Separation and identification of differentially expressed nuclear matrix proteins in breast carcinoma forming

Abstract

Background. Breast carcinoma is one of most prevalent malignant tumors occurring in women. Short of prevention, detection of breast carcinoma at an early, still curable stage would offer the best route to decrease its mortality rates. This highlights the urgent need for suitable biomarkers for early diagnosis and a better understanding of the disease pathogenesis. Material and methods. NMPs were extracted from normal human breast tissue (Group I), from hyperplastic mammary tissue specimens (Group II), from atypical epithelial hyperplasia specimens (Group III), and from breast carcinoma (Group IV) tissue. Differential proteome profiles were established and analyzed by means of immobilized pH gradient-based two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS). The different NMPs were analyzed in the National Center for Biotechnology Information (NCBI) database with Mascot software. Results. Well-resolved, reproducible 2-DE profiles of human breast tissues were obtained. Average protein spots were 904 ± 58, 912 ± 51, 931 ± 63, 944 ± 70 in Group I, Group II, Group III, and Group IV, respectively. Several different proteins were analyzed using mass spectrometry and bioinformation. Of these, 12 were well characterized. Compared to Group I, three proteins were up-regulated in Groups II, III, and IV, including Hsp27, prohibitin, and laminA/C. Upregulation was confirmed using Western blotting and immunohistochemical analysis. The correlation of prohibitin expression with clinicopathological features was also investigated. Discussion. The proteins identified in this study may potentially prove to be useful markers for breast carcinoma diagnosis.

Breast cancer imposes a significant healthcare burden on women worldwide. According to the American National Cancer Institute, this disease is estimated to be the most commonly diagnosed neoplasm in women in 2008 [1]. The 5-year survival rates associated with breast carcinoma decrease from 98% for localized disease to 26% for late stage disease. Hence, short of prevention, detection of breast carcinoma in the early, still curable stage would greatly help to decrease the mortality rate. However, since only 63% of breast carcinomas are still confined to the breast at the time of diagnosis, the currently applied diagnostic screening tools obviously are not sufficient for adequate breast carcinoma diagnosis. This highlights the urgent need for suitable biomarkers for early diagnosis, and a better understanding of the disease pathogenesis.

The nuclear matrix provides structural support for the nucleus and is involved in various nuclear functions including regulation of transcription, replication, and DNA repair. Nuclear matrix proteins (NMPs), which are tissue- and cell type-specific, are altered with transformation and state of differentiation [2]. They have been identified as informative markers of disease states. Informative NMPs have been identified for bladder, hepatocellular, colon, and prostate cancers [3–6].

Proteomic analysis is currently considered as a powerful tool for global evaluation of protein expression, and has been widely applied in studies of diseases, especially in the field of cancer research. Use of clinical tissue samples may be the most direct and persuasive way to find tumor-related proteins using a proteomic approach. However, in the case of NMP studies on breast cancer, most previous proteomic evaluations have employed cultured cell lines as a research model [7,8]. In the present study, we compared normal human breast tissue with hyperplastic mammary tissue specimens, atypical epithelial hyperplasia specimens, and breast carcinoma tissues. We used two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) and matrix-assisted laser desorption/ionization tandem time-of-flight mass spectrometry (MALDI-TOF/TOF-MS), together with online database searching, to investigate differential NMPs in these four tissue types. Confirmatory studies using Western blotting and immunohistochemistry validated the data obtained by the proteomic method. The proteins identified are expected to provide some clues for further study of carcinogenic mechanisms and also may have uses as molecular markers for diagnosis or prognosis of breast carcinoma.

Material and methods

Patients and tissue samples

Tissue specimens used in this study were obtained with the approval of the Human and Ethical Committee for Medical Research at Xi’an Jiaotong University School of Medicine. The experiment was conducted in two individual laboratories; with the specimens included accordingly divided into two groups. All specimens were collected immediately after surgical dissection. A total of eight normal human breast tissue specimens (Group I), eight hyperplastic mammary tissue specimens (Group II), eight atypical epithelial hyperplasia specimens (Group III), and eight breast carcinoma specimens (Group IV) were obtained. The proteomic comparative analysis was carried out in the proteomic lab of the University of Xi’an Jiaotong. Other experiments, including immunohistochemistry (IHC), a confirmative Western blot assay, and correlation analysis of prohibitin expression with clinicopathological features were performed using another group of samples at the Second Affiliated Hospital of Xi’an Jiaotong University. The total 56-case sample group contained 12 normal breast tissues, 13 hyperplastic mammary tissue specimens, and 31 breast carcinoma specimens. The 31 breast carcinoma specimens consisted of 19 infiltrating special breast carcinomas and 12 infiltrating non-special breast carcinomas. Seven cases were in stage I, 15 cases in stage II, and nine cases in stage III. Six cases were in low differentiation, 12 were in moderate differentiation, and 13 were well differentiated. All samples were histologically confirmed by two independent pathologists.

