Physiol Genomics
5: 129–136, 2001.
Insights into a plasma membrane signature
SHERYL HARVEY, YAN ZHANG, FRANCE LANDRY,
COLEEN MILLER, AND JEFFREY W. SMITH
Program on Cell Adhesion, Cancer Research Center,
The Burnham Institute, La Jolla, California 92037
Received 27 October 2000; accepted in final form 8 February 2001
proteome; expression profiling; breast cancer; metastasis
therapeutic intervention, making the PM a rich source
of drug targets. Many of the molecules at the PM are
also exposed to the extracellular milieu, providing an
additional therapeutic advantage. As recognized more
than 25 years ago by Garth Nicolson, there is a great
need to define the PM as a system: “It would be convenient if these (biological) properties were definable by
unique cell surface characteristics which were readily
identifiable so that highly specific molecular approaches to fighting cancer could be developed” (18).
Presented with this long-recognized but unmet need,
the present study was aimed at gaining some basic
information on the PM proteome of mammary carcinoma. We compared the PM proteomes of mammary
carcinoma to that of normal human fibroblasts. Indeed,
the study reveals large groups of proteins that can be
considered signatures of cell lineage. There are obvious
and striking differences between the PM proteomes of
fibroblasts and mammary carcinoma. In addition, we
are able to identify a smaller set of proteins that is
unique to the PM proteome of MDA-MB-435 tumor
cells, a highly metastatic mammary carcinoma line.
These proteins may represent part of the metastatic
signature.
MATERIALS AND METHODS
as 1953, Nordling (19) suggested that the
clinical manifestations of cancer result from the cumulative effect of multiple genetic mutations. The multifactorial nature of tumor progression precludes its explanation by any single factor or even a small group of
factors. It has yet to be determined how many alterations must occur to convert a normal mammary epithelial cell to carcinoma or how many additional
changes are necessary for the carcinoma to progress to
invasive disease. Answers to these questions require
global examinations of biological systems.
This study focuses on the plasma membrane (PM) as
a system. The PM regulates many aspects of tumor
progression including cellular differentiation, cell proliferation, adhesion and migration, pericellular proteolysis, and even the escape from immune surveillance.
Each of these events represents a point for potential
AS EARLY
Article published online before print. See web site for date of
publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: J. W.
Smith, The Burnham Institute, 10901 North Torrey Pines Road, La
Jolla, CA 92037 (E-mail: [email protected]).
Cells
All cell lines were obtained from American Type Culture
Collection (ATCC), except the MDA-MB-435 cells, which
were obtained from Janet Price (University of Texas). MDAMB-435 cells were cultured in MEM Earle’s (Irvine Scientific) with 10% fetal bovine serum (Irvine Scientific), 2 mM
L-glutamine (GIBCO-BRL), 1 mM sodium pyruvate (Irvine
Scientific), MEM vitamins (GIBCO-BRL), and nonessential
amino acids (Irvine Scientific). All other cells were cultured
according to specifications set forth by ATCC. Cells were
grown in 15-cm dishes at 37°C with 5% CO2. Cells were
harvested at 80–90% confluence.
PM Isolation
PMs were prepared using a slightly modified version of a
procedure described by Chaney and Jacobson (4). Briefly,
cells were harvested from the culture plates with phosphatebuffered saline containing 200 ␮M EDTA. The cells were
suspended in coating buffer (20 mM MES, 150 mM NaCl, 400
mM sorbitol, pH 5.3) and were added dropwise through a
21-gauge needle onto a gently swirling solution of 10% cationic colloidal silica (Aldrich). After removal of excess silica
by washing, the cells were diluted in coating buffer and were
1094-8341/01 $5.00 Copyright © 2001 the American Physiological Society
129
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 16, 2017
Harvey, Sheryl, Yan Zhang, France Landry, Coleen
Miller, and Jeffrey W. Smith. Insights into a plasma
membrane signature. Physiol Genomics 5: 129–136,
2001.—The plasma membrane (PM) is an organized biological system that serves as a structural barrier and communication interface with the extracellular environment. Many
basic questions regarding the PM as a system remain unanswered. In particular, we do not understand the scope of
similarity and differences in protein expression at the PM.
This study takes an initial step toward addressing these
questions by comparing the PM proteomes of fibroblasts and
mammary carcinoma cells. Three sets of proteins were revealed by the study. The first set comprises between 9 and
23% of all proteins at the PM and appears to be common to
both fibroblasts and mammary carcinoma. A second group of
proteins, comprising ⬃40% of the proteins at the PM, is
tightly linked to cell lineage. The third set of proteins is
unique to each cell line and is independent of cell lineage. It
is reasonable to hypothesize then, that this third group of
proteins is responsible for unique aspects of cell behavior. In
an effort to find proteins linked to the metastatic phenotype,
we identified several proteins that are uniquely expressed at
the PM of the metastatic MDA-MB-435 cells. These proteins
have functions ranging from cell adhesion to the regulation of
translation and the control of oxidant stress.
