compared to acidic solutions, we anticipated to detect different peptides from the urea extract. In this study, we use a combination of the multiple extraction protocols combined with high mass accuracy MS (LTQOrbitrap) and LTQ-ETD to characterize a large number of known as well as novel neuropeptides from rat brain tissue.

Experimental Procedures Extraction of the Neuropeptides. Experimental procedures were in accordance with the European directives (86/609/EEC) and approved by the Commission on Laboratory Animal Experiments of the University Medical Center Utrecht. Juvenile C57Bl/6J male mice were the offspring of C57Bl/6J mice obtained from The Jackson Laboratories (Bar Harbor, ME). Mice were decapitated without prior anesthesia and pituitary glands were dissected and directly boiled for 5 min in 150 µL

research articles

of milliQ water (Millipore, The Netherlands). After boiling, the water was decanted and 150 µL of either 0.25% acetic acid (HAc) or 8 M urea was added (freshly prepared to avoid carbamoylation). The pituitary tissue was homogenized in the extraction solution using microtip sonication (Labsonic M, Sartorius AG, Germany). Both the water fractions and the urea or HAc fractions were centrifuged at 14 000g for 30 min at 4 °C. For each pituitary gland, the water and the urea or HAc fraction were combined and passed through a 10 kDa molecular weight cutoff (MWCO) filter (biomax-10, Millipore, Bedford, MA) at 14 000g for 45 min at 4 °C. The MWCO filters were washed beforehand, by passing through ethanol twice, to remove polymers from the filter. The filtrate was stored until use at -20 °C. For the reduction of disulfide bridges, 10 mM dithiothreitol (DTT) was added just before MS analysis. NanoLC-MS/MS. The tissue extract was analyzed by nanoLCLTQ-Orbitrap-MS (Thermo, San Jose, CA) and by nanoLC-LTQXL-MS (Thermo, San Jose, CA). An Agilent 1100 series LC system was equipped with a 20 mm Aqua C18 (Phenomenex, Torrance, CA) trapping column (packed in-house, i.d., 100 µm; resin, 5 µm) and a 250 mm ReproSil-Pur C18-AQ (Dr. Maisch, GmbH, Ammerbuch, Germany) analytical column (packed inhouse, i.d., 50 µm; resin, 3 µm). Trapping was performed at 5 µL/min for 10 min, and elution was achieved with a gradient of 0-40% B in 35 min, 40-100% B in 2 min, 100% B for 2.5 min (with solvent B being 0.1 M acetic acid in 80% ACN and solvent A being 0.1 M acetic acid). The flow rate was passively split from 0.4 mL/min to 100 nL/min, as described previously.17 For samples extracted in 8 M urea, an additional washing step, with 100% solvent A, was implemented on the trap column at 5 µL/min for 20 min. Nanospray was achieved using a distally coated fused silica emitter (New Objective, Cambridge, MA) (o.d., 360 µm; i.d., 20 µm, tip i.d. 10 µm) biased to 1.7 kV. In the case of the LTQ-Orbitrap, the mass spectrometer was operated in the data dependent mode to automatically switch between MS and MS/MS. Survey full scan MS spectra were acquired from m/z 350 to m/z 1500 in the FT-Orbitrap with a resolution of R ) 60 000 at m/z 400 after accumulation to a target value of 500 000 in the linear ion trap with lock-mass. The two most intense ions were fragmented in the linear ion trap using collisionally induced dissociation at a target value of 10 000. In the case of the LTQ-XL, the mass spectrometer was operated in the data dependent mode to automatically switch between MS and MS/MS ETcaD and MS/MS CID. Survey full scan MS spectra were acquired from m/z 350 to m/z 1500 in the LTQ after accumulation to a target value of 100 000 in the linear ion trap. The two most intense ions were fragmented in the linear ion trap at a target value of 30,000. Data Processing. Spectra were processed with Bioworks 3.4.0 (Thermo, Bremen, Germany) and the subsequent data analysis was carried out using the Mascot (version 2.2.0) software platform (Matrix Science, London, U.K.). LTQ-Orbitrap and LTQ-XL analyses were searched against the IPI mouse protein database (version 3.36) and the SwePep precursor, SwePep peptides and SwePep predicted databases (version 1.0). The development and use of the SwePep databases is described by Fa¨lth et al.8 and they are freely available from www.swepep.org. The searches were conducted without the use of an enzyme and with acetyl (N-term), amidated (C-term), deamidated (NQ), Gln-pyro-Glu (N-term, Q), oxidation (M) and phosphorylation (ST) as variable modifications. For the LTQ-Orbitrap, the peptide tolerance was set at 5 ppm with 1+, 2+ and 3+ peptide charges and the MS/MS tolerance to 0.9 Da, and for the LTQJournal of Proteome Research • Vol. 8, No. 2, 2009 871

research articles

Altelaar et al.

