| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institut National de la Santé et de la Recherche Médicale, Equipe Mixte INSERM-Universitaire 0018, Laboratoire de Biochimie et Biologie Moléculaire (O.B., F.S., P.Ro., V.R., Y.M., P.Re.), Service d’Endocrinologie, Nutrition et Médecine Interne, Centre Hospitalier Universitaire (P.Ro., V.R.), Angers F-49033, France; Département d’Oncologie Moléculaire, Institut Paoli-Calmettes and Unité 119, Institut National de la Santé et de la Recherche Médicale (V.N., F.B., D.B.), Marseille F-13273, France; Laboratoire Technologies Avancées pour le Génome et la Clinique/Institut National de la Santé et de la Recherche Médicale, Equipe de Recherche et d’Innovation Technologique et Methodologique 206 (B.L., S.Gr., R.H.), Marseille F-13009, France; Laboratoire d’Anatomie Pathologique, Centre Hospitalier Recherche Universitaire (S.Gu.), Tours F-37044, France; and Laboratoire d’Anatomie Pathologique, Hôpital A. Paré (B.F.), Boulogne F-92104, France
Address all correspondence and requests for reprints to: Dr. Olivier Baris, Institut National de la Santé et de la Recherche Médicale, Equipe Mixte INSERM-Universitaire 0018, Laboratoire de Biochimie et de Biologie Moléculaire, Centre Hospitalier Universitaire, 4 rue Larrey, Angers F-49033, France. E-mail: olivier.baris{at}med.univ-angers.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The clinical behavior of thyroid oncocytoma and the distinction between benign and malignant tumors have given rise to considerable controversy. The mortality rate due to oxyphilic carcinoma appears to be higher than that due to other follicular thyroid carcinomas. This may be related to the poor responsiveness of oxyphilic cell carcinoma to radioiodine therapy. The classification of oncocytoma as either adenoma or carcinoma using the conventional criterion of vascular or capsular invasion is generally used as an indicator of malignancy in oncocytoma (4). However, the percentage of oncocytic tumors with capsular or vascular invasion is higher than that of nononcocytic follicular tumors showing signs of malignancy. The high prevalence of malignancy and the putative higher aggressiveness of oncocytic carcinoma contrasts with the trend toward benignity or low malignancy of most oxyphilic tumors in other organs (5).
The relationship between mitochondrial proliferation and the pathogenesis of oncocytic tumors is still unknown. It has been suggested that mitochondrial proliferation may represent a compensatory mechanism against mitochondrial defects (6). This hypothesis is supported by the fact that increased mitochondrial numbers are frequently reported in mitochondrial diseases associated with respiratory chain defects. However, the greater histochemical activity of the respiratory chain complexes is not associated with improved cellular performance (7), suggesting a defect in the energy-producing machinery of the affected cells. No mitochondrial DNA (mtDNA) deletions appear to be specifically involved in the pathogenesis of oxyphilic tumors, and the common 4977-bp mtDNA deletion, which has been described in oncocytoma, has also been shown to be involved in inflammation and ageing (8).
The considerable difference between the mtDNA and the mtRNA content in thyroid oncocytoma suggests a defect in the coordination of mitochondrial DNA transcription and replication (8, 9). mtDNA encodes for the 13 polypeptides of the respiratory chain complexes as well as the 12S and 16S ribosomal RNAs and the 22 transfer RNAs required for mitochondrial protein synthesis. As the whole mitochondrial proteome is estimated to contain approximately 800-1000 proteins, the majority of mitochondrial proteins must be nuclear encoded. Only a few nuclear factors involved in mitochondrial biogenesis have been identified, such as the mitochondrial transcription factor A, the nuclear respiratory factors (NRF-1 and NRF-2), and members of the peroxisome proliferator-activated receptor 
coactivator 1 family, such as PGC-1
, PGC-1ß, and PRC (10).
cDNA array technology allows quantitative measurement of mRNA expression for thousands of genes simultaneously (11). Among its multiple applications, the molecular classification of tumors has led to major advances in understanding the mechanisms underlying cancer. For example, distinct molecular subgroups were identified for the first time by gene expression profiling reflecting two differentially proliferative types of tumor (12). Gene profiling in breast cancer has allowed the identification of new tumor subgroups with differential clinical outcomes, thereby improving the histological criteria of tumor classification (13, 14). Comprehensive molecular analysis may therefore be expected to provide new insights into oncocytic neoplasms and lead to better understanding of the mechanisms involved in mitochondrial biogenesis and nuclear-mitochondrial cross-talk.
