Silencing of the SEC62 gene inhibits migratory and invasive potential of various tumor cells

Authors


Abstract

Sec62 is part of the protein translocation apparatus in the membrane of the endoplasmic reticulum (ER). In yeast, Sec62 participates in the post-translational translocation of proteins into the ER, but its function in mammals remains elusive. Previously we described the amplification and over-expression of the SEC62 gene in prostate cancer cell lines and the protein has been described as a potential target gene in prostate cancer. In the current study we show that in the tumor tissue of prostate cancer patients Sec62 protein levels are elevated compared with tumor-free tissue derived from the same patients or from prostates of control group patients and that the higher Sec62 protein content correlates with an increasing de-differentiation of the cells. Therefore, up-regulation of Sec62 protein content indeed is a phenomenon associated with prostate cancer progression. Analysis of a multi-tissue tumor array showed that in addition to prostate cancer, overproduction of Sec62 is observed in various other tumors, most significantly in tumors of the lung and the thyroid. To examine the tumor-related functions of Sec62, we silenced the SEC62 gene in the prostate cancer cell-line PC3 as well as in a set of other tumor cell-lines with two different siRNAs. In general, after silencing of SEC62 the cell migration and the invasive potential of the cells was blocked or at least dramatically reduced while cell viability was hardly affected. Thus, the SEC62 gene may indeed be considered as a target gene in the therapy of various tumors.

In yeast, Sec62 is part of the tetrameric Sec62/Sec63-subcomplex of the Sec-complex, and functions as a docking site for post-translational protein transport to the membrane or lumen of the endoplasmic reticulum (ER).1 A contribution to the secretory pathway for the mammalian ortholog has yet to be shown. As in yeast, mammalian Sec62 is also associated with the Sec61-complex, the main pore for protein translocation in the ER membrane, and with Sec63, an ER membrane protein containing a luminal J-domain.2 In contrast to the yeast protein, the mammalian protein can interact with ribosomes via two conserved peptide motifs in the N-terminal cytosolic domain.3, 4 These functional elements allow the protein to regulate translation, influence protein translocation to the ER, and to indirectly interact with the key regulator of the unfolded protein response (UPR), BiP (Grp78, HspA5), thereby affecting cells' responses to ER stress inducers. We previously characterized SEC62 as a probable target gene in prostate cancer because the copy number of the SEC62 gene is increased in DU145, DU145MN1 and PC3 cells; Sec62 mRNA is the most frequently over-expressed mRNA in prostate cancer samples; and the protein content level of Sec62 is increased in at least DU145 and DU145MN1 cells.5

In the present study, we analyzed Sec62 protein levels in more detail by comparing prostate cancer tissue to normal tissue from the same patient, and to benign prostate tissue from glands without prostate cancer by western blot analysis and immunohistochemical detection. We also used a multitumor tissue array to gain insight into SEC62 expression in other malignancies. Because our current results demonstrate Sec62 protein overproduction to be associated with tumor progression in prostate cancer, we analyzed putative tumor-related functions of the protein and found a contribution of Sec62 to cell migration, not only of prostate cancer cells, but also of cells from other tumor entities, indicating a general role for Sec62 in the ability of tumor cells to migrate.

Material and methods

Cell culture and tissue samples

Cell lines PC3 (DSMZ no. ACC 465), TX3868 (kindly provided by E. Meese),6 A549 (DSMZ no. ACC 107) and HT1080 (DSMZ no. ACC 315) were cultured at 37 °C in DMEM-medium (Gibco Invitrogene, Karlsruhe, Germany) containing 10% FBS (Biochrom, Berlin, Germany) and 1% penicillin/streptomycin (PAA, Pasching, Austria) under humidified environment with 5% CO2. The cell lines DU145 (DSMZ no. ACC 261), BPH01 (DSMZ no. ACC 143) and H1299 (ATCC no. CRL-5803D) were cultured in RPMI1640-medium (PAA) containing the same supplements. Tissue samples were obtained from prostate cancer patients or non prostate cancer patients (control group). Following radical prostatectomy the tissue was dissected by a pathologist, snap frozen and stored at –80°C. Only specimens containing more than 90% tumor in the histological review of the counterparts were used as tumor tissue in the western blot analysis. SEC62 expression in the tumor tissue was compared to the tumor-free tissue from the same prostate and to samples from the control group. Only patient samples with signed informed consent were used after approval of local ethic board.