130
PLASMA MEMBRANE SIGNATURE
Two-Dimensional Gel Electrophoresis
Proteins associated with PMs were analyzed by two-dimensional (2D) electrophoresis. Solubilized PM protein, 100
␮g in 200 ␮l of 8 M urea, 2 M thiourea, 2% CHAPS, and 13
mM DTT was used to rehydrate an 11-cm immobilized pH
gradient (IPG) strip (Immobiline DryStrip, 4-7L; Amersham
Pharmacia Biotech). The strip was rehydrated for 24 h at
ambient temperature under oil. The proteins in the strip
were focused using a Multiphor II apparatus (Amersham
Pharmacia Biotech) for a total of 19,250 V-h. IPG strips were
subsequently equilibrated in 6 M urea, 30% (wt/vol) glycerol,
2% (wt/vol) SDS, 1% DTT, and 0.05 M Tris 䡠 HCl, pH 8.8, for
15 min before the proteins were separated in the second
dimension. Second dimension electrophoresis was performed
for 15 min at 190 V and then for an additional 22 h at 65 V
using a 12% acrylamide gel. Gels were silver stained under
uniform and carefully controlled conditions according to the
procedure of Shevchenko et al. (28). Silver-stained gels were
scanned at 200 dpi using a Sharp JX-330 desktop scanner
(Amersham Pharmacia Biotech).
Image Analysis
Three to four 2D gels were prepared for each cell line. A
representative gel was chosen for image analysis, and the
remaining gels were stored in 1% acetic acid and used for
protein identification by mass spectrometry. Protein spots
were selected using the ImageMaster Analysis software (v.
2.01, Amersham Pharmacia Biotech), and each spot was
verified by the operator. The volume of each spot was quantified using the ImageMaster algorithm. These values were
summed for all spots within the gel. The percentage volume
that each spot contributed to the entire gel was determined
by dividing the individual spot volume by the total spot
volume on the gel. Based on visual inspection of the gels,
spots that represented less than 0.04% of the total protein on
the gel were not reproducibly detected by the software. Consequently, we constrained our analysis to spots with a volume above this value. When considering all eight test cell
lines, we observed an average of 360 ⫾ 57 (mean ⫾ SD) spots
with a volume above this cutoff. Protein spot patterns within
the gels were compared using the ImageMaster analysis
software. Gels were compared in a pairwise manner so that
every cell line was compared directly to every other cell line
in the study. The degree of similarity between gels was
determined by calculating the percentage of protein spots
with an identical migration pattern.
Cell Surface Iodination
To track the yield of cell-surface proteins through the
electrophoresis procedure, PM proteins were labeled by iodination, using procedures we have reported previously (32).
125
I-labeled cells were diluted 1:4 with unlabeled cells, and
PM protein was prepared as described above. A quantity of
100 ␮g of PM protein containing ⬃37,000 cpm was loaded
into each of several IPG strips, and the entire 2D PAGE
procedure was carried out. At various points in the procedure, samples were subjected to gamma counting.
Identification of Proteins by Mass Spectrometry
The identity of several polypeptide spots was determined
by peptide mass fingerprinting using methods that we have
previously described (13). Proteins were identified by comparing observed mass fingerprints to National Center for
Biotechnology Information database using a Bayesian algorithm (http://prowl.rockefeller.edu/cgi-bin/ProFound).
Five polypeptides unique to the highly metastatic cell line,
MDA-MB-435, were examined by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) and mass
spectrometry-mass spectrometry (MS-MS) sequencing (15,
16) by Protana (http://www.protana.com).
RESULTS
Cell Lines
Four breast carcinoma lines and four fibroblast lines
were used to assess similarities among PM proteomes
(Table 1). The analysis included the MCF7, MDA-MB231, and MDA-MB-435 mammary carcinoma lines, all
of which have been studied extensively. Each of these
lines is also part of the NCI60, a group of cancer lines
maintained for study by the National Cancer Institute’s Developmental Therapeutics Program (29). The
MDA-MB-435 cell line is especially important to the
present study, because this cell line is one of the only
breast carcinoma lines to exhibit extensive metastasis
from the mouse mammary fat pad (25).
