©2005 FASEB
The FASEB Journal express article 10.1096/fj.04-2970fje. Published online April 27, 2005.
Maspin alters the carcinoma proteome
Emily I. Chen,* Laurence Florens,*,1 Fumiko T. Axelrod,† Edward Monosov,†
Carlos F. Barbas III,‡ John R. Yates III,* Brunhilde Felding-Habermann,¶ and Jeffrey W. Smith†
*Department of Cell Biology, The Scripps Research Institute, CA 92037; †The Cancer Research
Center, Program on Cell Adhesion at The Burnham Institute, La Jolla, CA 92037; ‡Department
of Molecular Biology, The Scripps Research Institute, CA 92037; ¶Department of Molecular &
Experimental Medicine, The Scripps Research Institute, CA 92037
Corresponding author: Emily I. Chen, Department of Cell Biology, The Scripps Research
Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037. E-mail: emilyc@scripps.edu
1
Present address: Stowers Institute for Medical Research, Kansas City, MO 64110
ABSTRACT
Maspin, a member of the serine protease inhibitor (serpin) family, is a tumor suppressor in breast
and prostate cancer. To address molecular mechanisms underlying maspin’s activity, we restored
its expression in invasive carcinoma cells and analyzed the resulting changes by shotgun
proteomics. Using a mass spectrometry-based multidimensional proteomic method, we observed
changes to the expression of ~27% of the detectable proteome. In particular, we noted changes to
the expression of proteins that regulate cytoskeletal architecture, cell death, and protein turnover.
In each case, changes in protein expression were accompanied by measurable changes in tumor
cell phenotype. Thus, maspin-expressing cells exhibit a more prominent actin cytoskeleton, a
reduced invasive capacity, an increased rate of spontaneous apoptosis, and an altered proteasome
function. These observations reveal for the first time the far reaching effects of maspin on
multiple protein networks and a new hypothesis of maspin function based on the regulation of
proteasome function.
Key words: cancer proteomics • MudPIT • tumor suppressor • metastasis • protein profiling
M
aspin was discovered as a human mammary tumor suppressor in 1994 (1). Maspin’s
primary effect on tumor phenotype is a reduction in invasive capacity. In fact, the
reintroduction of maspin into invasive MDA-MB 435 mammary carcinoma cells
abolishes their metastatic capacity in vivo (1). A number of other studies have confirmed that
down-regulation of maspin correlates with tumor progression and invasive phenotype in humans
(2–4). The antitumor effects of maspin are not limited to mammary carcinoma, as similar
observations have been made in prostate carcinoma (5), and maspin also appears to inhibit
angiogenesis (6).
The mechanisms by which maspin elicits its antitumor and antimetastatic effects are still the
subject of intense study. Maspin is structurally homologous to members of the serine protease
Page 1 of 25
(page number not for citation purposes)
inhibitor (serpin) superfamily, like plasminogen activator inhibitors 1 and 2 (PAI-1 and PAI-2)
and α1-antitrypsin (7). The homology of maspin to serpins led to the hypothesis that its tumor
suppressor function might be attributed to its ability to inhibit proteolysis. This idea is supported
by the fact that maspin was reported to inhibit the activity of tissue-type plasminogen activator
(tPA) (8) and the observation that maspin can mediate the inhibition of urokinase-type
plasminogen activator (uPA) on the surface of prostate carcinoma cells (9).
Despite these studies, other work argues that maspin lacks classical serpin activity. For example,
maspin fails to undergo a “stressed to relaxed” conformational transition that is characteristic of
virtually all serpins (10). In addition, a highly rigorous study showed that maspin lacks the
ability to inhibit either tPA or uPA, two of its presumed targets, under conditions in which the
activity of these proteases can be verified by their ability to activate plasminogen (11). Even
though maspin may lack the ability to inhibit serine proteases, its biological function can be
attributed to the reactive loop. Recent work shows that subsitution of maspin’s reactive site loop
into ovalbumin endows that protein with proadhesive and antiinvasive function (12). Moreover,
the loop appears to mediate binding to a cell surface receptor, that ultimately promotes cell
adhesion to type I collagen and fibronectin. Other functions for maspin cannot be excluded
especially because histologic analysis of maspin shows prominent nuclear and cytoplasmic
staining (13, 14). These observations indicate that maspin could have functions that are therefore
unappreciated.
To gain insight into the molecular mechanisms of maspin’s tumor suppressor function, we
assessed the nature and scope of maspin’s effects on the tumor cell proteome. This was done
through shotgun proteomics, using multidimensional protein identification technology
(MudPIT), which couples tandem mass spectrometry with multiple liquid chromatography steps
(15). Surprisingly this analysis revealed that the expression of maspin has widespread effects on
the tumor cell proteome. The effects are observed in nearly one quarter of the detectable
proteome and span multiple functional categories. In most cases, protein expression was affected
without changes in mRNA levels, indicating that maspin has a significant influence on
posttranscriptional regulation of protein levels. This observation is probably a result of maspin’s
effects on the expression and activity of the proteasome, which may be central to maspin’s tumor
suppressor activity.
MATERIALS AND METHODS
Tissue culture
The MDA-MB 435 cell line was obtained from Dr. Jane E. Price of the M.D. Anderson Cancer
Center, Houston, TX. Tissue culture reagents were purchased from Invitrogen Life Technologies
(Carlsbad, CA) unless specified. The cells were maintained in minimal Eagle’s media (MEM)
Earl’s salts, 2 mM L-glutamine, MEM vitamins, nonessential amino acids, and antibiotics
(Omega Scientific, Tarzana, CA). Maspin-transfected cells were maintained in the same media
with addition of G418 neomycin.
Page 2 of 25
(page number not for citation purposes)
Establishing stable maspin-transfected MDA-MB 435 cells
Maspin cDNA was PCR amplified from total RNA extracted from human mammary epithelial
cells and cloned into a mammalian expression vector, pCDNA3 (Invitrogen Life Technologies).
The resulting construct was used to transfect MDA-MB-435 breast cancer cell using the DOTAP
reagent (Roche, Indianapolis, IN). Stable maspin MDA-MB-435 transfectants were selected by
800 µg/ml G418, and the expression of maspin was confirmed by both PCR and Western blot.
Positive clones were maintained in MEM media plus 400 µg/ml G418.
Matrigel invasion assay
The matrigel invasion assay was performed according to the manufacturer’s protocol (BD
Bioscience, San Jose, CA). Briefly, parent, mock-transfected, and maspin-transfected MDA-MB
435 cells were grown overnight into serum-free media containing 0.1% BSA. The next day, cells
were harvested with Versene (0.5 mM EDTA in PBS), resuspended into serum-free MEM media
containing 0.1% BSA, and plated in the top compartment of an invasion chamber (BD
Bioscience) with a 0.8 µm pore membrane coated with matrigel. MEM media with 5% fetal
bovine serum (FBS) was added to the lower chamber, and the cells were allowed to invade
through matrigel overnight at 37°C. The next day, the membranes were recovered, fixed, and
stained with DiffQuick (IMEB, San Marcos, CA). Cells that had invaded through matrigel were
counted at the underside of the membrane in ×20 high-power fields. Triplicates of each cell line
were seeded for each invasion experiment, and two independent experiments were performed.
The results represent average numbers of two experiments.
SCID mouse model of metastasis
The model of spontaneous metastasis used here was modified from the model originally
described by Price et al. (16). Briefly, female SCID mice aged 5–6 wk (The Jackson
Laboratories, Bar Harbor, ME) were anesthetized with ketamine (50 mg/kg) in xylazine
administered i.m. Tumor cells were injected into the surgically exposed second mammary fat pad
(m.f.p.) of the animal’s left side (5×105 cells in 50 µl of Hank’s balanced salt solution) using a
0.5 ml insulin syringe. Incisions were closed with surgical staples. Tumors were palpable 2 wk
after injection into the m.f.p. Primary tumors were allowed to grow for 8 wk, when they were
surgically removed under anesthesia. Metastases in lungs were evident 4–5 wk later and were
quantified (see below).