We present a global gene expression profile for sporadic thyroid oncocytoma on the basis of the analysis of 76 samples from various conditions, including benign or malignant thyroid oncocytomas, and samples from two human thyroid cell lines. Our aim was to identify the gene expression patterns of oncocytomas to understand the mitochondrial proliferation in oncocytes and to improve tumor diagnosis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Samples were taken from 22 benign and seven malignant oxyphilic thyroid tumors. We also examined samples from seven follicular thyroid adenomas, two follicular thyroid carcinomas, six papillary thyroid carcinomas, four Grave’s disease thyroids, one renal oncocytoma, one parathyroid oncocytoma, and one carotid paraganglioma. Samples from 25 paired normal thyroids served as controls. All of these cases were diagnosed between 1994 and 2001 at the Ambroise Paré Hospital (Paris, France; 29 cases) and the University Hospital (Angers, France; 47 cases). All of the samples used were rendered anonymous before the study by the deletion of all patient identifiers. The diagnoses were made according to the World Health Organization classification. The distinction between oxyphilic thyroid adenomas and carcinomas was made on the basis of vascular or capsular invasion. The patients consisted of 15 men and 42 women (mean age, 49 yr; range, 18–75 yr). The average tumor size was 26.1 ± 11.0 mm (mean ± SD; range, 10–55 mm). Normal thyroid samples were taken at a sufficient distance from the tumors to avoid contamination. All of the samples were stored immediately after resection in liquid nitrogen until required for total RNA extraction.
Cell cultures
Two human thyroid cell lines were used: the XTC.UC1 cell line derived from an oxyphilic thyroid carcinoma, and the B-CPAP cell line derived from a papillary thyroid cancer (15, 16). The growth medium of XTC.UC1 cells consisted of DMEM supplemented with 10% fetal calf serum (Seromed, Biochrom AG, Berlin, Germany), 100 U/ml penicillin, 100 µg/ml streptomycin, and 0.25 µg/ml fungizone. The B-CPAP cells were cultured in RPMI 1640 medium with 10% fetal calf serum, 100 U/ml penicillin, 100 µg/ml streptomycin, and 0.25 µg/ml fungizone. All of the products were obtained from Life Technologies, Inc. (Paisley, UK) unless otherwise stated. Both cell lines were cultured with 10 mU/ml TSH (Sigma-Aldrich Corp., Lyon, France), and without TSH over three generations to take into account the effect of TSH on the expression of the different genes studied.
Total RNA isolation
Total RNA was isolated from tissue samples using a standard guanidium isothiocyanate protocol (TRIzol reagent, Life Technologies, Inc., Gaithersburg, MD) and from cultured cells using the RNeasy kit (Qiagen GmbH, Hilden, Germany). RNA integrity was determined using a BioAnalyzer 2100 (Agilent Technologies, Waldbronn, Germany).
cDNA arrays
Gene expression was analyzed by hybridization of nylon cDNA arrays with radioactive probes. The arrays contained spotted PCR products from 6720 selected IMAGE (MRC Rosalind Franklin Centre for Genomics Research, GeneService, Cambridge, UK) human cDNA clones and control clones. Clones were selected on the basis of the following criteria: the 3' location of the corresponding mRNA sequences, the same cloning vector (pT3T7), the same host bacteria, and approximately the same insert size. The majority of clones were selected such that they contained genes with a proven or putative implication in cancer or immune reactions (full list available soon at http://tagc.univ-mrs.fr/pub). The IMAGE clones consisted of approximately 84% genes and 16% established sequence tags (ESTs). The control clones consisted of three differently sized polyadenylated sequences and pT7T3D cloning vectors (negative controls). PCR amplification and robotic spotting of PCR products onto Hybond N+ membranes (Amersham Pharmacia Biotech, Little Chalfont, UK) were performed according to protocols described previously (17).
