QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 15, Issue 6, 2907-2919, June 2004

This Article Abstract Full Text (PDF) Supplemental Table A correction has been published All Versions of this Article: E04-02-0101v1 15/6/2907    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wu, C. C. Articles by Howell, K. E. Search for Related Content PubMed PubMed Citation Articles by Wu, C. C. Articles by Howell, K. E.

Organellar Proteomics Reveals Golgi Arginine Dimethylation

Christine C. Wu ^*, Michael J. MacCoss ^* , Gonzalo Mardones , Claire Finnigan , Soren Mogelsvang , John R. Yates, III ^* , and Kathryn E. Howell   ^||

^* Department of Cell Biology, The Scripps Research Institute, La Jolla, California 92037; Department of Cell and Developmental Biology, University of Colorado Health Sciences Center, Denver, Colorado 80262

Submitted February 4, 2004; Revised March 11, 2004; Accepted March 12, 2004
Monitoring Editor: Benjamin Glick

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
The Golgi complex functions to posttranslationally modify newly synthesized proteins and lipids and to sort them to their sites of function. In this study, a stacked Golgi fraction was isolated by classical cell fractionation, and the protein complement (the Golgi proteome) was characterized using multidimensional protein identification technology. Many of the proteins identified are known residents of the Golgi, and 64% of these are predicted transmembrane proteins. Proteins localized to other organelles also were identified, strengthening reports of functional interfacing between the Golgi and the endoplasmic reticulum and cytoskeleton. Importantly, 41 proteins of unknown function were identified. Two were selected for further analysis, and Golgi localization was confirmed. One of these, a putative methyltransferase, was shown to be arginine dimethylated, and upon further proteomic analysis, arginine dimethylation was identified on 18 total proteins in the Golgi proteome. This survey illustrates the utility of proteomics in the discovery of novel organellar functions and resulted in 1) a protein profile of an enriched Golgi fraction; 2) identification of 41 previously uncharacterized proteins, two with confirmed Golgi localization; 3) the identification of arginine dimethylated residues in Golgi proteins; and 4) a confirmation of methyltransferase activity within the Golgi fraction.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Organelles are membrane-bound compartments that function by interacting with cytoplasmic and luminal soluble proteins making the protein composition of each organelle dynamic. The Golgi complex is the central organelle of the secretory pathway and functions to posttranslationally modify newly synthesized proteins and lipids and sort them for transport to their sites of function (Palade, 1975). However, other Golgi functions require interactions that are less clearly understood. The Golgi interacts with the cytoskeleton to maintain its perinuclear localization within the cytoplasm and to facilitate its dispersal during cell division (Lowe et al., 1998; Allan et al., 2002; Colanzi et al., 2003). Interactions with the endoplasmic reticulum (ER) are reported at both the entry and exit faces of the organelle (Ladinsky et al., 1999; Marsh et al., 2001; Ward et al., 2001). Furthermore, many signaling pathways intersect at the Golgi, and signaling molecules may transiently associate with Golgi membranes (Donaldson and Lippincott-Schwartz, 2000; Van Lint et al., 2002). These unresolved questions regarding nonclassical functions of the Golgi have stimulated several organellar proteomic analyses with the goal of comprehensively identifying its protein components (Taylor et al., 2000; Wu et al., 2000; Bell et al., 2001).

The number of genes that are expressed within a given cell type is estimated to be 10,000 (Huber et al., 2003). However, proteins are further covalently modified to mediate the complex functional interactions, thus increasing the complexity of proteomic analyses. To simplify the analysis of cellular compartments, enrichment strategies should precede proteomic analysis. Organellar proteomics combines subcellular fractionation with mass spectrometry and provide a powerful approach to identify the protein complement of each subcellular compartment. Classically, organellar proteomics has used gel electrophoresis technology (both one-dimensional and two-dimensional) followed by the identification of resolved proteins by mass spectrometry. However, technical limitations resulting in reduced identifications of transmembrane proteins and posttranslational modifications have led to the development of alternative "nongel" strategies (Wu and Yates, 2003). Previously reported Golgi proteomes reflect these limitations resulting in 73 (Taylor et al., 2000), 45 (Wu et al., 2000), and 81 (Bell et al., 2001) protein identifications and no information on posttranslational modifications. Recent progress in proteomic technology has enabled more comprehensive high-throughput profiling strategies of enriched organellar fractions, resulting in hundreds of proteins identifications, some of which were previously uncharacterized (Andersen et al., 2003; Mootha et al., 2003). These studies, although impressive, also lack the identification posttranslational modification sites. Clearly, proteomic strategies capable of identifying both soluble and membrane proteins as well as some posttranslational modifications will provide improved insights into organellar function.

Multidimensional protein identification technology (Mud-PIT) facilitates the goal of establishing more comprehensive organellar proteomes (Washburn et al., 2001). MudPIT has been optimized for the analysis of covalent modifications (MacCoss et al., 2002a; Wu et al., 2003) as well as membrane proteins (Washburn et al., 2001; Wu et al., 2003). MudPIT minimizes the bias against particular classes of proteins by first digesting the proteins into an even more complex mixture of peptides. These peptides are separated by microcapillary multidimensional chromatography interfaced directly with a tandem mass spectrometer by using electrospray ionization (Link et al., 1999). The peptide sequences and posttranslational modifications are determined by comparing experimentally acquired fragmentation spectra against theoretical spectra predicted from protein or nucleotide sequence information by using SEQUEST (Eng et al., 1994). Using DTASelect (Tabb et al., 2002), the resulting peptide sequences are reassembled back into protein identifications. This approach of identifying proteins and modifications is robust and facilitates proteomic profiling studies (Florens et al., 2002; Schirmer et al., 2003; Westermann et al., 2003).

