Identification of Differentially Expressed Proteins in Triple-Negative Breast Carcinomas Using DIGE and Mass Spectrometry Daniela M. Schulz,† Claudia Bo ¨ llner,† Gerry Thomas,‡ Mike Atkinson,§,| Irene Esposito,†,⊥ Heinz Ho ¨ fler,†,⊥ and Michaela Aubele*,†,⊥ Institute of Pathology, Helmholtz Zentrum München, German Research Center for Environmental Health, Ingolsta¨dter Landstrasse 1, 85764 Neuherberg, Germany, Molecular Pathology, Imperial College London, Hammersmith Hospital, Du Cane Road, London W12 0NN, United Kingdom, Institute of Radiation Biology, Helmholtz Zentrum München, German Research Center for Environmental Health, Ingolsta¨dter Landstrasse 1, 85764 Neuherberg, Germany, Klinik und Poliklinik fu ¨ r Strahlentherapie, Klinikum rechts der Isar, Technische Universita¨t Mu ¨ nchen, Ismaninger Strasse 22, 81675 Mu ¨ nchen, Germany, and Institut fu ¨ r Pathologie, Technische Universita¨t Mu ¨ nchen, Trogerstrasse 18, 81675 Mu ¨ nchen, Germany Received January 29, 2009
We compared the protein expression pattern of triple-negative breast carcinomas (HER2-, ER-, PR-) versus those being positive for HER2 and negative for the hormone receptors (HER2+, ER-, PR-) by 2-D DIGE and mass spectrometry. We obtained differential expression patterns for several glycolytic enzymes (as for example MDH2, PGK1, TKT, Aldolase1), cytokeratins (CK7, 8, 9, 14, 17, 19), further structure proteins (vimentin, fibronectin, L-plastin), for NME1-NME2, lactoferrin, and members of the Annexin family. Western blot analysis and immunohistochemistry were conducted to verify the results. The identified marker proteins may advance a more detailed characterization of triple-negative breast cancers and may contribute to the development of better treatment strategies. Keywords: proteomics • breast cancer • triple-negative • 2D-DIGE • MALDI-TOF/TOFMS
Introduction Breast cancer is the most common cancer diagnosed among women in the western world and is the leading cause of female cancer death.1 Determination of hormone receptor status (estrogen (ER) and progesterone receptor (PR)) has become standard practice in the management of invasive breast cancers and is useful as a prognostic and predictive factor.2 ER positivity predicts response to endocrine therapy such as antiestrogen administration or ovarian suppression. Similarly, human epithelial growth factor receptor 2 (HER2, c-erbB-2) positivity is useful for selecting targeted therapy with the monoclonal antibody against HER2.2 Triple-negative breast cancers are defined by a lack of expression of estrogen, progesterone, and HER2/neu receptors, and accounts for about 15% of all breast cancers. This subtype is associated with aggressive phenotype, poor prognosis, and unresponsiveness to usual endocrine therapies.2–6 Because of the absence of specific treatment guidelines for this group of patients, triple-negative breast cancers are managed with * Corresponding author: Dr. M. Aubele, Helmholtz Zentrum Mu ¨ nchen, Institut fu ¨ r Pathologie, Ingolsta¨dter Landstrasse 1, 85764 Neuherberg, Germany. E-mail:
[email protected]. † Institute of Pathology, German Research Center for Environmental Health. ‡ Imperial College London. § Institute of Radiation Biology, German Research Center for Environmental Health. | Klinik und Poliklinik fu ¨ r Strahlentherapie, Klinikum rechts der Isar, Technische Universita¨t Mu ¨ nchen. ⊥ Institut fu ¨ r Pathologie, Technische Universita¨t Mu ¨ nchen.
3430 Journal of Proteome Research 2009, 8, 3430–3438 Published on Web 06/02/2009
standard adjuvant chemotherapy, which, however, seems to be less effective in those cancers7 and leaves them associated with a high rate of local and systemic relapse.6 In population-based studies using survival analysis, it was shown that triple-negative tumors are diagnosed at later stage and are more aggressive,2 and that patients classified as triplenegative have a poor prognosis.3 Further, it could be shown in multivariate survival analysis that assessment of the androgen receptor status, in addition to the established prognostic parameters lymph node status and tumor size, can be useful to select high-risk and low-risk triple-negative patients at the time of primary surgery, thus, providing valuable information on treatment options in these tumors.8 Until now, however, there is no specific treatment for patients with triple-negative breast cancer, and there is a need to develop new therapeutic approaches. The pathways that drive proliferation of these tumors are still poorly understood.6 Potential targets for treatment could concern, for example, surface receptors like EGF-R, protein kinase components of the mitogen activated protein (MAP)-kinase pathway and of the protein kinase B (Akt) pathway.6 Microarray profiling of invasive breast carcinomas has identified five major subtypes of breast cancers according to their RNA expression pattern,4,9 which showed different prognosis:3,5,9 basal-like type (ER-, PR-, HER2-, the majority of triple-negative (TN) breast cancers belong to this subgroup), normal breast-like subtype, luminal A and B (ER+/PR+), and a HER2 overexpressing subtype.10 It has been demonstrated that this classification scheme has prognostic significance and 10.1021/pr900071h CCC: $40.75
2009 American Chemical Society
research articles
Proteomics in Triple-Negative Breast Cancers 10
implications with respect to response to therapy. Those analyses also suggested that triple-negative breast cancers formsat least in transcriptional termssa quite homogeneous group.4,6 Proteomic strategies have been used to identify cancer specific protein markers. From a number of reports, it is apparent that 2-DE (two-dimensional electrophoresis) in combination with MS (mass spectrometry) is a basic technology to separate and identify proteins.11 However, conventional 2-DE has a drawback in that the number of protein spots and the abundance of proteins vary between gels, even between identical samples. The 2-D DIGE (two-dimensional difference gel electrophoresis) has been developed to simplify the analysis and to compare protein expression levels quantitatively.12 In this comparative proteomic study, we aimed toward characterizing triple-negative breast cancers using 2D-DIGE and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS). In contrast to HER2positive or hormone receptor positive breast cancers, for which targeted therapies do exist, there is no targeted therapy option for triple-negative breast cancer, which underlines the necessity for further characterization of the molecular basis underlying TN mamma carcinomas. By comparing the protein expression patterns of triple-negative cancers versus those of HER2 positive/hormone receptor negative breast cancer subtypes, we identified differentially expressed proteins in triple-negative tumors. These may represent new therapy targets and may help to develop novel treatment strategies for this subtype of breast cancer.