Hybridization of cDNA arrays
cDNA arrays were first hybridized with a labeled vector oligonucleotide to determine the precise amount of cDNA accessible for hybridization in each spot. After stripping, each array was hybridized with a complex target prepared from 5 µg total RNA by simultaneous RT and [-33P]deoxy-CTP labeling as described previously (http://tagc.univ-mrs.fr/pub/cancer/). Each sample was hybridized on an individual array. After washing, hybridization images were obtained by scanning with an imaging plate device (Fuji BAS 5000, Raytest, Paris, France). Signal intensities were quantified using ArrayGauge software (Fujifilm Medical Systems, Stanford, CT).
cDNA array analysis
Complex target measurement was first corrected for the amount of spotted DNA measured by the oligovector hybridization. Spots exhibiting weak oligovector hybridization (low amounts of spotted DNA) were considered missing values. Genes with an expression similar to the background expression and genes with missing values greater than 25% of the samples were withdrawn from the analysis. This finally allowed us to retain a set of 1626 genes with significant expression in most of the samples. Normalization between arrays was obtained by dividing each gene measurement by the median value of the array. Data were log transformed, and classifications were obtained by hierarchical clustering using the uncentered correlation distance and average linkage method provided by Cluster software, and visualized using Tree View (18). Supervised analysis was performed using a discriminating score (DS) calculated as follows. DS = (µ1 - µ2)/(
1 + 2), where µ1 and 1 represent the mean and SD, respectively, of the expression levels of a given gene in sample subgroup 1, and µ2 and 2 represent the mean and SD, respectively of the expression levels of the same gene in sample subgroup 2 (19). Random permutations of the samples in the two subgroups were used to calculate the significance level at a risk of 0.01%, giving less than one false positive gene (number of false positive genes, 0.32) (20).
Immunohistochemistry
To validate our microarrays data for five genes, we looked for their protein expression by immunohistochemistry. We selected five oncocytoma/normal thyroid pairs randomly chosen among the samples previously tested with microarrays and 10 independent oncocytoma/normal thyroid pairs. After morphological examination of hematoxylin- and eosin-stained sections, the corresponding 3-µm sections of the paraffin blocks were prepared for the detection of proteins encoded by several genes that discriminate between oncocytic and normal thyroids. We used five monoclonal IgG1 antibodies against BCL-2, cathepsin-B (Ab-2), apolipoprotein D, microtubule-associated protein 2 (all four from Calbiochem, Biosciences, Inc., Darmstadt, Germany), and a complex IV subunit of the respiratory chain (clone 113-1, from Biogenex Laboratories, Inc., San Ramon, CA). Immunostaining was performed using the standard avidin-biotin peroxidase technique with antigen retrieval. For negative control slides, the primary antibody was either omitted or replaced by a suitable concentration of normal IgG of the same species.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Each of the 87 samples was hybridized on a microarray containing cDNA for 6720 known genes and ESTs. After normalization, 1626 genes were selected for further analysis. Data were clustered using a hierarchical algorithm and were represented in a color-coded matrix (18). The complete cluster diagram is shown in Fig. 1
. The tissue samples were separated into two groups (group I, 50 samples; group II, 26 samples), as shown in Fig. 1A. Significantly, all of the samples in group II (except one, which was a thyroid papillary carcinoma) were oncocytic tumors. Similarly, several samples in group I belonging to the same histological class (such as papillary carcinoma or normal thyroid) were clustered together, confirming the presence of specific gene clusters. The cell line samples were separated into two distinct groups according to their origin (Fig. 1A, XTC-UC1 and B-CPAP). An enlarged view of several interesting gene clusters in the complete diagram (Fig. 1B, colored bars) is shown in Fig. 1C.