Analysis of PM Proteomes
Although proteomic profiling has been applied to
identify proteins expressed by mammary carcinoma
cell lines (31) and sections of mammary tumors (2), no
study has focused on the PM. We reasoned that a
Table 1. Cell lines used for comparison of PM proteins
Cell Line
Description
MCF7
MDA-MB-231
SK-BR-3
Mammary carcinoma, tumorigenic, non-metastatic
Mammary carcinoma, tumorigenic, forms
microscopic metastases
Mammary carcinoma, tumorigenic, forms
macroscopic metastases
Mammary carcinoma, tumorigenic
Hs 58.Fs
HE-SK
CCD-1090Sk
Detroit 551
Normal;
Normal;
Normal;
Normal;
MDA-MB-435
skin;
skin;
skin;
skin;
newborn foreskin
fetus
adult
embryo
The plasma membrane (PM) proteomes of four mammary carcinoma cell lines (top) and four fibroblast cultures (bottom) were
compared in this study.
http://physiolgenomics.physiology.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 16, 2017
added dropwise through a 21-gauge needle onto a gently
swirling solution of 10 mg/ml polyacrylic acid (Fluka). After
washing to remove any excess of the anionic polymer, the
cells were resuspended in 2.5 mM imidazole, pH 7.0, containing a protease inhibitor cocktail (Sigma) and 1 mM vanadate
(Sigma). After a 30-min incubation on ice, the cells were lysed
by Dounce homogenization using multiple strokes with a size
B pestle. Sheets of PM were then separated from the cellular
contents by ultracentrifugation (60,000 g, 30 min) through a
gradient of Nycodenz (Sigma). The supernatant was aspirated, and the PM sheets, made dense by the silica and
polyacrylic acid, were collected from the bottom of the tubes.
After several washes in 2.5 mM imidazole, proteins were
solubilized from the membranes with 8 M urea, 2 M thiourea,
2% CHAPS, and 13 mM dithiothreitol (DTT). Protein concentration was determined using a micro-BCA protein assay
(Pierce) with bovine serum albumin as a standard.
131
PLASMA MEMBRANE SIGNATURE
Table 2. Tracking 125I-labeled cell surface proteins
through the 2D PAGE procedure
Procedural Step
Activity,
cpm
Overall Yield,
%
Input
Remaining in IPG after rehydration
Remaining in IPG after IEF
Entering 2D PAGE gel
37,268
23,044
13,854
6,962
61.8
37.2
18.7
comparison of the PM proteome between breast carcinoma and fibroblasts would provide the first information on the degree of similarity and distinction within
the PM proteomes of different cell types.
We isolated PMs from cells using a method established by Chaney and Jacobson (4, 10). This method
enriches for PMs by first tagging the cell surface with
cationic colloidal silica and then separating the PM
from the remainder of the cell by a density gradient.
We found that the 2D polypeptide spot patterns of PM
proteins isolated in this manner were highly reproducible within each cell line, with 2D gels from separate
membrane isolations yielding spot patterns that are
superimposable. In several comparisons of PM proteomes, we found the spot similarity within a cell line
to be greater than 90%. We also found that 2D gels of
Identification of PM Signatures
The PM proteomes of four fibroblast cell lines (Fig. 1,
A–D) and four breast cancer cell lines (Fig. 1, E–H)
Fig. 1. Comparison of plasma membrane (PM) proteins by two-dimensional
(2D) PAGE. PMs were isolated from
four fibroblast (A–D; Hs 58.Fs, HE-SK,
CCD-1090Sk, and Detroit 551) and
four mammary carcinoma cell lines (E–
H; MCF7, MDA-MB-231, MDA-MB-435,
and SK-BR-3). Following membrane
isolation, proteins were solubilized
with a mixture of chaotropic and reducing agents and detergent as described in MATERIALS AND METHODS. A
total of 100 ␮g of protein from this PM
preparation was analyzed by 2D
PAGE. Samples were focused in immobilized pH gradient (IPG) strips with a
linear pH range of 4–7 and then were
separated on a 12% acrylamide gel.
Gels were silver stained and scanned,
and the resulting image was analyzed
using ImageMaster Analysis software.
Only spots that constituted greater
than 0.04% of the total staining density within the gel were considered for
analysis. All of the protein spots used
for comparative analysis are shaded
dark. Each gel is representative of at
least three replicate gels.
http://physiolgenomics.physiology.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 16, 2017
MDA-MB-435 cells were subjected to cell surface iodination as
described in MATERIALS AND METHODS. Following extensive washing,
the PM of the cells was isolated by coating with cationic colloidal
silica and then isolating PM sheets on density gradients (4). PM
proteins were solubilized, and a portion of the protein (100 ␮g; 37,000
cpm) was used to track cell surface proteins through each step of the
two-dimensional (2D) PAGE procedure. The analysis was performed
twice, yielding virtually identical results. IPG, immobilized pH gradient; IEF, isoelectric focusing.
the PM proteome are substantially different from the
spot patterns of corresponding whole cell lysates. For
example, out of an average of 423 total spots, only 103
migrated to identical positions on the 2D gels when PM
proteins were compared with a whole cell lysate (data
not shown). On the basis of these findings, we elected to
enrich for PM proteins by using this purification strategy.
