Valproic acid inhibits growth, induces apoptosis, and modulates apoptosis-regulatory and differentiation gene expression in human thyroid cancer cells.

BACKGROUND: Among the most promising new therapies for thyroid cancer are the histone deacetylase inhibitors. Valproic acid (VA) is an anticonvulsant that inhibits histone deacetylase activity at nontoxic concentrations. We hypothesized that VA would have antineoplastic effects on human thyroid cancer cells. METHODS: We treated 1 papillary and 3 follicular thyroid cancer cell lines with VA (0.5-2 mmol/L) for 24 to 72 hours. Cell proliferation was measured with a cell proliferation assay kit. Annexin V-fluorescein isothiocyanate was used to quantitate cells that were undergoing apoptosis. Quantitative polymerase chain reaction was used to measure expression of apoptosis-regulatory and differentiation genes. RESULTS: VA inhibited growth in all cell lines by 26% to 59% at 48 hours and up to 77% at 72 hours. Nineteen percent to 30% of VA-treated cells underwent apoptosis, compared with 4% to 8% of the control cells. Expression of pro survival genes bcl-2 and bcl-xl was down-regulated by 10% to 60%; expression of the proapoptosis gene bax was up-regulated by 23% to 85%. Sodium-iodide symporter and thyroglobulin messenger RNA expression were up-regulated by 93% to 370% in follicular cell lines but remained unchanged in the papillary cell line. CONCLUSION: VA inhibits growth, induces apoptosis, and modulates apoptosis-regulatory and differentiation gene expression in thyroid cancer cells. These findings suggest that VA may be useful clinically for patients with thyroid cancers of follicular cell origin.

Troglitazone, the peroxisome proliferator-activated receptor-gamma agonist, induces antiproliferation and redifferentiation in human thyroid cancer cell lines.

Troglitazone is a potent agonist for the peroxisome proliferator-activated receptor-gamma (PPARgamma) that is a ligand-activated transcription factor regulating cell differentiation and growth. PPARgamma may play a role in thyroid carcinogenesis since PAX8-PPARgamma1 chromosomal translocations are commonly found in follicular thyroid cancers. We investigated the antiproliferative and redifferentiation effects of troglitazone in 6 human thyroid cancer cell lines: TPC-1 (papillary), FTC-133, FTC-236, FTC-238 (follicular), XTC-1 (Hürthle cell), and ARO82-1 (anaplastic) cell lines. PPARgamma was expressed variably in these cell lines. FTC-236 and FTC-238 had a rearranged chromosome at 3p25, possibly implicating the involvement of the PPARgamma encoding gene whereas the other cell lines did not. Troglitazone significantly inhibited cell growth by cell cycle arrest and apoptotic cell death. PPARgamma overexpression did not appear to be a prerequisite for a response to treatment with troglitazone. Troglitazone also downregulated surface expression of CD97, a novel dedifferentiation marker, in FTC-133 cells and upregulated sodium iodide symporter (NIS) mRNA in TPC-1 and FTC-133 cells. Our investigations document that human thyroid cancer cell lines commonly express PPARgamma, but chromosomal translocations involving PPARgamma are uncommon. Troglitazone, a PPARgamma agonist, induced antiproliferation and redifferentiation in thyroid cancer cell lines. PPARgamma agonists may therefore be effective therapeutic agents for the treatment of patients with thyroid cancer that fails to respond to traditional treatments.

GCMB gene, a master regulator of parathyroid gland development, expression, and regulation in hyperparathyroidism.

BACKGROUND: The glial cell missing gene, GCMB , encodes a transcription factor, which is a master regulator of parathyroid development. We postulated that the GCMB gene might play a role in parathyroid tumorigenesis in hyperparathyroidism. METHODS: We used real-time quantitative reverse transcriptase polymerase chain reaction to study GCMB mRNA expression in parathyroid tissue: normal (n = 3), hyperplastic (n = 16), adenomas (n = 19), and cancers (n = 8). In primary parathyroid culture, the effect of CaCl 2 on parathyroid hormone secretion and GCMB mRNA expression was studied by using enzyme-linked immunosorbent assay and reverse transcriptase polymerase chain reaction, respectively. RESULTS: GCMB mRNA expression was lower in normal (0.4 +/- 0.1, mean +/- standard error of mean) parathyroid glands as compared to adenoma (3.5 +/- 1.7), hyperplasia (3.2 +/- 1.3 primary hyperparathyroidism [n = 11] and 7.6 +/- 4.8 secondary hyperparathyroidism [n = 5]), and cancer (3.6 +/- 1.3) ( P = .001). There was no difference in the level of GCMB mRNA expression between parathyroid adenoma, hyperplasia, and cancer. In primary culture of parathyroid adenoma (n = 9) and hyperplasia (n = 2), parathyroid hormone secretion was increased 2- to 15-fold with low calcium concentration (0.5 to 4.0 mmol/L CaCl 2 from 2 to 6 hours, P < .005). The level of GCMB mRNA expression was down-regulated with lower extracellular CaCl 2 concentration ( P < .005). CONCLUSIONS: GCMB expression is upregulated in abnormal parathyroid glands of hyperparathyroidism and decreases in response to hypocalcemia. The GCMB transcription factor might mediate the effect of calcium on parathyroid cell parathyroid hormone expression/secretion.