Protein quantification of Sec62 in prostate tissue and cell culture by Western blot

From each frozen sample three 20 μm slices were suspended in 200 μl lysis-buffer (20 mM Tris-HCl pH 7.4, 150 mM NaCl, 1.5 mM EDTA, 3% glycerol, 1% NP-40) and incubated on ice for 30 min. Then the tissue was pounded with a pistil, centrifuged at 16,800g for 10 min at 4°C and a titration of 20, 50 and 100 μg total protein of each sample was used for western-blot analysis employing an affinity purified rabbit antipeptide antibody directed against the COOH terminal undecapeptide of human Sec62 protein (plus an aminoterminal cysteine)2 and an anti-GAPDH antibody from rabbit (Santa Cruz Biotechnology FL-335). To quantify Sec62 in lysate from cultured cells 2 × 105 cells were used for western-blot analysis employing the described primary antibodies. The primary antibodies were visualized with an ECLTM Plex goat-anti-rabbit IgG-Cy5 conjugate (GE-Healthcare, Munich, Germany) and the Typhoon-Trio imaging system (GE-Healthcare) in combination with the Image Quant TL software 7.0 (GE-Healthcare). The ratio of Sec62 protein to GAPDH was determined. We note that the anti-Sec62 antibody is specific for Sec62 under denaturing as well as native conditions (i.e., western blot and fluorescence microscopy-signals were quenched after silencing of the SEC62 gene; Supporting Information Fig. 1A and 1B).

Figure 1.

Quantification of Sec62 in human prostate tissue. (a) The quantification of Sec62 protein level was performed by SDS-PAGE and western blot analysis. The amount of Sec62 and GAPDH (loading control) was quantified and a ratio of Sec62/GAPDH was calculated. The mean values of these ratios were compared between a control group (n = 33, see Supporting Information Table 2), histological normal tissue from prostate with confirmed cancer and tumor tissue from this tissue (>90% tumor, n = 32, see Supporting Information Table 1). A statistic analysis using an unpaired double-sided t-test showed a significant difference between the groups. (b) Statistical correlation of Sec62 protein content with tumor progression. The tumor patients shown in a, were grouped with respect to the main Gleason value (left panel) and the tumor grading (right panel) and the numbers of patients with a Sec62 overproduction in the tumor vs. normal tissue (≥2) was compared to the number of patients without Sec62 overproduction (<2). The detailed values are given in Supporting Information Table 1 and 2. (c) Immunohistochemical detection of Sec62 in prostate cancer tissue. Comparing the immunohistochemical signal intensity of Sec62 in increasing Gleason growth patterns, declining weak signal intensity and a tendency towards a medium or strong signal intensity could be observed. (d) Demonstration of the distribution of color intensity of Sec62 in tissue microarrays according to different Gleason patterns. Primary and secondary frequent Gleason growth pattern of samples from 225 patients were analyzed by Immunohistochemistry with respect to Sec62 protein content. The signal intensity of Sec62 tumor cells was classified as weak, medium or strong.

Table 1. Multitissue TMA: Sec62 protein content in normal and neoplastic tissues
inline image

Silencing of gene expression by siRNA

For gene silencing 1.2 × 105 PC3 cells per well of 12-well plates or 2.4 × 105 cells per well of 6-well plates well were seeded in normal culture medium. The cells were transfected with Sec62 UTR siRNA (CGUAAAGUGUAUUCUGUACtt, Ambion, TX), Sec62 siRNA (GGCUGUGGCCAAGUAUCUUtt, Ambion), ERj1 siRNA (CCUCAAUAUUUCUACGUCAtt, Ambion), Sec63 siRNA (GGGAGGUGUAGUUUUUUUAtt, Ambion) or control siRNA (AllStars Neg. Control siRNA, Qiagen, Hilden, Germany) using HiPerFect Reagent (Qiagen, Hilden, Germany) following the manufacturer's instructions. After 24 h, the medium was changed and the cells were transfected a second time. Silencing was evaluated by western-blot analysis. The maximum silencing effect was seen 72 h after the first transfection.