Extraction of nuclear proteins and NMPs

NMPs were prepared according to a procedure previously reported [9]. Nuclear proteins were isolated following the protocol of Somasekhar et al. [10].

Two-dimensional gel electrophoresis

First-dimension isoelectric focusing was carried out using an Ettan IPGphor isoelectric focusing system, following the manufacturer's instructions (GE Amersham, USA). Samples containing up to 100 μg protein were added to 13 cm IPG strips (pH 3–10). The strips were then applied by rehydration at 30 V for 12 h, and 1 h at 500 V. A gradient was then applied at 1000 V for 1 h, 8000 V for 8 h and 500 V for 4 h. The temperature throughout this process was maintained at 20°C. Separation in the second dimension was carried out at a current setting of 15 mA/gel for the initial 30 min and 30 mA/gel until the bromophenol blue front reached the bottom of the gels. The gels were subsequently stained with silver nitrate and/or Coomassie blue for image analysis and/or protein spot identification, by standard procedures. Molecular weight markers were also run in the second-dimensional gels to facilitate molecular weight calibration of protein spots. The stained gels were then scanned with an Image Scanner (Amersham Biosciences) operated by the software program LabScan 3.00 Image analyses were refined with Image Master 2D Elite software 5.0 (Amersham Biosciences).

In-gel protein digestion

In-gel digestion of proteins was carried out as follows. Briefly, the protein spots were excised from the gels, destained in 50 mM NH4HCO3/30% ACN, and dried in a vacuum centrifuge. Digestion was carried out with 12 μg/ml sequencing grade-modified trypsin (Promega, USA) in 50 mM NH4HCO3 buffer at 37°C for 12–16 h. Peptides were recovered by extraction with 50% ACN/5%TFA, dried in a vacuum centrifuge and resuspended in 15 μl of solution containing 50% ACN and 2.5% TFA. The extracted solutions were mixed and dried in a vacuum centrifuge.

MALDI-TOF MS analysis

The peptide mixtures were solubilized with 5 μL of 1% formic acid, then vortexed and centrifuged. Samples (2 μL) of the supernatant were spotted on the target and dried. In total, 0.5 μL of saturated acyano-4-hydroxy-transcinnamic solution was spotted on top of the dried samples and air dried. Spots were then analyzed in a Bruker-Daltonics AutoFlex TOF-TOF LIFT mass spectrometer (Bruker, Germany). The running parameters were as follows: positive ion-reflector mode, accelerating voltage 20 kV. A trypsin-fragment peak served as an internal standard for mass calibration. A list of the corrected mass peaks was obtained as Peptide Mass Fingerprinting (PMF). Protein identification was performed by searching NCBInr database (20080718) using MASCOT search engine (http://www.matrixscience.com). The searching parameters were set up as follows: the fragment mass tolerance was ± 0.5 Da, peptide mass tolerance was ± 100 ppm, the number of missed cleavage sites was allowed up to one, the cysteine residue was modified as carbamidomethyl-cys, species selected was Homo sapiens (human), the peptide ion was [M + H]⁺, and the monoisotopic mass was used.

Western blotting analysis

The proteins were separated by 10% SDS-PAGE, electrotransferred to nitrocellulose membranes (Millipore, USA), and blocked with 5% non-fat milk in Tris Buffered Saline with Tween (TBST). The membranes were immunoblotted with anti-histone H3 (1:200 dilution; BIOS), anti-DNA topoisomerase IIa (1:300 dilution; BIOS), anti-Hsp27 (1:200 dilution; Santacruz), anti-prohibitin (1:200 dilution; Neomarker), or anti-laminA/C (1:200 dilution; Santacruz). Secondary antibodies conjugated to horseradish peroxidase (BIOS, China) were applied at a dilution of 1:2000. The signals were detected by an enhanced chemiluminescent (ECL) detection reagent (Pierce, USA).

