Original Paper
|
|
|
file on Synergy |
Acta Biochim Biophys
Sin 2008, 40: 5570
doi:10.1111/j.1745-7270.2008.00374.x
dataset of the plasma membrane
proteome of nasopharyngeal carcinoma cell line HNE1 for uncovering protein
function
Lijun Zhang1#,
Xiaofang Jia1#,
Xiaohui Liu1,
Tingting Sheng1,
Rui Cao1,
Quanyuan He1,
Zhen Liu1,
Xia Peng1,
Jixian Xiong1,
Pengfei Zhang2,
Ni Shi1,
and Songping Liang1*
1 College of Life Sciences, Hunan Normal
University, Changsha 410081, China
2 Key Laboratory of Cancer Proteomics of the
Chinese Ministry of Health, Xiangya Hospital, Central South University,
Changsha 410081, China
Received: August 1,
2007
Accepted: September
18, 2007
This work was
supported by the grants from the National 973 Project of China (No.
2001CB5102), the Chinese Human Liver Proteome Project (No. 2004 BA711A11), the
National Natural Science Foundation of China (Nos. 30000028 and 30240056), and
the Program for Changjiang Scholars and Innovative Research Team in University
(No. IRT0445)
#
These
authors contributed equally to this work
*Corresponding
author: Tel, 86-731-8872556; Fax, 86-731-8861304; E-mail, [email protected]
Nasopharyngeal
carcinoma (NPC) is a commonly occurring tumor in southern China and Southeast
Asia. The current study focused on developing an extensive analysis method for
the peripheral and integral proteins of NPC cell line HNE1. The peripheral
membrane proteins were extracted by biotinylated enrichment, 0.1 M Na2CO3, and H2O. Integral or total plasma membrane fractions were prepared
using 30% Percoll density grade centrifugation with or without 0.1 M Na2CO3 treatment
and evaluated by Western blot analysis. The proteins were subjected to
two-dimensional electrophoresis combined with tandem mass spectrometry, sodium
dodecyl sulfate-polyacrylamide gel electrophoresis combined with tandem mass
spectrometry, and shotgun analysis. We identified 371, 180, and 702 proteins from
peripheral, integral, and total plasma membrane fractions, respectively. In
all, 848 non-redundant proteins (534 groups) were identified. Binding,
catalytic, and structural molecules were the major classes. In addition to the
known cell surface markers of NPC cells, the analysis revealed 311 proteins
involved in multiple cell-signaling pathways and 25 proteins in disease
pathways that are characteristic of cancer cells. By searching the
Differentially Expressed Protein Database (http://protchem.hunnu.edu.cn/depd/index.jsp),
199 proteins were found to be differentially expressed in previous cancer
proteome research. A 671 protein-protein interaction network was obtained,
including 178 identified proteins in this work. The plasma membrane
localization of five proteins was confirmed by immunological techniques,
validating this proteomic strategy. Our study could offer some help for
understanding the molecular mechanism of NPC.
Keywords nasopharyngeal
carcinoma; plasma membrane; proteome; protein-protein interaction; immunocytochemistry
Nasopharyngeal carcinoma (NPC) is a
malignancy with high incidence in southern China and Southeast Asia [1].
Patients with NPC tend to present at an advanced stage of disease because the
primary anatomical site of tumor growth is located in a silent area, and the
tumors show a high metastatic potential [1–3]. Previous studies [3–6] have found that NPC might be related to
Epstein-Barr virus infection, certain environmental conditions, some genetic
factors, and diet (e.g., nitrosamine and nitrite intake). However, the
molecular basis for NPC is not fully understood. Therefore, it is very
important to study the molecular mechanism and look for a rapid diagnostic
assay for early detection and treatment of NPC.
NPC cell lines have long been used by
researchers as model systems for understanding the disease process itself
[6,7]. In proteomic research, direct measurement of protein expression and regulation
in cancer cells has become a goal, the successful attainment of which could
lead to a more fundamental understanding of the factors that lead to the onset
and progression of nasopharyngeal cancer, thus to more effective diagnostic
procedures and the identification of potential therapeutic targets. HNE1, one
of the widest-used NPC cell lines bearing wild-type p53 [8], was used as
a model for proteomics research in this work.
The cell surface membrane is of substantial
interest with regard to various aspects of disease, from molecular diagnosis to
therapeutics. Numerous cell surface proteins represent therapeutic targets. For
example, the discovery that the gene for a growth factor receptor (HER2)
is amplified in breast tumors and its protein product is consequently
overexpressed at the cell surface has led to an effective form of therapy for
breast cancer using an antibody that targets HER2 [9]. Another example
is annexin A1, a calcium-regulated membrane-binding protein known to be
overexpressed in lung cancer cell plasma membrane (PM). Radio-immunotherapy to
annexin A1 destroys tumors and increases animal survival [10].