Here, we describe an organellar proteomic analysis in which a stacked Golgi fraction was profiled using MudPIT. More than 400 proteins (421) were identified with a minimum of five independent peptide identifications per protein (>99% empirical confidence). Use of stringent criteria allowed us to identify 1) abundant proteins within the enriched fraction, including those with known and unknown functions and 2) sites of arginine dimethylation on proteins with high sequence coverage. Golgi localization was confirmed for two of the unknown membrane proteins, and interestingly, one of these is a predicted methyltransferase and is dimethylated on R230. Methyltransferase activity was subsequently confirmed in the Golgi fraction, and multiple Golgi and ER proteins were found to be arginine methylated in vitro. This organellar profiling study resulted in the generation of a new hypothesis regarding the role of methylation in the Golgi and serves to facilitate a better understanding of Golgi functions.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Materials
All chemicals were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise noted. Enzymes were purchased from Roche Applied Science (Indianapolis, IN). The arginine-dimethylated synthetic peptide LEWQPPPFR^*WLPVGPH (^* designates the site of arginine dimethylation) was purchased from Global Peptide Services (Fort Collins, CO). Male Sprague-Dawley rats (6 wk old, 250 g) were purchased from Harlan (Indianapolis, IN). All methods involving animals were approved by the institutional Animal Research Committee, accredited by the American Association for Accreditation of Laboratory Animal Care.

Isolation of Stacked Golgi Fraction
Rats were treated with cycloheximide to reduce total proteins in transit through the Golgi. Cycloheximide (50 mg/kg) was administered intraperitoneally 4 h before being sacrificed by halothane inhalation. Livers were removed from groups of 20 rats, and enriched Golgi fractions were prepared as described previously (Taylor et al., 1997). Briefly, livers were finely minced and resuspended at a ratio of 1 g of minced liver:1 ml of homogenization buffer (0.5 M phosphate-buffered sucrose containing 100 mM KH₂PO₄/K₂HPO₄, pH 6.8, 5 mM MgCl₂, and 4 µg each of proteolytic inhibitors chymostatin, leupeptin, antipain, and pepstatin). The sample was homogenized using a Polytron PT10/35 (Brinkmann, Westbury, NY) with one pass for 45 s moving from the top of the tube slowly to the bottom of the tube. The homogenate was centrifuged at low speed (1500 x g for 10 min at 4°C). The resulting postnuclear supernatant (PNS) was loaded in the middle of a sucrose step gradient (steps of 1.3 and 0.86 M sucrose were overlaid with the PNS, followed by a 0.25 M layer). The gradient was centrifuged at high speed (100,000 x g for 1 h at 4°C). The SII fraction (collected at the 0.5/0.86 M interface) was adjusted to 1.15 M sucrose, placed at the bottom of a second step gradient, and overlaid with steps 1.0, 0.86, and 0.25 M. The enriched Golgi fraction was collected at the 0.86/0.25 M interface. Protein concentrations of fractions were determined using DC protein assay (Bio-Rad, Hercules, CA).

Sample Digestion
The enriched Golgi fraction was digested to peptides using two different protocols. One protocol was the CNBr/formic acid method: Golgi samples (1 mg) were pelleted at 16,000 x g for 30 min at 4°C. The supernatant was discarded and the pellet was resuspended in 50 µl of 500 mg/ml CNBr in 90% formic acid and incubated in the dark in the fume hood overnight (Washburn et al., 2001). The pH of the sample was adjusted to 8.5 and then adjusted to 8 M urea, reduced (solution was adjusted to 25 mM dithiothreitol and incubated at 55°C for 20 min), and alkylated (solution was cooled to room temperature and adjusted to 100 mM iodoacetamide and incubated in the dark for 20 min). Endoproteinase Lys-C was added at a 1:500 (mass:mass) enzyme:substrate ratio and incubated at 37°C overnight in a Thermomixer (Brinkmann). The sample was then adjusted to 4 M urea and 1 mM CaCl₂. Modified trypsin was added at a 1:100 (mass:mass) enzyme:substrate ratio and incubated at 37°C overnight in a Thermomixer. The second protocol was the high pH/Proteinase K method: Golgi samples (1 mg) were pelleted at 16,000 x g for 30 min at 4°C. The supernatant was discarded, and the pellet was homogenized in 1 ml 0.2 M Na₂CO₃, pH 11 with five passes through an insulin syringe and incubated on ice for 1 h. The membrane sample was then adjusted to 8 M urea, reduced (solution is adjusted to 25 mM dithiothreitol and incubated at 55°C for 20 min), and alkylated (solution was cooled to room temperature and adjusted to 100 mM iodoacetamide and incubated in the dark for 20 min). Proteinase K (5 µg) was added to the sample and incubated at 37°C for 3 h in a Thermomixer. An additional aliquot of Proteinase K (5 µg) was added and incubated at 37°C for 1.5 h. The reaction was quenched with formic acid to 5% final concentration and microfuged at 16,000 x g at 4°C for 15 min to remove any insoluble particulates.

