Abstract
Prothymosin α (ProTα) is a nuclear polypeptide of great biological and, possibly clinical, importance, because its expression levels have been associated with early diagnosis/prognosis of human cancer. It is therefore interesting to raise easily available and cost-effective antibodies that would be applied to develop reliable ProTα immunodiagnostics. In this study, New Zealand white rabbits and laying hens were parallel immunized against intact ProTα or the synthetic fragments ProTα[1-28], ProTα[87-109], and ProTα[101-109], all conjugated to keyhole limpet hemocyanin (KLH). The corresponding antibodies G and Y were immunochemically evaluated in parallel with ELISA and Western blot systems and applied to fluorescence immunocytology experiments using various cancer cell lines and normal cells. The antibody G raised against ProTα[101-109]/KLH had excellent functional characteristics in the Western blot and immunocytology experiments, where the fluorescent signal was almost exclusively shown in the cell nucleus independently of the cells assayed. The above antibody has been applied to preliminary IHC staining of human cancer prostate tissues, leading to a high percentage of clearly and intensively stained nuclei in the adenocarcinoma tissue; this antibody can be further used in cancer tissue immunostaining and in research concerning the role of ProTa in tumorigenesis.
Keywords
P
Although its biological role has not been completely elucidated, literature points toward a dual role for ProTα: an extracellular one, associated with cell-mediated immunity (Cordero et al. 1997; Skopeliti et al. 2006), and an intracellular one, related to cell proliferation and apoptosis (Bustelo et al. 1991; Rodriguez et al. 1998; Jiang et al. 2003). Increased intracellular expression of ProTα, which has been thought to be an oncoprotein (Karapetian et al. 2005; Kobayashi et al. 2006), has been observed in several types of human cancer. Recently, several groups have applied IHC techniques and reported that ProTα is overexpressed in various cancers, e.g., gastric (Leys et al. 2007), prostate (Suzuki et al. 2006), and thyroid (Letsas et al. 2005) cancer, and might therefore be considered as an intracellular tumor biomarker.
Elucidation of the putative diagnostic and/or prognostic significance of ProTα in human cancer would be greatly facilitated by an easily available and cost-effective antibody for the polypeptide that could be further applied to develop reliable ProTα–in vitro immunodiagnostics for routine use. On the other hand, before its application in disease immunodiagnosis, any antibody for ProTα should be carefully characterized in terms of its specificity and its ability to recognize ProTα fragments as well, because intact ProTα might be processed inside the cell by various enzymes, such as asparaginyl endopeptidase (Sarandeses et al. 2003) or caspases (Enkemann et al. 2000a; Evstafieva et al. 2000, 2003), leading to different intracellular ProTα fragments that may retain ProTα immunoreactivity but may differ in their biological functions from the intact polypeptide.
In this study, we developed different antibodies for ProTα using different immunogens and different host animals. More specifically, we developed antibodies against intact, native ProTα of bovine origin or against the synthetic fragments ProTα [1-28], ProTα[87-109], and ProTα[101-109], all conjugated to the carrier protein keyhole limpet hemocyanin (KLH). The N-terminal fragment ProTα[1-28] and the C-terminal fragments ProTα[87-109] and ProTα[101-109] were selected as antigens, because the antigenic determinants of a polypeptide molecule are usually located in its N and/or C termini. Moreover, the N-terminal fragment ProTα[1-28] is identical to the bioactive peptide Tα1, which might be an endogenous biomolecule resulting from the intracellular proteolysis of ProTα by asparaginyl endopeptidase (Sarandeses et al. 2003). On the other hand, the peptides ProTα[87-109] and ProTα[101-109] are located in the C-terminal area ProTα[88-108], which presents relatively high hydrophilicity and mobility indices and is therefore likely to have high antigenicity, according to a previous theoretical study (Costopoulou et al. 1998). Another, practical reason for selecting these peptides as candidate antigens is that they both have a lysine residue in their N terminus, and it is therefore expected that, using the glutaraldehyde method (Avrameas 1969), they will be linked to the carrier protein KLH mainly through the Nα- and the N∊-amino groups of their N termini, thus leaving their C termini to be exposed unmodified to the B lymphocytes of the immunized animal.