Another consideration in the analysis is the ability of
2D gels to resolve integral membrane proteins. We
followed methods recently described by Rabilloud et al.
(26) and Pasquali et al. (22), which show that the
inclusion of thiourea in the extraction buffer significantly expands the range of integral membrane proteins that can be analyzed by 2D PAGE. Our preliminary work confirmed that thiourea dramatically
increased the number of spots visible on 2D gels of PM
preparations (data not shown). However, initial mass
fingerprinting of a number of spots on the gels identified few integral membrane proteins. By quantitative
tracking of proteins tagged by cell surface iodination,
we determined that less than 20% of the proteins on
the external side of the membrane are actually resolved onto the second dimension gel (Table 2). Consequently, the first part of our analysis of the PM proteome was performed with the realization that the
proteomes being compared consist primarily of proteins associated with the PM and that integral membrane proteins are likely to be under-represented.
132
PLASMA MEMBRANE SIGNATURE
estimated by dividing its spot density by the combined
spot density of all proteins on the gel. Differences in
the relative abundance of individual proteins among
the cell lines were generally small, falling within a twoto fourfold range (data not shown). This suggests that
most of the proteins in the PM signature of mammary
carcinoma are expressed at similar levels.
Identification of Proteins in the Metastatic Signature
Because the MDA-MB-435 cells are far more invasive than the MDA-MB-231 cells and the MCF7 cells,
we sought to identify proteins unique to the PM proteome of the MDA-MB-435 cells. The mammary carcinoma cell line, SK-BR-3, was not included in this
analysis because its metastatic properties are not defined. We reasoned that this unique set of proteins
might be linked to the metastatic phenotype. We found
82 polypeptide spots in the PM proteome of the MDAMB-435 cells that were not detected in the other two
mammary carcinoma cells lines (Fig. 3). The fact that
these spots are only observed in the MDA-MB-435 cells
indicates that they are either uniquely expressed by
Fig. 2. Similarity in the PM proteomes of fibroblasts and mammary carcinoma cells. To assess the similarities
among the PM proteomes of fibroblasts (Fib) and mammary carcinoma cells (MCC), the polypeptide spot pattern
of each 2D gel was compared with that of every other gel in a pairwise fashion. Computer-assisted comparisons
were made with ImageMaster analysis software as described in MATERIALS AND METHODS. The degree of similarity
between two cell lines is reported as a percentage (% Identity), calculated by dividing the number of matching spots
by the average number of spots on each gel pair.
http://physiolgenomics.physiology.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 16, 2017
were analyzed by 2D gel electrophoresis. The polypeptide spot patterns of these gels, which contained an
average of 360 spots per gel, were compared using
computer-assisted spot matching. Similarity between
the PM proteomes was quantified by determining the
percentage of matching spots between each gel pair.
This pairwise comparison showed that cells of the same
lineage exhibited remarkable similarity (Fig. 2), with
similarity among the mammary carcinoma cell lines
ranging from 38% to 55%. The similarity among the
fibroblasts ranged from 18% to 40%, but most of the
lines were nearly 30% identical. Conversely, cells of
different lineage had far less similarity, with values
ranging from only 9% to 23%. Altogether, these findings strongly suggest the presence of a PM “signature”
that is tightly linked to cell lineage and that the signature of the mammary carcinoma cells constitutes a
little under one-half of the PM proteome.
As an additional assessment of the relatedness
among cell lines, the relative abundance of a set of 83
proteins common to the PM of the mammary carcinoma lines, but absent from the fibroblasts, was compared. The relative abundance of each protein was
PLASMA MEMBRANE SIGNATURE
133
these cells or that the MDA-MB-435 cells express a
uniquely processed variant of each protein. Only a few
of these unique polypeptides were present at sufficient
levels for identification by peptide mass fingerprinting
(Fig. 3, A–E). Nevertheless, the results from mass
fingerprinting of these proteins (Table 3) reveal some
interesting observations. The potential function of each
of the identified proteins in tumor progression is discussed below.
The p47 subunit of eukaryotic initiation factor 3. Spot
B in Fig. 3 was found to be the p47 subunit of eukaryotic initiation factor 3 (eIF3). This protein migrated
very near its predicted Mr and isoelectric point (pI),
indicating that its presence in the PM proteome is due
to upregulation in the MDA-MB-435 cells or to a
unique association with the PM in these cells. The eIFs
control protein synthesis by regulating the interaction
between mRNA, tRNA, and ribosomes (21). eIF3 is a
multiprotein complex that binds to the 40S ribosomal
subunit and is believed to regulate its association with
the 60S ribosome (1). Interestingly, the p47 subunit of
the eIF3 complex was only recently cloned and was
found to be homologous to a number of other proteins
including Mov-34, a component of the 26S proteosome
(1). The function of p47 is not known, but its relationship to Mov-34 suggests that it may be involved in
regulating subunit interactions within eIF3. Our identification of p47 in PM preparations also suggests that
it could have a role outside of the eIF3 complex. Such a
novel function could explain its prevalence in PM fractions.