Quantification of tumorigenicity and metastasis
After careful excision, primary tumors were weighed. Lungs were fixed in Bouin’s solution, and
the number of macroscopic metastases was assessed by counting nodules at the surface of the
lung under a dissecting microscope. For consistency, the upper right lobe of each lung was used
for quantification. In all cases, this lobe represented metastases to the whole lung.
Fractionation and digestion of proteins
Three cell lines, MDA-MB-435 and two stable maspin MDA-MB-435 transfectants (M27 and
M59) were seeded at the same cell density in five 150-mm tissue culture dishes per cell line, and
the cell pellets were harvested when the cells reached ~80% confluency. From each cell line,
Page 3 of 25
(page number not for citation purposes)
5×106 cells were used for each set of MudPIT analysis. Two sets of sample preparation were
made in duplicates. In the first set of sample preparation, cell pellets were suspended in 100 µl of
10 mM Tris, pH 8.0, and incubated on ice for 1 h. Then, cells were spun down and the
supernatant was transferred to a fresh microcentrifuge tube (soluble fraction). The cell pellets
were resuspended in 100 µl of 0.1 M NaCO3, pH 11.0, and incubated on ice for 1 h (insoluble
fraction). Solid urea was added to 8 M. Proteins were reduced with 5 mM Tris(2carboxyethyl)phosphine hydrochloride (TCEP, Roche Applied Science, Palo Alto, CA) and
alkylated with 20 mM iodoacetamide (IAM, Sigma, St. Louis, MO). Proteins were digested
either following the trypsin protocol (15) or the high pH proteinase K protocol (17). In the
second set of sample preparation, cell pellets were resuspended in phosphate sucrose buffer and
digested with either trypsin or proteinase K (external fraction). Then, the remaining cell pellets
were further digested with proteinase K (internal fraction).
Multidimensional chromatography and tandem mass spectrometry
Peptide mixtures were pressure-loaded onto a 250 µm inner diameter (i.d.) fused-silica capillary
packed first with 4 cm of 5 µm strong cation exchange material (Partisphere SCX, Whatman,
Florham Park, NJ), followed by 2 cm of 5 µm C18 reverse phase (RP) particles (Aqua,
Phenomenex or Polaris 2000, Varian, Palo Alto, CA). Loaded and washed microcapillaries were
connected via a 2 µm filtered union (UpChurch Scientific, Oak Harbor, WA) to a 100 µm i.d.
column, which had been pulled to a 5 µm i.d. tip using a P-2000 CO2 laser puller (Sutter
Instruments, Novato, CA), then packed with 10 cm of RP particles and equilibrated in 5%
acetonitrile and 0.1% formic acid (buffer A). This split-column was then installed in-line with a
Quaternary Agilent (Foster City, CA) 1100 series HPLC pump. Overflow tubing was used to
decrease the flow rate from 0.1 ml/min to ~200–300 nl/min. Fully automated 12-step
chromatography runs were carried out. Three different elution buffers were used: 5% acetonitrile
and 0.1% formic acid (buffer A); 80% acetonitrile and 0.1% formic acid (buffer B); and 0.5 M
ammonium acetate, 5% acetonitrile, and 0.1% formic acid (buffer C). In such sequences of
chromatographic events, peptides are sequentially eluted from the SCX resin to the RP resin by
increasing salt steps (increase in buffer C concentration), followed by organic gradients (increase
in buffer B concentration). The last chromatography step consists of a high-salt wash with 100%
buffer C followed by acetonitrile gradient. The application of a 2.5 kV distal voltage
electrosprayed the eluting peptides directly into an LCQ-Deca ion trap mass spectrometer
equipped with a nano-LC electrospray ionization source (ThermoFinnigan, San Jose, CA). Full
MS spectra were recorded on the peptides over a 400–1600 m/z range, followed by three tandem
mass (MS/MS) events sequentially generated in a data-dependent manner on the first, second,
and third most intense ions selected from the full MS spectrum (at 35% collision energy). Mass
spectrometer scan functions and HPLC solvent gradients were controlled by the Xcalibur data
system (ThermoFinnigan).
Interpretation of MS/MS datasets
PEP_PROBE (18) was used to match MS/MS spectra to peptides in a database containing human
sequences downloaded from NCBI on March 2, 2004. PEP_PROBE is a modified version of
SEQUEST (19) and uses a hypergeometric probability model to calculate the confidence for a
match to be nonrandom. The validity of peptide/spectrum matches was hence assessed using the
SEQUEST-defined parameters, cross-correlation score (XCorr), and normalized difference in
Page 4 of 25
(page number not for citation purposes)
cross-correlation scores (DeltaCn), as well as the PEP_PROBE-defined parameters, probability
and confidence for a match to be nonrandom. Spectra/peptide matches were only retained if they
had a DeltaCn of at least 0.08 and minimum XCorr of 1.8 for +1, 2.5 for +2, and 3.5 for +3
spectra. An 85% confidence for the peptide/spectrum matches not to be random (as defined by
PEP_PROBE) was used as cut-off. In addition, the minimum sequence length was seven amino
acid residues. DTASelect (20) was used to select and sort peptide/spectrum matches passing this
criteria set. Peptide hits from multiple runs were compared using CONTRAST (20). Proteins
were considered detected if they were identified by at least three spectra passing all of the
selection criteria or with at least 10% sequence coverage. To generate a dendrogram, an arbitrary
minimum value was used to represent the relative protein abundance when the protein was not
detectable in the cell line.
Western blot analysis
Whole cell protein extracts were prepared using TotalProteinExtraction kit (Biochain, Hayward,
CA), and protein concentration was determined by BCA assay from Pierce (Rockford, IL). Total
cell lysate (50 µg) from each cell line were then used to compare protein expression.
Immunoblottting assays were carried out by standard procedures using commercially available
antibodies (see supplementary material for detailed list). Anti-β-actin antibody (Sigma-Aldrich
Corp., St. Louis, MO) was used for equal loading control. Bands were detected using horseradish
peroxidase-labeled secondary antibodies and enhanced chemiluminescence detection kit
(Amersham Pharmacia, Piscataway, NJ).
Real-time quantitative PCR
Changes in the expression of 56 target genes were examined with real-time PCR (Roche
Diagnostic Corporation). The full-length cDNA sequences for genes of interest were obtained
from the National Library of Medicine (http://www.ncbi.nlm.nih.gov/UniGene/). Primers were
designed to amplify the human target genes using a web-based software
(http://www.genome.wi.mit.edu/cgibin/primer/primer3_www.cgi), and then purchased from
Integrated DNA Technologies (Coralville, IA). RNA (5 µg) from each tested sample were
reverse-transcribed with Superscript II reverse transcriptase (Invitrogen Life Technologies,
Carlsbad, CA). The resulting cDNA was diluted 20-fold before PCR amplification. Real-time
PCR was performed using a LightCycler (Roche Applied Science, Palo Alto, CA). Reactions
were performed using LightCycler-DNA Master SYBR Green I mix (Roche Applied Science). A
typical protocol involved a 30 s denaturation step at 95°C, 40 cycles with annealing at 55°C for 5
s and extension at 72°C for 5 s. An automated melting curve analysis was used to verify that all
primers yielded a single PCR product.
Fluorescent staining of actin filaments
Cells were grown on Lab-Teck glass chamber slides (Fisher Scientific, Pittsburgh, PA) the day
before the experiment. The next day, cells were rinsed once with the serum-free media and fixed
in 4% formaldehyde (freshly prepared from p-formaldehyde) in 1× PBS, pH 7.3, for 15 min at
room temperature.
Page 5 of 25
(page number not for citation purposes)
Fixed cells were washed a few times in 1× PBS and permeabilized in 0.1% Triton X-100 in the
same buffer for 7 min, at room temperature, washed again with 1× PBS, and incubated with
Alexa Fluor 594 phalloidin (1:100) (Molecular Probes, Eugene, OR). Stained samples were
washed in 1× PBS to eliminate unbound fluorecent conjugates, infused with VectoShield
antifadding media (Vector Laboratories, Burlingame, CA), coverslipped and sealed with
CytoSeal 60 (Fisher Scientific). Specimens were studied under inverted microscope Nikon
TE300, with 60×, N.A. 1.4, oil immersion objective lens and Texas Red/Cy3,5 epi-fluorescent
cube (Chroma Technology, Rockingham, VT).