The above immunogens were parallel administered to New Zealand white rabbits and laying hens, so that antibodies G and Y were developed, respectively. Laying hens were previously immunized against ProTα/KLH (Klimentzou et al. 2006) in an effort to apply the so-called “IgY technology” (Schade et al. 2001) to ProTα, because the avian immune system is expected to react better against a highly conserved mammalian polypeptide (Tini et al. 2002; Schade et al. 2005), such as ProTα; in this study, a whole series of immunogens (i.e., ProTα/KLH, ProTα[1-28]/KLH, ProTα[87-109]/KLH, ProTα[101-109]/KLH) was used for raising the corresponding antibodies Y. The antibodies G and Y thus developed were parallel evaluated in ELISA and Western blot systems and then applied to fluorescence immunocytology experiments. The antibody showing the highest efficiency in Western blot and especially in fluorescence immunocytology (anti-ProTα[101-109]/KLH) was further applied to preliminary IHC staining of human cancer prostate tissues.
Materials and Methods
Materials
ProTα (generous gift of J. Czarnecki, PhD, Vienenburg, Germany, and M. Pesic, MD, Bad Harzburg, Germany) was isolated from bovine thymus. Thymosin β4 (Tβ4) was synthesized, purified, and characterized as previously described (Zikos et al. 2003). All laboratory solvents and chemicals were purchased from Merck (Darmstadt, Germany) or Sigma (St. Louis, MO) and they were analytical grade.
Peptide Synthesis
ProTα[1-28] (=Tα1), ProTα[87-109], and ProTα[101-109] were synthesized by the Fmoc-solid phase peptide synthesis following a protocol that has been previously described (Zikos et al. 2003). The peptides synthesized were purified on a Waters 600E HPLC System (Waters; Milford, MA) by semipreparative reverse phase-high performance liquid chromatography (RP-HPLC) using a 250 × 12.7-mm (ID) Nucleosil 7 C18 column (Macherey Nagel; Dueren, Germany). A solvent system consisting of 0.05% trifluoroacetic acid (TFA) in water (Solvent A) and 60% CH3CN in solvent A (Solvent B), a flow rate of 8 ml/min, and linear gradients from 100% A to 55% A in 52 min (ProTα[1-28]) and from 100% A to 60% A in 47 min (ProTα[87-109]) was applied. Crude ProTα[101-109] was of high purity, and further purification was not necessary. The purified peptides were characterized by analytical RP-HPLC on a Waters 616 (996 PDA detector) HPLC system using a 250 × 4.6-mm (ID) LiChrospher RP C18 column (5-μm particle size; Merck). A solvent system consisting of 0.05% TFA in 0.1 M NaCl (Solvent A) and 0.05% TFA in CH3CN (Solvent B), a flow rate of 1.0 ml/min, and linear gradients from 100% to 40% A in 19 min (ProTα[1-28]) and from 100% to 80% A in 19 min (ProTα[87-109], ProTα [101-109]) was applied. Peaks were detected spectrophotometrically (220 nm). The purified peptides were also characterized by electrospray ionization-mass spectrometry (ESI-MS) on an Esquire 3000 plus ion trap mass spectrometer (Bruker-Daltonics; Bremen, Germany). Better overall yield and higher purity were obtained in the order ProTα[101-109], ProTα[87-109], ProTα[1-28].
Immunization of Animals
New Zealand white rabbits were intradermally or subcutaneously injected at many sites on their back (Vaitukaitis 1981) with ProTα, ProTα[1-28], ProTα [87-109], or ProTα[101-109] conjugated to KLH (ProTα/KLH, ProTα[1-28]/KLH, ProTα[87-109]/KLH, or ProTα[101-109]/KLH). Conjugation was performed according to the glutaraldehyde method, as previously reported for the N-terminal fragment [1-14] of Tβ4 (Livaniou et al. 1992). The immunogens were injected as emulsions (1:1, v/v) in complete Freund's adjuvant. The animals were boosted initially after 6 weeks and subsequently every 4 weeks. Blood was collected 2 weeks after each booster injection. Antisera (source of antibodies G) were obtained with low-speed centrifugation of whole blood and stored at 235C.