MudPIT Analysis
Protein digests were pressure-loaded onto a fused silica capillary desalting column containing 5 cm of 5-µm Aqua C18 material (Phenomenex, Ventura, CA) and washed as described previously (Wu et al., 2003). The desalted peptides were then eluted onto the back-end of a triphasic chromatography column consisting of 7 cm of 5-µm Aqua C18 material (Phenomenex), 3 cm of 5-µm Partisphere strong cation exchanger (Whatman, Clifton, NJ), and 3 cm of 5-µm hydrophilic interaction chromatography material (PolyLC). The column was then placed in-line with a Surveyor quaternary high-performance liquid chromatography (HPLC) pump (ThermoElectron, San Jose, CA) and analyzed using a 12-step separation described previously (Wu et al., 2003). The HPLC pump was operated at a flow rate of 100 µl/min and was split to obtain flow through the column of 100-400 nl/min. As peptides eluted from the microcapillary column, they were electrosprayed directly into an LCQ-Deca mass spectrometer (ThermoElectron) with the application of a 2-kV spray voltage applied distally to the waste of the HPLC split as described by Martin et al. (2000). A cycle of one full-scan mass spectrum (400-1400 m/z) followed by three data-dependent tandem mass spectrometry (MS/MS) spectra at a 35% normalized collision energy was repeated continuously throughout each step of the multidimensional separation. Application of the mass spectrometer scan-functions and HPLC solvent gradients were controlled by the Xcaliber data system (ThermoElectron).

Data Analysis
MS/MS spectra were analyzed using the following software analysis protocol. 2to3 determined the charge state (+2 or +3) of multiply charged peptide spectra and deleted poor quality spectra. Each MS/MS spectrum after 2to3 was searched against the RefSeq protein database containing the RefSeq rat, mouse, and human sequences concatenated into a single fasta file using SEQUEST (Eng et al., 1994). The MS/MS spectra were then researched to consider modifications of 1) +14 on R (methylation) and 2) +28 on R (dimethylation). All searches were parallelized and run on either the Yates Lab Beowulf computer cluster consisting of 34 1.2-GHz Athlon computer nodes or The Scripps Research Institute SGI cluster. All searches were performed without any enzyme specificity. The program DTASelect was used to filter peptide sequences from +1, +2, and +3 charged peptide precursors with normalized SEQUEST XCorr scores >0.3 (MacCoss et al., 2002b) and C_n > 0.1, to assemble the peptide sequences into proteins and to remove redundant protein sequences (Tabb et al., 2002). To minimize false positives and to identify abundant proteins within the Golgi fraction, only proteins with five or more peptides exceeding the peptide filters were considered. The MS/MS spectra from the modified peptides were filtered with DTASelect and manually evaluated using criteria reported previously (Link et al., 1999; Wu et al., 2003).

Plasmids
cDNAs (corresponding to #1/gi 27229118 and #2/gi 21703704 in Table 2) were acquired from Open Biosystems (Huntsville, AL) and sequenced using the in-house facility at the University of Colorado Health Sciences Center. Both full-length open reading frames were cloned into pEGFP-N2 (BD Biosciences Clontech, Palo Alto, CA). The #2/gi 21703704 open reading frame also was cloned into pGEX-6P-1, and the fusion protein was expressed and isolated using Bulk glutathione S-transferase purification module (Pharmacia, Peapack, NJ). The expressed unknown protein was cleaved from glutathione S-transferase by using PreScission Protease and collected for in-house immunization of rabbits.

Methyltransferase Assay
The in vitro methylation assay was carried out according to Lin et al., 2000. Briefly, 45 µg of stacked Golgi fraction in the presence and absence of 0.2% TX-100, was incubated with 4 µCi of [³H]S-adenosylmethionine (SAM) in 25 mM Tris, pH 7.5, 1 mM EDTA, 1 mM EGTA at a final volume of 60 µl for 60 min at 37°C in the presence of a cocktail of protease inhibitors. The reaction was stopped by addition of SDS-PAGE sample buffer. Proteins were resolved on a 12% polyacrylamide gel. The gels were infiltrated with Amplify (Amersham Biosciences, Piscataway, NJ), dried, and exposed to Kodak X-OMAT for 30 d.

In-Gel Trypsin Digestion
In-gel digests were performed as described previously (Taylor et al., 2000). Briefly, gel slices were excised and washed with 100 mM ammonium bicarbonate for 20 min. Proteins were reduced with 3 mM dithiothreitol/100 mM ammonium bicarbonate for 20 min at 55°C. After cooling to room temperature, iodoacetamide was added to 6 mM final concentration and incubated in the dark for 15 min at room temperature. The aqueous solution was discarded, and the gel slice was washed in 50% acetonitrile/100 mM ammonium bicarbonate for 20 min. Each gel slice was cut into 1-mm³ pieces, dried, and reswelled with 0.2 µg of modified trypsin/25 mM ammonium bicarbonate overnight at 37°C. Peptides were extracted from the gel with 100 µl of 60% acetonitrile/0.1% trifluoroacetic acid for 20 min. The supernatants were lyophilized and the peptides were reconstituted in 10 µl of 5% formic acid immediately before analysis by mass spectrometry.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Proteomic Survey of an Enriched Stacked Golgi Fraction
Stacked Golgi fractions were enriched from rat livers and characterized as described previously (Taylor et al., 1997). Golgi samples were prepared for proteomic analysis by using two methods reported to optimize for membrane proteins (Washburn et al., 2001; Wu et al., 2003). The peptide mixtures resulting from the digestions were analyzed using MudPIT. Briefly, the complex peptide samples were resolved and mass analyzed by multidimensional chromatography coupled to MS/MS. The acquired tandem mass spectra were then searched against a RefSeq protein database containing rat, mouse, and human sequences by using SEQUEST (Eng et al., 1994). Protein identifications required five or more peptides to minimize false positives and to identify abundant proteins within the Golgi fraction. Using these stringent criteria, 421 proteins were identified. Most of these proteins are currently annotated in databases and were sorted into categories based on reported localizations and functions. Figure 1A provides an overview of all identified proteins in the Golgi fraction by subcellular localization and/or functional classifications. Bona fide Golgi and ER proteins are listed in Table 1, whereas proteins belonging to other classifications are available in Supplementary Table 1. Although not included in Table 1, many of the proteins listed in Supplementary Table 1 are essential for Golgi function, for example, cytosolic and cytoskeletal proteins. Others are molecules in transit, either to be secreted or en route to other organelles. Because it is well documented that no cell fractionation technique produces a "pure" preparation, proteins from other organelles such as mitochondria and perixosomes may be present as minor contaminants of the Golgi fraction. These would not be detectable by other less sensitive techniques. Therefore, because verification of the functional significance of these proteins in the Golgi requires both localization studies and functional assays, to be conservative, we listed these proteins in their most widely accepted categories.

