Search

A correlation has previously been demonstrated between the presence of the three HMGI proteins (HMGI, HMGY, and HMGI-C) and the expression of a highly malignant phenotype in epithelial and fibroblastic rat thyroid cells; this being subsequently extended to experimental thyroid, lung, prostate, mammary, and skin carcinomas. Recently, we have demonstrated that expression of HMGI and HMGY proteins, coded for by the HMGI(Y) gene, is associated with the malignant phenotype of human thyroid neoplasias. Here, we show that HMGI(Y) gene expression is present both at the RNA and protein level in human colorectal carcinoma cell lines and tissues examined in this study. Conversely, no HMGI(Y) proteins were detected in normal intestinal mucosa. Therefore, these results suggest an involvement of HMGI and HMGY proteins overexpression in colorectal tumorigenesis.

High Mobility Group I (HMGI) proteins are nuclear proteins involved in the regulation of chromatin structure and function. Elevated expression of the HMGI proteins (HMGI, HMGY and HMGI-C) has been correlated with the presence of a highly malignant phenotype in epithelial and fibroblastic rat thyroid cells, and in several experimental carcinomas. Here, we demonstrate that HMGI and HMGY proteins are expressed in human thyroid carcinomas and thyroid carcinoma cell lines, but not in adenomas, goiters, normal thyroid tissues and cells. These results indicate a correlation between HMGI and HMGY expression and the malignant phenotype of thyroid neoplasias, suggesting that these proteins may be used as markers in thyroid cancer.

Rearrangements of the HMGI-C gene have frequently been detected in human benign tumors of mesenchymal origin, including lipomas. The HMGI-C protein has three AT-hook domains and an acidic COOH-terminal tail. The HMGI-C modifications consist in the loss of the C-tail and the fusion with ectopic sequences. Recent results show that the loss of the COOH-terminal region, rather than the acquisition of new sequences, is sufficient to confer to HMGI-C the ability to transform NIH3T3 cells. Therefore, transgenic mice carrying a HMGI-C construct (HMGI-C/T), containing only the three AT-hook domains, were generated. The HMGI-C/T mice showed a giant phenotype, together with a predominantly abdominal/pelvic lipomatosis, suggesting a pivotal role of the HMGI-C truncation in the generation of human lipomas.

fra-1 gene overexpression has been shown to represent a general event in thyroid cell transformation in vitro and in vivo. Moreover, inhibition of FRA-1 protein synthesis by stable transfection with a fra-1 antisense construct significantly reduces the malignant phenotype of the transformed thyroid cells, indicating a pivotal role of the fra-1 gene product in the process of cellular transformation. In the attempt to define the potential use of FRA-1 protein detection in the diagnosis of thyroid diseases, we analyzed Fra-1 expression by a combination of immunohistochemistry and reverse transcription-PCR (RT-PCR) assay in 174 samples of thyroid nodules (22 nodular hyperplasias, 102 follicular adenomas, 34 papillary carcinomas, 12 follicular carcinomas, and 4 anaplastic carcinomas) representative of the spectrum of thyroid tumor pathology. FRA-1 protein was abundant in all of the carcinoma samples (50/50, 100\%), with an intense staining in the nucleus and the cytoplasm. Positive staining was also found in most of the adenomas (90 of 102; 88\%), but in this case, the staining was restricted to the nucleus. Similar results were obtained from the analysis of thyroid goiters; however, the number of positive cases is lower than adenomas (8 of 22; 36\%); moreover, the staining was not observed in all of the cells. Conversely, no FRA-1 protein was detectable in 12 normal thyroid tissue samples used as controls. RT-PCR analysis confirmed a higher fra-1 expression in papillary and follicular carcinomas compared with goiters and adenomas. fra-1 expression was also analyzed on 10 fine needle aspiration biopsy (FNAB) samples by RT-PCR. fra-1-specific mRNA was detected in seven of the eight FNABs corresponding to thyroid nodules that were eventually diagnosed as adenomas (three of four) and carcinomas (four of four) after surgery. Conversely, no fra-1 gene expression was observed in two FNABs derived from normal thyroid. Further studies are required before suggesting FRA-1 protein detection as a useful tool for the diagnosis of hyperplastic and neoplastic disorders of the thyroid gland.

The HMGI protein family includes three proteins, named HMG-I, HMG-Y and HMGI-C. The first two proteins are coded for by the same gene, HMGI (Y), through an alternative splicing mechanism. Their expression is elevated in neoplastic tissues and cells and this overexpression has a causal role in the process of cellular neoplastic transformation. We demonstrate that the HMGI (Y) gene is expressed at very low levels in normal adult tissues, whereas in embryonic tissues it is expressed at high levels comparable to those detected in neoplastic tissues. Specifically, a very high expression of the HMGI (Y) gene was detected in all embryonic tissues at 8.5 dpc. Then in the following days, even though the gene is expressed essentially in all tissues, an abundant gene expression was restricted to some tissues. These results indicate an important role of the HMGI (Y) gene in development.