Figure 1. Total Ion Chromatograms of the extracted endogeneous peptides from the murine pituitary tissue using (A) urea extraction and (B) HAc extraction. The identity of some of the more abundant peptides, as identified by using LC MS/MS, is indicated. Table 1. Number and Nature of Endogenous Peptides Identified in Pituitary Tissue Using the Two Used Extraction Proceduresa known endogenous peptides

previously reported potential endogenous peptides

novel potential endogenous peptides

cumulative numbers

43

23

21

87

28

14

13

55

18

14

12

44

Urea fraction HAc fraction Overlap a

The numbers contain sequences with and without post-translational modification. Details on the peptide sequences are given in Table 2 and Supporting Information Tables 1 and 2.

XL, the peptide tolerance was set at 1 Da with 1+, 2+ and 3+ peptide charges and the MS/MS tolerance to 0.9 Da. The significance threshold for the identifications was set to p < 0.05. Manual validation was performed on the assignment of the endogenous peptides.

Results and Discussion Improving Peptide Extraction. With the aim of a more comprehensive and less biased extraction of peptides from pituitary tissue, we explored a dual solvent system method using either acetic acid (HAc) or freshly prepared 8 M urea. Acidic solvents like HAc are commonly used for the extraction of endogenous peptides, often in combination with heating or boiling of the tissue to liberate the peptides.18 However, the analysis of neuropeptides is often hampered by postmortem degradation of the protein and peptide content. Rapid heating of the tissue by microwave irradiation can prevent this degradation by denaturation of active enzymes, after which the peptides are extracted using an acidic solvent.19 However, Che et al.20 indicated that this ‘hot’ acidic extraction may also result in an increase of degradation products, particularly from cytosolic proteins, which can obscure identification of endogenous peptides. To limit possible degradation as much as possible, we placed the pituitary tissue in boiling milliQ water for 5 min directly following dissection. After this heating step, the water fraction was decanted and cold HAc was added followed by microtip sonication on ice (to prevent heating of the sample) to extract the endogenous peptides. Subsequently, 872

Journal of Proteome Research • Vol. 8, No. 2, 2009

the water and acid fraction were combined for further sample preparation steps. In proteomic experiments, 8 M urea is often used to lyse cells and dissolve the protein content.21 We argued that, similarly, a high concentration of urea can solubilize tissue, efficiently liberating and dissolving the peptide content. Therefore, we repeated the above-described peptide extraction protocol but replaced HAc with 8 M urea. The urea was prepared fresh just before extraction since it is known that over time ammonium cyanate can be formed, which can induce unwanted carbamoylation of peptide amino groups by isocyanic acid. Following the pooling of the HAc or urea fraction with their respective water fractions from the boiling step, the samples are passed through a 10 kDa filter in order to retrieve the endogenous peptides. These fractions are analyzed by reversedphase, nanoLC-MS/MS using either an LTQ-Orbitrap or an LTQ-ETD instrument. Figure 1 shows an overview of the LCMS traces from the urea extraction (Figure 1A) and the HAc extraction (Figure 1B). The two traces show similarities originating from a few abundant peptides (e.g., vasopressin and oxytocin), but by eye, one can already see clear differences. For instance, the J-peptide of the POMC precursor protein is much more abundant in the urea extract. Each extract was subjected to sequencing by either CID MS/ MS on a LTQ-Orbitrap or by ETD on an LTQ linear ion trap. Applying a significance threshold of p < 0.05 followed by manual validation, 55 peptides could be identified in the HAc-

Identification of Endogenous Peptides from Murine Nervous Tissue

research articles

Figure 2. CID and ETD fragmentation spectra of the 7B2 precursor protein’s endogenous C-terminal neuropeptide SVPHFSEEEKEAE. Observed fragment ions are indicated in the sequence given in the inset. The ETD spectra provide here significantly more sequence indicative fragment ions.