Apoptosis and cell proliferation assays
Spontaneous apoptosis in all cell lines was measured using a Cell Death Detection ELISAPLUS
assay (Roche Applied Science). This is a photometric enzyme-immunoassay to detect and
quantitate cytoplasmic histone-associated-DNA-fragments (mono- and oligonucleosomes). Cell
proliferation in all cell lines was measured using a Cell Proliferation ELISA, which measures the
BrdU incorporation over time (Roche Applied Science). All procedures were performed
according to the manufacturer’s instructions. Briefly, sample cells were trypsinized, resuspended
in culture media, and seeded in triplicate wells (1×104 cells/well) in 96-well tissue culture plates.
Cells were incubated at 37°C and 5% CO2 for 4 days before measuring apoptosis and
proliferation. For the cell death assay, cells were resuspended in 200 µl lysis buffer following by
30 min incubation at room temperature. After incubation, 20 µl of lysate from each well were
transferred to the streptavidin-coated microtiter plate. 80 µl of immunoreagent were added to
each well, and the plate was incubated for 2 h at room temperature with gentle shaking (~300
rpm). Next, the plate was washed three times with the incubation buffer before the addition of
the ABTS solution, and the extent of apoptosis was measured by absorbance at 405 nm in a
microplate reader. Each experiment was repeated twice to confirm the trend, and one set of the
representative data was presented in the result section.
For the cell proliferation assay, the media was replaced with the BrdU labeling media 16 h
before the analysis of cell proliferation. After BrdU labeling, FixDenat solution was added to the
cells and removed after 30 min incubation at room temperature. Then antiBrdU POD was added
to each well and incubated for 30–120 min at room temperature before the addition of the
substrate solution. Once the desired intensity of color was developed, the reaction was stopped
by 1 M H2SO4 solution. The absorbance of each well was measured at 450 nm.
Fluorometric 20S proteasome assay
Cells were harvested after four passages in culture and homogenized in the lysis buffer from the
Biochain total protein extraction kit (Biochain). Cellular 20S proteasome activity was assayed
using a 20S proteasome assay kit (BIOMOL International LP, Plymouth Meeting, PA) according
to the manufacturer’s standard protocol. Fluorogenic peptide substrates Suc-LLVY-AMC
(chymotrypsin activity) and Z-LLE-AMC (PGPH activity) were obtained from Calbiochem (San
Diego, CA). Fluorogenic peptide substrate Boc-LRR-AMC (trypsin activity) was obtained from
BIOMOL International LP. Briefly, 50 µg of cell lysate were added to 1× reaction buffer diluted
from the 20× reaction buffer (500 mM HEPES, 10 mM EDTA, pH 7.6) so that the final reaction
volume was 90 µl. Triplicate reactions were set up for each cell lysate. The diluted lysates were
incubated at 37°C for 5 min for equilibration at assay temperature. During incubation, 20×
Page 6 of 25
(page number not for citation purposes)
substrate solution was made by diluting the fluorogenic peptide substrate stocks in reaction
buffer. To each reaction, 10 µl of the appropriate 20× substrate stock solution were added, and
the fluorescence activity (ëex: 380 nm; ëem: 460 nm) was measured in a microplate fluorescence
reader (Gemini XPS from Molecular Device, Sunnyvale, CA). Data shown in the result section
represented one of the replicate experiments. The final substrate concentration was 10 µM for the
chymotrypsin-like and trypsin-like activity assays and 100 µM for the PGPH activity assay. MG132 (Calbiochem, San Diego, CA), a chymotryptic proteasome inhibitor, was used to show the
specificity of the chymotrypsin-like proteolytic activity.
Analysis of high molecular weight ubiquitin-protein conjugates
Cell lysates were generated as described in the Western blot analysis section. Cell lysates were
separated in a 10% SDS-PAGE, transferred to a nitrocellulose membrane, and blocked with 5%
nonfat milk. A monoclonal anti-ubiquitin antibody (1:100) (EMD Bioscience, San Diego, CA)
was incubated with the blot for 2 h at room temperature. Then, the blot was washed three times
with wash buffer (PBS+0.05% Tween 20). A HRP-conjugated anti-mouse (1:10,000) antibody
(Bio-Rad, Hercules, CA) and an enhanced chemiluminescence detection kit (GE HealthAmersham Bioscience, Piscataway, NJ) were used to detect the presence of ubiquitin as well as
ubiquitin modified proteins.
RESULTS
Proteomic profiling of parental and maspin-transfected carcinoma cells
The anti-metastatic properties of maspin were originally demonstrated in the metastatic MDAMB 435 mammary carcinoma cells (1). To address the effects of maspin on the tumor proteome,
we recapitulated this system by establishing four stable sublines of the MDA-MB 435 cells that
were transfected with maspin cDNA. The expression of maspin was verified for all of the
transfected clones (Fig. 1A). As previously reported, transfection of the MDA-MB 435 cells with
maspin reduced cell invasion in vitro and metastatic spread in vivo (Fig. 1B, 1D). As previously
reported by Zou et al. (1), some of the clones transfected with maspin also had reduced
tumorigenicity (Fig. 1C, clones M27 and M60). However, reductions in primary tumor growth
do not appear to be necessary for reductions in metastasis because all maspin-transfected clones
showed nearly complete ablation of metastasis in vivo. We were most interested in identifying
changes linked to reductions in metastatic spread, so we chose to compare protein expression in
the parental MDA-MB 435 cells with that of the M27 and M59 lines. We expected this
comparison to reveal changes common to the nonmetastatic phenotype, without bias toward
proteins mechanistically involved in altering tumorigenicity.
The MDA-MB 435 cells and the M27 and M59 lines were subjected to MudPIT analysis. In
total, we detected 30,000 peptides that corresponded to 2315 different proteins (Table 1,
supplementary material). To obtain a semiquantitative comparison of protein expression levels,
we used two parameters: the number of spectra collected per protein (spectra count) (21) and the
percentage of protein sequence covered by detected peptides (sequence coverage) (22). To
minimize false positives, the initial list of 2315 proteins was filtered to eliminate proteins that
were identified by fewer than three spectra or <10% sequence coverage. These filters produced a
list of 1124 proteins for further analysis (Fig. 1, supplementary material). We compared the
Page 7 of 25
(page number not for citation purposes)
expression of each protein on this list in the M27, the M59, and parental MDA-MB 435 cells. An
expression ratio of 1.5-fold was arbitrarily set as a threshold for significance. Using this
criterion, we found 113 proteins to be up-regulated and 185 proteins to be down-regulated in
both of the maspin-transfected cell lines. Based on these values, maspin led to alterations in the
expression level of ~27% of the detected proteome.
Independent approaches were used to validate the changes in protein levels indicated by MudPIT
analysis. Based on known relevance to cancer biology and the availability of antibodies, we
chose 22 differentially expressed proteins for analysis by Western blot. The direction of the
change in protein expression observed by Western blot correlated exceptionally well with
MudPIT analysis. Twenty out of 22 proteins exhibited altered expression in the direction
indicated by the differential proteomic analysis (Fig. 2A, 2B).
In addition, we compared the mRNA level of 55 candidate proteins to determine if changes in
protein expression could be attributed to changes in gene transcription (Table 1). Interestingly,
only 17 of 55 genes exhibited changes that mirrored the alterations in protein levels. Therefore,
maspin elicits changes to both gene and protein expression levels, but the majority of changes
detected in this study can be attributed to posttranscriptonal regulation of protein expression.
Maspin expression affects proteins of the actin cytoskeleton network
Using the web-based data analysis software, Genesifter (www.GeneSifter.net), we clustered the
proteomic dataset based on the Gene Ontology (GO) classification. From this analysis we
recognized that maspin expression affected three protein networks that are crucial in tumor
progression and metastasis: cytoskeleton organization, apoptosis, and protein turnover.