|
, yellow bar) was clearly up-regulated in thyroid oncocytomas and in XTC-UC1 cells as well as in renal and parathyroid oncocytomas compared with the other samples. This cluster was rich in nuclear encoded genes of the five respiratory chain complexes (ATP5B, ATP5C1, ATP5I, COX4I1, COX5A, COX5B, COX6A1, COX7A2, COX7B, CYC1, HCS, NDUFA4, NDUFA8, NDUFS4, NDUFS8, SDHA, SDHB, UQCR2, UQCRB, and UQCRFS1). Also included were some nuclear genes encoding proteins located in the matrix and on the outer mitochondrial membrane (PMPCB and BZRP) as well as some energy metabolism genes (GPI and MDH1).
A second set of genes (Fig. 1, blue bar) was overexpressed in thyroid oncocytic tumors. This thyroid oncocytic tumor cluster was composed of approximately 200 genes, including mitochondrial genes (ANT2, COX7RP, HSPC051, ME3, MTND4, MTCYB, MTATP6, MTCO3, NDUFB9, MRPL49, NRF1, and PPOX) involved in various mitochondrial functions, such as heme biosynthesis, pyruvate metabolism, ATP transport, and oxidative phosphorylation. We also observed overexpression of these genes in both parathyroid and renal oncocytoma samples. One notable finding was the overexpression of NRF-1, which is known to be a transcription factor regulating the expression of many mitochondrial proteins. This cluster also included genes involved in protein metabolism, DNA replication and maintenance, DNA transcription, and the cell cycle, thus reflecting the proliferative status of these tumors. Incidentally, several muscle-specific genes (MYH2, TNNT1, TPM2, TNNI1, MB, and MYH1) were clustered together, possibly representing the myofibroblastic content of these tumors, as myofibroblasts are known to express muscle-specific genes (21).
Seven clustered genes were specific to both thyroid papillary and oncocytic tumors (Fig. 1, green bar). Their strongly correlated expression suggests a common involvement in the same cellular process. These genes are involved in proteolysis (CTSB and CSTB), signaling (MST1R, IGF2R, and IGFBP6), DNA repair (PRKDC), and regulation of cell proliferation (RBL2).
We also identified two clusters reflecting differential proliferative status. The first, a proliferation cluster (Fig. 1, turquoise bar), was overexpressed in both XTC.UC1 and B-CPAP cell lines, reflecting their highly proliferative status. This set was rich in cytoskeleton genes, DNA maintenance and replication genes, cell cycle genes, transcription factors, and protein metabolism genes. It also included the PCNA gene, which is frequently used as a marker of proliferation. The second, an immediate response cluster (Fig. 1, pink bar), was overexpressed in normal thyroid samples and was composed of genes involved in growth, proliferation, and differentiation stimuli (ATF3, EGR1, FOS, JUN, JUNB, JUND, and SRF). A similar gene cluster has been identified in normal ovarian tissue compared with tumoral tissue (22). This might indicate the differential of proliferative status between normal and malignant tissues, because overexpression of ATF3 and EGR1 results in slower growth rates (23, 24).
Lastly, an immune response cluster (Fig. 1, brown bar) was identified, mainly reflecting T and B lymphocyte populations and immune response mechanisms such as antigen presentation or interferon signaling. Overexpression of this cluster was concordant with the lymphocytic infiltrates revealed by histological examination of the corresponding tissue sections. Moreover, the four Graves’ disease samples showed increased expression of the Ig genes, reflecing the humoral autoimmune origin of Graves’ disease. Among the six oncocytomas that were not clustered in the branch corresponding to group II (Fig. 1A), histology confirmed the presence of extensive lymphocytic infiltrates in two of the samples. The withdrawal of the immune response cluster genes followed by two-dimensional reclustering resulted in the inclusion of both samples in group II (data not shown).