Immunohistochemical assay

Immunoperoxidase staining of formalin-fixed, paraffin-embedded tissue sections was performed using the streptavidin-peroxidase (SP) method. After deparaffinization and rehydration in a descending alcohol dilution, the sections were heated in an 800 W microwave oven at maximal power for 5 min in 0.01 M citrate buffer (pH 6.0) for antigen retrieval. After cooling to room temperature, the endogenous peroxidase activity in the treated sections was blocked by 3% hydrogen peroxide in methanol for 10 min at room temperature. Following blocking to reduce nonspecific binding, the sections were incubated with the primary antibodies anti-Hsp27 (1:50 dilution; Santacruz), anti-prohibitin (1:100 dilution; Neomarker), or anti-laminA/C antibody (1:50 dilution; Santacruz) at 37°C for 2 h. The subsequent steps were performed using an ultrasensitive SP kit (Maixin Biological, China) according to the manufacturer's protocols. At the completion of the procedure, the sections were counterstained with haematoxylin. A negative control was run by replacing the primary antibody with PBS.

Statistic analysis

One-tailed Student's t-test was used for statistically analyzing the Western blotting analysis data. A value of p<0.05 was considered significant.

Results

Evaluation of NMP quality

The NMPs extracted from normal breast tissue and breast carcinoma specimens was evaluated by Western blot; the aim was to monitor whether the samples were intact or contaminated by other non-NMP components. Previous studies have proved that DNA topoisomerase II was the major component of NMPs [11,12], but the histone in nucleosomes was the potential contaminating protein of NMPs [13]. In our study, 170 kDa DNA topoisomerase II was detected in the NMPs of normal breast tissue and breast carcinoma samples at the same position. Moreover, the histone H3 band (15KDa) was not detected in NMPs (Figure 1), illustrating that the majority of histone was removed during NMP extraction.

Figure 1. Western blot analysis of DNA topoisomerase II (A) and histone (B). Lane 1, nuclear protein of normal breast. Lane 2, nuclear matrix protein of normal breast. Lane 3, nuclear protein of breast carcinoma. Lane 4, nuclear matrix protein of breast carcinoma. β-actin as used as an internal loading control.

Proteomic analysis of differentially-expressed proteins

Figure 2 shows the representative proteome profiling for four groups. About 900 spots were detected in each gel using Image Master Software. The average spots in Group I, II, III, and IV were respectively 904 ± 58, 912 ± 51, 931 ± 63, and 944 ± 70; most spots were distributed in the gel area of pI 4–9. A comparison of normalized spot volume was made between two groups. From the silver stained 2-DE gel, 27 expressed differentially protein spots were selected, excised and digested in-gel with trypsin. The PMF maps were obtained by MALDI-TOF-MS and calibrated with a TPCK-trypsin auto-degraded peak (Figure 3). The resulting protein was determined by comprehensively considering the corresponding experimental pI, Mr, the number of matched peptides, and the sequence coverage. Of 27 spots analyzed, 12 proteins were successfully identified. Compared to Group I, three proteins were up-regulated in Groups II, III, and IV, including Hsp27, prohibitin, and laminA/C (Table I).

Figure 2. Representative 2-D gel analyses of NMPs from (A) normal breast, (B) hyperplastic mammary, (C) atypical epithelial hyperplasia and (D) breast carcinoma.

Figure 3. MALDI-TOF mass spectrum maps of Hsp27 (A), prohibitin (B) and laminA/C (C).

Table I. Differentially-expressed proteins in the different groups.

Spot. no.	Accession. no	Protein name-e	Score	MW(Da)	PI no	Peptide coverage (%)	Protein	Fold change*
								a	b	c
1	g|662841	heat shock protein 27	143	22427	7.83	11	48	+1.59	+2.08	+2.81
2	gi|33150528	osteoglycin OG	129	34243	5.33	9	18	−1.62	−1.60	−2.15
3	gi|28375587	unnamed protein product	85	43225	5.9	10	23	−1.53	−2.33	−2.04
4	gi|193787261	unnamed protein product	68	113269	5.54	8	9	−1.57	−1.78	−3.21
5	gi|55957496	LaminA/C	110	74311	6.44	13	28	+1.56	+1.65	+2.06
6	gi|38045913	Nucleoside diphosphate kinase	81	19869	5.42	6	32	+1.54	+4.02	+3.56
7	gi|4505773	Prohibitin	116	29843	5.57	9	38	+2.11	+3.56	+4.41
8	gi|34527739	unnamed protein product	66	51876	8.66	7	16	−1.85	−3.42	−3.64
9	gi|5453597	F-actin capping protein alpha-1 subunit	67	33073	5.45	5	24	−1.62	−1.61	−2.03
10	gi|127798841	ATP synthase	70	59785	9.07	9	22	+2.16	+2.12	+2.73
11	gi|28207863	unnamed protein product	75	41593	5.14	9	27	−1.54	−1.76	−2.11
12	gi|113812151	MYL6 protein	93	15923	4.60	7	42	+1.53	+1.56	+2.68

*Fold change: a: Group I/Group II; b: Group I/Group III; c: Group I/Group IV; + up-regulated;—down-regulated.