In maspin-transfected cells, considerable changes in the expression of proteins linked to
cytoskeletal organization were detected (Fig. 3A). Changes to several of the proteins were
confirmed by Western blot (Fig. 2A, 2B,) see α-actin, plectin 1, actinin-4, caldesmon 1, and
profilin 1). Immunostaining of the maspin-expressing cells revealed parallel and thick actin
filaments, which were different from the parental and mocked-transfected MDA-MB 435 cells
(Fig. 3B). The actin filaments of the maspin transfectants resemble that of nonmigratory cells,
whereas the thinner and less dense actin cables of the parental cells (Fig. 3B) are more consistent
with migratory cells. These observations are generally consistent with results from the proteomic
analysis, where proteins known to restrict cell migration are up-regulated in the maspintransfected cells (e.g., α actinin-4, tropomyosin 1, caldesmon 1, coronin, desmuslin), while
proteins known to promote cell migration were down-regulated in response to maspin expression
(e.g., actin-related protein, α-tubulin, leupaxin, F-actin capping protein). These results are
consistent with maspin’s antiinvasive activity.
Maspin expression increases the spontaneous apoptosis in carcinoma cells
Proteins involved in apoptosis were also affected by maspin. Differential proteomic analysis
showed that proteins known to promote apoptosis were up-regulated in maspin-expressing cells
(e.g., reticulon 4, tyrosyl tRNA ligase, Acinus, SON DNA binding protein), while antiapoptotic
proteins were down-regulated (e.g., α-synuclein, secreted phosphoprotein 1) (Fig. 4A). In
agreement with these changes, we found an increased rate of spontaneous cell death in all four
Page 8 of 25
(page number not for citation purposes)
maspin-transfected clones compared with the parental or mock-transfected MDA-MB 435 cells
(Fig. 4B). No significant differences in the rate of proliferation were observed (Fig. 4B). This
proapoptotic phenotype may contribute to the reduced metastatic behavior of maspin-transfected
cells in vivo.
Maspin expression inversely correlates with the chymotrypsin-like activity of the 26S
proteasome
Most of the changes elicited by maspin on protein expression appear to be regulated at the
posttranscriptional level, and our differential proteomic analysis revealed changes to important
components in the ubiquitin-proteasome system (Fig. 5A). Therefore, we performed a more
comprehensive analysis of this protein network. First, we compared the activity level of the
proteasome in maspin transfectants. The 20S proteasome has three major proteolytic activities: a
chymotrypsin-like, a trypsin-like, and a peptidylglutamylpeptide hydrolyzing activity (PGPH)
(23, 24, 25). Each of these can be assessed in cell lysates using fluorescent substrates that reflect
their catalytic specificity (26, 27). We found the trypsin-like and PGPH catalytic activities are
essentially identical in the maspin transfectants and the parental MDA-MB 435 cells (data not
shown). However, maspin-expressing cells showed reduced chymotrypsin-like activity compared
with the parental and mock-transfected cells (Fig. 5B). The β5 subunit of the proteasome plays a
crucial role in chymotrypsin-like activity of proteasome (28), but it was not detected in the
MudPIT analyses perhaps due to its low abundance. Therefore, we compared its protein level
among the parental and maspin-transfected MDA-MB 435 cells using Western blot. Indeed, we
found that the β5 subunit was down-regulated in all maspin-transfected cells compared with the
parent cells (Fig. 5C). Consequently, reduced chymotrypsin-like activity observed in maspintransfected cells can partly be attributed to down-regulation of β5 subunit of the proteasome.
In concert with the reduced chymotrypsin-like activity of the proteasome, we also found
accumulation of high molecular weight ubiquitin-protein conjugates in maspin-transfected cells
(Fig. 5D). This accumulation may result from an overall reduction in proteasome activity. The
different catalytic sites of the proteasome act in a hierarchical manner (29), with the
chymotrypsin-like activity normally determining the rate of protein breakdown (23). Thus, a
reduction in chymotryptic activity is likely to change the regulation of protein turnover in
maspin-expressing cells. Recent work showed that reductions in overall proteasome activity can
lead to the accumulation of high molecular weight ubiquitin conjugates (30). Consequently, our
results suggest an inverse correlation between the expression of maspin and proteasome activity.
DISCUSSION
It is not entirely clear whether maspin’s tumor suppressor activity is a result of its function as a
protease inhibitor (8, 9) or as a ligand for some cell surface receptor (12), or both. This study was
an attempt to jump downstream of these points of mediation and understand the biochemical
basis for maspin’s effects on tumor phenotype. To accomplish this, we used shotgun proteomics
to compare the proteomes of maspin-deficient and maspin-expressing tumor cells. The general
conclusions of this study are 1) the effects of maspin on the tumor cell proteome are widespread,
with changes in expression evident in about one quarter of all detected proteins; 2) maspin elicits
changes in the expression of proteins associated with the actin cytoskeleton that predict a less
motile and invasive phenotype, a finding that is consistent with the less invasive phenotype of
Page 9 of 25
(page number not for citation purposes)
maspin in vitro, and reduced metastatic spread in vivo; 3) maspin is correlated with changes to
proteins associated with apoptosis that predict a greater sensitivity to cell death, a phenotype that
is consistent with a higher rate of spontaneous apoptosis of maspin-expressing tumor cells, and
that may contribute to the inability of maspin-expressing cells to metastasize; 4) maspin appears
to have an effect on the ubiquitin-proteasome pathway, indicating that the proteasome could be a
key downstream regulator of tumor progression; and 5) MudPIT is capable of providing highfidelity comparisons of relative protein expression, which, in the vast majority of cases, can be
confirmed by Western blotting.
Importantly, the expression level of maspin in each transfected cell line appears to exceed the
threshold for a complete biological response. While there are variations in maspin levels among
the four transfected clones, their invasiveness in vitro and metastatic spread in vivo is essentially
ablated in all of the clones (Fig. 1). Similarly, even though the level of chymotryptic proteasome
activity in each transfected line is reduced compared with controls, there are no substantive
differences in this activity among the maspin-transfected lines (Fig. 5B). Together, these
bioassays indicate that the system used in this study is suitable for identifying trends in protein
expression correlated to maspin and its antimetastatic effects. However, the system is probably
not suitable for testing for correlations between changes to a proteins expression vs. a “dose
response” of maspin. With this limitation in mind, several correlations that we identified are
worthy of further discussion.
One hallmark of malignant transformation is the enhanced motility of cancer cells and their
invasion into normal tissues (31). The actin network and its associated proteins are necessary to
support this enhanced migratory ability of invasive cancer cells. Our results show that maspin
causes significant alterations in proteins associated with the actin network. In most cases the
changes in these cytoskeletal proteins are predicted to result in a reduction in cell motility. For
example, epithelial protein lost in neoplasm (EPLIN) is a member of cytoskeleton-associated
proteins that was found up-regulated in maspin-transfected breast cancer cells. EPLIN has been
shown to be down-regulated in transformed cells (32, 33). It promotes the formation of stable
actin filament structures such as stress fibers at the expense of more dynamic actin filament
structures such as membrane ruffles (34). Hence, reduced expression of EPLIN may contribute
to the motility of invasive tumor cells. Another example of a cytoskeleton-associated protein that
has a profound influence on cell motility is gelsolin. Opposite of EPLIN, gelsolin was found
down-regulated in maspin-transfected breast cancer cells. Gelsolin is one of a family of actin
binding proteins involved in controlling the organization of the actin cytoskeleton in cells (35).
Cells deficient in gelsolin exhibit defective chemotaxis and wound healing (36); conversely,
overexpression of gelsolin increases membrane ruffling and chemotaxis (37). These changes are
consistent with the fact that maspin-transfected breast cancer cells contain bundles of actin in the
form of stress fibers, and exhibit reduced motility, invasiveness, and metastatic potential.