Diagnostic and prognostic value of fas and telomeric-repeat binding factor-1 genes in adrenal tumors.

It is often difficult to distinguish histologically between an adrenal cortical cancer and a benign adenoma, or to predict the prognosis of patients with adrenal cortical cancers. In this investigation, we examined whether apoptosis-regulating genes, bcl-xL and fas, and a telomere-related gene, telomeric-repeat binding factor-1 (TRF-1), differ between adrenal cortical cancers and benign adrenal tumors. Tissues from 4 adrenal cortical cancers were compared with 7 normal adrenal tissues, 17 cortical adenomas, 4 cortical hyperplasias, and 20 pheochromocytomas for expressions of bcl-xL and fas by RT-PCR, and for expressions of TRF-1 by real-time quantitative RT-PCR. All benign adrenal tissues expressed both the antiapoptosis gene, bcl-xL, and proapoptosis gene, fas, but the adrenal cortical cancers expressed only bcl-xL and not fas. TRF-1 increased by more than 30-fold in the adrenal cortical cancers, compared with benign adrenal tissues, and inversely correlated with the prognosis of patients with the adrenal cortical cancers. This lack of expression of fas in adrenal cortical cancer may help to distinguish it from benign adrenal tumors. The level of TRF-1 expression may be helpful prognostically for patients with adrenal cortical cancers.

Expression of vascular endothelial growth factor-C in benign and malignant thyroid tumors.

In contrast to vascular endothelial growth factor (VEGF), which stimulates angiogenesis, VEGF-C is thought to stimulate lymphangiogenesis. The role of VEGF-C in thyroid cancer pathogenesis has not been clarified. One might expect a different pattern of VEGF-C expression in the various types of thyroid cancer because of their different means of metastases. In this investigation, we determined whether the differential expression of VEGF-C might explain the different propensity to lymph node metastasis in thyroid cancers. One hundred eleven normal and neoplastic thyroid tissues were analyzed by real-time quantitative PCR. Papillary thyroid cancers had a higher VEGF-C expression than other thyroid malignancies (P < 0.0005 ANOVA). Among the normal thyroid tissues from patients with malignant or benign thyroid diseases, there was no significant difference in VEGF-C expression. Paired comparison of VEGF-C expression between thyroid cancers and normal thyroid tissues from the same patients showed a significant increase of VEGF-C expression in papillary thyroid cancer (1.10 +/- 0.41 vs. 0.70 +/- 0.13; P = 0.001) and a significant decrease of VEGF-C expression in medullary thyroid cancer (0.11 +/- 0.13 vs. 0.78 +/- 0.29; P = 0.001). In contrast, there was no significant difference of VEGF-C expression between cancer and normal tissues in other types of thyroid cancer. In summary, VEGF-C expression is increased in papillary thyroid cancer, compared with paired normal thyroid tissues, but not in other thyroid cancers that are also prone to lymph node metastasis. The lymphangiogenic role of VEGF-C in thyroid cancers therefore appears to be complex and other factors are likely to be also involved.

The heat shock protein 90-binding geldanamycin inhibits cancer cell proliferation, down-regulates oncoproteins, and inhibits epidermal growth factor-induced invasion in thyroid cancer cell lines.