Laying hens were immunized with ProTα/KLH, ProTα[1-28]/KLH, ProTα[87-109]/KLH, or ProTα[101-109]/KLH (same products as above) as previously described for ProTα/KLH (Klimentzou et al. 2006). The antibodies Y were isolated from egg yolk as previously reported (Klimentzou et al. 2006). Care of the animals was in accordance with the corresponding European legislation.
Characterization of Antibodies With ProTα-ELISAs
Titration of antibodies was performed in a titer-ELISA system as previously described (Costopoulou et al. 1998; Klimentzou et al. 2006). The ability of antibodies to discriminate between native ProTα and synthetic fragments of it was performed by comparing the corresponding ELISA displacement curves, as previously described for antibody Y (Klimentzou et al. 2006).
Application of Antibodies in Western Blot Analysis of Cell and Nuclear Lysates
Cell Culture: Preparation of Total Cell and Nuclear Lysates
HeLa cells, PANC-1 cells, and human fetal fibroblasts were cultured in D-MEM (PAA Laboratories; Pasching, Austria), supplemented with 10% FCS (Biochrom KG; Berlin, Germany), at 37C in a 5% CO2 incubator.
The culture medium was discarded, and the cell monolayers were washed with 0.9% NaCl. Trypsin/EDTA [0.05%/0.02% (w/v) in PBS (0.01 M PBS, pH 7.4) free of Ca2+ and Mg2+; Biochrom AG] was added, and the cells were incubated for 10 min at 37C. The cells were resuspended in 0.9% NaCl and centrifuged (1500 × g for 10 min). For preparing (total) cell lysates, the supernatant was discarded, and the cells were resuspended in single-detergent cell lysis buffer (0.05 M Tris-HCl, pH 8.0, 0.9% NaCl, 0.02% NaN3, 100 μg/ml PMSF, 1 μg/ml aprotinin, 1% NP-40), incubated on ice for 10 min, and centrifuged (1500 × g for 10 min). The supernatant was transferred to a microtube and stored at 4C. For preparing nuclear lysates, after resuspension in 0.9% NaCl and centrifugation, the supernatant was discarded, and the cells were resuspended in hypotonic solution (0075 M KCl) and incubated on ice for 15 min. Samples of the suspension were checked on microscope to ensure that the cell membrane was disrupted and the suspension was centrifuged (1500 × g for 10 min). The supernatant was discarded, the cell nuclei were resuspended in the cell lysis buffer, and the above described procedure was followed.
Western Blot Analysis of Cell and Nuclear Lysates
Cell lysates (aliquots corresponding to 60 or 30 μg total protein), as well as nuclear lysates (60 or 30 μg total protein) from HeLa cells, PANC-1 cells, and human fetal fibroblasts, along with a ProTα solution (10 ng) as positive control, were subjected to SDS-PAGE and electrotransferred on a nitrocellulose membrane (ProTran BA83; Schleicher and Schuell, Dassel, Germany), which was preactivated according to a previously described protocol (Papamarcaki and Tsolas 1994). Before the electrotransfer, the SDS-PAGE gel and the preactivated membrane were equilibrated for at least 90 min in the electrotransfer buffer (0.02 M CH3COONa, pH 4.5). After electrotransfer, the membrane was blocked with blocking buffer A (1% BSA, 20% FCS, and 0.05% Tween 20 in PBS) for 1 hr at room temperature on a shaker, washed twice with PBS (for 5 min each time), and incubated with rabbit antibody G (anti-ProTα/KLH, anti-ProTα[1-28]/KLH, anti-ProTα[87-109]/KLH or anti-ProTα[101-109]/KLH antiserum, diluted 1:200 with diluting buffer A: 0.2% BSA, 2% FCS, and 0.05% Tween 20 in PBS) or a solution of hen antibody Y (anti-ProTα/KLH, 25 μg/ml in diluting buffer A) for 1 hr at room temperature. The membrane was washed three times with PBS (for 5 min each time) and incubated for 30 min at room temperature with goat anti-rabbit IgG/horseradish peroxidase (HRP; Sigma), diluted 1:2000 with diluting buffer A, or rabbit anti-hen IgY/HRP (Chemicon; Temecula, CA), diluted 1:3000 with diluting buffer A, respectively. The membrane was washed three times with PBS (for 5 min each time), and the protein bands were visualized with chemiluminescence (ECL kit; Amersham Biosciences, Uppsala, Sweden), according to the manufacturer's instructions.