Thyroid oncocytoma-specific profile
To further characterize thyroid oncocytoma, we compared all thyroid oncocytoma samples with the normal thyroid tissue samples that served as controls. For each gene we calculated a DS that enabled us to determine significantly distinct gene expression profiles for two groups of samples. The first group comprised the 29 thyroid oncocytomas, and the second group comprised the 25 normal thyroid samples. We performed 100 random permutations of the samples in each group to select the genes that discriminated between these two groups at a risk of significance of 0.01%. Using this approach we identified 163 genes that discriminated, to a statistically significant degree, between the two groups. Figure 2 shows a hierarchical clustering diagram of the 163 genes and the 54 samples analyzed. Thirty-seven genes were underexpressed, and 126 genes were overexpressed in thyroid oncocytoma compared with normal thyroid tissue. The robustness of our approach was illustrated by the fact that 26 of the 29 oncocytomas were grouped in the same cluster branch. The three remaining oncocytoma samples were histologically considered atypical. Despite the usual presence of granular cytoplasm and mitochondria stained with a cytochrome oxidase antibody, the three samples also exhibited atypical nuclear or structural aspects. These samples were among the thyroid oncocytic samples that clustered in group I of the unsupervised hierarchical clustering (Fig. 1A). In addition, we tried to identify the genes that discriminate best between thyroid oncocytic adenomas and carcinomas, but, unfortunately, obtained no statistically significant results.
|
) were involved in various cellular processes, such as lipid metabolism (APOD), inflammation (PTGS2 and TNFAIP3), transcription (FOS, JUN, JUNB, CHD2, and CREB1), adhesion (GJA1), signaling (DUSP1 and NBL1), and membrane structure (CAV1). The lowest expression levels were observed for 11 genes: APOD, FOS, JUN, CAV1, EPB41L2, EGR1, JUNB, POLD2, IFITM1, MATN2, and RAF1. As expected, the 126 up-regulated genes (Table 2) included several mitochondrial genes and, in particular, the nuclear genome-encoded genes, COX5B, COX6A1, SDHA, CYBA, CYC1, COX5A, COX7B, ATP5B, NDUFA4, HCS, and COX4I1, together with the mitochondrial genome-encoded genes, MTATP6 and MTND4. The proliferative status of thyroid oncocytic cells was illustrated by overexpression of the genes involved in DNA replication (RPA1, RFC4, and POLD1), the cell cycle (CCNA2, CCNE1, CCNG2, etc.), or protein synthesis (RPS8, RPS18, RPS19, and EIF2S2). Also included were genes involved in cell adhesion, cytoskeleton formation, cell cycle regulation, proteolysis, DNA repair, and transcription. Among these 126 up-regulated genes, 66 displayed a 2- to 6-fold increase in expression. The highest expression levels in thyroid oncocytomas were observed for 11 genes: ZNF42, ADA, HINT1, IGF2R, DTR, CYBA, RPS18, MRLP49, MTND4, PRKDC, and POLD1.
|
|
|
). The results obtained were in agreement with the corresponding mRNA expression values, indicating that these proteins are good tumoral markers for thyroid oncytoma (Table 3). Our COX antibody represents a positive control; in fact, it is commonly used to determine the proportion of oncocytes in thyroid tumors to confirm the diagnosis of oncocytomas. The overexpression of MAP2, a microtubule-associated protein, which binds to the outer membrane of mitochondria (25), suggests that mitochondrial proliferation in oncocytoma is accompanied by a modification of the expression of mitochondria-associated cytoskeleton proteins. In normal thyroid samples, cathepsin B was localized apically in the thyroid cells, as has been reported previously (26). However, in the oncocytoma samples, we found a homogeneous cytoplasmic distribution of the protein in the oncocytes, even when these cells were still polarized.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The gene profiling of thyroid oncocytic tumors revealed specific expression patterns that at the molecular level correspond to the mitochondrial proliferation histologically observed in oncocytomas. Two gene clusters were specifically overexpressed in 25 of the 29 thyroid oncocytomas, as well as in the two other nonthyroidal oncocytoma samples (kidney and parathyroid). One cluster, the mitochondrial cluster, composed mainly of genes coding for subunits of the five complexes of the respiratory chain, was overexpressed in oncocytic tumors and in the oncocytoma-derived cell line. The other cluster, the thyroid oncocytic tumor cluster, was composed of approximately 200 genes and included mitochondrial genes involved not only in oxidative phosphorylation, but also in various other mitochondrial functions. The underexpression of this cluster in XTC-UC1 cells probably reflects either the behavior of oncocytic cells in a tumoral environment or the loss of certain features in highly proliferative cultured cells. In addition, the mtDNA-encoded as well as the nuclear-encoded subunits of the respiratory chain were found in these two clusters, suggesting coordinated regulation of their expression in oncocytomas, as described by Heddi et al. (27).