Heat shock protein 90 (HSP90) serves as a chaperone protein and plays a critical role in tumor cell growth and/or survival. Geldanamycin, a specific inhibitor of HSP90, is cytotoxic to several human cancer cell lines, but its effect in thyroid cancer is unknown. We, therefore, investigated the effect of geldanamycin on cell proliferation, oncoprotein expression, and invasion in human thyroid cancer cell lines. We used six thyroid cancer cell lines: TPC-1 (papillary), FTC-133, FTC-236, FTC-238 (follicular), XTC-1 (Hürthle cell), and ARO (anaplastic). We used the dimethyl-thiazol-diphenyltetrazolium bromide assay, a clonogenic assay, an apoptotic assay, and a Matrigel invasion assay. We evaluated oncoprotein expression using Western blots and flow cytometry. After 6 d of treatment with 50 nM geldanamycin, the percent inhibition of growth was 29.4% in TPC-1, 97.5% in FTC-133, 96.7% in FTC-236, 10.8% in FTC-238, 70.9% in XTC-1, and 45.5% in ARO cell lines. In the FTC-133 cell line, geldanamycin treatment decreased clonogenicity by 21% at a concentration of 50 nM; geldanamycin induced apoptosis and down-regulated c-Raf-1, mutant p53, and epidermal growth factor (EGF) receptor expression; geldanamycin inhibited EGF-stimulated invasion. In conclusion, geldanamycin inhibited cancer cell proliferation, down-regulated oncoproteins, and inhibited EGF-induced invasion in thyroid cancer cell lines.

Modulation of tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis by chemotherapy in thyroid cancer cell lines.

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) induces apoptosis in many human cancer cells but not in normal cells. Thyroid cancer cells, however, appear to be relatively resistant to TRAIL-induced apoptosis. We therefore investigated the effect of chemotherapy on TRAIL-induced apoptosis in thyroid cancer cells. We used six thyroid cancer cell lines: TPC-1, FTC-133, FTC-236, FTC-238, XTC-1, and ARO82-1. We used flow cytometry to measure apoptosis, dimethyl-thiazol-diphenyltetrazolium bromide (MTT) assay to measure antiproliferation effects and Western blot to determine the expression of Bcl family proteins. Troglitazone, paclitaxel, geldanamycin, and cycloheximide were used for pretreatment. We used the Student's t test and analysis of variance (ANOVA) for statistical analysis. All thyroid cancer cell lines, except the TPC-1 cell line, were resistant to TRAIL, and growth inhibition was less than 20% at concentration of 800 ng/mL of TRAIL. In both TPC-1 (TRAIL-sensitive) and FTC-133 (TRAIL-resistant) thyroid cancer cell lines, pretreatment with troglitazone, cycloheximide, and paclitaxel enhanced TRAIL-induced cell death significantly but pretreatment with geldanamycin did not. There were no significant changes in Bcl-2, Bcl-xl, and Bax protein expression after troglitazone treatment. In conclusion, TRAIL in combination with troglitazone, paclitaxel, and cycloheximide induces apoptosis in thyroid cancer cells at suboptimal concentrations that cannot be achieved using TRAIL alone.

Chromosomal aberrations by comparative genomic hybridization in hürthle cell thyroid carcinomas are associated with tumor recurrence.

Hürthle cell thyroid neoplasms are classified as variants of follicular neoplasms, but they have distinct clinicopathological features. Chromosomal aberrations by comparative genomic hybridization (CGH) are common in Hürthle cell neoplasms. However, there is currently only limited information concerning the relationship between the chromosomal aberrations by CGH and tumor behavior. We, therefore, investigated chromosomal aberrations in primary Hürthle cell neoplasms (13 carcinomas and 15 adenomas) using CGH and correlated the aberrations identified with tumor node metastasis (TNM) stage, tumor differentiation, capsular invasion, and tumor recurrence. Chromosomal aberrations were found in 62% (8 of 13) of carcinomas and 60% (9 of 15) of adenomas. Overall, common chromosomal gains were found on 5p (29%), 5q (36%), 7 (29%), 12p (14%), 12q (21%), 17p (29%), 17q (32%), 19p (32%), 19q (25%), 20p (21%), 20q (29%), and 22q (18%). Common chromosomal losses were found on 2q (18%) and 9q (18%). Thirty-eight percent (5 of 13) of carcinomas were TNM stage III, 31% (4 of 13) were moderately to poorly differentiated, and 46% (6 of 13) were intermediately to widely invasive. Recurrence occurred in 38% (5 of 13). Carcinomas that subsequently recurred had a greater number of chromosomal gains (9.0 vs. 1.3; <0.005) and had more frequent chromosomal gains on 12q, 19q, and 20p (<0.001), 5p, 7, 19p, and 20q (<0.005), and 12p (<0.01) than those that did not recur. Five of the eight (63%) patients with aberrations developed recurrence, whereas none of the five patients without aberrations developed recurrence. In conclusion, chromosomal gains by CGH on 5p, 7, 12p, 12q, 19p, 19q, 20p, and 20q in Hürthle cell carcinomas are associated with tumor recurrence. Such chromosomal aberrations may be predictive for recurrent disease in patients with Hürthle cell thyroid carcinoma.