extracted sample, whereas, remarkably, 87 peptides were successfully sequenced from the urea extract. Table 1 shows the number of endogenous peptides identified from mouse pituitary tissue in the different fractions. The most likely rationale for the observed increase in detected peptides from the urea extract is improved cell lysis and solubility, possibly accompanied by decreased degradation of endogenous peptides. The cumulative numbers in Table 1 indicate that many peptides appear in both pools, but both extraction methods do provide unique identifications and are complementary. For example, the peptides R-MSH and β-endorphin from the POMC precursor protein were observed only in the HAc extract, whereas several variants of the chromogranin-A precursor (CMGA), like WE-14 and β-granin, were only detected in the urea extract. Improving Peptide Identifications by High Mass Accuracy and ETD. To increase confidence in the assignment of the peptide sequences, we used a combination of high mass accuracy MS with CID fragmentation in an LTQ-Orbitrap and, in a second experiment, ETD in an LTQ ion trap instrument. ETD allows better fragmentation of higher charged peptides (>2+) than CID.22 Figure 2 provides an example of a 3+ peptide subjected to CID and ETD analysis. After electron transfer dissociation, a range of fragments are observed allowing almost complete deduction of the sequence. As expected, cleavages adjacent N-terminal to proline residues are missing due to the cyclic nature of proline, which requires breaking of two bonds.23 In the CID spectrum, however, the fragments on the N-terminal side of the proline residue are very abundant since they are products of a favored fragmentation pathway. The c3 and z3 fragments in the ETD spectrum of Figure 2 are very low

compared to the other fragments which can be rationalized. Although ETD can produce highly homogenized fragmentation, observation of these fragments depends upon their ability to sequester a proton and become ions. The poor signal of the c3 and z3 ions can be likely related to them lacking basic residues in their sequence, which are needed to compete for the remaining protons.24-26 The endogenous peptide from secretogranin-1 in Figure 3 was not identified at all in the CID analysis of the HAc extract but was confidently assigned in the corresponding ETD analysis. Here CID fragmentation only results in the identification of four, mainly doubly charged, peptide fragment ions, while the analogous ETD spectrum allows complete assignment of the peptide sequence. Although ETD provided some complementary analysis of pituitary peptides, the combination of high mass accuracy and improved dynamic range of the LTQ-Orbitrap with CID fragmentation resulted in the identification of the majority of peptides. High mass accuracy allows reduction of the number of candidate sequences that can be matched to the peptide molecular weight and, when combined with the CID spectra, can lead to a confident identification even when the sequence information is incomplete. To identify endogenous peptides using database search strategies, the entire database needs to be searched against, without specification of a specific protease activity, allowing cleavages of proteins between every pair of amino acids. In such a search, the number of potential candidates grows rapidly with decreasing mass accuracy, increasing the number of false positive identifications. The two peptides vasopressin and oxytocin are known to be highly abundant in pituitary tissue and one would expect them to be easily detected in the MS analysis. Remarkably, this was Journal of Proteome Research • Vol. 8, No. 2, 2009 873

research articles

Altelaar et al.

Figure 3. CID and ETD fragmentation spectra of the endogenous peptide LLDEGHYPVRESPIDTA from secretogranin-1. Observed fragment ions are indicated in the sequence given in the inset. The ETD spectra provide here significantly more sequence indicative fragment ions.

Figure 4. CID fragmentation spectra of the endogenous peptide vasopressin with and without reduction of its disulfide bonds. Observed fragment ions are indicated in the sequence given in the inset. The CID spectra are reduction of the disulfide bridge provide here significantly more sequence indicative fragment ions.

initially found not to be the case. Investigation of our data indicated these peptides were present in their natural oxidized form. The measured masses were 2 Da lower than expected from their linear sequence, and the mass accuracy of the Orbitrap instrument allowed us to assign this mass loss to the 874

Journal of Proteome Research • Vol. 8, No. 2, 2009

loss of two hydrogen atoms. Vasopressin and oxytocin contain two cysteines in their sequences (CYFQNCPRG and CYIQNCPLG respectively) that are known to form cyclic structures bound by disulfide bridges. Fragmentation of these cyclic peptides was found to be rather poor, even by ETD (Supporting

research articles

Identification of Endogenous Peptides from Murine Nervous Tissue

Figure 5. Schematic workflow of the approach used to analyze and identify endogenous peptides from murine pituitary tissue.