Material and Methods Clinical Tumor Samples. We examined 34 frozen breast tumor tissues (15 triple-negative (Table 1, cases 1-15) and 19 HER-2/neu-positive/hormone receptor-negative tumors (cases 16-34)) obtained from the Technische Universita¨t Mu ¨ nchen, and the Wales Cancer Bank (authorized by the National Research Ethics Service of the U.K., www.walescancerbank. com). All of them were invasive ductal carcinomas. The patients’ age ranged from 35 to 96 years (mean 60.1 ( SD 15.2), and none of the patients received any treatment prior to surgery. Immunohistochemistry. A hematoxylin- and eosin-stained section from each tissue sample was used to verify histology. Subsequent sections from the frozen tumor tissues were used for immunohistochemistry (IHC) to confirm the receptor status (estrogen, progesterone, HER2/neu). Further subsequent 10 µm-sections were used to confirm the expression levels of cytokeratins 7 and 14 described in the Results. The following primary antibodies were used: HER2 (c-erbB-2 oncoprotein, A0485; DAKO GmbH, Hamburg, Germany; dilution 1:700); estrogen receptor (EI629C01), progesterone receptor (PI633C01, Innovative Diagnostik-Systeme, Hamburg, Germany), cytokeratin 7 (sc-53264, 1:100), and cytokeratin 14 (sc-53253, undiluted, Santa Cruz Biotechnologies, Heidelberg, Germany). Pretreatment of sections and incubation of the primary antibody were in general performed as described previously,13 with staining and counterstaining by an automated immunostainer (Ventana Medical System, Tucson, AZ).13 The staining levels were scored by a research scientist (M.A.) and two pathologists (H.H. and I.E.) independently. Only tumors negative for the hormone receptors, and negative for the HER2 receptor (triplenegative group) or positive for the HER2 receptor (HER2 positive group), respectively, were involved in the study.
Table 1. Characterization of Breast Cancer Casesa no. of age at tumor size positive case no. diagnosis histology [mm] lymph nodes
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
37 36 56 75 36 35 96 66 43 64 68 49 69 66 65 46 52 74 70 73 71 78 70 45 68 70 46 52 75 75 69 61 36 50
IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC IDC
10 35 13 20 28 10 60 35 22 22 33 22 13 14 22 12 23 15 15 25 20 30 25 21 30 60 30 27 28 23 20 15 30 22
2 2 3 0 0 0 4 0 0 1 6 2 0 0 0 3 0 0 0 0 0 0 0 0 0 1 0 33 0 21 0 1 10 1
HER2
ER/PR
negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative positive positive positive positive positive positive positive positive positive positive positive positive positive positive positive positive positive positive positive
negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative negative
a Cases 1-15 are triple-negative (TN), 16-34 are HER2 positive tumors; IDC ) invasive ductal caricnoma; ER ) estrogen receptor; PR ) progesterone receptor.
Sample Preparation. Between 12 and 58 mg of tumor tissues was available for protein extraction. Fifty micrometer thick sections were cut and the tumor tissue was microdissected to achieve samples containing >70% tumor cells. The microdissected tissue was mechanically disrupted with a mini-pestle and total soluble proteins were extracted in 400-600 µL of buffer (depending on amount of tissue) containing 7 M urea, 2 M thiourea, 15 mM Tris (pH 8.5), and 4% CHAPS (3-(3Cholamidopropyl)dimethylammonio-1-propanesulfonate). The lysate was centrifuged at 15 000 rpm for 20 min to obtain the supernatant. Protein concentrations were measured with the Bradford assay14 and the extracted proteins were stored at -80 °C until analysis. Protein concentrations ranged between 2.6 and 19.6 µg/µL extracted from 12 to 58 mg of tissue. Equal amounts of proteins extracted from frozen tumor tissue of 19 HER2 positive/hormone receptor negative tumors were pooled as the HER2 positive group, and likewise, equal amounts of proteins extracted from 15 triple-negative tumors were pooled as the triple-negative group. These pooled samples were used for the comparative proteomic analyses described below. 2D-DIGE and Data Analysis. Pooled protein extracts were labeled with Cy Dyes (GE Healthcare, Freiburg, Germany): Cy3 for the TN group, Cy5 for the HER2 positive group, and Cy2 for the internal standard. Fifty micrograms of proteins was labeled with 300 pmol of the respective dye. The strips were rehydrated with 7 M urea, 2 M thiourea, 2% CHAPS, 0.5% Journal of Proteome Research • Vol. 8, No. 7, 2009 3431
research articles
Schulz et al.
ampholytes, and DTT (dithiothreitol) for 14 h. Destreak reagent (GE Healthcare, Freiburg, Germany) was used for the 7-10 IPG strips. Combined samples (150 µg per strip) were separated in the first dimension (isoelectric focusing, IEF) on immobilized pH gradient (IPG) strips with ranges between pH 4-7 and pH 7-10 (18 cm, BioRad, Munich, Germany). The sample was applied by anodic cup-loading and IEF was run in three steps: 300 V, 1000 V, and 10 000 V (gradient in between the steps) with a total focusing time of 54 kV · h on an Ettan IPGphor Isoelectric Focusing System (GE Healthcare, Freiburg, Germany). After IEF, the IPG strips were first equilibrated for 15 min with the equilibration buffer (6 M urea, 30% glycerol, 2% SDS, and 50 mM Tris, pH 8.8) containing 100 mg DTT, and then equilibrated for 15 min with equilibration buffer containing 250 mg iodoacetamide. Strips were then transferred onto vertical 12.5% SDS-PAGE gels and sealed with 0.5% low melting point agarose. The second-dimension molecular weight separation was carried out using a PROTEAN II xi Cell system (BioRad, Hercules, CA). For each pH range, four technical replicates were prepared. Visualization of protein spots was achieved with the Typhoon 9400 fluorescence scanner (GE Healthcare, Freiburg, Germany) scanning at the Cy Dyes’ respective wavelengths. Spot detection and matching, and quantification of spot intensity were performed using the DeCyder software (GE Healthcare, Freiburg, Germany). To correct for differences in gel staining, spot volumes relative to the sum of the volume of all spots on each gel (%Vol) were calculated by the software. The differences in expression between the TN and the HER2 positive samples were analyzed by the Student’s t test; p-values e0.05 were considered significant, p e 0.01 as highly significant, and p e 0.001 as very highly significant. Only those spots showing at least a 1.3-fold change in spot intensity and those spots which produced consistent results in the quadruplet gels were selected for protein identification by MS. For spot picking for subsequent MS analysis, the gels were silver-stained according to standard protocols.15 Protein Identification by MALDI-TOFMS and MS/MS. Identification of proteins from gel spots was performed by MALDITOF/TOF tandem mass spectrometry following trypsin digestion. Briefly, the excised gel spots were destained with 30 mM potassium ferricyanide and 100 mM sodium thiosulphate (1: 1), dried in a SpeedVac, reswelled with 10-20 µL of digestion solution containing 20 mM ammonium bicarbonate and 20 ng/ µL sequencing grade trypsin (Promega, Madison, WI) and incubated at 37 °C overnight. The resulting peptides were extracted and desalted with ZipTip C18 columns (Millipore Corporation, Bedford, MA). The eluate was spotted onto the MALDI target using the dried droplet method with R-cyano4-hydroxy cinnamic acid matrix (Bruker Daltonik, Bremen, Germany), prepared as a 2.5 mg/mL solution in 70% acetonitrile/0.1% trifluoroacetic acid. Duplicate samples were applied to the target plate. Peptides were analyzed using the 4700 Proteomics Analyzer MALDI-TOF/TOF mass spectrometer (Applied Biosystems, Framingham, MA) in the positive ion reflector mode. The subsequent MS/MS analysis was performed in a data-dependent manner and the eight most abundant ions were subjected to CID (collision-induced dissociation) analysis. For the database search, known contamination peaks such as those of keratin and autoproteolytic products were excluded for peptide mass fingerprint database search with the Mascot server (www.matrixscience.com) in the NCBInr database (human 3432
Journal of Proteome Research • Vol. 8, No. 7, 2009
Figure 1. H&E-stained section (A) and immunohistochemistry (B-D), performed for case number 33 (see also Table 1) with strong positivity (3+) for HER2 (B) and negative staining for ER (C) and PR (D).