Immunohistochemical analysis of SEC62 expression

For immunohistochemical detection of Sec62 tissue microarray (TMA) samples were arranged with the Tissue Arrayer Model-2Booster (AlphaMetrix Biotech GmbH, Rödermark, Germany) according to the manufacturers' instructions. The TMA layout was designed using TMA Designer 1.6.8 Software (Alphelys, Plaisir, France) defining a 20 × 3 matrix. In total samples from 225 patients were transferred onto blank paraffin wax blocks via punches 1.5 mm in diameter. Each block consisted of preassigned formaldehyde-fixed and paraffin wax-embedded samples from a maximum of 20 patients each, with both primary and secondary frequent Gleason growth pattern and one nonmalignant area diagnosed on the basis of Hemalaun–Eosin stains. Afterwards, the prepared TMA blocks were finished at 37°C and sections of 3 μm thickness were produced using a standard microtome and transferred onto microscope slides (SuperFrost, Menzel-Gläser, Braunschweig, Germany). The slides were dried at 37°C overnight in the incubator. Immunohistochemistry was performed according to standard protocols for detection of Sec62. Briefly, sections were covered with a 1:600 dilution of the primary antibody and incubated at 4°C overnight. Visualisation of antigenic sites was performed using the Dako EnVision™ kit (Hamburg, Germany). The signal intensity of Sec62 tumor cells was classified as weak, medium or strong.

Tissue microarray

A tissue array was created out of 2071 tissue samples from the archives of the Institute of Pathology, University of Basel (Basel, Switzerland). Samples included 1939 primary tumors from 55 different tumor types and subtypes, and 132 samples contained 18 different normal tissues.

Viability, necrosis and proliferation assays

Cell proliferation, viability and necrosis were analyzed with commercially available assays following the manufacturer's instructions. A BrdU-Assay (Roche Diagnostics GmbH, Mannheim, Germany) was used in proliferation assays performed in a 96-well format with 2.5 × 104 cells of the respective cell-line. Viability was measured in 12-well plates with 1 × 105 cells of the respective cell line using the WST-Assay (Roche Diagnostics GmbH), and necrosis was detected with a LDH+-assay (Roche Diagnostics GmbH). All assays were done with the same cells used for the invasion assay (48 h).

Quantitative realtime PCR

Cells were cultured in the medium mentioned above, harvested and mRNA was isolated by using Nucleospin RNA II-Kit (Machery and Nagel, Düren, Germany). Reverse Transcription was performed with Superscript II (Invitrogene) and olgio-dT Primer (Eurofins MWG Operon, Ebersberg, Germany). TaqMan® Gene Expression Assays (Applied Biosystems, Carlsbad, CA) were used for quantitative real-time PCR of Sec61α1 (Hs00273698_m1), Sec61β (Hs00606455_m1), Sec62 (Hs00541815_s1), Sec63 (Hs00273093_m1), GRP170 (Hs00197328_m1) and BiP (Hs99999174_m1) in HT7900 system (Applied Biosystems). Δct-values were calculated using TBP (Hs00427620_m1) as a standard and the values were then normalized to BPH01 (Fig. 4) or untreated PC3-cells (Supporting Information Fig. 3).

Invasion assay

Invasion was tested on a Tumor Invasion System (BD BioCoat, BD, NJ) in matrigel coated 24-well inserts. A sample of 2.5 × 104 cells treated with control siRNA or Sec62 siRNAs was placed on this system in DMEM-medium containing 0.5% FCS. The inserts were set into DMEM-medium with 10% FCS as an attractant. After 48 h, the cells were stained with propidiumiodid and analyzed by fluorescence microscopy. To quantify invasion the described system was used on a 6-well scale. In this case 3.5 × 105 cells were loaded into the inserts. After 48 h invasive cells were stripped away with trypsin/EDTA and counted.

Migration potential analysis

Migration was tested with the BD Falcon FluoroBlok system (BD, NJ) in 24-well inserts. A total of 2.5 × 104 cells treated with control siRNA, Sec62 siRNAs, ERj1 siRNA or Sec63 siRNA were loaded into this system in DMEM-medium containing 0.5% FBS. The inserts were placed in DMEM-medium with 10% FBS as an attractant. After 24, 48, or 72 h the cells were fixed with methanol and stained with propidiumiodid or DAPI and migrating cells were analyzed on the backside of the membrane by fluorescence microscopy. The findings from this approach were compared with the newer technique of real time migration monitoring using the CIM-devices and the xCELLigence DP system (Roche Diagnostics GmbH). In this system, 1.0 × 104 or 2.0 × 104 cells treated either with siRNA (Sec62 or control) or left untreated, then seeded in the upper chamber in the normal culture medium of the respective cell line without FBS. This upper chamber was then placed on the lower part of the CIM-device containing growth medium supplemented with 10% FBS as an attractant or without FBS (negative control). Migration of the cells was followed over a time period of up to 18 h by changes of the impedance signal in a CIM-plate (Roche Diagnostics GmbH) measured on the backside of the membrane and cell growth was monitored in a 16-well e-plate (Roche Diagnostics GmbH) as described for the xCELLigence SP system (Roche Diagnostics GmbH).7