Thus, comprehensive profiles of PM proteins
in HNE1 cells will facilitate our understanding of their critical roles in
biological processes such as cell-to-cell adhesion, cell signaling, and ion
transport. Additionally, profiling cell surface proteins will increase our
understanding of the biological process of NPC and facilitate target
identification for developing biomedical therapeutics. Comprehensive profiles
of cell surface proteins in NPC, however, are not yet available.
In the present work, to investigate the
cell surface proteome, one- and two-dimensional electrophoresis combined with
tandem mass spectrometry, and shotgun methods were used simultaneously.
Together, 848 non-redundant proteins were identified, of which 371 proteins
were from cell peripheral PM, 702 from PM, and 180 from integral PM. Analysis
of identified proteins indicated that HNE1 cells express a wide variety of cell
surface markers, cancer-related proteins, and signaling molecules (such as
receptors, transporters, and cell adhesion molecules). A complex network was
constructed according to the interaction of proteins in the cell surface.
Moreover, PM localization of five proteins was shown by immunocytochemistry.
This study serves four objectives: (1) to
provide a set of methods to study cell surface proteins, (2) to characterize
the HNE1 cell surface membrane proteome and identify many new proteins that
potentially play a critical role in NPC biogenesis and function, (3) to
construct the network of NPC cell surface proteins, and (4) to uncover the
protein functions involved in disease or cancer pathways and processes.
Materials and Methods
Materials
RPMI 1640, trypsin, and
penicillin/streptomycin were obtained from Invitrogen (Carlsbad, USA). Fetal
bovine serum was purchased from Tianjing Blood Institute (Tianjing, China).
EZ-LinkTM sulfo-NHS-LC-biotin and ImmunoPure Monomeric Avidin
kits were from Pierce (Rockford, USA). Proteomics Sequencing Grade trypsin,
dithiothreitol, iodacetamide, trifluoroacetic acid, HEPES,
1-(4-trimethylammoniumphenyl)-6-phenyl-1,3,5-hexatriene p-toluenesulfonate, and
Na2CO3
were obtained from Sigma-Aldrich (St. Louis, USA). Acrylamide, bis-acrylamide,
urea, glycine, Percoll, Tris and sodium dodecyl sulfate (SDS) were from GE
(Solon, USA). Centricon YM-3 columns and Immobilon-P polyvinylidene difluoride
membranes were purchased from Millipore (Bedford, USA). Bio-Rad DC protein
assay kit was from Bio-Rad Laboratories (Hercules, USA). Anti-NADH ubiquinol
oxydoreductase 39, anti-fibrillarin, anti-valosin-containing protein (VCP),
and anti-non-metastatic 23 (nm23) were from abcam
(Cambridge, UK). Anti-keratin 8 and anti-heat shock protein 70 were from Lab
Vision (Fremont, USA). Anti-human galectin-1 was from Cytolab (Rehovot,
Israel). Anti-flotillin immunoglobulin G (IgG) monoclonal antibody and
horseradish peroxidase-conjugated anti-mouse IgG were obtained from BD
Biosciences (San Jose, USA). LumiGLO Chemiluminescent Substrate and fluorescein-isothiocyanate-conjugated
or tetramethylrhodamine isothiocyanate-conjugated goat anti-rabbit or mouse IgG were from KPL (Gaithersburg, USA).
HPLC-grade acetonitrile and acetone were from the Chinese National Medicine
Group, Shanghai Chemical Reagent Company (Shanghai, China). Water was obtained
from an Aquapro purification system (Chongqing, China). All other reagents were
of analytical grade.
Cell culture
The HNE1 cell line was provided by the
Cancer Research Institute of Xiangya Medicine College, Central South
University, Changsha, China. The cells were cultured in RPMI 1640 medium
supplemented with 10% fetal bovine serum, 200 U/ml penicillin, and 100 U/ml
streptomycin in a water-saturated, 5% CO2 atmosphere at 37 ?C in 75 cm2 flasks.
Preparation of peripheral
proteins
Preparation of peripheral
proteins
Immunoaffinity enrichment and direct
extraction were used to extract cell peripheral proteins. The immunoaffinity
enrichment method presented here consisted of: (1) in situ biotinylation
of surface proteins on intact cells using the membrane-impermeable reagent
sulfo-NHS-LC-biotin, (2) extraction of peripheral proteins, and (3) affinity
capture of the biotinylated proteins with avidin [1113]. Direct extraction was
carried out as follows. Cells in 75 cm2
flasks were washed twice with RPMI 1640, and twice with phosphate-buffered
saline (PBS), and then the cell surface proteins were extracted directly with
0.1 M Na2CO3
or H2O. The change in cell morphology was
observed microscopically. The extraction was stopped when the cell membranes
were about to break.