taxonomy). One missed tryptic cleavage was considered and a mass accuracy of 100 ppm was used for the searches and within 0.1 u for CID experiments. Searches were performed without constraining protein molecular weight or pI, and allowed for carbamidomethylation of cysteine (fixed modification) and partial oxidation of methionine residues. MOWSE scores (Molecular Weight Search, a measure of probability) of greater than 75 were considered as significant (p e 0.05). Western Blot. Pooled protein of both the TN and the Her2positive group and individual samples were analyzed by Western blot. Twenty micrograms of protein per sample was separated by 4% stacking/10% resolving SDS-PAGE and resolved proteins were transferred onto nitrocellulose membranes (0.2 µm) by semidry Western blotting at 25 V for 40 min. Antibodies (CK7, sc-53264; CK19, sc-6278; PGK1, sc-130335; and GAPDH, sc-25778) were obtained from Santa Cruz Biotechnologies (Heidelberg, Germany). Secondary antibodies were purchased from GE Healthcare (Freiburg, Germany).
Results Immunohistochemistry of Receptor Status. Only tumors diagnosed as invasive ductal carcinoma, with a percentage of tumor cells being more than 70% of the total number of cells present in the sample, were included in this study. The HER2, estrogen receptor, and progesterone receptor status of the tissues, as determined during diagnostic assessment, was confirmed by immunohistochemistry. The key clinicopathological features of the cases used in this study are listed in Table 1 and representative examples of immunohistochemistry are given in Figure 1. Identification of Differentially Expressed Proteins. Pooled triple-negative (TN) breast tumors (15 cases) were compared to a pool of HER2 positive breast cancer samples (19 cases). DIGE analysis was performed for assessing quantitative data on protein expression levels between these two breast cancer subtypes. For better protein separation, the samples were applied to two complementary pH ranges (pH 4-7 and 7-10 on 18 cm IPG strips) and gels of each pH range were run in quadruplet. Protein spots exhibiting at least a 1.3-fold change
research articles
Proteomics in Triple-Negative Breast Cancers a
Table 2. Down-Regulated Proteins in the Triple-Negative Group spot
identified protein
ratio
t testb
accession no. in NCBI
MW [kDa]c
pIc
MOWSE score
assigned peptides
sequ. cov. [%]
A1
Cytokeratin 7 (CK7, Sarcolectin) Cytokeratin 8 (CK8) Cytokeratin 9 (CK9) Cytokeratin 19 (CK19) Beta-actin Fructose 1,6-Bisphosphate Aldolase Aldolase A Transketolase (TKT) Sorbitol dehydrogenase Phosphoglycerate kinase 1 (PGK1) GAPDH Glyceraldehyde-3phosphate dehydrogenase mitochondrial malate dehydrogenase 2 (MDH2) Annexin A2 Annexin A2 NME1-NME2 protein Peroxiredoxin 1 Electron-transferflavoprotein, beta polypeptide isoform 1 Electron-transferflavoprotein, beta polypeptide isoform 2 Chain B, Deoxy Hemoglobin Chain A, Deoxy Rhb1.2 (Recombinant Hemoglobin) delta globin hemoglobin beta subunit variant, beta globion
1.84
***
gi|20178293
51443
5.5
313
25
42
1.38 1.66 1.38 1.43 1.64
*** *** *** *** *
gi|62913980 gi|119581148 gi|90111766 gi|194375299 gi|4557976
41082 57754 44065 37666 39720
4.94 6.95 5.04 5.49 8.39
228 104 249 224 160
19 13 27 15 10
49 33 65 35 37
1.51 1.57 1.64 1.40
*** *** ** ***
gi|4557305 gi|388891 gi|1583520 gi|4505763
39851 68527 38900 44985
8.3 7.89 7.56 8.3
205 142 172 151
15 11 11 14
49 24 49 37
1.38 1.47
*** **
gi|67464043 gi|89573929
36482 24775
8.58 8.68
110 78
10 8
34 40
1.44
***
gi|89574129
32404
8.1
466
20
66
1.30 1.35 1.49 1.99 1.32
** * *** *** **
gi|18645167 gi|194388544 gi|66392203 gi|55959888 gi|4503609
38779 21827 30345 10726 28054
7.57 5.98 9.06 8.79 8.24
354 78 302 103 96
20 6 11 3 7
53 37 58 32 34
1.49
***
gi|62420877
37753
6.78
177
14
41
1.78
**
gi|27574248
16017
6.75
128
6
48
1.57
***
gi|9256890
30475
8.91
121
3
28
1.40 1.55
** **
gi|4504351 gi|6003534
16159 16086
7.85 6.75
254 218
10 15
76 75
A2 A3 A4 A5 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 B13
B14
B15 B16 B17 B18
a Gel spots A and B refer to gels with pH range 4-7 (Figure 2) and 7-10 (Figure 3), respectively. b p-values e0.05 were considered significant (*), p e 0.01 as highly significant (**), and p e 0.001 as very highly significant (***). c The listed molecular weights and pI values correspond to the protein with the highest MOWSE score obtained from duplicate measurements. This explains why in few cases they do not exactly match the positions in the gels, because protein with the highest score may not necessarily reflect that of full-length protein or that of isoforms/splice variants.