View larger version (47K): [in this window] [in a new window]   Figure 1. Functional classification of the Golgi proteome. (A) Overall profile of organelle and functional categories of protein identified from a stacked Golgi fraction. (B) Golgi-specific proteins fall in eight different functional classes and include 64% predicted membrane proteins.

Of the 421 proteins identified in the Golgi fraction, 110 are previously documented resident Golgi proteins (Table 1). Of these, 70 proteins (64%) are predicted transmembrane proteins (HMMTOP software, version 2.0). Multiple families of Golgi proteins are well represented (Figure 1B). Importantly, previous Golgi proteomes reported very few Golgi transferases (for glycosylation and sulfation) (Taylor et al., 2000; Wu et al., 2000; Bell et al., 2001). In this study, 23 glycosylation (21%) and six sulfation (5%) enzymes were identified. Furthermore, three of the transferases involved in glycosylation were later shown to be arginine dimethylated (see below; Table 3).

Identification of Novel Putative Golgi Proteins
Of the proteins identified in the Golgi fraction, 41 were proteins with no previously reported functions (Table 2). To establish whether these were Golgi residents or proteins from other organelles, it was necessary to determine their intracellular localization. For practical purposes, we selected two of the unknown predicted transmembrane proteins (protein #1/Q9DD20/putative methyltransferase and protein #2/Q8VCS2) from Table 2. Sequence analysis of Q9DD20 suggested a 20-amino acid N-terminal transmembrane domain and a domain with high sequence similarity to the SAM binding domain KOG4300 in the Conserved Domain Database at National Center for Biotechnology Information (Figure 3A, yellow and green shaded areas, respectively). SAM is the donor molecule for most methyltransferases. Therefore, this domain prediction suggested that the protein may be a putative methyltransferase (Schubert et al., 2003). Corresponding cDNAs were acquired for both unknown proteins, fused with GFP and expressed in NRK cells. When expressed in NRK cells, colocalizations with various markers were used to determine their subcellular localization. Both proteins localized to the cis-Golgi judged by their colocalization with known Golgi markers (Figure 2). The top panels show that the presumptive EGFP-tagged methyltransferase (MethT-EGFP, Q9DD20) colocalizes with the cis-Golgi protein GM130. The bottom panels show that an antibody against Q8VCS2 colocalizes this protein with the cis-Golgi marker giantin. These results confirm that two of the 41 unknown proteins are Golgi localized.

View larger version (36K): [in this window] [in a new window]   Figure 3. Identification of arginine dimethylation on putative methyltransferase. The protein identified by gi 27229118 NP_082129 [GenBank] . RIKEN cDNA 061000 was selected from the list of proteins of unknown function for further analysis. (A) Protein sequence is displayed in boxed text. Shaded regions indicate coverage by identified peptides (gray), the predicted TMD (yellow), and the predicted SAM binding domain (green). Peptides are displayed below the protein sequence in blue text. Modified peptides are displayed in bold blue text, and modification site is indicated on the protein sequence with an arrow. The modified peptide tagged with ** was synthesized to confirm the modification site. (B) Spectrum of the dimethylarginine modified peptide acquired from Golgi MudPIT analysis is displayed on the left. Spectrum of the synthetic dimethylarginine-modified peptide is displayed on the right. Spectra are annotated using Roepstorff and Fohlman nomenclature (Roepstorff and Fohlman, 1984).

View larger version (18K): [in this window] [in a new window]   Figure 2. Subcellular localization of proteins of unknown function identified by subcellular proteomics. Fluorescence detection of GFP and antibody staining was performed in NRK cells. Top, immunofluorescence shows the localization of cis-Golgi protein GM130 (red, left). The fluorescence from the expression of MethT-EGFP (protein #1 in Table 2/Q9DD20) (green, middle) is shown to overlap with GM130. Bottom, double immunofluorescence shows the colocalization of cis-Golgi marker giantin (red, left) and Q8VCS2 (protein #2 in Table 2) (green, middle). Both unknown proteins are highly concentrated in the perinuclear region.

Identifications of Arginine Dimethylated Proteins in the Golgi Proteome
To search for methylation of arginine residues in the proteome, the MS/MS spectra were researched against the subset database to consider modifications of 1) +14 on R (methylation) and 2) +28 on R (dimethylation). A subset database including only rat sequences was used to expedite SEQUEST differential modification searches. Eighteen proteins were identified to be arginine dimethylated in the Golgi proteome and of these, 10 were confirmed Golgi proteins, including the methyltransferase Q9DD20 (Table 3). Protein arginine methylation is a posttranslational modification that results in the addition of monomethyl or dimethyl groups to the guanidino group of arginine. As expected, most of the proteins represented are the abundant proteins with high sequence coverage. The methylated proteins are not all Golgi localized but reside in multiple subcellular organelles. The methylated residues are localized on protein domains reported to be either lumenal or cytoplasmic. This implies that there are at least two methyltransferases necessary for these modifications and that a SAM transporter is present in the Golgi membrane.