Although the eIFs are not considered classic oncogenes, there is now an expanding body of literature
indicating that misregulation of individual components
of eIF2, eIF3, and eIF4 can lead to neoplastic transformation (6, 7, 17, 20, 27). Although p47 of eIF3 has not
been extensively studied, one might hypothesize that it
could play a similar role in tumor progression.
An uncharacterized protein, p25. We also observed a
25-kDa protein (spot C, Fig. 3) that is unique to the PM
proteome of the MDA-MB-435 cells. Peptide mass fingerprinting showed this protein to be identical to three
nucleotide sequences that were independently deposited in the databases (GenBank accession nos.
AAD27784, BAA76626, and AAD40193). The predicted
mass of the protein derived from each of these genes
http://physiolgenomics.physiology.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 16, 2017
Fig. 3. Toward the identification of a
metastatic proteome. Spot comparisons
were performed among the PM proteomes of the mammary carcinoma
cells to identify polypeptide spots
unique to the 2D gel of the highly metastatic MDA-MB-435 cells. For the
purpose of illustration, the position of
the unique spots is shown on the 2D gel
of MDA-MB-435 cells (in red) compared
with a gel of the MDA-MB-231 cells.
Further detail of the boxed areas is
shown in the insets, with arrows referencing the comparable regions of the
two gels. The identities of spots A–E
were determined by mass spectrometry
(Table 3).
134
PLASMA MEMBRANE SIGNATURE
Table 3. Identification of proteins from the MDA-MB-435 cells by mass spectrometry
Spot ID
(Fig. 3)
Protein
Identity
Residues Identified By:
Database
Accession No.
eIF3 p47
O00303
C
Not named
AAD27784
BAA76626
AAD40193
D
␣-Synuclein
P37840
E
Galectin-1
P09382
not identified
94–108
109–115
193–210
215–226
239–246
9–16
21–31
39–52
73–86
19–28
37–48
64–73
MS-MS
262–278
279–297
298–306
307–317
87–100
140–155
141–155
177–199
1–8
21–31
39–52
73–87
35–43
61–80
61–81
205–218
66–80
81–96
81–97
100–107
112–127
Five polypeptide spots unique to the MDA-MB-435 cells (Fig. 3, spots A–E) were analyzed by mass spectrometry. Proteins were identified
by peptide mass fingerprinting using MALDI-TOF or by MS-MS sequencing of tryptic peptide fragments. The position of each tryptic peptide
identified by mass spectrometry is shown. MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; MS-MS, mass spectrometry-mass spectrometry
matches the migration position of p25 on 2D gels.
Expressed sequence tags encoding the protein are
found in several tissues. However, to our knowledge
this is the first demonstration that the protein product
of the gene is expressed. This protein lacks homology to
any other known protein or gene in the databases and
has no known function.
␣-Synuclein. Spot D in Fig. 3 was found to be
␣-synuclein. This protein also migrated close to its
predicted Mr and pI in the 2D gels from MDA-MB-435
cells. Synucleins are small cytoplasmic proteins of
⬃130 amino acids. Although their precise biological
function is still poorly understood, the synucleins (␣, ␤,
and ␥) contain lipid binding helices that bind to acidic
phospholipids in vitro (5). The prevailing notion is that
synucleins have some function at the inner leaf of the
PM. ␣-Synuclein was originally identified as the precursor protein for the non-␤ amyloid component of
Alzheimer’s disease amyloid plaques (30), and a variant of ␣-synuclein appears to be responsible for a rare
genetic from of Parkinson’s disease (24). However, the
mechanistic role of these proteins in the pathology of
these diseases is unknown. Synucleins are not expressed in normal breast epithelia, but both ␤- and
␥-synuclein are expressed in the vast majority of ductal
carcinomas of the breast (3). The role of synucleins in
tumor progression remains unclear, but it is worth
noting that transfection of the MDA-MB-435 cells with
␥-synuclein was reported to increase the metastatic
potential of this cell line (11). Our finding that
␣-synuclein is unique to the PM proteome of the MDAMB-435 cells underscores the need to understand the
function of this family of proteins in metastatic spread.
Galectin-1. Galectin-1, a mammalian lectin that
binds to ␤-galactoside, was also found to be unique to
the PM proteome of MDA-MB-435 cells (spot E, Fig. 3).
Galectin-1 migrated at its predicted Mr and pI in the
2D gel. Galectin-1 can be secreted from cells and then
is able to bind back to glycosylated proteins on the PM.