and passing statistical criteria (t test e0.05) were selected for analysis by MALDI-TOFMS and tandem mass spectrometric analysis (TOF/TOF). For MS analysis, the protein spots of the four technical replicates were pooled prior to trypsin digestion to enhance signal intensity and therefore protein identification. Furthermore, each sample was spotted onto the MALDI target in duplicate and only the higher MOWSE scores for protein identifications are given in Tables 2 and 3. The molecular weights and pI’s of the identified proteins listed in the Tables match the proteins’ respective positions in the gels (Figures 2 and 3). All identified proteins are listed in Tables 2 (downregulated in TN) and 3 (up-regulated in TN). Identified Proteins. 1. Cytokeratins. Not surprisingly, since cytokeratins (CK) are known to be involved in many important cellular functions, for example, cell motility, cell signaling and cell division, we identified a number of differentially expressed CKs (Tables 2 and 3). While CKs 7, 8, 9, and 19 are down-regulated in the TN group in comparison to the HER2 positive group, we found CKs 14 and 17 to be more highly expressed in the TN pool. CK7 (also known as sarcolectin) was identified from spot A1 (Figure 2) (1.84-fold down-regulation). CK8 was identified from protein spot A2 (Figure 2, 1.38-fold down-regulation). Down-regulation in the TN group was also found for CK9 (Figure 2, spot A3 (1.66-fold)) and CK19 (Figure 2, spot A4 (1.38-fold)). Among the identified CKs, elevated
expression levels of CK8 and 19 are known to be indicative of HER2 positivity, explaining the relative down-regulation in the TN versus the HER2 positive group. On the other hand, two CKs that are known to be highly expressed in basal-like breast cancers, to which the majority of TN breast cancers belong, were found to be up-regulated in the TN group. CK17 was identified in spot A7 (Figure 2 (2.00-fold regulation)), while CK14 was identified in spot A6 (Figure 2 (2.05-fold regulation)). CK14 is used as one of the marker proteins that are used to define the basal-like phenotype of breast cancer. As such, the identification of the above-mentioned cytokeratins that are known to be differentially expressed between HER2 positive and TN phenotypes proves the feasibility of our experimental setup. 2. Other Structure Proteins. Besides cytokeratins, we identified a number of other differentially expressed proteins (Tables 2 and 3) that are involved in the structural organization of the cell, as for example vimentin, fibronectin, and L-plastin. Vimentin is up-regulated in the TN group (Figure 2, spot A12, 1.80-fold regulation). We identified fibronectin from two protein spots (Figure 2, spots A8 and A9, 1.32- and 1.35-fold upregulation, respectively), which has been shown to be involved in carcinoma development and to stimulate phosphatidylinositol 3-kinase (PI3K).16 Journal of Proteome Research • Vol. 8, No. 7, 2009 3433
research articles
Schulz et al. a
Table 3. Up-Regulated Proteins in the Triple-Negative Group spot
A6 A7 A8 A9 A10 A11 A12 B19 B20 B21 B22
B23 B24
MW [kDa]c
pIc
MOWSE score
assigned peptides
sequ. cov. [%]
gi|12803709 gi|4557701 gi|47132549
51905 48361 243078
5.09 4.97 5.6
101 117 79
16 18 21
30 33 14
***
gi|119590943
259928
5.53
191
39
22
1.43
***
gi|114651523
70815
5.2
191
28
37
1.63
***
gi|87196339
109602
5.26
100
22
18
1.80
***
gi|167887751
49680
5.19
171
21
53
1.34 2.96 2.69 1.41
** *** *** *
gi|4502101 gi|193787568 gi|157831799 gi|160286045
38918 78585 77900 23307
6.57 8.39 8.4 8.61
120 132 151 76
13 21 21 7
39 28 29 40
1.39
***
gi|10334619
40579
8.3
71
7
15
1.37
***
gi|149673887
23664
6.97
205
14
49
identified protein
ratio
t testb
keratin 14 keratin 17 fibronectin 1 isoform 6 preproprotein fibronectin 1 isoform 4 preproprotein L-plastin variant collagen, type VI, alpha 1 precursor Vimentin variant 3 Annexin A1 Apolactoferrin Apolactoferrin Chain L, Crystal Structure Of A Recombinant Ige Fab immunoglobulin heavy chain immunoglobulin light chain
2.05 2.00 1.31
*** *** **
1.35
accession no. in NCBI
a Gel spots A and B refer to gels with pH range 4-7 (Figure 2) and 7-10 (Figure 3), respectively. b p-values e0.05 were considered significant (*), p e 0.01 as highly significant (**), and p e 0.001 as very highly significant (***). c The listed molecular weights and pI values correspond to the protein with the highest MOWSE score obtained from duplicate measurements. This explains why in few cases they do not exactly match the positions in the gels, because protein with the highest score may not necessarily reflect that of full-length protein or that of isoforms/splice variants.
Figure 2. DIGE gel with pH range 4-7. Arrows indicate identified protein spots (A1-A12).
Furthermore, we observed an up-regulation (1.43-fold) for L-plastin (Figure 2, spot A10), which is an actin-binding protein that is known to be involved in migration, invasion and metastasis. Beta-actin (Figure 2, spot A5) was 1.43-fold downregulated in the TN group. 3. Glycolytic Enzymes. Seven enzymes involved in glycolysis were identified and all of them showed decreased expression levels in the TN group compared to the HER2 positive group: mitochondrial malate dehydrogenase (MDH2, Figure 3, spot B8, 1.44-fold), glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Figure 3, spots B6 and B7, 1.38- and 1.47-fold), sorbitol dehydrogenase (SDH, Figure 3, spot B4, 1.64-fold), transketolase 3434
Journal of Proteome Research • Vol. 8, No. 7, 2009
Figure 3. DIGE gel with pH range 7-10. Arrows indicate identified protein spots (B1-B24).