Apoptosis is another critical cellular function that is frequently dysregulated in tumor cells (38–
40), particularly in those with increasing metastatic propensity (41). A proapoptotic effect of
maspin has been reported by Sheng’s group in breast- and prostate cancer cells (42, 43). They
reported that endogenous maspin expression sensitizes breast carcinoma cells to staurosporine
(STS)-induced apoptosis in vitro and Bax seems to play an essential role in the proapoptotic
effect of maspin. Our data further demonstrate that increased apoptosis plays an important role in
maspin function. For example, one of the apoptosis regulatory proteins, protein phosphatase 2
Page 10 of 25
(page number not for citation purposes)
(PP2A), was found up-regulated in maspin-transfected breast cancer cells. Recently it has been
shown that ceramide activates protein phosphatase 2 and can regulate apoptosis by a mechanism
involving dephosphorylation of the antiapoptotic molecule Bcl2 (44). The nonphosphorylatable
(i.e., PP2A-resistant) gain-of-function S70E mutant Bcl2 can protect cells from ceramideinduced apoptosis (44). Since ceramide production is a nearly universal component of apoptosis,
it is possible that such a mechanism may exit in response to other apoptotic stimuli. Another
apoptosis regulatory protein, Acinus, was found up-regulated in maspin-transfected breast cancer
cells. Using an in vitro system, Tsujimoto’s group showed that Acinus is essential in the
induction of apoptotic chromatin condensation in cells after cleavage by caspase-3 without
inducing DNA fragmentation (45). These changes are consistent with our observation that
maspin-transfected breast cancer cells show signs of increased spontaneous apoptosis.
The ubiquitin proteasome pathway is a highly conserved intracellular pathway for intracellular
protein degradation, and our results strongly suggest that the function of this pathway is
influenced by the expression of maspin. In this study, we showed that there is a negative
correlation between maspin expression and the chymotrypsin-like activity of the proteasome.
The reduction in chymotrypsin-like activity appeared to result from the down-regulation of the
β5 subunit of the proteasome in all maspin-transfected cells. This newly found connection raises
the interesting possibility that the tumor suppressor function of maspin may be mediated through
the proteasome.
This idea is supported by three results. First, most of the changes in protein abundance that we
detect occur at the post-transcriptional level. Since the proteasome is a major regulator of protein
turnover, it stands to reasons that it plays a role in enacting many of the changes in protein
abundance that we observe. Second, we also find direct evidence that the abundance of several
subunits of the proteasome are altered by maspin, including the β5 subunit (Fig. 5C), as well as
the α6 and α7 subunits (Fig. 6). Third, we observed that maspin-expressing cells exhibit
reductions in the chymotrypsin-like activity of the proteasome (Fig. 5B). Altogether, these
observations indicate that subunit composition and activity of the proteasome are altered in
maspin-expressing cells.
Interestingly, there are a number of examples in the literature where changes in subunit
composition of the proteasome result in phenotypic changes in cells. For example, proteasomes
isolated from different tissues are comprised of different combinations of subunits, and have
different activities toward small peptide-based substrates (46, 47). Changes in proteasome
subunit composition and activity are also observed in Drosophila during development (48) where
dominant-negative mutations in individual subunits of the proteasome have been shown to
regulate cell fate (49). Perhaps the best known phenotypic change resulting from proteasome
subunit switching is the interferon-γ-induced expression of the immunoproteasome, which has a
key role in the presentation of antigen peptides by MHC class I molecules (50). Interestingly, the
immunoproteasome has a distinct substrate specificity by substituting three of the proteasome
subunits with interferon-γ inducible subunits (51–54). The results presented here show
interesting parallels with subunit switching in the case of the immunoproteasome, and suggest
the possibility that maspin enacts a change in proteasome composition and activity, which
ultimately plays a role in determining the metastatic phenotype.
Page 11 of 25
(page number not for citation purposes)
Maspin could potentially be involved directly or indirectly in the regulation of proteasome
activity. To prove an interaction of maspin with the proteasome, we immuno-precitated the 20S
and 26S proteasomes with antiproteasome α-subunits (BIOMOL International LP) and anti19S
antibodies (EMD Bioscience) but we did not detect maspin in the pulled-down complexes (data
not shown). The result indicated either no direction interaction or weak interaction between the
proteasome and maspin. Furthermore, we found that maspin expression in breast cancer cells
caused the accumulation of high molecular weight ubiquitin-protein conjugates as determined by
immunoblotting with ubiquitin-specific antibody (Fig. 5D). Maspin was detected in the enriched
ubiquitinated protein fraction from the cell lysate (data not shown). There was no higher
molecular weight maspin detected, suggesting that maspin is probably not ubiquitin-conjugated.
It is known that catalytic properties of the proteasome can vary widely, depending on its
association with regulatory proteins (55). An example of this paradigm is inhibition of the
proteasome activity by insulin. In HepG2 cells, insulin increases ubiquitin-conjugate protein
accumulation by 80% and decreases ubiquitin-mediated proteasomal activity (56). Furthermore,
it was shown that the displacement of insulin degradation enzyme (IDE) by insulin from the 20S
proteasome provided a mean by which insulin may inhibit the ubiquitin-dependent degrading
activity of the 26S proteasome (56). Our preliminary results of maspin and the proteasome
regulation seem to resemble the paradigm of reducing the proteasome activity by cofactor
displacement such as the insulin and insulin degradation enzyme system described above.
Therefore, it raises the possibility of reducing the proteasome activity by displacing a yet
unknown proteasome cofactor by maspin.
In summary, proteome analysis provided valuable insights into maspin’s antimetastasis
mechanism. By investigating the proteome of cancer cells affected by maspin expression, we
identified potential molecular pathways for the observed phenotypic changes in maspintransfected cancer cells: structural alterations of actin cytoskeleton, and increased apoptosis. We
also revealed an inverse correlation between maspin and 20S proteasome activity. Deregulation
in the expression of 20S proteasome subunits seemed to accompany the decreased cellular
proteasome activity in the maspin-transfected cancer cells. As we continue to learn about the
aberrant regulation of the ubiquitin-proteasome pathway in cancer cells and the resulting changes
in its housekeeping functions, we can speculate that alteration in proteasome activity might be a
key step toward metastasis. Based on the present study, we speculate that many effects that have
been reported about maspin could be attributed to maspin’s function in keeping the delicate and
specific regulation of the ubiquitin-proteasome pathway, the net result of which could influence
metastasis-related events. Although we have only little clues about the detail mechanism of how
maspin can modulate proteasome activity directly or indirectly, our data point toward a new
hypothesis of the molecular mechanism of maspin, and insinuate a potential contribution of the
ubiquitin-proteasome pathway in the regulation of tumor metastasis.
ACKNOWLEDGMENTS
This study was supported by NIH grants CA69306, U54 RR020843-01, and CA82713, the
D.O.D. Breast Cancer Center of Excellence grants DAMD17-02-0693 and W81XWH-04-10515, and grant 10WB-0083 from the California Breast Cancer Research Program to JWS.
Additional support was derived from the Cancer Center grant to the Burnham Institute
(CA30199). EC was supported by NIH grants 5 R01 CA086258, the GCRC NIH grant (M01
Page 12 of 25
(page number not for citation purposes)
RR00833), and the NIH training grant T32 HL 07695. LF was supported by Office of Army
Research (ONR) grant N00014-01-2-003 and NIH/NIAID subcontract grant UCSD/R21AI53781
(formerly UTMB/R21AI53781). Additional support was by NIH grant CA95458 and GCRC
NIH grant (M01 RR00833) to BFH.
REFERENCES
1.
Zou, Z., Anisowicz, A., Hendrix, M. J., Thor, A., Neveu, M., Sheng, S., Rafidi, K., Seftor,
E., and Sager, R. (1994) Maspin, a serpin with tumor-suppressing activity in human
mammary epithelial cells. Science 263, 526–529
2.
Hojo, T., Akiyama, Y., Nagasaki, K., Maruyama, K., Kikuchi, K., Ikeda, T., Kitajima, M.,
and Yamaguchi, K. (2001) Association of maspin expression with the malignancy grade and
tumor vascularization in breast cancer tissues. Cancer Lett. 171, 103–110
3.