Control SDS-PAGE gels were run in parallel with those used in the electrotransfer; protein bands in these gels were stained with Coomassie blue R250.
Application of Antibodies to Fluorescent Immunocytology
HeLa, PANC-1, MCF-7, MDA-MB-231 cells, and human fetal fibroblasts were grown on special object slides and fixed with 4% paraformaldehyde for 30 min at room temperature. The slides were washed twice with PBS (for 2 min each time), incubated with Triton X-100 (0.5% in PBS) for 10 min at room temperature, and washed three times with PBS (for 5 min each time). Blocking was performed by incubating with blocking buffer A for 1 hr at room temperature. The slides were washed three times with PBS (for 5 min each time) and incubated with rabbit antibody G (anti-ProTα/KLH, anti-ProTα[1-28]/KLH, anti-ProTα[87-109]/KLH, or anti-ProTα[101-109]/KLH antiserum, diluted 1:400 with diluting buffer A) or hen antibody Y (anti-ProTα/KLH, 30 μg/ml, in diluting buffer A) for 1 hr at room temperature. The slides were washed three times with PBS (for 5 min each time) and incubated either with donkey anti-rabbit-IgG/Cy3 (Fa. Dianova; Hamburg, Germany), diluted 1:500 with diluting buffer A, for 1 hr at room temperature, or with rabbit anti-hen IgY/biotin (Chemicon), diluted 1:2500 with diluting buffer A, for 1 hr at room temperature and then, after washing, with streptavidin-FITC, 0.5 μg/ml in diluting buffer A, for 30 min at room temperature, respectively. The slides were washed three times with PBS (for 5 min each time), incubated with an appropriate solution for DNA staining [YOPRO and TOPRO, respectively (Molecular Probes; Leiden, The Netherlands), diluted 1:2000 and 1:500 with diluting buffer A] for 5 min at room temperature, and washed once more with PBS. Finally, they were mounted with MOWIOL (Calbiochem; San Diego, CA) and observed with a confocal microscope (Leica TCS SP; Leica Microsystems, Bensheim, Germany). Preimmune rabbit serum or hen immunoglobulins were used as a negative control instead of the specific antibody G or Y, respectively, at the same dilution. Moreover, anti-ProTα[101-109]/KLH antiserum (1:400) preincubated (2 hr) with an aqueous solution of ProTα (10 μg/ml, final volume) was used as a negative control in “blank” experiments.
Application of the Antibody G Against ProTα[101-109]/KLH to IHC Staining of Cancer Prostate Tissues
Tissue sections (5 μm) cut from paraffin blocks were dewaxed by incubation in xylene for 30 min at 55-60C (waterbath) and then for 5 min at room temperature. After immersion in absolute ethanol, any endogenous peroxidase activity was blocked by incubation with a solution of 1% H2O2/30% MeOH for 30 min; the tissue sections were rehydrated by passing through a graded series of ethanol and water mixtures (96-70%, 2 min) and distilled water (5 min). Microwave pretreatment for reactivating antigenicity was carried out in 0.01 M citrate buffer, pH 6.0, at 180 (4 min), 360 (4 min), and 480 W (2 min). After treatment with blocking solution B (0.2% BSA and 20% FCS in PBS-T, i.e., 0.01% Triton X-100 in PBS), tissue sections were incubated with the anti-ProTα[101-109]/KLH antiserum, diluted 1:100 with diluting solution B (0.2% BSA and 2% FCS in PBS-T), for 1 hr at room temperature, washed twice with PBS-T (for 5 min each time), and incubated with goat anti-rabbit IgG-biotin (Sigma), diluted 1:300 with diluting solution B, for 40 min at room temperature. Tissue sections were washed twice with PBS-T (for 5 min each time), incubated with streptavidin-HRP (Sigma), 1 μg/ml in PBS for 40 min at room temperature, and washed twice with PBS (for 5 min each time), followed by the addition of the chromogen (diaminobenzidine; Sigma). Finally, the slides were counterstained with Harry's hematoxylin, dehydrated by passing through a graded series of ethanol and water mixtures (70%, 96%, absolute ethanol, 2 min), cleared with xylene (2 min), and mounted with coverslips using Entellan. Preimmune serum, diluted 1:100 with diluting solution, as well anti-ProTα[101-109]/KLH antiserum (1:100) preincubated (2 hr) with an aqueous solution of ProTα (10 μg/ml final volume), were used as negative controls in “blank” experiments.