(TKT, Figure 3, spot B3, 1.57-fold), phosphoglycerate kinase 1 (PGK1, Figure 3, spot B5, 1.40-fold), and fructose 1,6-bisphosphate aldolase (aldolase A, Figure 3, spots B1 and B2, 1.64and 1.51-fold). 4. Proteins of the Annexin Family. Two members of the annexin family of calcium- and phospholipid-binding proteins, namely, annexin A1 (ANXA1) and annexin A2 (ANXA2) (Tables 2 and 3), were identified. Compared to the HER2 positive group, we found that ANXA1 (Figure 3, spot B19) was up-regulated (1.34-fold) in the triple-negative group. ANXA2 was identified from two protein spots (Figure 3, spots B9 and B10) displaying
research articles
Proteomics in Triple-Negative Breast Cancers
Figure 4. 3D-view of identified differentially expressed proteins. (A and C) Down-regulated protein spots in TN; (A) CK7 identified in pH 4-7 gel (Figure 2, spot A1), (C) NME1-NME2 protein identified in pH 7-10 gel (Figure 3, spot B11). (B and D) Up-regulated protein spots in TN; (B) Fibronectin identified in pH 4-7 gel (Figure 2, spot A8), (D) Annexin A1 identified in pH 7-10 gel (Figure 3, spot B19).
comparable molecular weight, but differing in their pI values. This is likely due to post-translational modifications known for ANXA2 like for example phosphorylation and N-terminal acetylation. The ANXA2-mediated plasmin generation is known to be critical for invasion and metastasis. ANXA2 was found to be down-regulated (spot B9, 1.30-fold; spot B10, 1.35-fold) in the TN group. 5. NME, PRX1, ETF, and LF. Further, the following proteins were identified to be down-regulated in TN tumors: NME1NME2 protein (Figure 3, spot B11, 1.49-fold), peroxiredoxin (PRX1, Figure 3, spot B12, 1.99-fold), and electron-transfer flavoprotein (ETF, Figure 3, spots B13 and B14, 1.32- and 1.49fold). PRX1 is a cytosolic peroxidase, thus, displaying antioxidant activity. Compared to the HER2 positive group, we observed less expression of PRX1 in the TN group (Figure 3, spot B12, 1.99-fold down-regulation). We observed an enhanced expression of apolactoferrin (lactoferrin, LF) in the TN group (Figure 3, spots B20 and B21, -2.96 and -2.69-fold up-regulation, respectively). Among other functions of LF, it is involved in the regulation of cellular growth and differentiation and seems to have also an effect on mitogenactivated protein kinase (MAPK) and the nuclear factor-kB (NFkB) pathways, resulting in altered gene expression.17–21 Furthermore, three protein spots were identified as immunoglobulins (Table 3, Figure 3, spots B22, B23, B24). All three protein spots were up-regulated in the TN group (1.41-, 1.39-, and 1.37-fold, respectively). Four protein spots in the lowmolecular weight range of the pH 7-10 gel (Table 2, Figure 3, spots B15, B16, B17, and B18) were down-regulated (1.78-, 1.57-, 1.40-, and 1.55-fold, respectively) in the TN group and were identified as hemoglobin. Immunohistochemistry of Cytokeratins 7 and 14. Tenmicrometer sections from frozen tumor tissues of two triplenegative cases and two Her2-positive/HR-negative cases were immunohistochemically stained for cytokeratins 7 and 14, two of the proteins identified by 2D-DIGE and subsequent MS analysis to be differentially expressed between the tumor groups. Figure 5 shows IHC for cases 5, 9, 18, and 21, with case numbers corresponding to those given in Table 1 and Figure 6. CK7 (upper row) in TN cases shows less staining intensity compared to CK14 staining in TN (Figure 5A,B), and also compared to CK7 staining in Her2+ cases (Figure 5C,D). The same patterns of intensities were observed by Western blot analysis of CK7 (Figure 6). Staining for CK14 is more intense
in the TN cases (Figure 5A,B) and corresponds to the finding that CK14 was more highly expressed in TN compared to Her2+ breast cancer. Western Blot Analysis of Identified Proteins. To assess the expression levels of protein extractions from individual breast cancer cases, we performed Western blot analysis of four differentially expressed proteins (CK7, GAPDH, PGK1, and CK19). The Western blots displayed in Figure 6 show the result for the pooled proteins of the TN and Her2+ groups, as well as that of the individual samples. Protein extracts of cases 27, 32, and 33 are not shown, because of the low amount of protein that was obtained from these tissues. All the proteins shown in Figure 6 are down-regulated in TN cancer. The band intensities for the pooled proteins, that were applied as controls, are identical for blots of the TN and Her2+ groups (lanes 2 and 3, respectively), thus, allowing for direct comparison of the blots. The differences in protein abundance between the TN and Her2+ groups were in most of the cases below 2-fold. These differences are clearly reflected in the band intensities of the pooled samples. As expected, we observed interindividual differences of protein expression levels, especially for CK7 and CK19; however, the band intensities of GAPDH and PGK1 are identical and in average the bands of the TN samples are less intense compared to those of the Her2+ samples. Band intensities of the CK7 blot reflect the IHC staining intensities (Figure 5, upper row), with moderate staining for case no. 5, none for case no. 9, and strong staining for case nos. 18 and 21. Since we wanted to correct for individual variabilities, we decided to use pooled samples in order to identify up-/downregulations representative for the two groups.
Discussion In this study, we identified differentially expressed proteins in TN tumors versus HER2 positive/hormone receptor negative tumors using 2D-DIGE and mass spectrometry. We identified 13 up-regulated proteins and more than 20 down-regulated proteins in TN breast cancers. Among these proteins, there are cytokeratins and other structural proteins, glycolytic proteins, annexins, lactoferrin, peroxiredoxin, NME, and others. Cytokeratins. Besides their structural function as intermediate filament proteins, cytokeratins (CKs) play also an important role in signal transduction and apoptosis.22 It is known that the cytoskeleton is subject to remodelling processes, thus, explaining the involvement of CKs in cell motility, cell signaling and cell cycle regulation. Among the identified CKs, elevated expression levels of CK8 are known to be indicative of HER2 positivity,22 and an enhanced expression of CK19 in HER2 positive breast cancer was also described by Zhang et al.23 explaining the relative down-regulation in our TN tumors compared with the HER2 positive tumors. Furthermore, CK8 is a substrate for c-Jun N-terminal kinase.24 On the other hand, CKs 14 and 17 that are known to be highly expressed in basallike breast cancers (to which the majority of TN breast cancers belong25) were up-regulated in the TN group. CK14 is one of the marker proteins that are used to define the basal-like phenotype of breast cancer.25 Laakso et al.26 showed that expression levels of CKs 5 and 14 were elevated in steroid receptor negative tumors and that, furthermore, elevated expression levels were inversely associated with gene amplification of HER2. Our results are in perfect agreement with what is known from the literature and thus clearly prove the feasibility of our experimental setup. Journal of Proteome Research • Vol. 8, No. 7, 2009 3435
research articles
Schulz et al.
Figure 5. Representative examples of CK7 (upper row) and CK14 IHC (lower row) on 10 µm sections from frozen tumor tissues. CK7 expression is down-regulated, whereas CK14 expression is up-regulated in TN tumors (A and B). CK7 expression is up-regulated, whereas CK14 expression is down-regulated in TN tumors (C and D).