Maass, N., Teffner, M., Rosel, F., Pawaresch, R., Jonat, W., Nagasaki, K., and Rudolph, P.
(2001) Decline in the expression of the serine proteinase inhibitor maspin is associated with
tumour progression in ductal carcinomas of the breast. J. Pathol. 195, 321–326
4.
Yasumatsu, R., Nakashima, T., Hirakawa, N., Kumamoto, Y., Kuratomi, Y., Tomita, K., and
Komiyama, S. (2001) Maspin expression in stage I and II oral tongue squamous cell
carcinoma. Head Neck 23, 962–966
5.
Sheng, S., Carey, J., Seftor, E. A., Dias, L., Hendrix, M. J., and Sager, R. (1996) Maspin
acts at the cell membrane to inhibit invasion and motility of mammary and prostatic cancer
cells. Proc. Natl. Acad. Sci. USA 93, 11669–11674
6.
Zhang, M., Volpert, O., Shi, Y. H., and Bouck, N. (2000) Maspin is an angiogenesis
inhibitor. Nat. Med. 6, 196–199
7.
Gettins, P. G. (2002) Serpin structure, mechanism, and function. Chem. Rev. 102, 4751–
4804
8.
Sheng, S., Truong, B., Fredrickson, D., Wu, R., Pardee, A. B., and Sager, R. (1998) Tissuetype plasminogen activator is a target of the tumor suppressor gene maspin. Proc. Natl.
Acad. Sci. USA 95, 499–504
9.
McGowen, R., Biliran, H., Jr., Sager, R., and Sheng, S. (2000) The surface of prostate
carcinoma DU145 cells mediates the inhibition of urokinase-type plasminogen activator by
maspin. Cancer Res. 60, 4771–4778
10. Pemberton, P. A., Wong, D. T., Gibson, H. L., Kiefer, M. C., Fitzpatrick, P. A., Sager, R.,
and Barr, P. J. (1995) The tumor suppressor maspin does not undergo the stressed to relaxed
transition or inhibit trypsin-like serine proteases. Evidence that maspin is not a protease
inhibitory serpin. J. Biol. Chem. 270, 15832–15837
11. Bass, R., Fernandez, A. M., and Ellis, V. (2002) Maspin inhibits cell migration in the
absence of protease inhibitory activity. J. Biol. Chem. 277, 46845–46848
Page 13 of 25
(page number not for citation purposes)
12. Ngamkitidechakul, C., Warejcka, D. J., Burke, J. M., O'Brien, W. J., and Twining, S. S.
(2003) Sufficiency of the reactive site loop of maspin for induction of cell-matrix adhesion
and inhibition of cell invasion. Conversion of ovalbumin to a maspin-like molecule. J. Biol.
Chem. 278, 31796–31806
13. Mohsin, S. K., Zhang, M., Clark, G. M., and Craig Allred, D. (2003) Maspin expression in
invasive breast cancer: association with other prognostic factors. J. Pathol. 199, 432–435
14. Sood, A. K., Fletcher, M. S., Gruman, L. M., Coffin, J. E., Jabbari, S., Khalkhali-Ellis, Z.,
Arbour, N., Seftor, E. A., and Hendrix, M. J. (2002) The paradoxical expression of maspin
in ovarian carcinoma. Clin. Cancer Res. 8, 2924–2932
15. Washburn, M. P., Wolters, D., and Yates, J. R., III (2001) Large-scale analysis of the yeast
proteome by multidimensional protein identification technology. Nat. Biotechnol. 19, 242–
247
16. Price, J. E., Polyzos, A., Zhang, R. D., and Daniels, L. M. (1990) Tumorigenicity and
metastasis of human breast carcinoma cell lines in nude mice. Cancer Res. 50, 717–721
17. Wu, C. C., MacCoss, M. J., Howell, K. E., and Yates, J. R. (2003) A method for the
comprehensive proteomic analysis of membrane proteins. Nat. Biotechnol. 21, 532–538
18. Sadygov, R. G., and Yates, J. R., III (2003) A hypergeometric probability model for protein
identification and validation using tandem mass spectral data and protein sequence
databases. Anal. Chem. 75, 3792–3798
19. Eng, J. K., MaCormack, 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. Journal of
American Society of Mass Spectrometry 5, 976-989.
20. Tabb, D. L., McDonald, W. H., and Yates, J. R., III (2002) DTASelect and Contrast: tools
for assembling and comparing protein identifications from shotgun proteomics. J. Proteome
Res. 1, 21–26
21. Liu, H., Sadygov, R. G., and Yates, J. R., III (2004a) A model for random sampling and
estimation of relative protein abundance in shotgun proteomics. Anal. Chem. 76, 4193–4201
22. Florens, L., Washburn, M. P., Raine, J. D., Anthony, R. M., Grainger, M., Haynes, J. D.,
Moch, J. K., Muster, N., Sacci, J. B., Tabb, D. L., et al. (2002) A proteomic view of the
Plasmodium falciparum life cycle. Nature 419, 520–526
23. Kisselev, A. F., Akopian, T. N., Castillo, V., and Goldberg, A. L. (1999) Proteasome active
sites allosterically regulate each other, suggesting a cyclical bite-chew mechanism for
protein breakdown. Mol. Cell 4, 395–402
24. Ciechanover, A. (1998) The ubiquitin-proteasome pathway: on protein death and cell life.
EMBO J. 17, 7151–7160
Page 14 of 25
(page number not for citation purposes)
25. DeMartino, G. N., and Slaughter, C. A. (1999) The proteasome, a novel protease regulated
by multiple mechanisms. J. Biol. Chem. 274, 22123–22126
26. Orlowski, M. (1990) The multicatalytic proteinase complex, a major extralysosomal
proteolytic system. Biochemistry 29, 10289–10297
27. Rivett, A. J. (1989) The multicatalytic proteinase. Multiple proteolytic activities. J. Biol.
Chem. 264, 12215–12219
28. Heinemeyer, W., Fischer, M., Krimmer, T., Stachon, U., and Wolf, D. H. (1997) The active
sites of the eukaryotic 20 S proteasome and their involvement in subunit precursor
processing. J. Biol. Chem. 272, 25200–25209
29. Jager, S., Groll, M., Huber, R., Wolf, D. H., and Heinemeyer, W. (1999) Proteasome betatype subunits: unequal roles of propeptides in core particle maturation and a hierarchy of
active site function. J. Mol. Biol. 291, 997–1013
30. Meiners, S., Heyken, D., Weller, A., Ludwig, A., Stangl, K., Kloetzel, P. M., and Kruger, E.
(2003) Inhibition of proteasome activity induces concerted expression of proteasome genes
and de novo formation of Mammalian proteasomes. J. Biol. Chem. 278, 21517–21525
31. Parish, R. W., Schmidhauser, C., Schmidt, T., and Dudler, R. K. (1987) Mechanisms of
tumour cell metastasis. J. Cell Sci. Suppl. 8, 181–197
32. Chang, D. D., Park, N. H., Denny, C. T., Nelson, S. F., and Pe, M. (1998) Characterization
of transformation related genes in oral cancer cells. Oncogene 16, 1921–1930
33. Maul, R. S., and Chang, D. D. (1999) EPLIN, epithelial protein lost in neoplasm. Oncogene
18, 7838–7841
34. Maul, R. S., Song, Y., Amann, K. J., Gerbin, S. C., Pollard, T. D., and Chang, D. D. (2003)
EPLIN regulates actin dynamics by cross-linking and stabilizing filaments. J. Cell Biol. 160,