Statistical analysis

For statistical analysis (Fig. 1a) an unpaired double-sided t-test was used. For statistical analysis (Figs. 1b and 1c), the commercially available software Statistical Package for the Social Sciences v. 17.0 (SPSS, Chicago, IL) was used. Comparisons between tumors of different stage and grade were made using the two-sided Fisher's exact test; p values <0.05 were considered statistically significant.

Results

Increased Sec62 protein level is a frequent finding in prostate cancer

In a first step we analyzed Sec62 protein levels by western-blot in prostate tissue from patients without known prostate cancer and compared this to histologically normal prostate tissue from a prostate with confirmed cancer as well as to tumor regions of the same prostate (Fig. 1a, Supporting Information Table 1 and 2 and Fig. 2).The mean values of all groups were found to be significantly different indicating the tumor related up-regulation of Sec62 protein content as well as a possible predisposition of the histologically normal prostate tissue from prostates with confirmed cancer (Fig. 1a). Analyzing the expression data with respect to Gleason score, we found an increased number of patients with at least two-fold Sec62 content in the tumor compared to the histologically normal tissue in more de-differentiated tumors (Gleason score ≥7; 54%; n = 14) as compared to tumors with Gleason score <7 (17%; n = 6). This association became even more evident, when the predominant Gleason pattern in a tumor was used, reaching statistical significance when comparing SEC62 over-expression between tumors with a primary Gleason pattern of 3 vs. >3 (p = 0.03, Pearson's Chi square test, Fig. 1b left panel). A differentiation with respect to tumor grade showed a tendency towards more SEC62 over-expression in pT3b tumors compared with a combined group of pT2c and pT3a tumors, but did not reach statistical significance (p = 0.12, Pearson's Chi spare test, Fig. 1b right panel). These results could also be confirmed by immunohistochemical staining of tumor tissue. Here, too, the intensity of the Sec62 signal increased with higher Gleason pattern (Fig. 1d and 1e). Thus, our data indicate a strong correlation between increasing SEC62 expression level and de-differentiation of the cells although SEC62 expression was inhomogeneous within all groups we observed.

Figure 2.

Effects of silencing of the SEC62 gene in the prostate cancer cell-line PC3. (a) PC3 cells where transfected with Sec62 UTR siRNA or control siRNA in 12-well plates. After 24 h transfection was repeated and after another 24 h the cells were used for the invasion assay. Invasion of siRNA treated cells was tested on a BD BioCoat Tumor Invasion System in matrigel coated 6-well inserts. After 48 h cells were stripped away by trypsin/EDTA and counted. The amount of Sec62 was measured after SDS-PAGE and western-blotting of an aliquot taken from the cells that were seeded on the matrigel inserts. GAPDH was used as a loading control. Cell proliferation was tested with a BrdU-Assay and, viability was tested with WST-Assay. The assays were performed at the same time periods as the invasion assay (48 h). (b) Effects of silencing of the SEC62 gene on Invasion. Invasion was tested on the same system as described above in 24-well inserts. PC3 cells either untreated or treated with siRNAs as described and after 48 h cells were stained with propidiumiodid and analyzed by fluorescence microscopy. (c) Effects of silencing of the SEC62 gene on Migration. Migration was tested in the BD Falcon FluoroBlok system in 24-well inserts. PC3 cells either untreated or treated with siRNAs as described were loaded on this system. After 48 h cells were fixed with methanol and stained with DAPI and analyzed by fluorescence microscopy. (d) Real time migration analysis after SEC62 gene silencing. Cells were pretreated with siRNAs as described in Material and Methods or left untreated and seeded to a CIM-plate or ePlate. Migration was monitored for 24 h in the xCELLigence DP system.