Preparation of PM and integral
PM
The PM was isolated as previously described
[14,15]. All steps were carried out at 4 ?C. Briefly, adherent cells (2?108)
were washed three times with PBS, scraped using a plastic cell lifter, and
broken using a glass homogenizer. For integral PM proteins, the post-nuclear
supernatant was treated with 0.1 M Na2CO3 for 30 min, then the post-nuclear
supernatant with or without 0.1 M Na2CO3 was diluted with 100% Percoll to make a
30% Percoll solution, then centrifuged at 84,000 g for 30 min using an
SW-41 rotor (Beckman, USA). Fractions of 1.0 ml (typically 13 fractions in
total) were collected from the top of the gradient. To characterize the
contents of these subcellular fractions, all fractions (1 ml sample) were analyzed by western blot with antibodies against
known molecular markers for several organelles, flotillin for the PM, NADH
ubiquinol oxydoreductase 39 for the mitochondrial apparatus, and fibrillarin
for the nucleus. Fractions 2 and 3 (a visible band) containing flotillin but
less NADH ubiquinol oxydoreductase 39 or fibrillarin were precipitated with
acetone and represented the PM. The 0.1 M Na2CO3
treated sample was named the integral PM protein.
Gel separation, in-gel
digestion, and mass spectrometry analysis
These experiments were carried out
according to our previous reports [1621].
Data analysis and
bioinformatics
Perl software (http://www.perl.com/download.csp)
was written to pick up significant hits from Mascot output files (html files)
into tab-delimited data files suitable for subsequent data analysis as
described in a previous report [21]. The molecular weight, score, and peptides
matched were included. The protein location and function, such as the pathway,
were obtained through a Gene Ontology (GO) database search (http://www.geneontology.org/).
Furthermore, the protein accession numbers from Swiss-Prot (http://expasy.org/sprot/)
were searched against the Differentially Expressed Protein Database (DEPD; http://protchem.hunnu.edu.cn/depd/index.jsp)
(data not shown).
Protein-protein interaction
(PPI) network
To create the PPI network, we searched
against the local BIND database
(version 2.0) [22] using proteins we identified as input. The network file was
created by our Proteomics Profile Analyzer and drawn by Pajek software (version
1.21) (http://vlado.fmf.uni-lj.si/pub/networks/pajek).
Immunocytochemistry and
fluorescence microscopy
Cells were washed with PBS, fixed with 4%
paraformaldehyde/PBS for 15 min, and permeabilized with acetone/methanol (1:1)
at room temperature for 30 s. Cells were washed with PBS four times and blocked
by exposure to PBS buffer containing 5% normal goat serum for 30 min. The
cells were incubated with primary antibodies for 1.5 h at room temperature.
Dilutions of primary rabbit or mouse antibodies were: Na+/K+-ATPase,
nm23, human galectin-1 and VCP, 1:250; keratin 8, 1:150; and heat shock protein
70, 1:70. After rinsing with PBS, cells were incubated with fluorescein-isothiocyanate-conjugated
or tetramethylrhodamine isothiocyanate-conjugated goat anti-rabbit or mouse
IgG (KPL) diluted 1:1000 in 1% normal
goat serum in PBS. Some slides were stained by 1-(4-trimethylammoniumphenyl)-6-phenyl-1,3,5-hexatriene
p-toluenesulfonate. All slides were viewed with a Zeiss Axioskop2 Plus
fluorescence microscope (Carl Zeiss, Jena, Germany) and digitized images were
processed using AxioVision 3.1 software (Carl Zeiss).
Results
Profiling the peripheral PM
proteome enriched by biotinylation
According to the study by Peirce et al
[13], biotin could penetrate to the inner membrane and result in the
contamination of cytoplast proteins. So in this research, we collected cell
peripheral protein from intact cells after the cells were biotinylated [Fig.
1(A,B)].
Profiling the peripheral PM
proteome enriched by Na2CO3 or H2O treatment
Although the biotinylation-affinity method
is a classical method for extraction of cell surface proteins, it has some
limitations. Only those proteins with lysine residues exposed on the cell
surface can be labeled by the biotinylation reagent [23]. So, in our study, we
also explored two convenient ways to obtain more peripheral PM proteins
on the basis that these proteins dissolve easily in water and high salt
solution. We used 0.1 M Na2CO3 or H2O for extraction. It was found that they can extract
peripheral PM proteins with intact cells [Fig. 1(C–E)]; these proteins were used for proteomics analysis.
Preparation of PM
Thirty percent Percoll was used as the
medium for PM separation in our research. After ultracentrifugation, the
fractions were dotted to polyvinylidene difluoride membranes, and analyzed
using organelle-specific antibodies. Fractions 2 and 3 containing flotillin (a
PM-specific marker) but less NADH ubiquinol oxydoreductase 39
(mitochondrial-specific marker) and no fibrillarin (nucleus-specific protein),
were pooled for proteome analysis (Fig. 2). A relatively pure PM was
obtained and used for all subsequent proteomics investigations.