399–407
35. Sun, H. Q., Yamamoto, M., Mejillano, M., and Yin, H. L. (1999) Gelsolin, a multifunctional
actin regulatory protein. J. Biol. Chem. 274, 33179–33182
36. Witke, W., Sharpe, A. H., Hartwig, J. H., Azuma, T., Stossel, T. P., and Kwiatkowski, D. J.
(1995) Hemostatic, inflammatory, and fibroblast responses are blunted in mice lacking
gelsolin. Cell 81, 41–51
37. Cunningham, C. C., Stossel, T. P., and Kwiatkowski, D. J. (1991) Enhanced motility in NIH
3T3 fibroblasts that overexpress gelsolin. Science 251, 1233–1236
38. Ellis, R. E., Yuan, J. Y., and Horvitz, H. R. (1991) Mechanisms and functions of cell death.
Annu. Rev. Cell Biol. 7, 663–698
39. Hanahan, D., and Weinberg, R. A. (2000) The hallmarks of cancer. Cell 100, 57–70
Page 15 of 25
(page number not for citation purposes)
40. Steller, H. (1995) Mechanisms and genes of cellular suicide. Science 267, 1445–1449
41. Townson, J. L., Naumov, G. N., and Chambers, A. F. (2003) The role of apoptosis in tumor
progression and metastasis. Curr. Mol. Med. 3, 631–642
42. Jiang, N., Meng, Y., Zhang, S., Mensah-Osman, E., and Sheng, S. (2002) Maspin sensitizes
breast carcinoma cells to induced apoptosis. Oncogene 21, 4089–4098
43. Liu, J., Yin, S., Reddy, N., Spencer, C., and Sheng, S. (2004b) Bax mediates the apoptosissensitizing effect of maspin. Cancer Res. 64, 1703–1711
44. Ruvolo, P. P., Deng, X., Ito, T., Carr, B. K., and May, W. S. (1999) Ceramide induces Bcl2
dephosphorylation via a mechanism involving mitochondrial PP2A. J. Biol. Chem. 274,
20296–20300
45. Sahara, S., Aoto, M., Eguchi, Y., Imamoto, N., Yoneda, Y., and Tsujimoto, Y. (1999)
Acinus is a caspase-3-activated protein required for apoptotic chromatin condensation.
Nature 401, 168–173
46. Cardozo, C., Eleuteri, A. M., and Orlowski, M. (1995) Differences in catalytic activities and
subunit pattern of multicatalytic proteinase complexes (proteasomes) isolated from bovine
pituitary, lung, and liver. Changes in LMP7 and the component necessary for expression of
the chymotrypsin-like activity. J. Biol. Chem. 270, 22645–22651
47. Dahlmann, B., Ruppert, T., Kuehn, L., Merforth, S., and Kloetzel, P. M. (2000) Different
proteasome subtypes in a single tissue exhibit different enzymatic properties. J. Mol. Biol.
303, 643–653
48. Haass, C., and Kloetzel, P. M. (1989) The Drosophila proteasome undergoes changes in its
subunit pattern during development. Exp. Cell Res. 180, 243–252
49. Schweisguth, F. (1999) Dominant-negative mutation in the beta2 and beta6 proteasome
subunit genes affect alternative cell fate decisions in the Drosophila sense organ lineage.
Proc. Natl. Acad. Sci. USA 96, 11382–11386
50. Van den Eynde, B. J., and Morel, S. (2001) Differential processing of class-I-restricted
epitopes by the standard proteasome and the immunoproteasome. Curr. Opin. Immunol. 13,
147–153
51. Boes, B., Hengel, H., Ruppert, T., Multhaup, G., Koszinowski, U. H., and Kloetzel, P. M.
(1994) Interferon gamma stimulation modulates the proteolytic activity and cleavage site
preference of 20S mouse proteasomes. J. Exp. Med. 179, 901–909
52. Gaczynska, M., Goldberg, A. L., Tanaka, K., Hendil, K. B., and Rock, K. L. (1996)
Proteasome subunits X and Y alter peptidase activities in opposite ways to the interferongamma-induced subunits LMP2 and LMP7. J. Biol. Chem. 271, 17275–17280
Page 16 of 25
(page number not for citation purposes)
53. Gaczynska, M., Rock, K. L., Spies, T., and Goldberg, A. L. (1994) Peptidase activities of
proteasomes are differentially regulated by the major histocompatibility complex-encoded
genes for LMP2 and LMP7. Proc. Natl. Acad. Sci. USA 91, 9213–9217
54. Kuckelkorn, U., Frentzel, S., Kraft, R., Kostka, S., Groettrup, M., and Kloetzel, P. M. (1995)
Incorporation of major histocompatibility complex–encoded subunits LMP2 and LMP7
changes the quality of the 20S proteasome polypeptide processing products independent of
interferon-gamma. Eur. J. Immunol. 25, 2605–2611
55. DeMartino, G. N., and Slaughter, C. A. (1993) Regulatory proteins of the proteasome.
Enzyme Protein 47, 314–324
56. Bennett, R. G., Hamel, F. G., and Duckworth, W. C. (2000) Insulin inhibits the ubiquitindependent degrading activity of the 26S proteasome. Endocrinology 141, 2508–2517
Received September 8, 2004; accepted March 7, 2005.
Page 17 of 25
(page number not for citation purposes)
Table 1
Correlation of gene expression and the proteomic data by quantitative PCRa
>1.5-fold up-regulated protein
Real-time PCR
Protein ID
Gene ID
Gene Name
Maspin
M27/435 M59/435 M58/435 M60/435
1298
2992
4489
3914
gi4885049
NM_005159
α-Actin
1.8
2.2
2.4
2.1
gi4505773
NM_002634
Prohibitin
1.0
1.2
0.9
1.6
gi4507877
NM_003373
Vinculin
0.7
1.2
1.7
1.2
gi15149461
NM_033139
Caldesmon 1
1.0
1.0
1.0
2.4
gi4506345
NM_002859
Paxillin
1.6
2.0
1.7
2.0
gi7705373
NM_016357
EPLIN
1.1
1.6
2.6
1.7
gi4505877
NM_000445
Plectin 1
1.0
0.8
0.5
0.8
gi4757952
NM_001791
Cdc 42
0.9
1.4
0.8
1.6
gi27597085
NM_000366
Tropomyosin 1
2.5
1.3
4.1
1.9
gi4759302
NM_004738
Vesicle-associated membrane protein-associated protein B
0.8
1.8
1.2
1.9
gi5031877
NM_005573
Lamin B1
1.3
2.9
2.1
1.3
gi28460688
NM_175852
DKFZp451J0118
1.8
3.1
1.3
2.0
gi20357556
NM_138565
Cortactin
1.2
0.9
1.9
1.7
gi4502747
NM_001261
Cyclin-dependent kinase 9
1.0
1.8
1.2
2.9
gi21618349
NM_145109
Mitogen-activated protein kinase kinase 3
1.1
1.0
2.1
1.2
gi5713317
NM_004127
G protein pathway suppressor 1
1.2
1.5
2.4
1.5
gi8922652
NM_018209
ADP-ribosylation factor GTPase activating protein 1
0.8
0.5
1.1
0.9
gi19705428
NM_020350
Angiotensin II receptor-associated protein
0.8
0.8
0.9
1.1
gi5174431
NM_006013
Ribosomal protein L10
1.1
1.0
1.5
1.1
gi4504549
NM_002160
Tenascin C
1.6
3.1
4.0
2.2
>2 fold down regulated protein
Realtime PCR
Protein ID
Gene ID
Gene Name
435/M27 435/M59 435/M58 435/M60
gi4826898
NM_005022
Profilin 1
3.3
2.5
2.5
2.5
gi4758184
NM_004412
DNA (cytosine-5-)-methyltransferase 2
1.4
0.6
1.8
0.8
gi3801694
NM_000177
Gelsolin
1.5
5.1
5.3
5.1
gi19743813
NM_002211
Integrin, β1
1.3
1.2
1.2
1.3
gi4557761
NM_000251
MutS homolog 2
2.9
1.5
0.4
1.6
gi14195630
NM_030983
Microtubule-associated protein 4
1.2
1.4
1.3
1.3
gi5453740
NM_006471
Myosin regulatory light chain MRCL3
1.3
1.7
2.3
1.1
gi4503743
NM_002018
Flightless I homology
2.7
3.5
2.3
3.3
gi4758670
NM_004811
Leupaxin
1.5
1.1
1.6
1.1
gi5454102
NM_006342
Transforming, acidic coiled-coiled containg protein 3
0.7
0.5
0.7
2.0
gi18375505
NM_080649
APEX nuclease
1.3
1.4
1.4
1.4
gi7656985
NM_001845
Alpha 1 type IV collagen
0.7
0.8
0.4
1.3
Page 18 of 25
(page number not for citation purposes)
gi4757804
NM_004045
ATX1 antioxidant protein 1 homolog
1.7
0.7
0.9
1.1
gi5730095
NM_001561
Tumor necrosis factor receptor superfamily, member 9
1.7
1.6
1.3
2.1
gi4826774
NM_005101
5.6
2.0
3.2
3.4
gi4759274
NM_004786
Thioredoxin-like, 32kDa
1.7
0.6
1.8
0.7
gi15809016
NM_033546