Results
Characterization of Antibodies With ProTα-ELISAs
Best titers were obtained with the antibody G against ProTα[1-28]/KLH and ProTα[101-109]/KLH (titer of the corresponding antisera: 1:7500). The anti-ProTα/KLH and anti-ProTα[87-109] antisera gave lower titers, 1:4000 and 1:1500, respectively. The anti-ProTα/KLH antibody Y gave a titer of 1.7 μg/ml, whereas the anti-ProTα[1-28]/KLH, anti-ProTα[87-109]/KLH, and anti-ProTα[101-109]/KLH antibody Y gave a very low ratio of specific to background signal, and no titer value could be estimated for them.
The anti-ProTα[101-109]/KLH antibody G could recognize native ProTα and both of the C-terminal synthetic fragments tested, whereas it did not recognize Tα1 (Figure 1). Similar results were obtained with the anti-ProTα[87-109]/KLH antibody G. The antibody G against ProTα[1-28]/KLH could recognize both native ProTα and the N-terminal synthetic fragment but could not recognize—as expected—the C-terminal synthetic fragments. The antibodies G and Y raised against ProTα/KLH gave similar results with those previously reported (Klimentzou et al. 2006). As expected, none of the above antibodies cross-reacted with structurally irrelevant peptides, such as Tβ4, a peptide of the β-thymosin family.
ELISA displacement curves obtained with the anti-prothymosin a (ProTα)[101-109]/keyhole limpet hemocyanin (KLH) antibody G in the presence of increasing concentrations (100 ng/ml to 100 μg/ml) of ProTα, ProTα[1-28], ProTα[87-109], and ProTα[101-109]. Symbols corresponding to the intact polypeptide and the different synthetic fragments are shown as an inset.
Performance of Antibodies in Western Blot Analysis of Cell and Nuclear Lysates
Among all the antibodies applied to the Western blot experiments (i.e., antibody G against ProTα/KLH, ProTα[1-28]/KLH, ProTα[87-109]/KLH, or ProTα [101-109]/KLH, and antibody Y against ProTα/KLH), best results, i.e., strongly positive protein bands corresponding to a molecular mass of ∼11 kDa with negligible nonspecific signal, were obtained with the antibody G against ProTα[101-109]/KLH. Using the anti-ProTα [101-109]/KLH antibody, ProTα was detected in aqueous solutions (positive control) at amounts ranging from 1 μg to 10 ng. Moreover, ProTα could be detected in cell lysates of HeLa and PANC-1 cells (Figure 2), as well as in the corresponding nuclear lysates. According to the band intensities, nuclear lysates are enriched in ProTα, in comparison with the corresponding total cell lysates (Figure 2). HeLa cells seemed to contain the highest amounts of ProTα. No band corresponding to ProTα could be detected in the lysates of human fetal fibroblasts, which may probably be attributed to the comparatively lower amounts of ProTα in these cells.