Galectin-1 is upregulated in several cancers, and its
expression is often higher in more differentiated tumors (33, 34). Galectin-1 exists as a homodimer and
can facilitate intramolecular and intermolecular crossbridging. Such cross-bridging may facilitate cellular
signaling through growth factor receptors or may mediate cell-cell adhesion (23). Interestingly, galectin-1
has been shown to mediate the adhesion of MDA-MB435 cells to human endothelial cells, and this has been
proposed as a key route to metastatic spread (8). Our
observation that galectin-1 is unique to the PM proteome of MDA-MB-435 cells lends further support to
the idea that galectin-1 is connected to metastasis.
Thiol-specific antioxidant protein. Eighty polypeptide spots were observed to be present in the PM
preparations of MDA-MB-231 cells and MCF7 cells but
were undetectable in the PM preparation of the metastatic MDA-MB-435 cells. This finding suggests that
downregulation of large sets of proteins may make an
equally important contribution to the metastatic phenotype. One of these proteins was identified from the
MCF7 and MDA-MB-231 cells by mass fingerprinting.
It was found to be thiol-specific antioxidant protein
(TSA). TSA presumably has a role in protecting cells
against oxidative damage (35). This observation is consistent with a prior report showing that the TSA gene
is downregulated in MDA-MB-435 cells compared with
MDA-MB-231 and MCF7 cells (12).
DISCUSSION
The primary objective of this study was to obtain
basic information on the similarities and differences
among PM proteomes. One might hypothesize that
cells with similar lineage, or phenotype, would exhibit
a high degree of similarity at the PM. However, to our
knowledge, there has been no information published on
http://physiolgenomics.physiology.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 16, 2017
A
B
MALDI-MS
PLASMA MEMBRANE SIGNATURE
specific proteins appears to be immense and is consistent with the widely disparate phenotype of these
mammary carcinoma cell lines.
In the area of protein profiling there is a need to
perform large-scale comparisons that provide both protein identity and protein abundance. The present report illustrates some of the difficulties and limitations
encountered by a typical biochemistry laboratory in
meeting these hurdles. Although 2D gels represent the
best technology available for protein separation, they
clearly fall short in the separation of integral membrane proteins as well as true measurements of protein
abundance. Recent advances suggest that labeling of
proteins within a mixture with isotope tags can be
combined with mass spectrometry to obtain protein
identity and abundance on a broad scale (9). Independent work indicates that combining protein digestion
with highly sophisticated peptide mass analysis could
yield a similar level of information (14). Despite the
appeal of each of these strategies, they all require
significant technological expertise and resources. Given
the value of the profiling information to cell biology and
biochemistry, a strong argument for a national effort to
completely define the composition of cellular and subcellular proteomes can be made.
This study was supported by National Institutes of Health Grants
CA-69306 and HL-58925 (to J. W. Smith) and by Cancer Center
Support Grant CA-30199.
REFERENCES
1. Asano K, Vornlocher HP, Richter-Cook NJ, Merrick WC,
Hinnebusch AG, and Hershey JW. Structure of cDNAs encoding human eukaryotic initiation factor 3 subunits. Possible
roles in RNA binding and macromolecular assembly. J Biol
Chem 272: 27042–27052, 1997.
2. Bini L, Magi B, Marzocchi B, Arcuri F, Tripodi S, Cintorino M, Sanchez JC, Frutiger S, Hughes G, Pallini V,
Hochstrasser DF, and Tosi P. Protein expression profiles in
human breast ductal carcinoma and histologically normal tissue.
Electrophoresis 18: 2832–2841, 1997.
3. Bruening W, Giasson BI, Klein-Szanto AJ, Lee VM, Trojanowski JQ, and Godwin AK. Synucleins are expressed in
the majority of breast and ovarian carcinomas and in preneoplastic lesions of the ovary. Cancer 88: 2154–2163, 2000.
4. Chaney LK and Jacobson BS. Coating cells with colloidal
silica for high yield isolation of plasma membrane sheets and
identification of transmembrane proteins. J Biol Chem 258:
10062–100672, 1983.
5. Clayton DF and George JM. The synucleins: a family of
proteins involved in synaptic function, plasticity, neurodegeneration and disease. Trends Neurosci 21: 249–254, 1998.
6. De Benedetti A and Harris AL. eIF4E expression in tumors:
its possible role in progression of malignancies. Int J Biochem
Cell Biol 31: 59–72, 1999.
7. Donze O, Jagus R, Koromilas AE, Hershey JW, and Sonenberg N. Abrogation of translation initiation factor eIF-2 phosphorylation causes malignant transformation of NIH 3T3 cells.
EMBO J 14: 3828–3834, 1995.