Figure 6. Western blot analysis of identified differentially expressed proteins. Western blots with antibodies against CK7, GAPDH, PGK1, and CK19. First lane, MW standard; second lane, TN pool; third lane, Her2+ pool. The numbers of the individual samples correspond to case numbers listed in Table 1.
Other Structural Proteins. In this group of proteins, we summarized those, that are involved in the structural organization of the cell, as for example, vimentin, fibronectin, and L-plastin. Vimentin was up-regulated in the TN tumors in our study. We found vimentin to be up-regulated in the TN group, and it is known that it is expressed in many hormoneindependent breast cancer cell lines27 and is associated with invasiveness and chemoresistance.28 Fibronectin, which we identified to be up-regulated in TN tumors, has been shown to be involved in carcinoma development and to stimulate phosphatidylinositol 3-kinase (PI3K).16 Furthermore, we observed an up-regulation of the actinbinding protein L-plastin, which is known to be involved in migration, invasion and metastasis.29 L-plastin is absent in normal breast epithelial cells and its expression in breast cancer cell lines correlates with the degree of invasiveness.29 Collagen is the principal component of the extracellular matrix (ECM). It has been shown that collagen production by stromal fibroblasts was down-regulated in more advanced cancers.30 We found collagen to be up-regulated in the TN group. 3436
Journal of Proteome Research • Vol. 8, No. 7, 2009
Glycolytic Enzymes. Tumor cells exhibit an altered metabolism, characterized by an increased glucose uptake and elevated glycolysis.31 Consequently, differentially expressed glycolytic enzymes have been described for different tumor entities and cell lines.23,31,32 In comparison to HER2 positive tumors, we identified down-regulation of several glycolytic proteins in TN tumors, such as phosphoglycerate kinase 1(PGK1), glyceraldehyde dehydrogenase (GAPDH), transketolase 1, TPI (triosephosphate isomerase), and aldolase A. Phosphoglycerate kinase 1 (PGK1) functions as an enzyme in glycolytic metabolism, participates in the angiogenesis process, and is transcriptionally activated by hypoxia inducible factor-1.23 Further, it influences DNA repair in mammalian cells33 and was upregulated in invasive breast carcinomas in comparison with the corresponding normal tissues.34 This corresponds with our result of PGK1 down-regulation in triple-negative versus HER2 positive tumors. It was also observed that PGK1 expression was reduced by partially switching off HER2 signaling with Herceptin treatment.23
research articles
Proteomics in Triple-Negative Breast Cancers We further identified differential expression of two members of the annexin family, namely, annexin A1 (ANXA1) and annexin A2 (ANXA2). Annexins are known to be involved in diverse biological processes including signal transduction, mediation of apoptosis,35 in the regulation of cytokine production and in the MAPK/ERK signal transduction pathway.36 Both annexin A1 and A2 were differentially expressed, with ANXA1 being overexpressed and ANXA2 being down-regulated in our TN tumors. Kreunin et al.37 described that the expresion of ANXA1 and ANXA2 was associated with a nonmetastatic phenotype of breast cancer cell lines. The expression of ANXA1 in breast cancer is still a subject of controversial discussion. Overexpression of ANXA1 has been observed in various types of breast tumors, including noninvasive ductal carcinoma in situ (DCIS), invasive and metastatic breast tumors.38 The proteomic and gene expression results in the study of Shen39,40 suggested that ANXA1 is downregulated in breast cancer cells and only detected in the stromal cells of the tumors. A reduced ANXA1 expression was also found in myoepithelial cells in ductal carcinoma in situ lesions compared with the same cell population in either normal or hyperplastic lesions.39 These results suggest that suppressed ANXA1 expression in breast tissue is correlated with breast cancer development and progression.39 On the contrary, Cao et al.41 found that ANXA1 expression was lost in 79% of breast carcinomas, and that most ANXA1 negative tumors were positive for ER and PR, but negative for HER2. We found a reduced ANXA1 expression in TN tumors, which is in accordance with the findings of Cao.41 Annexin A2 (ANXA2) is a receptor for plasminogen and tPA (tissue-type plasminogen activator), which binds to plasminogen and converts it to plasmin. The latter is capable of degrading ECM (extracellular matrix),42 thus, facilitating cell invasion and migration. ANXA2 was undetectable in normal and hyperplastic ductal epithelial cells,42 but was consistently expressed in invasive breast cancer, indicating its involvement in carcinogenesis. In agreement with these findings, we identified an up-regulation of ANXA2 in TN versus HER2 positive tumors. The ANXA2 overexpression in chemoresistant MCF-7 breast cancer cell lines,43 as well as the expression of ANXA2 in metastatic vs nonmetastatic cell lines, indicates that ANXA2 may indirectly contribute to angiogenesis and metastasis. This makes it an attractive target for new antiangiogenic and antibreast cancer therapies.42 LF, PRX, NME, ETF. Lactoferrin (LF) is an iron binding protein that is present in exocrine fluids.44 It is known to have an effect on mitogen-activated protein kinase (MAPK) and the nuclear factor-kB (NF-kB) pathways, and thus functions as a regulator of gene expression.17–21 That LF activates MAPK signaling pathways45 and its up-regulation in our TN tumors suggests it to be a potential therapeutic target. Peroxiredoxins (PRXs) are a group of peroxidases containing high antioxidant efficiency and some of them also have effects on cell differentiation and apoptosis.46 There is growing evidence that oxidative stress is important not only for normal cell physiology, but also for many pathological processes including cancer.47–49 Karihtala et al.46 found that PRXs’ expression is increased in breast malignancy, suggesting the induction of PRXs as response to increased production of reactive oxygen species in carcinomatous tissue. NME1-NME2 protein is a nucleoside diphosphate kinase (NDPK). Decreased mRNA levels of NME were shown to be associated with tumor metastasis,50,51 which underlines its role
as metastasis suppressor. Youn et al.51 described NME1-NME2 protein as a key molecule in breast tumor angiogenesis that inhibits the metastatic potential of cancer cells through its interaction with proteins involved in cellular signaling. Therefore, decreased levels of NME1-NME2 protein fail to exert its role as suppressor of tumorigenisis and angiogenesis. We found a decreased expression of the NME1-NME2 protein in the TN tumors. This finding may suggest inhibition of angiogenesis as treatment strategy for TN breast cancers. We chose a proteomic approach (2D-DIGE and MS) to characterize protein expression patterns in breast cancer tissues. We used pooled protein extracts from two groups of frozen tumor tissues (TN and HER2+) to correct for interindividual variation and separated the proteins on two pH ranges (4-7 and 7-10) to achieve better resolution of the protein mixtures. With this approach, we identified a number of proteins differentially expressed in triple-negative breast cancers in comparison with HER2 positive/hormone receptor negative tumors. Immunohistochemistry and Western blot results supported the validity of our results. The identified proteins, as for example annexins A1 and A2, lactoferrin, NME1NME2 protein, fibronectin, and L-plastin, may represent potential targets, and, therefore, may lead to the development of new treatment regimes effective in those tumors that fail to express the known targets such as estrogen and progesterone or HER2 receptors.