Western blotting analysis

Western blotting was performed to verify three selected proteins, Hsp27, prohibitin and laminA/C, which may play important functional roles in tumorigenesis. The expression levels of these three proteins in 15 breast carcinoma specimens and six normal breast specimens were examined. Figure 4 displays the representative image of Western blotting. The band intensity was measured with ImageMaster 2D Elite 5.0 software, and the intensity ratio to corresponding β-actin band was calculated. Hsp27, prohibitin, and laminA/C increased obviously in 66.7% (10 of 15), 93.3% (14 of 15), and 60.0% (9 of 15) of cases, respectively, confirming the 2DE analysis results. The mean intensity of these proteins showed a statistical difference between tumor tissues and normal tissues. The altered expression ratio of these proteins matched well with the ratio detected in 2DE.

Figure 4. Western blot analysis of the Hsp27 (A), prohibitin (B) and laminA/C (C) in normal breast (Lane 1) and breast carcinoma (Lane 2). β-actin was used as an internal loading control.

Immunohistochemical analysis of proteins

To further confirm the altered expression, we examined the expression of Hsp27, prohibitin, and laminA/C in cancer and normal tissues by IHC studies. A negative control was set up by eliminating incubation with primary antibodies. IHC study further confirmed the increased expressions of Hsp27, prohibitin, and laminA/C in breast carcinoma specimens (Figure 5). These results were consistent with the results obtained by both 2D electrophoresis and Western blot analyses.

Figure 5. Representative immunohistologic features of Hsp27 (A), prohibitin (B) and laminA/C (C) in normal breast and breast carcinoma (original magnification 200).

Correlation of prohibitin expression with clinicopathological characteristics of breast carcinoma

The above experiments showed that prohibitin was one protein that was significantly up-regulated in breast carcinoma specimens. We therefore examined the possible correlation of prohibitin expression with the clinicopathological characteristics of breast carcinoma. Prohibitin expression in 56 clinical specimens, including normal breast tissues, hyperplastic mammary tissues, and breast carcinoma specimens was evaluated for correlation with the presence of breast carcinoma. The correlation of prohibitin expression level with tumor differentiation and clinical stage was also investigated. Prohibitin expression was closely correlated to the tumor status, as evidenced by a markedly higher expression of prohibitin in breast carcinomas than that in normal breast tissues and hyperplastic mammary glands specimens (p<0.05). In addition, the prohibitin expression level had no correlation with other clinicopathological features, including the histological type, clinical stage, and cell differentiation (Figure 6 and Table II).

Figure 6. Correlation analysis of prohibitin expression with clinicalpathological characteristics. (A and B) The expression of prohibitin was increased from normal breast (lane 1–2), hyperplastic mammary (HM) (lane 3), to breast carcinoma (BC) lanes 4–6). However, prohibitin expression was not correlated with the differentiation status of breast carcinoma (lanes 4–6 are low-, medium-, and high-differentiated breast carcinoma, respectively).

Table II. The correlation of prohibitin with clinicalpathological features.

Groups	Cases	Band intensity ratio	p-Value (mean±SD)
Tissues
Normal breast	12	0.59±0.19
Hyperplasia of mammary gland	13	0.69±0.17
Breast carcinoma	31	1.02±0.22	0.001*
Histological type
Infiltrating special breast carcinoma	19	1.03±0.16
Infiltrating non-special breast carcinoma	12	1.01±0.31	0.850
Differentiated status
Low-differentiated	6	1.06±0.14
Medium-differentiated	12	1.02±0.17
High-differentiated	13	1.00±0.29	0.559
Clinical stages
Stage I	7	1.04±0.14
Stage II	15	1.03±0.27
Stage III	9	1.00±0.19	0.657

*Asterisk indicates a significant difference in breast carcinoma group in comparison with normal breast group.