Interestingly, the putative methyltransferase, was identified to be dimethylated at Arginine-230 by 12 independent overlapping peptides (Figure 3A, bolded blue peptides). Arg-230 is located near the C terminus of the protein and the localization is predicted to be cytoplasmic. To verify the modification, one of the identified arginine dimethylated peptides, LEWQPPPFR^*WLPVGPH (modified residue labeled with an asterisk), was synthesized. Tandem mass spectra were collected from the synthetic peptide and compared against the spectrum acquired in the Golgi sample (Figure 3B). The spectra from the Golgi fraction and the synthetic peptide are nearly indistinguishable. Both spectra contain prominent y5, y12, and b11 fragment ions resulting from the favored fragmentation at the N-terminal side of proline residues.

Little is known about the functional and regulatory implications of arginine methylation of nonnuclear proteins, but recent evidence has indicated a role in Golgi function. The drug ilimaquinone, which inhibits the synthesis of the methyl donor SAM, vesiculates the Golgi complex and blocks secretion (Takizawa et al., 1993; Radeke et al., 1999). Addition of SAM can reverse the ilimaquinone effect (Casaubon and Snapper, 2001). Together with our proteomic evidence for arginine dimethylation of multiple Golgi proteins, these results suggest that the Golgi has endogenous methyltransferase activity. To test this prediction, the enriched stacked Golgi fraction was incubated with [³H]SAM plus/minus TX-100 (Lin et al., 2000). The resulting samples were resolved by SDS-PAGE, and the dried gels exposed to film. More than 10 bands were labeled (our unpublished data). These were excised from the dried gels and in-gel digested with trypsin for protein identification by mass spectrometry. Three proteins from the original list of dimethylated Golgi proteins (-manosidase 1/#1, cytochrome 450 2d2/#2, and Tmp23/#3) (Table 3) were identified in these bands (Figure 4). These data support the hypothesis that select Golgi and ER membrane proteins are modified by dimethylation of arginine residues.

View larger version (41K): [in this window] [in a new window]   Figure 4. Methyltransferase activity is detected in the stacked Golgi fraction. Stacked Golgi fractionated from rat liver was incubated with [3H]SAM in the presence and absence of 0.2% TrintonX-100 and resolved by SDS-PAGE. The gel was infiltrated with fluorographic reagent, dried, and exposed to film. Different labeled bands were detected in the presence and absence of detergent. The labeled bands were excised, digested with trypsin, and identified by mass spectrometry.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
We have performed a proteomic survey of an enriched stacked Golgi fraction from rat liver and have analyzed the results in the context of publicly available annotations. Our proteomic analysis is a survey of the abundant proteins within the Golgi fraction. All protein identifications have high confidence by using a minimum of five independent peptides for identification. Because the likelihood of obtaining multiple peptide identifications per protein is greater for highly abundant proteins (Gao et al., 2003), these criteria favor the analysis of the most abundant proteins within the fraction. This approach was used to optimally identify abundant proteins of unknown function within the fraction. Approximately 10% of the identified proteins in this Golgi fraction are unknowns (41 proteins listed in Table 2) and represent a rich source of candidates to explore novel functions or integrate known functions of this organelle.

Many known ER proteins were identified in the Golgi fraction. The ER physically adheres to multiple trans-Golgi cisternae (Novikoff, 1964; Ladinsky et al., 1999; Marsh et al., 2001), and this interaction is postulated to be the site of transfer of ceramide to sphingomyelin synthase, which facilitates the synthesis of sphingolipid for sorting and exit at the trans-Golgi (van Meer and Lisma, 2002; Munro, 2003; Hanada et al., 2003). As expected, the ER remains attached to the stacked Golgi during the fractionation protocol (our unpublished data), and ER proteins account for 22% of the total proteins identified. Other proteins fall into categories that are functionally linked to the Golgi or in transit through the Golgi. These proteins have been listed in separate categories. Twenty-six percent of the total proteome are bona fide, literature confirmed, Golgi proteins with the transferases forming the largest group. Because the Golgi proteome will vary in different cells and tissues and in different functional states, this study serves as a baseline to be compared in future studies.

A number of proteins in the isolated Golgi fraction were posttranslationally modified with dimethylation of arginine residues. One of these methylated proteins has a predicted SAM binding domain, suggesting that it is a methyltransferase. All methyltransferases are thought to directly transfer the methyl group from SAM to a substrate/acceptor and SAM binding domains have a high degree of homology (Schubert et al., 2003). Methylation of histones for chromatin regulation and gene silencing is a well-studied example. DNA is also methylated and abnormal methylation patterns on DNA are a nearly universal finding in cancer, making this an actively studied area today (Laird, 2003). Interestingly, from the point of view of our study, most methylated nuclear proteins are modified in the cytosol, and known methyltransferases are cytosolic (Friesen et al., 2001a,b). In contrast to methylated nuclear proteins, however, little is known about methylation of proteins elsewhere in the cell or the functional and regulatory implications of this modification (McBride and Silver, 2001). This protein seems to be a member of a novel methyltransferase family, and our data suggests that there are two methyltransferases (cytosolic and lumenal) as well as a SAM transporter in the Golgi membrane. Because little is known about methylation of ER and Golgi proteins, this represents an exciting new area of study.

The concept arising from the nuclear methylation data is that methylation (especially of arginines) promotes specific protein-protein interactions required in assembly of functional complexes (Friesen et al., 2001a,2001b). We know little about functional complexes formed with the proteins in the Golgi proteome. One of the big quandaries in the field is that arginine methylation seems to be irreversible. Methylases have never been identified, neither as an activity nor as an enzyme. However, most investigators believe that methylation is regulatory and that methylases will be uncovered (Bannister et al., 2002).