Preparation of integral PM
proteins
In order to enrich integral PM proteins, we
treated the post-nuclear supernatant with Na2CO3 at a
final concentration of 0.1 M. Similarly, PM-enriched fractions 2 and 3 were
pooled and used for integral PM protein analysis (data not shown).
Overview of peripheral,
integral, and total PM protein separation and identification
The peripheral, integral, and total PM
proteins of NPC cells were extracted into complementary protein populations,
separated and identified by SDS-polyacrylamide gel electrophoresis
(PAGE)-MS/MS, 2DE-MS/MS or shotgun, as depicted in Fig. 3.
Representative 1DE and 2DE are shown in Fig. 4. Such extensive
fractionation will aid the identification of proteins with low expression
levels and provide information relevant to function. In addition, we used different
strategies to improve the success rate of protein identification by MS.
Three complementary subproteomes were
generated, including: (1) a cell peripheral PM subproteome [Fig. 3(A)
and Fig. 4(A)], (2) the total PM proteome [Fig. 3(B) and Fig.
4], and (3) the integral PM proteome [Fig. 3(C) and Fig. 4].
A total of 848 non-redundant proteins were identified (data not shown), of
which 371 proteins were from peripheral PM fractions, 702 from total PM
fractions, and 180 from integral PM fractions. Only 76 proteins were identified
as present in all three kinds of fractions (Fig. 5). For the 371
peripheral PM proteins, 277 were from Na2CO3
extraction, 103 from H2O
extraction, and 83 from the fraction enriched by biotinylation. Among the 702
proteins identified from the PM fraction, 54 were identified by 2DE-MS (some
protein spots are marked in fig.
4), 354 by high capacity trap, and 518 by SDS-PAGE-MS (Table 1).
When a protein was identified by several methods, only the protein
identification information with the highest score was picked. This includes the
protein names, molecular weight, function, and location of the protein
according to GO annotation.
Although some proteins were identified by
only a single peptide, 72% of the proteins contributed by electronic spray
ionization quadrupole-time of flight mass spectrometer were identified by two
or more peptides at the 95% confidence level (p<0.05); for proteins identified by shotgun, 96% of proteins were identified by two or more peptides at the 95% confidence level (p<0.05). For those proteins identified by only one peptide, the MS/MS profile was checked manually; only proteins identified by peptides with continuous three Y or B ions were selected. For 54 proteins identified by matrix-assisted laser desorption ionization-time of flight-time of flight (MALDI-TOF-TOF), all were analyzed by PSD with at least one peptide. Most were overlapped with those from other methods, except that two proteins for only protein spots with high abundance were cut and identified in this work (data not shown).
Physicochemical
characteristics of the identified proteins
The analysis of identified proteins by the
TMHMM program (http://www.cbs.dtu.dk/services/TMHMM/TMHMM2.0b.guide.php)
predicted transmembrane (TM) segments for four proteins (1.1%) from the
peripheral fractions, 45 (6.3%) from the PM fractions, and 18 (10.0%) from the
integral fractions that had molecular characteristics typical of integral
membrane proteins. The number of TM domains in the molecules ranged from one
(41 proteins) to 10 (one protein) (data not shown). Because peripheral proteins
are soluble and loosely combined with integral membrane protein or
phospholipid, the methods for peripheral protein extraction presented here
yielded only 1.0% proteins with TM regions, indicating that the method was
efficient. Furthermore, because 20%–30%
of all open reading frames encoded by the genome have been predicted to be
integral membrane proteins [24,25], the method for integral membrane
extraction presented here concentrated potential TM proteins not very
efficiently.
To validate the present procedure, the
subcellular locations of the identified proteins were categorized according
to the universal GO cellular component annotation. Four hundred and eighty
(56.4%) proteins had a GO annotation for cellular component or cellular
locations, of which 168 (35%) were PM or PM-related proteins. The proteins
annotated as intermediate filament, cytoskeleton, and membrane were also
considered as PM or PM-related proteins. The remaining proteins were localized
primarily to intracellular region (20%) or cytoplasm (23%), as shown in Fig.
6. mitochondrial, and nuclear
proteins were also identified, perhaps due to other organelles in close contact
with the PM, or the existence of proteins at more than one site in the cell.
Of course, during sample preparation, some intracellular or cytoplasmic
proteins were extracted, with peripheral, PM, and integral PM proteins collected
by acetone precipitation.