Acknowledgment. The authors gratefully acknowledge the technical assistance of Ulrike Reich, Nina Weber, and Jaqueline Mu ¨ ller. This study was supported by the European Comission, ‘TRANSBIG’-project, no. LSHC-CT-2004-503426. References (1) Lund, M. J.; Butler, E. N.; Bumpers, H. L.; Okoli, J. High prevalence of triple-negative tumors in an urban cancer center. Cancer 2008, 113, 608–615. (2) Bauer, K. R.; Brown, M.; Cress, R. D.; Parise, C. A.; Caggiano, V. Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: a populationbased study from the California cancer Registry. Cancer 2007, 109, 1721–1728. (3) Haffty, B. G.; Kim, J. H.; Yang, Q.; Higgins, S. A. Concurrent chemoradiation in the conservative management of breast cancer. Int. J. Radiat. Oncol. Biol. Phys. 2006, 66, 1306–1312. (4) Perou, C. M.; Sorlie, T.; Eisen, M. B.; van de Rijn, M. Molecular portraits of human breast tumours. Nature 2000, 406, 747–752. (5) Sorlie, T.; Perou, C. M.; Tibshirani, R.; Aas, T. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 10869– 10874. (6) Cleator, S.; Heller, W.; Coombes, R. C. Triple-negative breast cancer: therapeutic options. Lancet Oncol. 2007, 8, 235–244. (7) Altundag, K.; Harputluoglu, H.; Aksoy, S.; Gullu, I. H. Potential chemotherapy options in the triple negative subtype of breast cancer. J. Clin. Oncol. 2007, 25, 1294-1295, author reply 12951296. (8) Rakha, E. A.; El-Sayed, M. E.; Green, A. R.; Lee, A. H. Prognostic markers in triple-negative breast cancer. Cancer 2007, 109, 25– 32. (9) Carey, C. M.; Kirk, J. L.; Ojha, S.; Kostrzynska, M. Current and future uses of real-time polymerase chain reaction and microarrays in the study of intestinal microbiota, and probiotic use and effectiveness. Can. J. Microbiol. 2007, 53, 537–550. (10) Reis-Filho, J. S.; Tutt, A. N. Triple negative tumours: a critical review. Histopathology 2008, 52, 108–118. (11) Smith, L.; Lind, M. J.; Welham, K. J.; Cawkwell, L. Cancer proteomics and its application to discovery of therapy response markers in human cancer. Cancer 2006, 107, 232–241. (12) Alban, A.; David, S. O.; Bjorkesten, L.; Andersson, C. A novel experimental design for comparative two-dimensional gel analysis:
Journal of Proteome Research • Vol. 8, No. 7, 2009 3437
research articles (13)
(14) (15)
(16)
(17)
(18)
(19)
(20)
(21) (22) (23)
(24)
(25) (26) (27)
(28)
(29)
(30) (31)
3438
two-dimensional difference gel electrophoresis incorporating a pooled internal standard. Proteomics 2003, 3, 36–44. Quintanilla-Martinez, L.; Davies-Hill, T.; Fend, F.; Calzada-Wack, J.; et al. Sequestration of p27Kip1 protein by cyclin D1 in typical and blastic variants of mantle cell lymphoma (MCL): implications for pathogenesis. Blood 2003, 101, 3181–3187. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 1976, 72, 248–254. O’Connell, K. L.; Stults, J. T. Identification of mouse liver proteins on two-dimensional electrophoresis gels by matrix-assisted laser desorption/ionization mass spectrometry of in situ enzymatic digests. Electrophoresis 1997, 18, 349–359. Kijima, T.; Maulik, G.; Ma, P. C.; Madhiwala, P. Fibronectin enhances viability and alters cytoskeletal functions (with effects on the phosphatidylinositol 3-kinase pathway) in small cell lung cancer. J. Cell. Mol. Med. 2003, 7, 157–164. Damiens, E.; El Yazidi, I.; Mazurier, J.; Duthille, I. Lactoferrin inhibits G1 cyclin-dependent kinases during growth arrest of human breast carcinoma cells. J. Cell. Biochem. 1999, 74, 486– 498. Dhennin-Duthille, I.; Masson, M.; Damiens, E.; Fillebeen, C. Lactoferrin upregulates the expression of CD4 antigen through the stimulation of the mitogen-activated protein kinase in the human lymphoblastic T Jurkat cell line. J. Cell. Biochem. 2000, 79, 583– 593. Grey, A.; Banovic, T.; Zhu, Q.; Watson, M. The low-density lipoprotein receptor-related protein 1 is a mitogenic receptor for lactoferrin in osteoblastic cells. Mol. Endocrinol. 2004, 18, 2268– 2278. Haversen, L.; Ohlsson, B. G.; Hahn-Zoric, M.; Hanson, L. A.; Mattsby-Baltzer, I. Lactoferrin down-regulates the LPS-induced cytokine production in monocytic cells via NF-kappa B. Cell. Immunol. 2002, 220, 83–95. Oh, S. M.; Pyo, C. W.; Kim, Y.; Choi, S. Y. Neutrophil lactoferrin upregulates the human p53 gene through induction of NF-kappaB activation cascade. Oncogene 2004, 23, 8282–8291. Brouillard, F.; Fritsch, J.; Edelman, A.; Ollero, M. Contribution of proteomics to the study of the role of cytokeratins in disease and physiopathology. Proteomics: Clin. Appl. 2008, 2, 264–285. Zhang, D.; Tai, L. K.; Wong, L. L.; Chiu, L. L. Proteomic study reveals that proteins involved in metabolic and detoxification pathways are highly expressed in HER-2/neu-positive breast cancer. Mol. Cell. Proteomics 2005, 4, 1686–1696. He, T.; Stepulak, A.; Holmstrom, T. H.; Omary, M. B.; Eriksson, J. E. The intermediate filament protein keratin 8 is a novel cytoplasmic substrate for c-Jun N-terminal kinase. J. Biol. Chem. 2002, 277, 10767–10774. Rakha, E. A.; Reis-Filho, J. S.; Ellis, I. O. Basal-like breast cancer: a critical review. J. Clin. Oncol. 2008, 26, 2568–2581. Laakso, M.; Loman, N.; Borg, A.; Isola, J. Cytokeratin 5/14-positive breast cancer: true basal phenotype confined to BRCA1 tumors. Mod. Pathol. 2005, 18, 1321–1328. Sommers, C. L.; Walker-Jones, D.; Heckford, S. E.; Worland, P. Vimentin rather than keratin expression in some hormoneindependent breast cancer cell lines and in oncogene-transformed mammary epithelial cells. Cancer Res. 1989, 49, 4258–4263. Korsching, E.; Packeisen, J.; Liedtke, C.; Hungermann, D. The origin of vimentin expression in invasive breast cancer: epithelialmesenchymal transition, myoepithelial histogenesis or histogenesis from progenitor cells with bilinear differentiation potential. J. Pathol. 2005, 206, 451–457. Lin, C. S.; Lau, A.; Yeh, C. C.; Chang, C. H.; Lue, T. F. Upregulation of L-plastin gene by testosterone in breast and prostate cancer cells: identification of three cooperative androgen receptor-binding sequences. DNA Cell Biol. 2000, 19, 1–7. Fenhalls, G.; Geyp, M.; Dent, D. M.; Parker, M. I. Breast tumour cell-induced down-regulation of type I collagen mRNA in fibroblasts. Br. J. Cancer 1999, 81, 1142–1149. Young, C. D.; Anderson, S. M. Sugar and fatsthat’s where it’s at: metabolic changes in tumors. Breast Cancer Res. 2008, 10, 202.