Application of Antibodies to Fluorescence Immunocytology
The antibodies G and Y against ProTα/KLH and the antibody G against ProTα[1-28]/KLH led to a diffused immunostaining of both the cell nucleus and the cytoplasm; however, a moderate but always present background signal did not allow us to come to solid conclusions concerning specificity of the immunostaining. Similar results were obtained with the antibodies G against ProTα[87-109]/KLH, except that a more profound immunostaining of the cell nucleus was observed in most of the slides. Excellent immunostaining data (i.e., very low background signal and very good precision—across samples and across days) were obtained with the antibodys G against ProTα[101-109]. As shown by using these antibodies in confocal microscopy (Figure 3), ProTα-like immunoreactivity was observed in all cells tested (HeLa, PANC-1, MCF-7, MDA-MB-231 cells, and human fetal fibroblasts); moreover, ProTα-like immunoreactivity was localized almost exclusively within the cell nucleus, independently of the cell type assayed.
Application of the Antibody G Against ProTα[101-109]/KLH to IHC Staining of Cancer Prostate Tissues
Fifteen tissue sections from eight radical prostatectomies for prostatic adenocarcinoma were selected, including the site of carcinoma and adjacent non-neoplastic tissue, normal or hyperplastic. The Gleason's grade ranged from 6 to 7. Cells were considered positive for ProTα immunoreactivity when their nucleus was stained brown, regardless of the stain intensity. Nuclear positivity was assessed in normal, benign hyperplastic, and malignant epithelial cells.
The basal and luminal cells of normal and hyperplastic glands were positive with diffuse, but alternating, nuclear positivity. Stain intensity of the malignant epithelial cell nuclei looked similar to that of the non-neoplastic epithelium. In all carcinomas, ProTα expression was observed in .50% of the cells in continuity, regardless of Gleason's grading (Figures 4A and 4B).
ProTα was also expressed in the nucleus of endothelial, smooth muscle cells, fibroblasts, and lymphocytes of the follicular center and the mantle zone of the lymphoid follicles in all sections studied.
Discussion
Intact ProTα and appropriately selected synthetic fragments of the molecule were conjugated to KLH and administered to rabbits and hens to develop antibodies G and Y, respectively, following previously presented methodologies (Costopoulou et al. 1998; Klimentzou et al. 2006). Characterization of antibodies G and Y raised against ProTα/KLH in ProTα-ELISAs and dotblots has been recently reported and commented on by our group (Klimentzou et al. 2006).
The antibody G raised against ProTα[1-28]/KLH, ProTα[87-109]/KLH, and ProTα[101-109]/KLH, as those previously developed against similar immunogens (Costopoulou et al. 1998), showed moderate titers for the ProTα molecule, which are, however, at least comparable to those thus far reported in the literature (Yialouris et al. 1988; Loidi et al. 1997). On the contrary, the antibody Y raised against ProTα[1-28]/KLH, ProTα[87-109]/KLH, and ProTα[101-109]/KLH showed a very low specific signal during titer determination. This experimental result, which indicates a further “limit” in our efforts to successfully apply the “IgY technology” to the development of antibodies with the highest possible titer and specificity for ProTα (Klimentzou et al. 2006), might be associated with differences in the ProTα epitopes that are recognized by hen and rabbit B lymphocytes (Carlander et al. 1999). This assumption is supported by our findings according to which the antibody Y raised against ProTα/KLH recognizes, to a much lesser extent, the N-terminal fragment ProTα[1-28] than the antibody G raised against the same immunogen (Klimentzou et al. 2006).
Western blot analysis of cell and nuclear lysates from HeLa cells, PANC-1 cells, and human fetal fibroblasts: PANC-1 cells, cell lysate, 60 μg total protein (Lane 1); PANC-1 cells, cell lysate, 30 μg total protein (Lane 2); HeLa cells, cell lysate, 60 μg total protein (Lane 3); HeLa cells, cell lysate, 30 μg total protein (Lane 4); human fetal fibroblasts, cell lysate, 30 μg total protein (Lane 5); ProTα, 10 ng (Lane 6); PANC-1 cells, nuclear lysate, 30 μg total protein (Lane 7); HeLa cells, nuclear lysate, 60 μg total protein (Lane 8).