Characterization of cell surface proteins Functional classification
The molecular functions of the proteins identified in this study were
classified according to the GO database (Fig. 7). Proteins with binding
activity were the largest subgroup in all fractions, consisting of 45%, 41%,
and 47% of total identified PM, peripheral, and integral proteins,
respectively. Compared with this ratio in all identified proteins (44%), the
ratio in integral PM fractions was 3% higher, and that in peripheral PM
fractions was 3% lower. This group included many important proteins such as
CD68 antigen variant, a widely used marker for cancer detection [26,27],
alkaline phosphatase, T-complex protein 1 (gamma subunit), putative S100
calcium-binding protein, and calmodulin. Proteins with catalytic activity were
the second-largest subgroup. The ratio of catalytic activity protein to total
protein was highest in the peripheral PM fraction of the three fractions (total
PM, peripheral, and integral PM). In all, 220 such proteins were identified,
including 4F2 cell surface antigen heavy chain (also named CD98 antigen),
splice isoform 1 of 3-hydroxyacyl-CoA dehydrogenase type II, and migration
inhibitory factor protein. The CD98 antigen involved in cell growth is
up-regulated in oral squamous cell carcinoma and might have an important role
in the early stages of multistep oral carcinogenesis [28]. Another major
category was structural proteins (24% of the 848 identified proteins). This study
also identified five kinds of other functional proteins, including enzyme
regulator activity proteins, transporters, and antioxidant molecules.
Approximately 10% (85 proteins) had no annotated functions and, therefore,
were classified as unknown. Through prediction, we found that many new
proteins might play important roles in cell function. For example, hypothetical
protein FLJ16459, International Protein Index (IPI) accession No. IPI00442122 (http://apr2005.archive.ensembl.org/IPI/),
is an unknown protein with 71% similarity to tropomyosin 2 beta that plays a
central role in the calcium-dependent regulation of vertebrate striated muscle
contraction. Hypothetical protein FLJ46846 (IPI00418700) is a protein with 78%
similarity to neuroblast differentiation-associated protein AHNAK, required for
neuronal cell differentiation. Hypothetical protein FLJ32377 (IPI00387164) is
84% similar to ubiquitin C that acts as an E3 ubiquitin-protein ligase and
accepts ubiquitin from specific E2 ubiquitin-conjugating enzymes. Hypothetical
protein (IPI00604713) is 89% similar to CKAP4 protein, a cell surface protein
with binding activity. Hypothetical protein FLJ27077 (IPI00442522) is 94%
similar to L-lactate dehydrogenase A chain, whose defects cause exertional myoglobinuria.
Cell-signaling molecules The 311 proteins identified are involved in
91 kinds of pathways. Of these, 126 non-redundant proteins have potential roles
in many cell-signaling pathways, including 37 cell communication molecules and
22 calcium-signaling pathway proteins (data not shown). Among these signaling
molecules, many proteins with roles in multiple pathways were found, such as
transforming protein Ras homolog gene family, member A variant, and titin. We
found four proteins of the CD44 gene (CD44 antigen, CD44 antigen precursor,
hypothetical protein DKFZp451K1918, and splice isoform CD44 of the CD44
antigen precursor). CD44 antigen is involved in the extracellular matrix
receptor interaction pathway. It is a protein up-regulated in many tumor cells
[29,30] and serves as a marker for many cancers.
Proteins involved in disease pathways The cell surface membrane is a subcellular
component of substantial interest in regard to various aspects of disease,
from molecular diagnosis to therapeutics. Through the KEGG search, we found
nine disease pathways (25 non-redundant proteins) including Huntington’s
disease (eight proteins), cholera infection (eight proteins), Parkinson’s
disease (four proteins), prion disease (three proteins), and neurodegenerative
disorders (three proteins) (Table 2). For example, clathrin heavy chain
1 (IPI00024067) is a very important PM protein involved in the Huntington’s
disease pathway. It is the major protein of the polyhedral coat of coated pits
and vesicles, and was found to have increased expression in a neck and head
cancer cell line [31]. Calmodulin (IPI00075248), another very important PM
protein, was found to be involved in Huntington’s disease. Calmodulin plays a
role in the calcium signaling pathway, the phosphatidylinositol signaling
system, and insulin signaling pathway, and was found to be expressed
differentially in many cancer cells [32,33].
Proteins found in DEPD
Apart from this general function research, we analyzed protein function
through DEPD, a database including 3000 differentially expressed proteins
manually extracted from published reports, largely from studies of serious
human diseases including lung cancer, breast cancer, and liver cancer [34]. A
total of 199 proteins differentially expressed in cancer were found (data not
shown), including proteins involved in breast, prostate, brain, liver, stomach,
colon, blood, kidney, and vein cancer. Many proteins were found to be
up-regulated in different cancers, for example, T-complex protein 1, delta
subunit (up-regulated in breast cancer), prohibitin (up-regulated in gliomas),
annexin V (up-regulated in colorectal, pituitary, colon, vein, stomach, and
skin cancer), and pyruvate kinase 3 isoform 1 variant (up-regulated in breast,
kidney, pancreas, colon, blood, and breast cancer). Splice isoform 1 of heat
shock cognate 71 kDa protein, found in total PM, peripheral PM, and integral PM
protein fractions, is a very important cell surface protein involved in many
cancers such as colon, breast, liver, thymus, blood, lung, prostate, and womb
cancer. Furthermore, other proteins were also found that were down-regulated
after treatment with anti-cancer drugs, such as tubulin-specific chaperone A
and phosphatidylethanolamine-binding protein. Some of these proteins identified
by 2DE-MS/MS are shown in Fig. 4.