Compelling data suggest that methylation of Golgi proteins is important and physiologically relevant. Many of the golgins, a group of matrix-like and tethering proteins, are direct targets in autoimmune disease (Shields and Arvan, 1999; Doyle and Mamula, 2002; Nozawa et al., 2002). The question to be addressed is are the golgins methylated when they become targets for the autoimmune reaction? The theory that methylation is a factor in autoimmune disease is based on numerous reports from the multiple sclerosis field and other diseases involving demyelination. At the outset of the disease, myelin basic protein becomes symmetrically arginine dimethylated and, in parallel, phosphorylation is greatly reduced or absent (Kim et al., 2003). Although many golgins are reversibly phosphorylated, there is no information available about their methylation or coordinated methylation/phosphorylation. In the histone field, it is postulated that coordinated dimethylation and phosphorylation regulate gene silencing. Allis and colleagues have proposed the hypothesis that binary switches of dimethylated and reversibly phosphorylated neighboring residues in defined cassettes can regulate protein-protein interaction (Fischle et al., 2003). These models provide a very compelling argument for studying methylation of Golgi proteins.

The data obtained from these studies will be further mined to obtain information on other posttranslational modifications. Goals for the future will be to make organellar proteomics a robust and high-throughput tool that will facilitate the understanding of global changes in protein expression and modification that occur with cellular function and disease states.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
We acknowledge support from the National Institutes of Health grants F32-AI54333 (to C.C.W.), F32-DK59731 (to M.J.M.), R01-MH067880 (to J.R.Y.), and R01-GM42629 (to K.E.H.).

	Footnotes

 
Article published online ahead of print. Mol. Biol. Cell 10.1091/mbc.E04-02-0101. Article and publication date are available at www.molbiolcell.org/cgi/doi/10.1091/mbc.E04-02-0101.

Online version of this article contains supporting material. Online version is available at www.molbiolcell.org.

Present address: Department of Genome Sciences, University of Washington, Seattle, WA 91895.

These authors contributed equally to this work.

^|| Corresponding author. E-mail address: kathryn.howell{at}uchsc.edu.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Allan, V.J., Thompson, H.M., and McNiven, M.A. (2002). Motoring around the Golgi. Nat. Cell Biol. 10, 236-242.

Andersen, J.S., Wilkinson, C.J., Mayor, T., Mortensen, P., Nigg, E.A., and Mann, M. (2003). Proteomic characterization of the human centrosome by protein correlation profiling. Nature 426, 570-574.[CrossRef][Medline]

Bannister, A.J., Schneider, R., and Kouzarides, T. (2002). Histone methylation: dynamic or static? Cell 109, 801-806.[CrossRef][Medline]

Bell, A.W., et al. (2001). Proteomics characterization of abundant Golgi membrane proteins. J. Biol. Chem. 276, 5152-565.[Abstract/Free Full Text]

Casaubon, R.L., and Snapper, M.L. (2001). S-adenosylmethionine reverses ilimaquinone's vesiculation of the Golgi apparatus: a fluorescence study on the cellular interactions of ilimaquinone. Bioorg. Med. Chem. Lett. 11, 133-136.[CrossRef][Medline]

Colanzi, A., Suetterlin, C., and Malhotra, V. (2003). Cell-cycle-specific Golgi fragmentation: how and why? Curr. Opin. Cell Biol. 15, 462-467.[CrossRef][Medline]

Donaldson, J.G., and Lippincott-Schwartz, J. (2000). Sorting and signaling at the Golgi complex. Cell 101, 693-696.[CrossRef][Medline]

Doyle, H.A., and Mamula, M.J. (2002). Posttranslational protein modifications: new flavors in the menu of autoantigens. Curr. Opin. Rheumatol. 1, 244-249.

Eng, J.K., McCormack, A.L., and Yates III, J.R. (1994). An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J. Am. Soc. Mass. Spectrom. 5, 967-989.

Fischle, W., Wang, Y., and Allis, C.D. (2003). Binary switches and modification cassettes in histone biology and beyond. Nature 425, 475-479.[CrossRef][Medline]

Florens, L., et al. (2002). A proteomic view of the Plasmodium falciparum life cycle. Nature 419, 520-526.[CrossRef][Medline]

Friesen, W.J., Massenet, S., Paushkin, S., Wyce, A., and Dreyfuss, G. (2001a). SMN, the product of the spinal muscular atrophy gene, binds preferentially to dimethylarginine-containing protein targets. Mol. Cell. 7, 1111-1117.[CrossRef][Medline]

Friesen, W.J., Paushkin, S., Wyce, A., Massenet, S., Pesiridis, G.S., Van Duyne, G., Rappsilber, J., Mann, M., and Dreyfuss, G. (2001b). The methylosome, a 20S complex containing JBP1 and pICln, produces dimethylarginine-modified Sm proteins. Mol. Cell. Biol. 21, 8289-8300.[Abstract/Free Full Text]

Gao, J., Opiteck, G.J., Friedrichs, M.S., Dongre, A.R., and Hefta, S.A. (2003). Changes in the protein expression of yeast as a function of carbon source. J. Proteome Res. 2, 643-649.[CrossRef][Medline]

Hanada, K., Kumagai, K., Yasuda, S., Miura, Y., Kawano, M., Fukasawa, M., and Nishijima, M. (2003). Molecular machinery for non-vesicular trafficking of ceramide. Nature 426, 803-809.[CrossRef][Medline]

Huber, L.A., Pfaller, K., and Vietor, I. (2003). Organelle proteomics: implications for subcellular fractionation in proteomics. Circ. Res. 92, 962-968.[Abstract/Free Full Text]

Kim, J.K., Mastronardi, F.G,. Wood, D.D., Lubman, D.M., and Moscarello, M.A. (2003). Multiple sclerosis: an important role for post-translational modifications of myelin basic protein in pathogenesis. Mol. Cell Proteomics 2, 453-462.[Abstract/Free Full Text]

Ladinsky, M.S., Mastronarde, D.N., McIntosh, J.R., Howell, K.E., and Staehelin. L.A. (1999). Golgi structure in three dimensions: functional insights from the normal rat kidney cell. J. Cell Biol. 144, 1135-1149.[Abstract/Free Full Text]

Laird, P.W. (2003). The power and the promise of DNA methylation markers. Nat. Rev. Cancer 4, 253-266.