Informaiton Figure 1), and identification could only be based on peptide fragments from outside the cyclic ring structure. For oxytocin, the identification was not even confident enough due to the absence of abundant peptide fragments. To tackle

this problem, we added 10 mM dithiothreitol (DTT) to the sample solution to reduce the disulfide bridges. Consequently, in the following LC-MS/MS analysis, both peptides with free cysteines were confidently assigned showing close to complete sequence coverage, as shown in Figure 4 for the endogenous peptide vasopressin. The Mouse Pituitary Peptidome. Combining the abovedescribed peptide extraction methods, the two MS fragmentation methods and the sample preparation procedures (Figure 5) lead to an optimized peptidomics approach, which resulted in a significantly more diverse population of known and previously reported peptides and peptide fragments (Supporting Information Table 1) compared to a cumulative list of previously reported peptides from peptidomic studies of the pituitary gland, wherein mainly the very abundant peptide products of the POMC precursor protein were found. In our approach, several unique peptides from different precursor proteins besides the expected POMC products are observed. This increased peptide diversity will be very helpful in future studies toward endogenous peptide controlled signaling events. Besides the known and previously published endogenous peptides and peptide fragments, also a number of peptides were identified which follow criteria for being considered potential endogenous, as previously described by Fa¨lth et al.8 (Supporting Information Table 2). Endogenous peptides are predominantly formed via cleavage of the precursor protein on the C-terminal side of two basic amino acids that are separated by 0, 2, 4 or 6 residues. After this initial proteolytic event, the C-terminal basic residue is removed by carboxypeptidase E leading to a putative ‘template’ for such peptides, that is, K/R-(X)m-K/R-(X)k-K/R-(X)n-K/R, where m, n ) 0, 2, 4 or 6, X ) any amino acid except Cys,27 and k) 3-50. After the second proteolytic event by carboxypepitdase E, the endogenous peptide sequence (X)k remains. Utilizing the template to the results obtained from our Mascot searches against the IPI mouse database (version 3.36) results in the identification of 24 novel peptides (Table 2), which, to the best of our

Table 2. Novel Potential Endogenous Peptides Identified Using the Double Extraction Methoda entry name

peptide name

peptide sequence

urea

HAc

POMC POMC POMC POMC POMC GNAS3 GNAS3 GNAS3 GNAS3 CMGA CMGA CMGA CMGA CMGA PDYN 7B2 SCG1 SCG1 SCG1 SCG2 NEC1 NEU2

Melanotropin beta (12-18) NPP (61-74) NPP (61-74), NQ-Deam Lipotropin gamma (1-25) Ac- Lipotropin gamma (1-25) Neuroendocrine secretory protein 55 (61--67) Ac-Neuroendocrine secretory protein 55 (61-67) Neuroendocrine secretory protein 55 (239-254) Neuroendocrine secretory protein 55 (240-257) WE14 (361-371) Chromogranin A (405-417) Chromogranin A (419-432) Chromogranin A (374-390) Chromogranin A (374-390), Met-ox Beta-neoendorphin-dynorphin Precursor (211-218) Secretogranin-5 (182-197) Secretogranin-1 (186-199) Secretogranin-1 (419-435) Secretogranin-1 (64-83), 2 NQ-Deam Secretogranin-2 (168-181) Propeptide (90-108) Copeptin (150-168)

R****R.WSNPPKD.KR R**R.FGPRNSSSAGSAAQ.RR R**R.FGPRNSSSAGSAAQ.RR KR.ELEGERPLGLEQVLESDAEKDDGPY.R****R KR.ELEGERPLGLEQVLESDAEKDDGPY.R****R RR.SFLNAHH.RS RR.SFLNAHH.RS RR.RDQSPESPPRKGPIPI.RR RR.DQSPESPPRKGPIPIRRH.R**R.MDQLAKELTAE.KR RR.GWRPSSREDSVEA.R******K R******R.SDFEEKKEEEGSAN.RR KR.LEGEDDPDRSMKLSFRT.R****R KR.LEGEDDPDRSMKLSFRT.R****R R**R.PKLKWDNQ.KR RR.SVNPYLQGKRLDNVVA.KK KK.HIEDSGEKPNTFSN.KR R**R.GRGREPGAHSALDTREE.KR KK.SGKEVKGEEKGENQNSKFEV.R**R K**R.FPLMYEENSRENPF.KR KR.LSDDDRVTWAEQQYEKERS.KR R****R.LVQLAGTRESVDSAKPRVY.-

X X X X X X X X X X X X X X X X X X

X X X

X X X

X

X X X X X X X X

X

a C-terminal amidated ) C-term-Am; deamidated NQ ) NQ-Deam; acetylated ) Ac; methionine oxidation ) Met-ox; conversion of Gln to pyroGlu ) Gln f pyro-Glu (N-term Q); X ) identified.