Our findings support the hypothesis of the up-regulation of mitochondrial biogenesis in oncocytic tumors. Indeed, NRF-1, a mitochondrial ribosomal protein (MRLP49), and a mitochondrial processing peptidase (PMPCB), all involved in mitochondrial biogenesis, were overexpressed in these tumors. Among the genes that were overexpressed in oncocytic tumors, we also found the gene encoding endothelial nitric oxide synthase (NOS3). Recently, it has been shown that endogenous nitric oxide can trigger mitochondrial biogenesis in various cell types through the induction of the PGC-1 (28). The up-regulation of both NOS3 and NRF-1 suggests that mitochondrial biogenesis in oncocytic tumors is mediated by endogenous nitric oxide through the induction of PGC-1 or related factors.
Our results also reveal a profound modification of energy metabolism in oncocytomas. Twenty-six genes coding for subunits of the respiratory chain enzymes, three genes coding for glycolytic enzymes (GPI, GAPD, and ENO1), and several other genes encoding other energy metabolism enzymes (MDH1, ME3, and PGM1) were overexpressed in oncocytic tumors. Up-regulation of genes coding for glycolysis, the tricarboxylic acid cycle, and oxidative phosphorylation enzymes suggests that thyroid oncocytic tumors produce energy through an aerobic pathway. Moreover, the lactate dehydrogenase A gene, frequently overexpressed in human cancers characterized by anaerobic glycolysis, was underexpressed in the thyroid oncocytoma samples compared with normal thyroid tissue. This strengthens our hypothesis of the existence of an aerobic glycolytic mechanism in thyroid oncocytoma. The overexpression of energy metabolism proteins has been reported in renal oncocytoma, with findings of high activity of the enzymes involved in glycolysis and the tricarboxylic acid cycle coupled with low lactate dehydrogenase activity (29). This suggests that oxidative energy metabolism is a common feature of oncocytic tumors. In a previous study we demonstrated that mitochondrial ATP synthesis is altered in oxyphilic cells, possibly by uncoupling of the proton gradient in mitochondria, as suggested by the overexpression of uncoupling protein-2 in thyroid oncocytoma (8). A high content of aerobic energy metabolism enzymes and increased expression of the regulators of mitochondrial biogenesis in these tumors suggest a close relationship between these two processes. The up-regulation of mitochondrial biogenesis may compensate for defective mitochondrial ATP production by a feedback mechanism, or the two pathways may be concomitantly dysregulated at some higher molecular level (Fig. 4).
|
To our knowledge, this is the first report of a global gene expression profile in thyroid oncocytoma. We have identified 163 dysregulated genes in thyroid oncocytoma compared with normal thyroid tissue. Molecular similarities were found among oncocytomas originating from three different tissues, i.e. thyroid, parathyroid, and kidney. Thus, a common mechanism may underlie the abnormal mitochondrial proliferation observed in oncocytic tumors. Nevertheless, further experiments including more samples from nonthyroidal oncocytic tumors are needed to confirm this hypothesis. We identified a thyroid oncocytic tumor cluster that highlights the intimate relationship between mitochondrial biogenesis and cellular energy metabolism in these tumors. Given the role of NOS3 on oxidative enzyme activities (34), we hypothesize that this gene plays a major role in thyroid oncocytomas by leading to an increase in mitochondrial content and promoting an oxidative energetic metabolism.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: DS, Discriminating score; EST, established sequence tag; mtDNA, mitochondrial DNA; NOS3, nitric oxide synthase; NRF-1, nuclear respiratory factor 1; PGC, peroxisome proliferator-activated receptor coactivator.
