Implication of intracellular localization of transcriptional repressor PLZF in thyroid neoplasms

The study conformed to the principles outlined in the Declaration of Helsinki, and was approved by the Ethics Committee of the Faculty of Medicine, Tottori University. All patients provided written informed consent. Thyroid samples were obtained at surgery from patients as follows: normal thyroid (N; n = 4); adenomatous lesion (AL; n = 5); follicular adenoma (FA; n = 2); papillary thyroid carcinoma (PTC; n = 20); anaplastic thyroid carcinoma (ATC; n = 3). We performed immunohistochemical (IHC) staining in all samples, whereas western blot analyses were carried out using four samples of N (cases 1–4), five samples of AL (cases 5–9), and 13 samples of PTC (cases 12–24). We performed IHC staining using four samples of N, five samples of AL, two samples of FA, 20 samples of PTC, and three samples of ATC. The characteristics of the analysed samples are listed in Table 1.

Human thyroid tissues were homogenized in 300 μl of RIPA buffer (1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulphate in PBS with a protease inhibitor cocktail (Roche Molecular Biochemicals, Indianapolis, IN)) and centrifuged at 15000 rpm for 20 min at 4°C. The supernatants were used as the total cell lysates. A aliquot (10 μg protein) of each lysate was subjected to 10% SDS-PAGE, and the separated proteins were electrophoretically transferred to nitrocellulose membranes. The membranes were blocked with Tris-buffered saline containing 0.05% Tween-20 and 5% non-fat dried milk, washed, and incubated with polyclonal primary antibodies against PLZF (sc-22839; Santa Cruz Biotechnology Inc., Santa Cruz, CA; 1:200 dilution) and β-actin (sc-1616; Santa Cruz Biotechnology Inc.; 1:500 dilution). The membranes were then exposed to an anti-rabbit secondary antibody (NA934; Amersham Pharmacia Biotech, Little Chalfont, UK; 1:5000 dilution). After incubation with the secondary antibody, detection of the bound antibodies was performed using enhanced chemiluminescence.

PLZF expression was analysed by IHC staining of thyroid samples. Paraffin-embedded tissue sections (4-μm thickness) were deparaffinized in xylene and rehydrated through a graded alcohol series to deionized water. After microwave antigen retrieval, the endogenous peroxidase activity was blocked with H₂O₂. The sections were then sequentially incubated with the anti-PLZF primary antibody (1:50 dilution) for 12 h at 4°C, and a peroxidase-labelled anti-rabbit secondary antibody (414141 F; Nichirei, Tokyo, Japan) for 30 min at room temperature. After colour development with 3,3′-diaminobenzidine and counterstaining with haematoxylin, the sections were scanned at various magnifications (×100 to × 400) using light microscopy.

The proportions of positively stained cells and the intensity scores were evaluated by two experienced professionals. The Allred scoring system [10] was used for PLZF staining interpretation. The proportion of positively stained cells was rated as follows: 0%, no positive staining; 1, between 0% and 1% positive staining; 2, between 1% and 10% positive staining; 3, between 10% and 33% positive staining; 4, between 33% and 66% positive staining; 5, between 66% and 100% positive staining. In addition to the proportion score, an intensity score was made on the basis of the average intensity of staining as follows: 0, negative; 1, weak; 2, intermediate; 3, strong. The intensity score and proportion score were added to obtain the total score, termed the IHC score, which is referred to as the Allred score [10] and is either 0 or between 2 and 8.

We estimated PLZF expression by western blot analyses. All samples were confirmed to express PLZF (Figure 1A and B). All samples of PTC showed high expression of PLZF (Figure 1B). Conversely, the PLZF expression levels were relatively low in N and AL compared with PTC (Figure 1A). To compare the PLZF expression levels between different membranes, we used the PLZF expression ratios, representing the densitometrically analysed data for each membrane divided by the mean value for N, which was used as an internal control (Figure 1C). The PLZF/actin expression ratio was significantly higher in PTC than in both N and AL (p < 0.05).

We estimated the intracellular distribution of PLZF by IHC staining. In N (Figure 2A), PLZF was mainly localized in the nucleus. The strength of its expression ranged from intermediate to strong. However, the expression of PLZF in the cytosol was faint or absent. The IHC scores were 4–7 in the nucleus and 0 or 2 in the cytosol. In AL (Figure 2B), PLZF was mainly localized in the nucleus, and its strength was intermediate or strong. In the cytosol, the PLZF expression was weak, but slightly stronger than in N. The IHC scores were 5–8 in the nucleus and 2 or 3 in the cytosol. In FA (Figure 2C), PLZF was mainly localized in the nucleus. In the cytosol, the PLZF expression was stronger than in N. The IHC scores were 7 or 8 in the nucleus and 4 in the cytosol.In almost all samples of PTC, the PLZF expression in the nucleus was absent or very weak. In general, the PLZF expression was strong in the cytosol. However, the proportion of positive cells and the intensity of staining varied according to each case. As shown in Figure 2D, the intensity of PLZF staining was very weak or absent and the proportion of positive cells was low. Conversely, as shown in Figure 2E, the intensity of PLZF staining was strong and the proportion was high, despite the same histotype (PTC). In all samples of ATC (Figure 2F), PLZF was localized in the cytosol and the intensity of staining was high, resembling the findings shown in Figure 2E.We clarified the staining patterns of PLZF between benign and malignant thyroid lesions by grading using the IHC scores. In benign lesions, such as N, AL, and FA, PLZF was localized in the nucleus (Figure 3A). In contrast, in malignant lesions, such as PTC and ATC, PLZF was localized in the cytosol (Figure 3B). There was a significant difference between benign and malignant thyroid lesions.Subsequently, we examined PTC in more detail because there were different proportions and intensities of IHC staining in the cytosol. We classified PTC according to CI and LN metastasis. As shown in Figure 4A, the IHC scores tended to gradually increase from CI(-)LN metastasis(-) to CI(+)LN metastasis(-), CI(+)LN metastasis(+), and ATC, albeit without statistical significance. The IHC scores in samples with CI were higher than those in samples without CI (Figure 4B), although the difference was not significant. With respect to LN metastasis, the PLZF expression levels in samples with LN metastasis were significantly higher than those in samples without LN metastasis (Figure 4C).