Journal of Proteome Research • Vol. 8, No. 2, 2009 875

research articles knowledge, have not been previously reported. The table also contains one peptide, LSAKPAPAKVDAKPK, from the precursor protein HMG-14, which does not completely follow the template in the sense that the C-terminus basic amino acids are separated by three additional amino acids. However, the similarity between this sequence and the sequence LSAKPAPPKPEPKPK, from the precursor protein HMG-17, which does follow the template, is interesting enough to consider it as a possible endogenous peptide. These novel endogenous peptides provide new avenues for investigation but will require further study to validate whether they are genuine bioactive peptides. Interestingly, indications in literature exist for possible roles for some of the identified peptides. For example, the peptides SFLNAHH and DQSPESPPRKGPIPIRRH are processed fragments from the neuroendocrine secretory protein 55 (NESP55). It is known that this protein (like other chromogranins) is processed into small peptides that are suggested to be involved in both regulated and constitutive secretion pathways.28,29 However, these breakdown products and their function are yet unknown and our current approach might help in elucidating the sequence of these fragments, and others, providing stronger indicators to their function.

Conclusion Combination of acetic acid and urea based extraction procedures in the analysis of endogenous peptides from murine pituitary tissue significantly increases the number of peptides identified. The use of 8 M urea, introduced in this work, allows the efficient extraction of endogenous peptides partly explaining the higher amount of identified endogenous peptides in this extraction. However, with both extraction procedures, specific peptides could be detected, and thus, only the combination of these different extraction procedures will result in the highest number of successfully identified endogenous peptides. The combination of LC-ETD MS/MS experiments with LCCID MS/MS on a high mass accuracy Orbitrap allows assignment of the peptides with high confidence. ETD is the preferred fragmentation technique for larger multiple charged endogenous peptides because of the better fragmentation when compared to CID. However, the use of high mass accuracy and improved dynamic range of the Orbitrap turned out to be the most crucial parameters in peptide identification of most of the novel peptides reported here. In this respect, the combination of ETD experiments with a high mass accuracy MS scan would be ideal in further peptidomics research. In summary, the here described use of multiplexed peptide extraction methods and multiplexed mass spectrometric analyses improves the number of identifications of endogenous peptides in the nervous tissue significantly, providing a list of putative novel endogenous peptides.

Acknowledgment. This work was supported by The NetherlandsProteomicsCentre(http://www.netherlandsproteomicscenter. nl), a program embedded in the Netherlands Genomics Initiative. Supporting Information Available: Supplementary Figure 1, ETD fragmentation spectra of the endogenous pep-

876

Journal of Proteome Research • Vol. 8, No. 2, 2009

Altelaar et al. tides vasopressin (A) and oxytocin (B). Supplementary Table 1, endogenous peptides identified using the double extraction method. Supplementary Table 2, previously reported potential endogenous peptides identified using the double extraction method. This material is available free of charge via the Internet at http://pubs.acs.org.