Received July 17, 2003.
Accepted November 5, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. Trojanowicz, A. Winkler, K. Hammje, Z. Chen, C. Sekulla, D. Glanz, C. Schmutzler, B. Mentrup, S. Hombach-Klonisch, T. Klonisch, et al. Retinoic acid-mediated down-regulation of ENO1/MBP-1 gene products caused decreased invasiveness of the follicular thyroid carcinoma cell lines J. Mol. Endocrinol., March 1, 2009; 42(3): 249 - 260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Menon, Q. Zhang, Y. Zhang, D. Fermin, N. Bardeesy, R. A. DePinho, C. Lu, S. M. Hanash, G. S. Omenn, and D. J. States Identification of Novel Alternative Splice Isoforms of Circulating Proteins in a Mouse Model of Human Pancreatic Cancer Cancer Res., January 1, 2009; 69(1): 300 - 309. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Fujarewicz, M. Jarzab, M. Eszlinger, K. Krohn, R. Paschke, M. Oczko-Wojciechowska, M. Wiench, A. Kukulska, B. Jarzab, and A. Swierniak A multi-gene approach to differentiate papillary thyroid carcinoma from benign lesions: gene selection using support vector machines with bootstrapping Endocr. Relat. Cancer, September 1, 2007; 14(3): 809 - 826. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Gasparre, A. M. Porcelli, E. Bonora, L. F. Pennisi, M. Toller, L. Iommarini, A. Ghelli, M. Moretti, C. M. Betts, G. N. Martinelli, et al. Disruptive mitochondrial DNA mutations in complex I subunits are markers of oncocytic phenotype in thyroid tumors PNAS, May 22, 2007; 104(21): 9001 - 9006. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Eszlinger, K. Krohn, A. Kukulska, B. Jarzab, and R. Paschke Perspectives and Limitations of Microarray-Based Gene Expression Profiling of Thyroid Tumors Endocr. Rev., May 1, 2007; 28(3): 322 - 338. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Rasbach and R. G. Schnellmann Signaling of Mitochondrial Biogenesis following Oxidant Injury J. Biol. Chem., January 26, 2007; 282(4): 2355 - 2362. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Nisoli and M. O. Carruba Nitric oxide and mitochondrial biogenesis J. Cell Sci., July 15, 2006; 119(14): 2855 - 2862. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Cerutti, F. R.M. Latini, C. Nakabashi, R. Delcelo, V. P. Andrade, M. J. Amadei, R. M.B. Maciel, F. C. Hojaij, D. Hollis, J. Shoemaker, et al. Diagnosis of Suspicious Thyroid Nodules Using Four Protein Biomarkers. Clin. Cancer Res., June 1, 2006; 12(11): 3311 - 3318. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Eszlinger, M. Wiench, B. Jarzab, K. Krohn, M. Beck, J. Lauter, E. Gubala, K. Fujarewicz, A. Swierniak, and R. Paschke Meta- and Reanalysis of Gene Expression Profiles of Hot and Cold Thyroid Nodules and Papillary Thyroid Carcinoma for Gene Groups J. Clin. Endocrinol. Metab., May 1, 2006; 91(5): 1934 - 1942. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Jacques, O. Baris, D. Prunier-Mirebeau, F. Savagner, P. Rodien, V. Rohmer, B. Franc, S. Guyetant, Y. Malthiery, and P. Reynier Two-Step Differential Expression Analysis Reveals a New Set of Genes Involved in Thyroid Oncocytic Tumors J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2314 - 2320. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Jarzab, M. Wiench, K. Fujarewicz, K. Simek, M. Jarzab, M. Oczko-Wojciechowska, J. Wloch, A. Czarniecka, E. Chmielik, D. Lange, et al. Gene Expression Profile of Papillary Thyroid Cancer: Sources of Variability and Diagnostic Implications Cancer Res., February 15, 2005; 65(4): 1587 - 1597. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