Increased Sec62 protein levels can also be found in other tumor entities

To study Sec62 protein content in other tumors we utilized a multi-tumor TMA containing a total of 2071 tissue samples from the archives of the Institute of Pathology, University of Basel (Basel, Switzerland). This array included 1939 primary tumors from 55 tumor types and subtypes. One hundred thirty two samples from 18 different normal tissues were stained with the Sec62 antibody and analyzed by fluorescence microscopy. Our findings were confirmed by Sec62 detection in about half of the prostate cancer samples on the array (Table 1). A striking finding was that Sec62 immunoreactivity was present in nearly all lung cancer samples on the TMA independent of their histological entity (93–97%), whereas none of the 10 normal lung tissue samples showed a Sec62 signal. Another tumor with a high number of Sec62-positive tumor samples was thyroid carcinoma. Eighty-seven percent of the follicular carcinoma samples and 100% of the papillary carcinoma samples had detectable Sec62 and only 19% of the nonmalignant adenoma samples. Nevertheless, we also saw tumors with a reduced Sec62 signal. For instance, we did not observe any Sec62 signal in testicular cancer samples whereas 88% of the normal testis tissues were Sec62 immunoreactive. A similar finding was seen for pancreas tissue with only 9% of the tumor but 88% of the normal tissue samples showing a Sec62 signal (Table 1). In general, the results of this multi-tissue tumor array suggest that a differential expression of Sec62 is a common finding in various malignancies.

The potential capability of cancer cells to migrate and invade is strongly reduced after SEC62 silencing

One hallmark of tumor cells is their potential to migrate and to grow invasively. Therefore, in our first experiment we compared the invasive behavior of PC3-cells transfected with control siRNA or with Sec62-specific siRNAs in the BD BioCoat Tumor Invasion System. Side-effects caused by siRNA were excluded by the use of two different Sec62 siRNAs in individual experiments. Remarkably, invasion of the cells was reduced to 13% for the Sec62-siRNA treated and 3% for the Sec62-UTR siRNA-treated cells at a residual Sec62 protein level of about 8–12%, respectively (Figs. 2a and 2b). In a next step we tested whether this effect was due to a reduced viability, reduced proliferation or increased rate of necrosis. The viability of the transfected cells was only slightly reduced within the time range of the experiment, proliferation was reduced to 57% (Sec62 siRNA) or 75% (Sec62-UTR siRNA) and necrosis remained within the range that was observed for the control transfected cells (Fig. 2a). Next we used the BD Falcon FluoroBlok Insert System without matrigel coating and found that cell migration was also affected by SEC62 silencing (Fig. 2c). The effect of siRNA mediated gene silencing on cell migration was also studied using a real-time method, the xCELLigence system. Silencing efficiency was evaluated by western-blot analysis. Under these experimental conditions, silencing of the SEC62 gene also led to a significant reduction of the migratory abilities of PC3 cells compared to untreated PC3-cells or PC3-cells transfected with control siRNA while cell viability was hardly affected (Fig. 2d).

Because Sec62 overproduction was also found in malignancies other than prostate cancer, we extended our studies to other cancer entities. We additionally tested HT1080 (fibrosarcoma), TX3868 (glioblastoma), A549 (lung carcinoma) and H1299 (nonsmall cell lung carcinoma) cells in the same tumor invasion system. We found that these cell-lines also showed a marked reduction in invasive growth after SEC62 silencing which again was not due to reduced viability, proliferation rate or a higher necrosis rate as compared to control siRNA transfected cells (Fig. 3a). The effect of siRNA-mediated gene silencing on cell migration was also studied in the xCELLigence system. Under these experimental conditions, silencing of the SEC62 gene also led to a significant reduction of the migratory abilities of all tested tumor cells, whereas cell viability was hardly affected (Fig. 3b). Summarizing these results, we can state that Sec62 is necessary for cell migration and invasion.

Figure 3.

Effects of SEC62 gene silencing in various cancer cell-lines. (a) Cells of different tumor entities were tested as described in the legend to Fig. 2a. (b) Real-time cell migration analysis was performed with cells of different tumor entities as described in the legend to Fig. 2d.