8. Glinsky VV, Huflejt ME, Glinsky GV, Deutscher SL, and
Quinn TP. Effects of Thomsen-Friedenreich antigen-specific
peptide P-30 on ␤-galactoside-mediated homotypic aggregation
and adhesion to the endothelium of MDA-MB-435 human breast
carcinoma cells. Cancer Res 60: 2584–2588, 2000.
9. Gygi SP, Rist B, Gerber SA, Turecek F, Gelb MH, and
Aebersold R. Quantitative analysis of complex protein mixtures
using isotope-coded affinity tags. Nat Biotechnol 17: 994–999,
1999.
http://physiolgenomics.physiology.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 16, 2017
this topic. The findings of the present study indicate
that there are indeed signatures of proteins at the PM.
In general, the proteins at the PM can be segregated
into three groups. One group of proteins is likely to be
common to most cell types. We found that between 9%
and 23% of all proteins at the PMs of fibroblasts and
mammary carcinoma cells migrate to the same position
on 2D gels and are likely to be identical. Given the
functional distinction between these two types of cells,
it seems reasonable to suggest that this common set of
PM proteins is widely expressed among most cell types.
A second set of PM proteins appears to be linked
with cell lineage. From our 2D gel study, we found that
between 38% and 55% of the proteins at the PM of
mammary carcinoma are identical. The similarity
among fibroblasts was somewhat lower, ranging from
18% to 40%. The lower degree of similarity among the
fibroblasts may be explained by the fact that fibroblasts are less differentiated and have functions that
are less specialized than mammary epithelial cells.
One might expect that this set of proteins confer the
overall character of the cell. For example, in the case of
mammary carcinoma cells, these proteins might be
responsible for epithelial structure and function. This
structure-function relationship, then, would clearly
distinguish these cells from fibroblasts.
A third group of proteins appears to be expressed
uniquely on each individual cell line and be independent of cell lineage. These proteins composed roughly
22% of all proteins in the PM preparations. When
considering this set of proteins, it is important to emphasize the distinction between cell lineage and cellular phenotype. Lineage, or the tissue source of a cell,
clearly has a significant impact on the profile of proteins expressed by a cell. However, cells of similar
lineage, like the MCF7 and MDA-MB-435 cells, can
exhibit distinct phenotypes. This is particularly evident with cancer cells, which can represent any number of stages of tumor progression. For example, the
MCF7 cells will only grow as a primary tumor in mice
when estrogen is supplemented. In contrast, the MDAMB-435 cells grow tumors independent of estrogen.
Furthermore, the MCF7 cells fail to metastasize,
whereas the MDA-MB-435 cells exhibit extensive macroscopic metastases to multiple organs (25). Although a
clear lineage-specific signature is common among these
two cell lines, it is not surprising that another set of
proteins is different between the two. Nearly onefourth of all PM proteins appeared to be unique to
individual cell lines. Our identification of a few of these
proteins from the MDA-MB-435 cells, which show wide
variation in their proposed functions, illustrates the
scope of biochemical pathways that could be influenced
by these differences. For example, compared with the
MCF7 and MDA-MB-231 cells, the MDA-MB-435 cells
have undetectable levels of TSA, a key regulator of
oxidant stress. The MDA-MB-435 cells also differentially express proteins that could dramatically influence the kinetics of large groups of mRNAs, as well as
proteins that could alter the character of the inner leaf
of the PM. Altogether, the functional scope of the cell-
135
136
PLASMA MEMBRANE SIGNATURE
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
petropoulos T, Johnson WG, Lazzarini AM, Duvoisin RC,
Di Iorio G, Golbe LI, and Nussbaum RL. Mutation in the
␣-synuclein gene identified in families with Parkinson’s disease.
Science 276: 2045–2047, 1997.
Price JE, Polyzos A, Zhang RD, and Daniels LM. Tumorigenicity and metastasis of human breast carcinoma cell lines in
nude mice. Cancer Res 50: 717–721, 1990.
Rabilloud T, Adessi C, Giraudel A, and Lunardi J. Improvement of the solubilization of proteins in two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 18:
307–316, 1997.
Raught B, Gingras AC, James A, Medina D, Sonenberg N,
and Rosen JM. Expression of a translationally regulated, dominant-negative CCAAT/enhancer-binding protein ␤ isoform and
up-regulation of the eukaryotic translation initiation factor 2␣
are correlated with neoplastic transformation of mammary epithelial cells. Cancer Res 56: 4382–4386, 1996.
Shevchenko A, Wilm M, Vorm O, and Mann M. Mass spectrometric sequencing of proteins silver-stained polyacrylamide
gels. Anal Chem 68: 850–858, 1996.
Stinson SF, Alley MC, Kopp WC, Fiebig HH, Mullendore
LA, Pittman AF, Kenney S, Keller J, and Boyd MR. Morphological and immunocytochemical characteristics of human
tumor cell lines for use in a disease-oriented anticancer drug
screen. Anticancer Res 12: 1035–1053, 1992.