PPI network Networks of
protein interactions mediate many cellular responses to environmental stimuli
and direct the execution of developmental programmers. Each protein typically
interacts and reacts with interacting partners to execute its functions. Two
essential questions that concern us are how proteins coming from other
subcellular organelles interact with PM proteins, and how PM proteins interact
with each other (such as integral membrane proteins with peripheral membrane
proteins). To investigate these issues, we searched the BIND database, the
largest database for protein interaction, against proteins we identified as
seeds. As shown in Fig. 8 and table
3, 178 seed proteins were involved in 671 interactions with 345 other
proteins or small molecules. In the network, all seed proteins are grouped into
seven categories based on the fractions of sample (peripheral and/or integral
and/or total PM), and are shown in different colors. The topology of the
network suggests that some proteins interact with many partners, such as
calmodulin (spot1, IPI00075248), MIF protein (spot39, IPI00293276), and
voltage-dependent anion-selective channel protein 1 (spot96, IPI00216308)
interacting with 28, 4, and 2 proteins, respectively.
As shown in Fig. 8, several proteins
identified from peripheral, integral, and PM fractions interact with paxillin
(spot51, gi|27735219), a cytoskeletal protein (http://www.expasy.org/uniprot/P49023),
and calnexin precursor (spot98, IPI00020984) interacts with calreticulin
precursor (spot191, IPI00020599) through another protein. These interacting
proteins usually have very important functions in the PM. For example,
voltage-dependent anion-selective channel protein 1 (spot96, IPI00216308)
interacts with two partners and forms a channel through the PM, which
participates in the formation of the permeability transition pore complex
responsible for the release of mitochondrial products that trigger apoptosis
[35]. Caveolin-1 (spot134, IPI00009236) interacts with nine partners, including
G-protein alpha subunits, and acts as a scaffolding protein within caveolar
membranes (www.expasy.org, Q03135).
In summary, this PPI analysis highlighted
many important membrane proteins as well as interactions between them.
Immunocytochemistry analysis
To further validate the present procedure,
the subcellular localizations of five proteins with annotated localization on
cytoplasm or nucleus were studied by immunocytochemistry in HNE1 cells. These
were galectin-1, nm23, and keratin 8, annotated to be cytoplasmic proteins, and
heat shock protein 70 and VCP, annotated to be nuclear and cytoplasmic
proteins. Immunofluorescence experiments using mouse anti-galectin-1,
anti-nm23, anti-keratin 8, anti-VCP, and anti-heat shock protein 70 showed that
they were widely distributed in PM and confirmed their identification (Fig.
9).
Discussion
Organelle proteomics, which is being used
increasingly to analyze cellular fractions, is advantageous in two respects. One
is the decreased complexity of the samples to be analyzed and the other is the
information provided on the subcellular localization of protein components.
Unfortunately, such analysis has never been reported in NPC research. In this
paper, we have attempted to characterize PM proteins based on the construction
of the PM and their extractability under three different treatment conditions.
One of our goals was to develop a methodology to categorize peripheral,
integral, and total PM proteome. In this work, peripheral proteins were
enriched by Na2CO3, H2O, and biotinylation. During their extraction, the
cells were intact by microscopic observation, so as to decrease contamination
by intracellular proteins. Highly purified PM and integral PM were obtained by
Percoll density gradient centrifugation and verified by Western blot analysis.
The proteins were separated and identified by different methods, including
2DE-MS/MS, 1DE-MS/MS, and shotgun. From our analysis, using a set of stringent
identification criteria (e.g., for MALDI, peptide mass fingerprinting, 50 ppm,
lift, 0.5 Da; and for ESI, MS, 0.6 Da; and for MS/MS, 0.3 Da) according to
previous reports [16–21], we identified
848 non-redundant proteins. This is the first comprehensive proteomic study of
NPC cell surface proteins.
In this study, approximately 36% of
proteins were annotated to be proteins from other organelles, in spite of the
stringent experiment control of PM preparations. To further validate our
results, five proteins that were annotated to locate on cytoplasm or nuclear
membrane, were confirmed to be at multiple locations, including PM, by
immunocytochemistry. So our results indicated that some of these annotated
contaminating proteins might potentially be cell surface components. However, it
was still difficult to distinguish them from contamination, because
intracellular proteins might penetrate to extracellular sites and be extracted
as peripheral proteins. So it was necessary to develop new methods to separate
and verify PM proteome.