The over-expression of SEC62 gene does not correlate with expression of genes coding for other ER-proteins or with the ER-stress response

To test, if the over-expression of SEC62 in some tumor cell lines was specific for this gene or due to a general ER expansion in these cell lines, we analyzed the mRNA content of prostate cancer (PC3, DU145), lung cancer (A549, H1299), glyoblastoma (TX3868) and fibrosarcoma (HT1080) cell lines and compared the content of mRNAs coding for several ER-membrane proteins (Sec61a1, Sec61b, Sec62, Sec63) or ER-luminal proteins (GRP170, BiP). There was at least twofold over-expression of SEC62 in DU145-, A549-, TX3868- and HT1080 cells whereas PC3- and H1299-cells showed a similar mRNA content compared to the nontumor BPH01-cells (Fig. 4a). Furthermore, none of the five other genes, including the UPR-controlled genes SEC61A1, SEC61B, GRP170 and BIP, showed a similar pattern of over-expression.

Figure 4.

Realtive mRNA content coding for different ER-membrane or ER-lumenal proteins in cell lines representing different tumor entities. (a) mRNA of the different cell lines was isolated, reverse transcribed and analyzed by quantitative real-time PCR. The mRNA content was calculated relative to TBP and then normalized to the non tumor cell line BPH01. (b) Effects of silencing of the ERJ1 and SEC63 gene on Migration. Migration was tested in the BD Falcon FluoroBlok system in 24-well inserts. PC3 cells treated with control siRNA, ERj1 siRNA or Sec63 were loaded on this system. Silencing of SEC62 was used as a positive controle in this assay. After 48 h cells were fixed with methanol and stained with DAPI and analyzed by fluorescence microscopy.

Although these data did not at all suggest a link between expression of the SEC62 gene and the ER-stress response, we asked if ER-stress leads to an over-expression of the SEC62 gene and if silencing of the SEC62 gene leads to ER-stress. We observed that the up-regulation of Sec62 protein content does not occur within the framework of UPR (Supporting Information Fig.3a) and that the silencing of SEC62 does not lead to an UPR induction in the cells (Supporting Information Fig. 3b). This leads to the conclusion that our findings on SEC62 over-expression in some tumor cells are based on a specific and functionally relevant phenomenon in some tumor entities and cannot be due to general ER expansion or UPR.

Gene silencing of ERJ1 or SEC63 does not influence the migration potential of PC3-cells

We also analyzed the influence of ERJ1- and SEC63-gene silencing on the ability of PC3-cells to migrate to see, whether the reduced migration after SEC62 silencing is specific for this gene. PC3-cells treated with ERj1 siRNA or Sec63 siRNA did not show a reduced migration compared to control siRNA transfected cells or untransfected cells (Fig. 4b), but again treatment with Sec62 siRNA dramatically reduced the migration potential of the cells. This shows that in contrast to ERj1 or Sec63, Sec62 indeed is necessary for cell migration. On the other hand, the over-expression of SEC62 alone is not sufficient to promote cell migration as DU145 showed the highest Sec62 mRNA content of all analyzed cell lines (Fig. 4a) but did not migrate in our assay (data not shown).

Discussion

Over-expression of the SEC62 gene in various tumors

In previous studies we identified a common amplification unit encompassing 3q25-q26 in about 30% of prostate carcinomas.5, 8, 9 Frequent involvement of chromosome 3q in prostate cancer was also reported by others, who discovered gains at 3q in 45% of androgen-independent carcinomas.10 Furthermore, a microarray-based study confirmed our results in primary prostate tumors.11 Determination of the copy number status of the relevant genes in 22 prostate cancer samples revealed the SEC62 gene to be most frequently increased in 50% of the samples.5 Comparison of the matched tumor and adjacent normal tissue samples showed a significant up-regulation of SEC62 expression in tumor tissues. Furthermore, it is of particular note that the histological normal prostate tissue from cancerous prostate glands also yielded a significantly higher SEC62 expression compared with normal prostate tissue from noncancerous prostate glands. Although it cannot be totally excluded that a microscopically undetected contamination of the normal samples by small cancerous foci may be responsible for this observation our findings underline the concept of a “field cancerization” in the prostate gland as was already indicated by other genetic changes in the prostate and other organ systems.12, 13 The putative relevance of Sec62 in prostate cancer development can now be confirmed at the protein level. Our results also demonstrated that the increased SEC62 expression is independent of the expression of genes, coding for other ER-membrane proteins as well as ER-stress indicators like BiP or GRP170. Furthermore, a significant over-expression of the SEC62 gene was also observed in other tumor types, such as the different subtypes of tumors of the lung and the thyroid gland. At least in the case of lung cancer, this observation is consistent with the fact that the 3q25-q26 region was previously linked to malignancies and that SEC62 over-expression was observed in microarrays for cancer versus normal samples (http://www.genecards.org).14