Ueda K, Fukushima H, Masliah E, Xia Y, Iwai A, Yoshimoto M, Otero DAC, Kondo J, Ihara Y, and Saitoh T.
Molecular cloning of cDNA encoding an unrecognized component
of amyloid in Alzheimer disease. Proc Natl Acad Sci USA 90:
11282–11286, 1993.
Williams K, Chubb C, Huberman E, and Giometti CS.
Analysis of differential protein expression in normal and neoplastic human breast epithelial cell lines. Electrophoresis 19:
333–343, 1998.
Wong NC, Mueller BM, Barbas CF, Ruminski P, Quaranta
V, Lin EC, and Smith JW. ␣v Integrins mediate adhesion and
migration of breast carcinoma cell lines. Clin Exp Metastasis 16:
50–61, 1998.
Xu XC, el-Naggar AK, and Lotan R. Differential expression of
galectin-1 and galectin-3 in thyroid tumors. Potential diagnostic
implications. Am J Pathol 147: 815–822, 1995.
Yamaoka K, Mishima K, Nagashima Y, Asai A, Sanai Y,
and Kirino T. Expression of galectin-1 mRNA correlates with
the malignant potential of human gliomas and expression of
antisense galectin-1 inhibits the growth of 9 glioma cells. J Neurosci Res 59: 722–730, 2000.
Yim MB, Chae HZ, Rhee SG, Chock PB, and Stadtman ER.
On the protective mechanism of the thiol-specific antioxidant
enzyme against the oxidative damage of biomacromolecules.
J Biol Chem 269: 1621–1626, 1994.
http://physiolgenomics.physiology.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 16, 2017
10. Jacobson BS, Stolz DB, and Schnitzer JE. Identification of
endothelial cell-surface proteins as targets for diagnosis and
treatment of disease. Nat Med 2: 482–484, 1996.
11. Jia T, Liu YE, Liu J, and Shi YE. Stimulation of breast cancer
invasion and metastasis by synuclein ␥. Cancer Res 59: 742–747,
1999.
12. Kirschmann DA, Seftor EA, Nieva DR, Mariano EA, and
Hendrix MJ. Differentially expressed genes associated with the
metastatic phenotype in breast cancer. Breast Cancer Res Treat
55: 127–136, 1999.
13. Landry F, Lombardo CR, and Smith JW. A method for
application of samples to matrix-assisted laser desorption ionization time-of-flight targets that enhances peptide detection.
Anal Biochem 279: 1–8, 2000.
14. Link AJ, Eng J, Schieltz DM, Carmack E, Mize GJ, Morris
DR, Garvik BM, and Yates JR III. Direct analysis of protein
complexes using mass spectrometry. Nat Biotechnol 17: 676–
682, 1999.
15. Mann M. A shortcut to interesting human genes: peptide sequence tags, expressed-sequence tags and computers. Trends
Biochem Sci 21: 494–495, 1996.
16. Mann M and Wilm M. Error-tolerant identification of peptides
in sequence databases by peptide sequence tags. Anal Chem 66:
4390–4399, 1994.
17. Nathan CA, Carter P, Liu L, Li BD, Abreo F, Tudor A,
Zimmer SG, and De Benedetti A. Elevated expression of
eIF4E and FGF-2 isoforms during vascularization of breast carcinomas. Oncogene 15: 1087–1094, 1997.
18. Nicolson GL. Trans-membrane control of the receptors on normal and tumor cells. II. Surface changes associated with transformation and malignancy. Biochim Biophys Acta 458: 1–72,
1976.
19. Nordling CO. A new theory on cancer-inducing mechanism.
Br J Cancer 7: 68–72, 1953.
20. Nupponen NN, Porkka K, Kakkola L, Tanner M, Persson
K, Borg A, Isola J, and Visakorpi T. Amplification and overexpression of p40 subunit of eukaryotic translation initiation
factor 3 in breast and prostate cancer. Am J Pathol 154: 1777–
1783, 1999.
21. Pain VM. Initiation of protein synthesis in eukaryotic cells. Eur
J Biochem 236: 747–771, 1996.
22. Pasquali C, Fialka I, and Huber LA. Preparative two-dimensional gel electrophoresis of membrane proteins. Electrophoresis
18: 2573–2581, 1997.
23. Perillo NL, Marcus ME, and Baum LG. Galectins: versatile
modulators of cell adhesion, cell proliferation, and cell death. J
Mol Med 76: 402–412, 1998.
24. Polymeropoulos MH, Lavedan C, Leroy E, Ide SE, Dehejia
A, Dura A, Pike B, Root H, Rubenstein J, Boyer R, Stenroos ES, Chandrasekharappa S, Athanassiadou A, Papa-