Lin, C.H., Hsieh, M., Li, Y.C., Li, S.Y., Pearson, D.L., Pollard, K.M., and Li, C. (2000). Protein N-arginine methylation in subcellular fractions of lymphoblastoid cells. J. Biochem. 128, 493-498.[Abstract/Free Full Text]

Link A.J., Eng, J., Schieltz, D.M., Carmack, E., Mize, G.J., Morris, R.D., Garvik, B.M., and Yates III, J.R. (1999). Direct analysis of protein complexes using mass spectrometry. Nat. Biotechnol. 17, 676-682.[CrossRef][Medline]

Lowe, M., Nakamura, N., and Warren, G. (1998). Golgi division and membrane traffic. Trends Cell Biol. 8, 40-44.[CrossRef][Medline]

MacCoss, M.J., et al. (2002a). Shotgun identification of protein modifications from protein complexes and lens tissue. Proc. Natl. Acad. Sci. USA 99, 7900-7905.[Abstract/Free Full Text]

MacCoss, M.J., Wu, C.C., and Yates III, J.R. (2002b). Probability-based validation of protein identifications using a modified SEQUEST algorithm. Anal. Chem. 74, 5593-5599.[Medline]

Marsh, B.J., Mastronarde, D.N., Buttle, K.F., Howell, K.E., and McIntosh, J.R. (2001). Organellar relationships in the Golgi region of pancreatic beta cell line, HIT-T15, visualized by high resolution electron tomography. Proc. Nat. Acad. Sci. USA 98, 2399-2406.[Abstract/Free Full Text]

Martin, S.E., Shabanowitz, J., Hunt, D.F., and Marto, J.A. (2000). Subfemtomole MS and MS/MS peptide sequence analysis using nano-HPLC micro-ESI Fourier transform ion cyclotron resonance mass spectrometry. Anal. Chem. 72, 4266-4274.[Medline]

McBride, A.E., and Silver, P.A. (2001). State of the Arg: protein methylation at arginine comes of age. Cell 106, 5-8.[CrossRef][Medline]

Mootha, V.K., et al. (2003). Integrated analysis of protein composition, tissue diversity, and gene regulation in mouse mitochondria. Cell 115, 629-640.[CrossRef][Medline]

Munro, S. (2003). Cell biology: earthworms and lipid couriers. Nature 42, 775-776.

Nozawa, K., Casiano, C.A., Hamel, J.C., Molinaro, C., Fritzler, M.J., and Chan. C. (2002). Fragmentation of Golgi complex and Golgi autoantigens during apoptosis and necrosis. Arthritis Res. 4, 1-9.[Medline]

Novikoff, A.B. (1964). GERL, its form and function in neurons of rat spinal ganglia. Biol. Bull. 127, 358

Palade, G.E. (1975). Intracellular aspects of the process of protein secretion. Science 189, 347-358.[Free Full Text]

Radeke, H.S., Digits, C.A., Casaubon, R.L., and Snapper, M.L. (1999). Interactions of (-)-ilimaquinone with methylation enzymes: implications for vesicular-mediated secretion. Chem. Biol. 9, 639-647.

Roepstorff, P. and Fohlman J. (1984). Proposal for a common nomenclature for sequence ions in mass spectra of peptides. Biomed. Mass. Spectrom. 11, 601.[CrossRef][Medline]

Schirmer, E.C., Florens, L., Guan, T., Yates III, J.R., and Gerace, L. (2003). Nuclear membrane proteins with potential disease links found by subtractive proteomics. Science 301, 1380-1382.[Abstract/Free Full Text]

Schubert, H.L., Blumenthal, R.M., and Cheng, X. (2003). Many paths to methyltransfer: a chronicle of convergence. Trends Biochem. Sci. 28, 329-335.[CrossRef][Medline]

Shields, D., and Arvan, P. (1999). Disease models provide insights into post-Golgi protein trafficking, localization and processing. Curr. Opin. Cell Biol. 11, 489-494.[CrossRef][Medline]

Tabb, D.L., McDonald, H.D., and Yates, III, J.R. (2002). DTASelect and Contrast: tools for assembling and comparting protein identification form Shotgun proteomics. J. Proteome Res. 1, 21-26.[Medline]

Takizawa, P.A., Yucel, J.K. Veit B., Faulkner, D.J. Deerinck, T., Soto, G., Ellisman, M., and V. Malhotra, V. (1993). Complete vesiculation of Golgi membranes and inhibition of protein transport by a novel sea sponge metabolite, ilimaquinone. Cell 73, 1079-1090.[CrossRef][Medline]

Taylor, R.S., Jones, S.M., Dahl, R.H., Nordeen, M.H., and Howell, K.E. (1997). Characterization of the Golgi complex cleared of proteins in transit and examination of calcium uptake activities. Mol. Biol. Cell 8, 1911-1931.[Abstract/Free Full Text]

Taylor, R.S., Wu, C.C., Hays, L.G., Eng, J.K., Yates, J.R. 3rd, and Howell, K.E. (2000). Proteomics of rat liver Golgi complex: minor proteins are identified through sequential fractionation. Electrophoresis. 16, 3441-3459.