Journal of Proteome Research • Vol. 8, No. 7, 2009
Schulz et al. (32) van den Bemd, G. J.; Krijgsveld, J.; Luider, T. M.; van Rijswijk, A. L. Mass spectrometric identification of human prostate cancerderived proteins in serum of xenograft-bearing mice. Mol. Cell. Proteomics 2006, 5, 1830–1839. (33) Popanda, O.; Fox, G.; Thielmann, H. W. Modulation of DNA polymerases alpha, delta and epsilon by lactate dehydrogenase and 3-phosphoglycerate kinase. Biochim. Biophys. Acta 1998, 1397, 102–117. (34) Kabbage, M.; Chahed, K.; Hamrita, B.; Guillier, C. L. Protein alterations in infiltrating ductal carcinomas of the breast as detected by nonequilibrium pH gradient electrophoresis and mass spectrometry. J. Biomed. Biotechnol. 2008, 2008, 564127. (35) Vishwanatha, J. K.; Salazar, E.; Gopalakrishnan, V. K. Absence of annexin I expression in B-cell non-Hodgkin’s lymphomas and cell lines. BMC Cancer 2004, 4, 8. (36) Alldridge, L. C.; Harris, H. J.; Plevin, R.; Hannon, R.; Bryant, C. E. The annexin protein lipocortin 1 regulates the MAPK/ERK pathway. J. Biol. Chem. 1999, 274, 37620–37628. (37) Kreunin, P.; Yoo, C.; Urquidi, V.; Lubman, D. M.; Goodison, S. Proteomic profiling identifies breast tumor metastasis-associated factors in an isogenic model. Proteomics 2007, 7, 299–312. (38) Ahn, S. H.; Sawada, H.; Ro, J. Y.; Nicolson, G. L. Differential expression of annexin I in human mammary ductal epithelial cells in normal and benign and malignant breast tissues. Clin. Exp. Metastasis 1997, 15, 151–156. (39) Shen, D.; Nooraie, F.; Elshimali, Y.; Lonsberry, V. Decreased expression of annexin A1 is correlated with breast cancer development and progression as determined by a tissue microarray analysis. Hum. Pathol. 2006, 37, 1583–1591. (40) Shen, D.; Chang, H. R.; Chen, Z.; He, J. Loss of annexin A1 expression in human breast cancer detected by multiple highthroughput analyses. Biochem. Biophys. Res. Commun. 2005, 326, 218–227. (41) Cao, Y.; Li, Y.; Edelweiss, M.; Arun, B. Loss of annexin A1 expression in breast cancer progression. Appl. Immunohistochem. Mol. Morphol. 2008, 16, 530–534. (42) Sharma, M. R.; Koltowski, L.; Ownbey, R. T.; Tuszynski, G. P.; Sharma, M. C. Angiogenesis-associated protein annexin II in breast cancer: selective expression in invasive breast cancer and contribution to tumor invasion and progression. Exp. Mol. Pathol. 2006, 81, 146–156. (43) Chuthapisith, S.; Layfield, R.; Kerr, I. D.; Hughes, C.; Eremin, O. Proteomic profiling of MCF-7 breast cancer cells with chemoresistance to different types of anti-cancer drugs. Int. J. Oncol. 2007, 30, 1545–1551. (44) Davidson, L. A.; Lonnerdal, B. Fe-saturation and proteolysis of human lactoferrin: effect on brush-border receptor-mediated uptake of Fe and Mn. Am. J. Physiol. 1989, 257, G930-934. (45) Oh, S. M.; Hahm, D. H.; Kim, I. H.; Choi, S. Y. Human neutrophil lactoferrin trans-activates the matrix metalloproteinase 1 gene through stress-activated MAPK signaling modules. J. Biol. Chem. 2001, 276, 42575–42579. (46) Karihtala, P.; Mantyniemi, A.; Kang, S. W.; Kinnula, V. L.; Soini, Y. Peroxiredoxins in breast carcinoma. Clin. Cancer Res. 2003, 9, 3418–3424. (47) Dhalla, N. S.; Temsah, R. M.; Netticadan, T. Role of oxidative stress in cardiovascular diseases. J. Hypertens. 2000, 18, 655–673. (48) Ambrosone, C. B. Oxidants and antioxidants in breast cancer. Antioxid. Redox Signaling 2000, 2, 903–917. (49) Klaunig, J. E.; Xu, Y.; Isenberg, J. S.; Bachowski, S. The role of oxidative stress in chemical carcinogenesis. Environ. Health Perspect. 1998, 106 (1), 289–295. (50) Florenes, V. A.; Aamdal, S.; Myklebost, O.; Maelandsmo, G. M. Levels of nm23 messenger RNA in metastatic malignant melanomas: inverse correlation to disease progression. Cancer Res. 1992, 52, 6088–6091. (51) Youn, B.; Kim, H. D.; Kim, J. Nm23-H1/nucleoside diphosphate kinase as a key molecule in breast tumor angiogenesis. Expert Opin. Ther. Targets 2008, 12, 1419–1430.
PR900071H