References (1) Ooi, G. T.; Tawadros, N.; Escalona, R. M. Mol. Cell. Endocrinol. 2004, 228, 1–21. (2) Che, F. Y.; Lim, J.; Pan, H.; Biswas, R.; Fricker, L. D. Mol. Cell. Proteomics 2005, 4, 1391–1405. (3) Hummon, A. B.; Richmond, T. A.; Verleyen, P.; Baggerman, G.; Huybrechts, J.; Ewing, M. A.; Vierstraete, E.; Rodriguez-Zas, S. L.; Schoofs, L.; Robinson, G. E.; Sweedler, J. V. Science 2006, 314, 647– 649. (4) Dowell, J. A.; Heyden, W. V.; Li, L. J. Proteome Res. 2006, 5, 3368– 3375. (5) Falth, M.; Skold, K.; Svensson, M.; Nilsson, A.; Fenyo, D.; Andren, P. E. Mol. Cell. Proteomics 2007, 6, 1188–1197. (6) Pritchard, L. E.; White, A. Endocrinology 2007, 148, 4201–4207. (7) Hokfelt, T.; Broberger, C.; Xu, Z. Q.; Sergeyev, V.; Ubink, R.; Diez, M. Neuropharmacology 2000, 39, 1337–1356. (8) Falth, M.; Skold, K.; Norrman, M.; Svensson, M.; Fenyo, D.; Andren, P. E. Mol. Cell. Proteomics 2006, 5, 998–1005. (9) Fricker, L. D.; Lim, J.; Pan, H.; Che, F. Y. Mass Spectrom. Rev. 2006, 25, 327–344. (10) Svensson, M.; Skold, K.; Nilsson, A.; Falth, M.; Nydahl, K.; Svenningsson, P.; Andren, P. E. Anal. Chem. 2007, 79, 14–21. (11) Olsen, J. V.; Macek, B.; Lange, O.; Makarov, A.; Horning, S.; Mann, M. Nat. Methods 2007, 4, 709–712. (12) Syka, J. E.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.; Hunt, D. F. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9528–9533. (13) Xia, Y.; Gunawardena, H. P.; Erickson, D. E.; McLuckey, S. A. J. Am. Chem. Soc. 2007, 129, 12232–12243. (14) Skold, K.; Svensson, M.; Kaplan, A.; Bjorkesten, L.; Astrom, J.; Andren, P. E. Proteomics 2002, 2, 447–454. (15) Drabik, A.; Bierczynska-Krzysik, A.; Bodzon-Kulakowska, A.; Suder, P.; Kotlinska, J.; Silberring, J. Mass Spectrom. Rev. 2007, 26, 432– 450. (16) Nylander, I.; Stenfors, C.; Tan-No, K.; Mathe, A. A.; Terenius, L. Neuropeptides 1997, 31, 357–365. (17) Meiring, H. D.; van der Heeft, E.; ten Hove, G. J.; de Jong, A. P. J. M. J. Sep. Sci. 2002, 25, 557–568. (18) Conlon, J. M. Nat. Protoc. 2007, 2, 191–197. (19) Svensson, M.; Skold, K.; Svenningsson, P.; Andren, P. E. J. Proteome Res. 2003, 2, 213–219. (20) Che, F. Y.; Zhang, X.; Berezniuk, I.; Callaway, M.; Lim, J.; Fricker, L. D. J. Proteome Res. 2007, 6, 4667–4676. (21) Pinkse, M. W.; Mohammed, S.; Gouw, J. W.; van Breukelen, B.; Vos, H. R.; Heck, A. J. J. Proteome Res. 2008, 7 (2), 687–697. (22) van den Toorn, H. W. P.; Mohammed, S.; Gouw, J. W.; van Breukelen, B.; Heck, A. J. R. J. Proteomics Bioinform. 2008, 1, 379– 388. (23) Mikesh, L. M.; Ueberheide, B.; Chi, A.; Coon, J. J.; Syka, J. E.; Shabanowitz, J.; Hunt, D. F. Biochim. Biophys. Acta 2006, 1764, 1811–1822. (24) Taouatas, N.; Altelaar, A. F.; Drugan, M. M.; Helbig, A. O.; Mohammed, S.; Heck, A. J. R. Mol. Cell. Proteomics 2009, 8 (1), 190–200. (25) Taouatas, N.; Drugan, M. M.; Heck, A. J.; Mohammed, S. Nat. Methods 2008, 5, 405–407. (26) Emory, J. F.; McLuckey, S. A. J. Am. Soc. Mass Spectrom. 2008, in press. (27) Scamuffa, N.; Calvo, F.; Chretien, M.; Seidah, N. G.; Khatib, A.-M. FASEB J. 2006, 20, 1954–1963. (28) Li, Y.; Fischer-Colbrie, R.; Dahlstrom, A. J. Comp. Neurol. 2008, 506, 733–744. (29) Ischia, R.; Lovisetti-Scamihorn, P.; Hogue-Angeletti, R.; Wolkersdorfer, M.; Winkler, H.; Fischer-Colbrie, R. J. Biol. Chem. 1997, 272, 11657–11662.

PR800449N