Sec62 protein as a potential diagnostic tool in pathology

Much work has been done in the past years to find new biomarkers to supplement or replace PSA.15 The major agreement from the published data on new biomarkers is that in future a combination of different markers will be used for screening.16 Although we observed increased levels of Sec62 in the tumor-containing regions of the prostate in cancer patients by western blot analysis as well as by immunohistochemistry, Sec62 does not appear to be a robust marker for the diagnosis of prostate cancer. One reason being that increased Sec62 was observed in only 50% of prostate cancer patients. Another reason is that the levels of Sec62 appear to be too variable in both the nontumor tissue of prostate cancer and in patients without cancers. This view is also supported by the results of the multi-tissue tumor array. Nevertheless, Sec62 protein content may turn out to be a useful candidate marker for prognostic purpose, since we found an increased number of patients with high Sec62 protein content in more de-differentiated tumor samples. Furthermore, based on the same array we found some indication that Sec62 protein may be a relevant marker in the diagnosis of lung or thyroid cancer and, therefore, we will evaluate these preliminary results in a future study.

SEC62 as a potential target gene in tumor therapy

Intriguingly, silencing the SEC62 gene led to a reduction in the migratory and invasive potential of the androgen-independent prostate cancer cell line PC3 as well as of various other tumor cell lines, irrespectively of whether or not they have a significantly overproduced Sec62. At the same time, viability of the tumor cells was largely unaffected, suggesting that SEC62 may not be an essential gene. The latter is consistent with the fact that the Sec62 protein's interaction partner the in the ER membrane, Sec63, was suggested to be absent from affected liver cells in patients with polycystic liver disease.17 The question is how Sec62 protein may play a role in cell migration and cell invasion. Although the function of Sec62 (and as a matter of fact Sec63) in mammalian cells is still elusive, the two proteins can be expected to play a role in the early steps of the biogenesis of certain secretory and plasma membrane proteins. Along these lines, a reduction in the level of certain plasma membrane or secretory proteins that are involved in processes such as migration and invasion would not be unexpected. However, alternative interpretations such as a direct or indirect role of Sec62 in the unfolded protein response (UPR) must be considered. Our current focus is to identify potential substrate proteins of Sec62 in proteomic analyses. As the therapeutic use of siRNAs comes into focus now18 and strategies for the delivery problem are developed19, 20 the effect of SEC62 silencing on tumor cells should be further investigated to possibly develop a novel gene silencing strategy for therapeutic intervention in androgen-independent prostate cancer and also cancers of the lung and the thyroid.

Sec proteins in oncology

Sec62 is not the only Sec protein linked to cancer. Most recently, Sec61γ was linked to glioblastoma.21SEC61γ is overexpressed in almost 80% of glioblastoma multiforme but not in lower grade gliomas. Furthermore, siRNA-mediated silencing of the SEC61γ gene in glioblastoma multiforme cells led to growth suppression and apoptosis. The latter may be attributed to the probability that the SEC61γ gene is essential. Thus, components of the ER's protein transport machinery may become limiting in certain tumor cell types under conditions of increased proliferation and therefore, may be essential for proliferation or more specialized activities, such as cell migration or stress resistance. On the other hand, the SEC63 gene was among the most frequently-mutated genes in hereditary nonpolyposis colorectal cancer-associated small-bowel cancer, and sporadic cancers with frequent microsatellite instability.22, 23 Here tumorigenesis appears to be associated with a loss of Sec63 function. Therefore, alternative explanations must be considered when investigating the link between Sec proteins and cancer, such as additional functions of the Sec proteins.21

Acknowledgements

The excellent technical assistance of Ms. Monika Barth, Ms. Anika Müller, Ms. Katharina Kopsch and Ms. Eva Schmitt is gratefully acknowledged. This work was supported by joint grants from the Deutsche Krebshilfe (107381/107390) and Stiftung Europrofession to G.U., R.Z. and B.W., by a HOMFOR-grant to G.U. and R.Z., and by grants from the Deutsche Forschungsgemeinschaft to J.D. (FOR967) and R.Z. (SFB 530, C1).

Ancillary