Immunolocalization of ProTα in HeLa (
The antibody G against ProTα/KLH, ProTα[1-28]/KLH, ProTα[87-109]/KLH, and ProTα[101-109]/KLH (as well as the antibody Y against ProTα/KLH) gave similar ELISA displacement curves, ranging roughly from 0.1 to 100 μg/ml, in the presence of increasing concentrations of ProTα in solution (Figure 1). In general, mainly because of their moderate titer values, such antibodies are difficult to apply to the development of sensitive immunoassay systems, at least those of the competitive type, in which they should be used at limited amounts. Affinity chromatography purification will not necessarily improve the antibody functional characteristics, because very often critical immunochemical features of low titer antibody molecules are destroyed during the purification process. On the other hand, monoclonal antibodies are usually characterized by rather low affinity for the corresponding antigen, and consequently, they are also difficult to apply to the development of sensitive competitive immunoassays. This is probably true for the few monoclonals for ProTα reported until now in the literature (Staehli et al. 1983; Sukhacheva et al. 2002); as a consequence, few immunoassays have been reported for α thymosins and even fewer have been applied to the analysis of biological samples (Panneerselvam et al. 1987; Tsitsiloni et al. 1994; Costopoulou et al. 1998; Mitani et al. 2000), whereas their quantitative experimental results seem to vary and therefore have been approached with caution by some researchers (Enkemann et al. 2000b). However, provided that their specificity is appropriate, carefully selected antibodies for ProTα may be of great value for certain clinical applications and especially for the immunostaining of tissue samples, in which they can be used in excess.
The antibody G against ProTα/KLH, ProTα[1-28]/KLH, ProTα[87-109]/KLH, and ProTα[101-109]/ KLH and the antibody Y against ProTα/KLH were evaluated in a Western blot system. Optimal specificity was obtained with the antibody G against ProTα[101-109]/KLH; using this antibody, ProTα was detected in aqueous solutions (positive control) at amounts as low as—at least—10 ng; moreover, ProTα could be detected in cell and cell nuclear lysates of HeLa and PANC-1 cells.
It should be noted here that the behavior of ProTα in Western blotting has been an issue of controversy among research groups for many years (Freire et al. 2002; Sukhacheva et al. 2002; Karetsou et al. 2004). According to previous results of ours (Klimentzou et al. 2006), by using antibodies G or Y against ProTα/KLH, no band corresponding to the molecular mass of ProTα (∼11 kDa) could be observed on the Western blot membrane, although a large number of different experimental protocols had been tried. However, by elongating the washing step of the SDS-PAGE gel and the pretreated membrane in the electrotransfer buffer to at least 90 min, so that complete equilibration was ensured before electrotransfer, we were able to visualize a band corresponding to ∼11 kDa, for the first time, in the frame of this study. We strongly believe that the ProTα molecule does behave in a tricky way in the Western blot (Freire et al. 2002; Lal et al. 2005), and the slightest changes in the experimental conditions may greatly affect its immobilization onto the membrane, which is probably the main reason for the often reported poor Western blotting results. However, under strictly adjusted conditions and careful handling, one can test the specificity of antibodies for ProTα with Western blot.
(
The antibody G against ProTα/KLH, ProTα[1-28]/KLH, ProTα[87-109]/KLH, and ProTα[101-109]/KLH (and the antibody Y against ProTα/KLH) was also evaluated in fluorescence immunocytology experiments, in which immunostaining was observed with confocal microscopy. Excellent results, i.e., negligible background signal and high reproducibility, were obtained with the antibody G against ProTα[101-109]/KLH. By using this antibody, ProTα-like immunoreactivity was located almost exclusively in the cell nucleus, where it appeared punctuate but widely dispersed, as previously reported by Enkemann et al. (2000b), who used COS-1 and NIH3T3 cells transfected with genes of epitopetagged ProTα proteins in combination with antibodies against the epitope tags. The intensity of the fluorescent signal was stronger in the cancer cells than in normal ones. Among the cancer cell lines studied, the strongest signal was observed in HeLa cells. This empirical observation was confirmed by the Western blot results. Nuclear localization of the immunostaining indicates that neither intact ProTα nor shorter C-terminal fragments of it are present in the cell cytoplasm, at least not at detectable levels under the experimental conditions used, whereas the presence of Tα1 in the cytoplasm cannot be excluded, because the antibody used does not cross-react with Tα1. On the other hand, it would be interesting to study the immunostaining pattern obtained with a larger number of cancer cell lines, differing, for example, in their invasion potential.