Myosin regulatory light chain MRLC2
0.7
0.6
1.1
0.6
gi19923356
NM_006449
Cdc42 effector protein 3
1.2
0.9
0.9
1.3
gi5453710
NM_006148
Lasp-1
1.9
1.2
1.6
1.9
gi4758988
NM_004161
RAB1A, member RAS oncogene
1.2
0.5
0.8
1.3
gi14165278
NM_032412
Putative nuclear protein ORF1-FL49
1.2
1.5
1.3
2.0
gi11024714
NM_018955
Ubiquitin B precursor
7.5
3.2
2.3
1.2
gi7657046
NM_014501
Ubiqutin carrier protein, E2-EPF
3.3
2.1
1.4
3.1
gi4507791
NM_003969
Ubiquitin-conjugating enzyme E2M
2.9
0.3
1.0
0.5
gi4506203
NM_002799
Proteasome beta 7 subunit
2.0
1.7
3.1
1.6
gi4506195
NM_002794
Proteasome beta 2 subunit
1.7
1.2
1.0
0.5
gi14150139
NM_032345
PYM protein
2.5
2.0
1.4
2.5
gi7657176
NM_014255
Transmembrane protein 4
0.7
0.7
1.4
0.6
gi4507781
NM_003343
Ubiquitin-conjugating enzyme E2G2
2.5
1.1
2.7
1.1
gi5031875
NM_005572
Lamin A/C isoform 2
1.7
1.5
1.0
0.8
gi5803123
NM_006814
Proteasome inhibitor subunit 1 (PI31)
1.7
0.8
0.8
0.8
gi20070255
NM_014570
ADP-ribosylation factor GTPase activating protein 3
1.7
1.4
1.2
2.4
gi4557675
NM_000210
Integrin, α6
0.8
1.4
1.3
2.3
gi38146097
NM_000582
Osteopontin
1.7
1.0
1.8
2.7
gi4885510
NM_005381
Nucleolin
1.5
2.0
1.1
1.6
Interferon, alpha-inducible protein
Protein with equal abundance
Protein ID
Gene ID
Gene Name
gi4507895
NM_003380
Vinmentin
435/M27 435/M59 435/M58 435/M60
1.0
1.0
1.0
a
1.0
Gene expression of 55 proteins affected by maspin expression as indicated by the proteomic data were analyzed by quantitative PCR. One protein
(Vinmentin) with equal abundance indicated by the proteomic comparison was used as control. Matched gene expression and proteomic data in at
least three out of four maspin-transfected cells were counted as confirmation. Differential gene expression of 17 proteins (italic) out of 55 total
proteins was correlated with the predicted trend as up- or down-regulated. Proteins analyzed by both gene and protein expression analysis are
highlighted in bold.
Page 19 of 25
(page number not for citation purposes)
Fig. 1
Figure 1. Functional validation of maspin expression in maspin-transfected MDA-MB 435 human breast cancer cells. A)
Overexpression of maspin was detected in all four maspin cDNA-transfected MDA-MB 435 cells by a polyclonal maspin
antibody. B) Reduced matrigel invasion was found in all four maspin-transfected clones. C) All four maspin-transfected
clones generated smaller primary tumors than the parent and mock-transfected MDA-MB 435 tumors (n=10). Two of the
clones, M27 and M60, showed 50% reduction in primary tumor size. D) All four maspin-transfected clones generated
significantly less lung metastases than the parent and mock-transfected MDA-MB 435 cells (n=10). See Materials and
Methods for details.
Page 20 of 25
(page number not for citation purposes)
Fig. 2
Figure 2. Correlation of protein levels to the proteomic data by Western blot. Western blot analysis of 22 proteins affected
by maspin expression in the breast cancer cells as indicated by the proteomic data was carried out in whole cell extracts of
parent and all four maspin-transfected MDA-MB 435 cell lines. Candidate proteins were counted as positives when the
predicted differential protein expression was confirmed in at least three out of four maspin-transfected cells. All 10 proteins
chosen from the 1.5-fold up-regulated list correlated well with the proteomic data. A) Western blot analysis of nine protein
candidates is shown. One candidate protein, nm23A, was confirmed on 2D gel electrophoresis (data not shown). Twelve
proteins were chosen from the 1.5-fold down-regulated list. Ten proteins out of total 12 from this list were confirmed
down-regulated in the maspin-transfected cells, and the Western blot is shown in B.
Page 21 of 25
(page number not for citation purposes)
Fig. 3
Figure 3. Significant alterations of actin and actin-associated proteins were found in maspin-transfected MDA-MB 435
cells. A) A list of actin and actin-associated proteins found in the proteomic data was presented in an expression boxplot.
B) Parent, mock-transfected, and two maspin-transfected MDA-MB 435 cells were grown in glass slide chambers to 60–
70% confluence, fixed, and stained for actin filament. Actin filaments were made visible by staining with Alexa Fluor 594
phalloidin (Molecular Probes, Eugene, OR). Cell images from at least three fields were acquired, and one representative
picture was used from each staining. Only one image representing 435 and 435-neo cell staining was shown in the result
section because no significant difference was found in all images acquired from these two cell lines.
Page 22 of 25
(page number not for citation purposes)
Fig. 4
Figure 4. Maspin-transfected MDA-MB 435 cells showed increased spontaneous apoptosis without significant difference
in cell proliferation. A) A list of apoptosis-associated proteins found in the proteomic data was presented in an expression
boxplot. B) An increase in spontaneous apoptosis was observed in all four maspin-transfected cells compared with the
parent and mock-transfected cells. An increase of apoptosis between maspin-transfected cells and control cells was
statistically significant (P<0.01); however, differences in cell proliferation was not (P>0.01).
Page 23 of 25
(page number not for citation purposes)
Fig. 5
Figure 5. Maspin-transfected MDA-MB 435 cells showed lower chymotrypsin-like 20S proteasome activity, reduced
expression of β5 proteasome subunit, and increased high molecular weight ubiquitin-protein conjugates. A) A list of
proteins associated with the ubiquitin-proteasome pathway found in the proteomic data was presented in an expression
boxplot. B) Significant reduction in chymotryptic activity of the 20S proteasome was observed in maspin-transfected cell
lysates (see Materials and Methods for details). A proteasome inhibitor targeting the chymotrypsin-like site, MG132, was
used to show the specificity of the chymotrypsin-like proteolytic activity in this assay. C) Reduced protein level of β5
proteasome subunit corresponding to the chymotrypsin-like proteolytic activity was found in all maspin transfectants by
Western blot. D) As the result of reduced chymotrypsin-like 20S proteasome activity, the accumulation of high molecular
weight ubiquitin-protein conjugates was observed in the cell lysate of maspin-transfected comparing to the control MDAMB 435 cells. The high molecular weight ubiquitin-protein conjugates were detected by an anti-ubiquitin antibody.
Page 24 of 25
(page number not for citation purposes)
Fig. 6
Figure 6. Western blot analysis of proteasome α and β subunits revealed dysregulation of several 20S proteasome
subunits. Protein levels of 20S proteasome subunits in maspin-transfected and control MDA-MB 435 cells were analyzed
by Western blot. Our results revealed down-regulation of α6 and α7 subunits.
Page 25 of 25
(page number not for citation purposes)