This research provides new directions for
studying the molecular mechanisms of NPC, including not only new experimental
methods, but also new methods for the discovery of protein function. The
protein functions were studied in five ways: (1) functional classification; (2)
cell-signaling molecules; (3) disease pathways; (4) DEPD searching; and (5) PPI
analysis. We profiled not only the functional distribution of cell surface
proteins, but also a complex PPI network. We found many important proteins,
including CD68 antigen variant and CD44 antigen. Through searching DEPD, we
found 199 proteins differentially expressed in cancer. Although DEPD is not a
comprehensive database, it is the only database available for differential
proteome research. In any case, many cancer-related PM proteins were found in
this research. To our knowledge, this study has constructed the first PPI
network of NPC cell surface proteins. These cell surface proteins are connected
to each other by shared components. The network that resulted is a topological
description of the cell surface proteome. It is expected that this might
provide drug discovery programs with a molecular context for the choice and
evaluation of drug targets.
In summary, we isolated three cell surface
fractions, peripheral, integral, and total PM proteins. For the first time, we
have provided a proteome-wide analysis of the PM in NPC by using a combination
of subcellular separation and proteome identification, and constructed the
first PPI network in NPC PM. A comprehensive understanding of the proteome of
NPC could help elucidate how cells fulfill their biological function in a
context of a biological system. Cell surface interaction mapping of NPC can
contribute to an understanding of how proteins carry out their functions at the
cell surface. Differences in protein expression patterns between different
cells as well as changes occurring during disease are of great interest in
experimental medicine for diagnosis and treatment. The technology described
here could help exploit such possibilities in the future.
References
1 Chan AT, Teo PM, Johnson PJ. Controversies in
the management of locoregionally advanced nasopharyngeal carcinoma. Curr Opin
Oncol 1998, 10: 219–225
2 Chan AS, To KF, Lo KW, Mak KF, Pak W, Chiu B,
Tse GM et al. High frequency of chromosome 3p deletion in histologically
normal nasopharyngeal epithelia from southern Chinese. Cancer Res 2000,
60: 5365–5370
3 Raab-Traub N. Epstein-Barr virus in the
pathogenesis of NPC. Semin Cancer Biol 2002, 12: 431–441
4 Ward MH, Pan WH, Cheng YJ, Li FH, Brinton LA,
Chen CJ, Hsu MM et al. Dietary exposure to nitrite and nitrosamines and
risk of nasopharyngeal carcinoma in Taiwan. Int J Cancer 2000, 86: 603–609
5 Jia WH, Feng BJ, Xu ZL, Zhang XS, Huang P,
Huang LX, Yu XJ et al. Familial risk and clustering of nasopharyngeal
carcinoma in Guangdong, China. Cancer 2004, 101: 363–369
6 Zhang B, Nie X, Xiao B, Xiang J, Shen S, Gong
J, Zhou M et al. Identification of tissue-specific genes in
nasopharyngeal epithelial tissue and differentially expressed genes in
nasopharyngeal carcinoma by suppression subtractive hybridization and cDNA
microarray. Genes Chromosomes Cancer 2003, 38: 80–90
7 Tan C, Li J, Wang J, Xiang Q, Zhang X, Dong
L, Shen S et al. Proteomic analysis of differential protein expression
in human nasopharyngeal carcinoma cells induced by NAG7 transfection.
Proteomics 2002, 2: 306-312
8 Yao KT, Zhang HY, Zhu HC, Wang FX, Li GY, Wen
DS, Li YP et al. Establishment and characterization of two epithelial
tumor cell lines (HNE-1 and HONE-1) latently infected with Epstein-Barr virus
and derived from nasopharyngeal carcinomas. Int J Cancer 1990, 45: 83–89
9 Slamon DJ, Leyland-Jones B, Shak S, Fuchs H,
Paton V, Bajamonde A, Fleming T et al. Use of chemotherapy plus a monoclonal
antibody against HER2 for metastatic breast cancer that overexpresses HER2. N
Engl J Med 2001, 344: 783–792
10 Oh P, Li Y, Yu J, Durr E, Krasinska KM, Carver
LA, Testa JE et al. Subtractive
proteomic mapping of the endothelial surface in lung and solid tumours for
tissue-specific therapy. Nature 2004, 429: 629–635
11 Jang JH, Hanash S. Profiling of the cell
surface proteome. Proteomics 2003, 3: 1947–1954
12 Shin BK, Wang H, Yim AM, Le Naour F, Brichory
F, Jang JH, Zhao R et al. Global profiling of the cell surface proteome
of cancer cells uncovers an abundance of proteins with chaperone function. J
Biol Chem 2003, 278: 7607–7616
13 Peirce MJ, Wait R, Begum S, Saklatvala J, Cope
AP. Expression profiling of lymphocyte plasma membrane proteins. Mol Cell
Proteomics 2004, 3: 56–65
14 Smart EJ, Ying YS, Mineo C, Anderson RG. A
detergent-free method for purifying caveolae membrane from tissue culture
cells. Proc Natl Acad Sci USA 1995, 92: 10104–10108
15 Lehner I, Niehof M, Borlak J. An optimized
method for the isolation and identification of membrane proteins.