The antibody G against ProTα[101-109]/KLH, which gave the best results in the fluorescence immunocytology experiments, was further used in preliminary IHC experiments. According to the literature, IHC studies of ProTα have been mainly performed by using antibodies raised against Tα1 (Dominguez et al. 1993), which may recognize both intact ProTα and Tα1, thus raising a priori a “specificity issue.” On the other hand, few IHC studies were performed with antibodies raised against C-terminal fragments of ProTα, including the fragment ProTα[101-109] (Costopoulou et al. 1997; Letsas et al. 2005); however, these studies have not provided pre-evaluation data on the specificity of the antiserum used with Western blotting or with single cell immunostaining.
In the frame of this study, the antibody G against ProTα[101-109]/KLH was used for the localization of ProTα in 15 prostate tissue sections obtained from patients undergoing total surgical prostatectomy. Before this, extensive negative control immunostaining was performed, in which anti-ProTα[101-109]/KLH antiserum preincubated with ProTα had been used, which confirmed the specificity of the antiserum (Figures 4C and 4D). Similar experiments with equally good results were also performed by immunocytology (data not shown).
In the only article—to our knowledge—reporting on the immunochemical localization of ProTα in prostatic tissue samples, Suzuki et al. (2006) studied a large number of normal prostate, benign prostatic hyperplasia, and prostate cancer samples. According to the results of that study, ProTα-like immunoreactivity was observed mainly in the cell nucleus, whereas a weak cytoplasmic staining was also observed; this may be attributed to the antibody the authors used (commercially available monoclonal antibody to ProTα, clone 2F11), which, if identical to the one first described by Sukhacheva et al. (2002), recognizes an epitope on the N terminus of the ProTα molecule (aa 1-31) and might, therefore, bind to cytoplasmic Tα1 as well. In addition, according to Suzuki et al. (2006), ProTα-like immunoreactivity increased from normal epithelium to adenocarcinomas and was positively correlated with Gleason's grade as well as with disease clinical stage. In our study, strong nuclear immunostaining was observed in the malignant cells of all adenocarcinomas. The intensity of the staining was similar to that of adjacent hyperplastic glands, but the percentage of positive nuclei was higher in adenocarcinomas. Nevertheless, because of the small number of samples tested, this study cannot directly support, at least at this time, that ProTα-like immunoreactivity increases in prostate cancer tissue compared with the adjacent benign prostate hyperplasia areas.
In summary, in this study, various antibodies G and Y for ProTα were developed and evaluated in an attempt to produce an easily available and cost-effective antibody of the highest possible specificity for the polypeptide. According to the results obtained, the antibody G against ProTα[101-109]/KLH, which could recognize intact ProTα and C-terminal fragments of the molecule and did not recognize Tα1, showed excellent specificity and overall efficiency in Western blot and fluorescence immunocytology experiments. This antibody was applied to the IHC staining of a small number of cancer prostate tissues, leading to a high percentage of clearly and intensively stained nuclei in the adenocarcinomas. This well-characterized, easily available, and cost-effective polyclonal antibody can be further used in cancer tissue immunostaining, aiming eventually at both the establishment of a reliable ProTα immunodiagnostics technique for potential routine use and the study of the role of prothymosin a in tumorigenesis.
Footnotes
Acknowledgements
This work was financially supported by the Programme “Excellence in the Research Institutes supervised by the Greek General Secretariat for Research and Technology.”
The authors thank R. Nordin for excellent technical assistance in the immunocytology experiments, Dr. C. Zikos for advice concerning peptide synthesis, and Dr. A. Beck for the ESI-MS analyses. They also thank Dr. O. Tsitsilonis, Dr. H. Kalbacher, and especially Dr. G.P. Evangelatos for useful discussions, helpful suggestions, and ideas.