Discussion

In contrast to most cancers, the incidence of breast carcinoma has been increasing in recent years. As discussed earlier, because of difficulties in treating late stage cancers, the most promising approach to breast carcinoma management is early detection. Currently, the screening methods for breast malignancies include mammography and serum marker screening with CA-153, or a combination of both. However, these strategies lack adequate sensitivity and specificity for early tumor detection. To better understand the pathogenesis of breast carcinoma and the identification of early diagnosis markers, many studies at the level of DNA and mRNA have been conducted [14–16]. Recently, the identification of breast carcinoma-associated proteins has also been reported. Some studies about NMPs have certainly provided considerable data. However, most studies employed cultured cell lines as a research model, which cannot reflect an in vivo microenvironment and lack of the influence of tumor-host interaction. Parvinderjit et al. [17] employed clinical samples as study materials. However, they only prepared nuclear matrices from normal and breast carcinoma tissues. In the current study, we collected breast tissues from actual patients as analytic materials, which allowed assessment of the real condition of disease development. The 2-DE profiles of human breast tissues among our different clinical groups allowed some differentially expressed NMPs to be identified. Of all 27 protein-spots selected for MS analysis, 12 proteins were characteristic of all groups. These proteins participate in important cellular events such as cell proliferation, apoptosis, and DNA replication, and many of these have been related to the tumorigenesis and progression of malignancies.

The heat shock proteins (Hsps) are highly conserved proteins present in nearly every organism. They act as cell chaperones to facilitate protein transport and polypeptide assembly in physiologic conditions. Several members of the Hsp family, such as Hsp27, are constitutively expressed at low levels in the cytosol. After induction they undergo rapid post-translation phosphorylation and move from the cytoplasm to the nucleus. These were up-regulated in malignant cells. Hsps have been shown to be closely associated with carcinogenesis and the development of many kinds of tumors, including breast and ovarian cancer [18,19]. The high expression of Hsp27 has been indicated to imply poor prognosis in breast cancer [20]. Regarding chemotherapy, Hsp27 over-expression has been correlated with a shorter disease-free survival in advanced breast cancer patients who received neoadjuvant chemotherapy [21].

In humans, the prohibitin gene PHB1 is located on chromosome 17q21, close to the ovarian and breast carcinoma susceptibility gene (BRAC1) 1. This gene encodes a highly conserved 30 kDa protein that plays a significant role in cell cycle control, differentiation, apoptosis, and senescence [22]. Although the prohibitin protein content is higher in primary human tumors, including breast cancer, the significance of increased prohibitin expression in cancer cells has not yet been clearly defined [23,24]. Fusaro et al. [25] have shown that over-expression of prohibitin in a human lymphoma cell line blocked apoptosis induced by the topoisomerase I inhibitor camptothecin. Recent studies [26] on the role of prohibitin in rat ovarian follicular development indicate that over-expression of prohibitin attenuates both staurosporine (STS) and serum withdrawal-induced granulosa cell apoptosis via the intrinsic apoptotic pathway, confirming that prohibitin is an effective antiapoptotic agent. Therefore, novel approaches that target prohibitin may lead to the development of new therapies that limit the development of resistance or enhance the sensitivity of cancer cells to current chemotherapeutic agents [27]. We found the expression of prohibitin was increased in breast carcinoma specimens. However, there was no statistical difference between carcinoma groups in different clinical stages, differentiation status, and histological type. These results suggest the increased level of prohibitin is a common marker for breast malignancies.

The nuclear lamina is a meshwork of 10 nm diameter filaments located on the inner aspect of the inner nuclear membrane. The lamina is composed of proteins termed lamins that are members of the intermediate filament family. The nuclear lamina are composed of two types of polypeptide filaments—type A and type B lamins. Type A lamin includes both types A and C, as they are related transcripts from a single gene [28]. Type B lamins are composed of sub-types B1 and B2. Lamin provides the structural support of the nucleus and is involved in various functions, including regulation of DNA replication and DNA synthesis. Moss et al. [29] have found reduced nuclear lamin expression to be an early and frequent finding in gastrointestinal cancer, often accompanied by aberrant cytoplasmic immunolabelling. In this study, a higher expression of laminA/C was found in breast cancer tissue. Although the cause of this phenomenon is currently unknown, our finding suggests that laminA/C may be useful for distinguishing normal breast tissue from a tumor.

In general, by using 2DE-based proteomic technology, we comparatively analyzed the NMP profiles of normal breast tissues, hyperplastic mammary gland specimens, atypical epithelial hyperplasia specimens, and breast carcinoma tissues, and identified several differently expressed NMPs. Although no breast carcinoma specific biomarker was found in this study, the successful identification of tumor-associated NMPs proved that the proteomic method was able to provide rich information that would be helpful for better understanding the cell malignant transformation in tumorigenesis. Monitoring of these differentially expressed NMPs may be useful in guiding more rationally designed diagnosis and treatment methods, which will hopefully translate into improved patient outcome.

Acknowledgments

We thank Jiang-jian Tao for his assistance in obtaining the pathological specimens. This work was supported by the National Nature and Science Foundation of China (No. 30500600).