Electrophoresis 2003, 24: 1795–1808
16 Washburn MP, Wolters D, Yates JR 3rd.
Large-scale analysis of the yeast proteome by multidimensional protein
identification technology. Nat Biotechnol 2001, 19: 242–247
17 Delahunty C, Yates JR 3rd. Protein
identification using 2D-LC-MS/MS. Methods 2005, 35: 248–255
18 Durr E, Yu J, Krasinska KM, Carver LA, Yates
JR, Testa JE, Oh P et al. Direct proteomic mapping of the lung microvascular
endothelial cell surface in vivo and in cell culture. Nat Biotechnol 2004,
22: 985–992
19 Zhang L, Liu X, Zhang J, Cao R, Lin Y, Xie J,
Chen P et al. Proteome analysis of combined effects of androgen and
estrogen on the mouse mammary gland. Proteomics 2006,6: 487–497
20 Zhang L, Xie J, Wang X, Liu X, Tang X, Cao R,
Hu W et al. Proteomic analysis of mouse liver plasma membrane: use of differential extraction to
enrich hydrophobic membrane proteins. Proteomics 2005, 5: 4510–4524
21 Cao R, Li X, Liu Z, Peng X, Hu W, Wang X, Chen
P et al. Integration of a two-phase partition method into proteomics
research on rat liver plasma membrane proteins. J Proteome Res 2006, 5:
634–642
22 Bader GD, Donaldson I, Wolting C, Ouellette
BF, Pawson T, Hogue CW. BIND – The
Biomolecular Interaction Network Database. Nucleic Acids Res 2001, 29:
242-245
23 Hurley WL, Finkelstein E, Holst BD.
Identification of surface proteins on bovine leukocytes by a biotinavidin
protein blotting technique. J Immunol Methods 1985,85: 195-202
24 Mouse Genome Sequencing Consortium, Waterston
RH, Lindblad-Toh K, Birney E, Rogers J, Abril JF, Agarwal P et al.
Initial sequencing and comparative analysis of the mouse genome. Nature 2002,
420: 520–562
25 Okazaki Y, Furuno M, Kasukawa T, Adachi J,
Bono H, Kondo S, Nikaido I et al. Analysis of the mouse transcriptome
based on functional annotation of 60,770 full-length cDNAs. Nature 2002,
420: 563–573
26 Vicandi B, Jim?nez-Heffernan JA, L?pez-Ferrer
P, Hardisson D, P?rez-Campos A, Gonz?lez-Peramato P, Viguer JM. Fine needle
aspiration cytology of mammary carcinoma with osteoclast-like giant cells.
Cytopathology 2004, 15: 321–325
27 Svec A, Velenska Z. Renal epithelioid
angiomyolipoma – a close
mimic of renal cell carcinoma. Report of a case and review of the literature.
Pathol Res Pract 2005, 200: 851–856
28 Kim DK, Ahn SG, Park JC, Kanai Y, Endou H,
Yoon JH. Expression of L-type amino acid transporter 1 (LAT1) and 4F2 heavy
chain (4F2hc) in oral squamous cell carcinoma and its precusor lesions.
Anticancer Res 2004, 24: 1671–1675
29 Oertl A, Castein J, Engl T, Beecken WD, Jonas
D, Melamed R, Blaheta RA. Endothelial adhesion of synchronized gastric tumor
cells changes during cell cycle transit and correlates with the expression
level of CD44 splice variants. World J Gastroenterol 2005, 11: 6243–6248
30 Karadag A, Fedarko NS, Fisher LW. Dentin
matrix protein 1 enhances invasion potential of colon cancer cells by bridging
matrix metalloproteinase-9 to integrins and CD44. Cancer Res 2005, 65:
11545–11552
31 Schmidt M, Schler G, Gruensfelder P, Hoppe F.
Differential gene expression in a paclitaxel-resistant clone of a head and neck
cancer cell line. Eur Arch Otorhinolaryngol 2006, 263: 127–134
32 Castagna A, Antonioli P, Astner H, Hamdan M, Righetti
SC, Perego P, Zunino F et al. A proteomic approach to cisplatin
resistance in the cervix squamous cell carcinoma cell line A431. Proteomics 2004,
4: 3246–3267
33 Zheng JH, Shi D, Zhao Y, Chen ZL. Role of
calcium signal in apoptosis and protective mechanism of colon cancer cell line
SW480 in response to 5-aminolevulinic acid-photodynamic therapy. Ai Zheng 2006,
25: 683–688
34 He QY, Cao J, Liu XH, Li MX, Liu YS, Xie JY,
Liang SP. DEPD: a novel database
for differentially expressed proteins. Bioinformatics 2005, 21: 3694–3696
35 Verrier F, Mignotte B, Jan G, Brenner C. Study
of PTPC composition during apoptosis for identification of viral protein
target. Ann NY Acad Sci 2003, 1010: 126–142