The high mobility group (HMG) proteins (HMGA1a, HMGA1b, and HMGA2) bind to DNA and interact with various transcriptional factors. Therefore, they play an important role in chromatin organization. HMGA protein expression is low in normal adult tissues, but abundant during embryonic development and in several experimental and human tumors. Blockage of HMGA expression inhibits the transformation of rat thyroid PC Cl 3 cells treated with oncogene-carrying retroviruses, thus implicating HMGA in rat thyroid transformation. To better understand the role of HMGA and to establish whether its up-regulated expression is sufficient to induce the transformed phenotype, we generated PC Cl 3 cells that overexpress the protein. We demonstrate that HMGA1b protein overexpression does not transform normal rat thyroid PC Cl 3 cells, but it deregulates their cell cycle: cells enter S-phase earlier and the G(2)-M transition is delayed. HMGA1-overexpressing cells undergo apoptosis through a pathway involving caspase-3 activation, probably consequent to the conflict between mitogenic pressure and the inability to proceed through the cell cycle. Using various HMGA1b gene mutations, we found that the third AT-hook domain and the acetylation site K60 are the protein regions required for induction of apoptosis in PC Cl 3 cells. In conclusion, although HMGA1 protein overexpression is associated with the malignant phenotype of rat and human thyroid cells, it does not transform normal thyroid cells in culture but leads them to programmed cell death.

We have recently reported that neoplastic transformation of two rat thyroid epithelial cell lines by retroviruses carrying the v-mos and v-ras Ki oncogenes is associated with a drastic increase of AP-1 activity. The most important effects were represented by the dramatic junB and fra-1 gene induction, which was abolished by the block of the transformation-induced HMGI-C protein synthesis. Here, we have further characterized the transformation-dependent AP-1 activity, by analysing the expression of different jun- and fos-related components, in rat thyroid cell lines transformed by several oncogenes, in human thyroid carcinoma cell lines, and in naturally occurring human thyroid tumours. A significant increase of Fra-1 and JunB protein levels was detected in all oncogene transformed rat thyroid cell lines. Fra-1 gene induction was demonstrated to occur also in human thyroid carcinoma cell lines and tissues. Conversely, c-Jun and JunD proteins, rather than JunB, accumulated in human thyroid carcinoma cell lines. An induction of AP-1 target genes was also detected both in rat and human thyroid transformed cell lines. Therefore, in vivo and in vitro thyroid cell transformation is associated with important compositional changes in the AP-1 complex and an increased transcriptional activity.

The expression of the high mobility group I (HMGI)-C chromatin component was shown previously to be essential for the establishment of the neoplastic phenotype in retrovirally transformed thyroid cell lines. To identify possible targets of the HMGI-C gene product, we have analyzed the AP-1 complex in normal, fully transformed and antisense HMGI-C-expressing rat thyroid cells. We show that neoplastic transformation is associated with a drastic increase in AP-1 activity, which reflects multiple compositional changes. The strongest effect is represented by the dramatic junB and fra-1 gene induction, which is prevented in cell lines expressing the antisense HMGI-C. These results indicate that the HMGI-C gene product is essential for the junB and fra-1 transcriptional induction associated with neoplastic transformation. The inhibition of Fra-1 protein synthesis by stable transfection with a fra-1 antisense RNA vector significantly reduces the malignant phenotype of the transformed thyroid cells, indicating a pivotal role for the fra-1 gene product in the process of cellular transformation.

Rearrangements of the high mobility group protein I-C (HMGI-C) gene, consisting in the loss of the carboxyl-terminal tail, have been frequently detected in benign human tumors of mesenchymal origin. We have previously demonstrated that transgenic (TG) mice carrying a truncated HMGI-C construct (HMGI-C/T) exhibit a giant phenotype together with a predominantly abdominal/pelvic lipomatosis. Here, we report that HMGI-C/T TG mice develop natural killer (NK)-T/NK cell lymphomas starting from 12 months of age. We found an increased expression of IL-2 and IL-15 proteins and their receptors in these lymphomas, and we demonstrate that HMGI-C/T protein positively regulates their expression in vitro . Therefore, the HMGI-C/T-mediated chronic stimulation of the IL-2/IL-15 pathway could be responsible for the onset of NK-T/NK cell lymphomas in HMGI-C/T TG mice.

The HMGA family is comprised of four proteins: HMGA1a, HMGA1b, HMGA1c and HMGA2. The first three proteins are products of the same gene, HMGA1, generated through an alternative splicing mechanism. The HMGA proteins are involved in the regulation of chromatin structure and HMGA DNA-binding sites have been identified in functional regions of many gene promoters. Rearrangements of the HMGA2 gene have been frequently detected in human benign tumors of mesenchymal origin including lipomas. 12q13-15 chromosomal translocations involving the HMGA2 gene locus, account for these rearrangements. The HMGA proteins have three AT-hook domains and an acidic C-terminal tail. The HMGA2 modifications consist in the loss of the C-terminal tail and fusion with ectopic sequences. A pivotal role of the HMGA2 rearrangements in the process of lipomagenesis is suggested by experiments showing that transgenic mice carrying a truncated HMGA2 gene showed a giant phenotype together with abdominal/pelvic lipomatosis. As HMGA2 null mice showed a great reduction in fat tissue, a positive role of the HMGA2 gene in adipocytic cell proliferation is proposed. More recently, similar alterations of the HMGA1 gene have been described. As the block of the HMGA1 protein synthesis induces an increase in growth rate of the pre-adipocytic cell line 3T3-L1, we suggest a negative role of the HMGA1 proteins in adipocytic cell growth and, therefore, we propose that adipocytic cell growth derives from the balance of the HMGA1 and HMGA2 protein functions.