Van Lint, J., Rykx, A., Maeda, Y., Vantus, T., Sturany, S., Malhotra, V., Vandenheede, J.R., and Seufferlein, T. (2002). Protein kinase D: an intracellular traffic regulator on the move. Trends Cell Biol. 12, 193-200.[CrossRef][Medline]

van Meer, G., and Lisma, Q. (2002). Sphingolipid transport: rafts and translocators. J. Biol. Chem. 277, 25855-25858.[Free Full Text]

Ward, T.H., Polishchuk, R.S., Caplan, S., Hirschberg. K., and Lippincott-Schwartz, J. (2001). Maintenance of Golgi structure and function depends on the integrity of ER export. J. Cell Biol. 155, 557-570.[Abstract/Free Full Text]

Washburn, M.P., Wolters, D. and Yates III, J.R. (2001). Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 19, 242-247.[CrossRef][Medline]

Westermann, S., Cheeseman, I.M., Anderson, S., Yates III, J.R., Drubin, D.G., and Barnes, G. (2003). Architecture of the budding yeast kinetochore reveals a conserved molecular core. J. Cell Biol. 163, 215-222.[Abstract/Free Full Text]

Wu, C.C., MacCoss, M.J., Howell, K.E., and Yates III, J.R. (2003). A method for the comprehensive proteomic analysis of membrane proteins. Nat. Biotechnol. 21, 532-538.[CrossRef][Medline]

Wu, C.C., and Yates III, J.R. (2003). The application of mass spectrometry to membrane proteomics. Nat. Biotechnol. 21, 262-267.[CrossRef][Medline]

Wu, C.C., Yates III, J.R., Neville, M.C., and Howell, K.E. (2000). Proteomic analysis of two functional states of the Golgi complex in mammary epithelial cells. Traffic 1, 769-782.[CrossRef][Medline]

This article has been cited by other articles:

				S. Y. Dejgaard, A. Murshid, A. Erman, O. Kizilay, D. Verbich, R. Lodge, K. Dejgaard, T. B. N. Ly-Hartig, R. Pepperkok, J. C. Simpson, et al. Rab18 and Rab43 have key roles in ER-Golgi trafficking J. Cell Sci., August 15, 2008; 121(16): 2768 - 2781. [Abstract] [Full Text] [PDF]

				T. P. J. Dunkley, S. Hester, I. P. Shadforth, J. Runions, T. Weimar, S. L. Hanton, J. L. Griffin, C. Bessant, F. Brandizzi, C. Hawes, et al. Mapping the Arabidopsis organelle proteome PNAS, April 25, 2006; 103(17): 6518 - 6523. [Abstract] [Full Text] [PDF]

				K. S Lilley and P. Dupree Methods of quantitative proteomics and their application to plant organelle characterization J. Exp. Bot., April 1, 2006; 57(7): 1493 - 1499. [Abstract] [Full Text] [PDF]

				F. Bachand, D. H. Lackner, J. Bahler, and P. A. Silver Autoregulation of ribosome biosynthesis by a translational response in fission yeast. Mol. Cell. Biol., March 1, 2006; 26(5): 1731 - 1742. [Abstract] [Full Text] [PDF]

				R. J. Thompson, H. C. S. R. Akana, C. Finnigan, K. E. Howell, and J. H. Caldwell Anion channels transport ATP into the Golgi lumen Am J Physiol Cell Physiol, February 1, 2006; 290(2): C499 - C514. [Abstract] [Full Text] [PDF]

				G. A. Mardones, C. M. Snyder, and K. E. Howell Cis-Golgi Matrix Proteins Move Directly to Endoplasmic Reticulum Exit Sites by Association with Tubules Mol. Biol. Cell, January 1, 2006; 17(1): 525 - 538. [Abstract] [Full Text] [PDF]

				M. Girard, P. D. Allaire, P. S. McPherson, and F. Blondeau Non-stoichiometric Relationship between Clathrin Heavy and Light Chains Revealed by Quantitative Comparative Proteomics of Clathrin-coated Vesicles from Brain and Liver Mol. Cell. Proteomics, August 1, 2005; 4(8): 1145 - 1154. [Abstract] [Full Text] [PDF]

				M. J. Roth, A. J. Forbes, M. T. Boyne II, Y.-B. Kim, D. E. Robinson, and N. L. Kelleher Precise and Parallel Characterization of Coding Polymorphisms, Alternative Splicing, and Modifications in Human Proteins by Mass Spectrometry Mol. Cell. Proteomics, July 1, 2005; 4(7): 1002 - 1008. [Abstract] [Full Text] [PDF]

				F.-M. Boisvert, C. A. Chenard, and S. Richard Protein Interfaces in Signaling Regulated by Arginine Methylation Sci. Signal., February 15, 2005; 2005(271): re2 - re2. [Abstract] [Full Text] [PDF]

				T. P. J. Dunkley, R. Watson, J. L. Griffin, P. Dupree, and K. S. Lilley Localization of Organelle Proteins by Isotope Tagging (LOPIT) Mol. Cell. Proteomics, November 1, 2004; 3(11): 1128 - 1134. [Abstract] [Full Text] [PDF]

				B. A. Wetmore and B. A. Merrick Invited Review: Toxicoproteomics: Proteomics Applied to Toxicology and Pathology Toxicol Pathol, October 1, 2004; 32(6): 619 - 642. [Abstract] [PDF]

This Article Abstract Full Text (PDF) Supplemental Table A correction has been published All Versions of this Article: E04-02-0101v1 15/6/2907    most recent Alert me when this article is cited