| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institut National de la Santé et de la Recherche Médicale, E0018, Laboratoire de Biochimie et Biologie Moléculaire, Centre Hospitalier Universitaire (C.J., O.B., D.P.-M., F.S., P.R., V.R., Y.M., P.R.), F-49033 Angers, France; Laboratoire d’Anatomie Pathologique, Hôpital Ambroise Paré (B.F.), F-92104 Boulogne, France; and Laboratoire d’Anatomie Pathologique, Centre Hospitalier Régional Universitaire (S.G.), F-37044 Tours, France
Address all correspondence and requests for reprints to: Dr. Caroline Jacques, Institut National de la Santé et de la Recherche Médicale, E0018, Laboratoire de Biochimie et Biologie Moléculaire, Centre Hospitalier Universitaire d’Angers, 4 rue Larrey, F-49033 Angers, France. E-mail: address: caroline.jacques{at}med.univ-angers.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A feedback mechanism of mitochondrial proliferation in response to some mitochondrial defect, similar to that observed in various pathologies involving mitochondrial disorders (4), has been postulated for thyroid oncocytoma (5, 6). Recently, a compensatory mechanism was proposed to explain mitochondrial proliferation in renal oncocytoma (7). A defect in the enzymatic activity and protein content of complex I was specifically found in renal oncocytoma as well as in the vicinity of the tumor. In thyroid oncocytomas, no defects of oxidative phosphorylation or any significant mutations in mitochondrial DNA (mtDNA) have been discovered to date (5, 8). However, we reported a deficit in ATP production in thyroid oncocytic tumors as well as in a thyroid oncocytic cellular model (XTC.UC1) (5, 9). The slight uncoupling was associated with the overexpression of uncoupling protein 2 (UCP2). Transcriptional microarray profiling demonstrated the coordinated up-regulation of mitochondrial respiratory chain genes as well as the coordinated regulation of most of the metabolic pathways involved in oxidative metabolism, e.g. overexpression of the genes involved in cytoplasmic glycolysis, the tricarboxylic cycle, and the respiratory chain, and down-regulation of lactate dehydrogenase A. These results suggest that oncocytomas, in contrast with most solid tumors, preferentially use an oxidative metabolism to produce ATP (6). We also found that a ubiquitous pathway of factors involved in the coordination of mitochondrial biogenesis was strongly up-regulated in thyroid oncocytoma (10).
The differential display, an RT-PCR-derived technique developed by Liang and Pardee (11), offers the advantage of revealing the differential expression of identified and unidentified genes. This method, based on the use of short arbitrary primers, allows the selection of reproducible, differentially expressed fragments in two sets of samples, controls and tests. Overexpressed, underexpressed, and isoexpressed fragments can all be visualized in a single experiment. However, although the differential display is considered to be a powerful technique (12), it may produce up to 80% false-positive results (13). Thus, differential expression of the fragments needs to be confirmed using an adequate independent method, such as macroarray analysis.
In the case of thyroid oncocytoma, very few gene expression profiles are currently available (6, 14). To reveal new genes specifically deregulated in oncocytoma, we screened about 60% of the entire transcriptome belonging to two thyroid oncocytic adenomas using the differential display RT-PCR technique. We confirmed the systematic up-regulation of 30 genes in these two samples by the macroarray analysis of four other thyroid oncocytic adenomas. Finally, quantitative RT-PCR and immunohistochemistry were used to validate the overexpression of several genes and proteins.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Two sporadic tumors, diagnosed, according to the World Health Organization classification (15), as a follicular adenoma with Hurtle cells and an oncocytic tumor of uncertain malignant potential, and two paired normal thyroid tissue samples were chosen to perform the differential display analysis. For the macroarray analysis, we used four other sporadic thyroid tumors, diagnosed as follicular oncocytic adenomas, and four paired normal thyroid tissues. For the quantitative RT-PCR analysis, 18 thyroid tumors, consisting of 13 oncocytic adenomas and five oncocytic carcinomas, were chosen. Normal thyroid samples were taken distant from the tumor and histologically tested as being appropriate for controls. All samples, which were rendered anonymous (i.e. with patient identifiers deleted before the study), were deep-frozen in liquid nitrogen immediately after surgery and conserved at –80 C.
For the immunohistochemistry study, we used 11 formalin-fixed, paraffin-embedded, thyroid oncocytic tumor samples and their normal counterparts: five oncocytic follicular carcinomas and six oncocytic follicular adenomas.
RNA preparation
RNA was isolated from frozen tissues using the guanidinium isothiocyanate procedure (TRIzol reagent, Invitrogen Life Technologies, Inc., Gaithersburg, MD). Quantification, degradation, and DNA contamination of RNA were assessed using an RNA 6000 Nano Assay (Agilent Technologies, Palo Alto, CA) following the manufacturer’s procedure.
Differential display analysis
The differential display analysis was performed using the RNAimage sets G501, G502, and G503 (RNAimage kit, GeneHunter Corp., Nashville, TN) (11) according to the manufacturer’s instructions. Briefly, these three sets allow 72 PCR amplifications per sample, using 24 different arbitrary primers. Statistically, this 72 PCR per sample analysis is estimated to cover between 56 and 71% of the cell transcriptome (see www.genhunter.com/support/index.html). The RTs and the PCR amplifications were performed on a PTC 200-MJ Research Thermocycler (MJ Research, Watertown, MA). The PCR amplifications of oncocytoma and paired normal thyroid samples obtained with the same primer pair were loaded in consecutive wells to visually select differentially expressed bands (Fig. 1
). The bands were cut and eluted from the acrylamide gels, and reamplified using the original set of primers. The sizes of all the reamplified PCR products were checked on a 2.5% agarose gel with the DNA Molecular Weight Marker IX (Roche, Mannheim, Germany).
|
After reamplification, the PCR products were cloned into the pCR 2.1-TOPO vector (TOPO TA Cloning kit, Invitrogen Life Technologies, Carlsbad, CA) following the manufacturer’s protocol. Plasmid DNA was extracted from three overnight subcultures of each cloned fragment, with the QIA Prep Spin Miniprep kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. The sequencing reactions were performed on 50-fmol plasmid DNA using the two insert flanking primers, the M13 reverse primer and the M13 forward primer (TOPO TA Cloning kit), and the CEQ DTCS-Quick Start kit (Beckman Coulter, Fullerton, CA) following the manufacturer’s recommendations. The sequencing reactions were analyzed on the capillary automated sequencer CEQ 8000 Genetic Analysis System (Beckman Coulter).
Macroarrays
The plasmids containing the 460 inserts obtained by differential display and two control cDNA clones (UCP2 and ANT2) were PCR-amplified in 96-well plates using the two M13 primers mentioned above. The amplification conditions were 95 C for 10 min, followed by 35 cycles at 95 C for 30 sec, 48 C for 30 sec, and 72 C for 30 sec, with a final step at 72 C for 10 min. Several wells containing DNA-free water were used as PCR controls. The PCR amplification products were spotted on Hybond TM-N+ membranes optimized for nucleic acid transfer (Amersham Biosciences, Piscataway, NJ) using a Microgrid II spotter (Biorobotics, Cambridge, UK). After denaturation (0.5 M NaOH and 1.5 M NaCl) and DNA fixation [2 h at 80 C, and UV cross-linking (UV Stratalinker 1800; Stratagene, La Jolla, CA)], the membranes were kept at room temperature until hybridization. The first hybridization was performed with an oligonucleotide corresponding to a small portion of the cloning plasmid (5'-CCCTATAGTGAGTCGTATTA-3') to quantify the amount of DNA available for hybridization. One microgram of the oligonucleotide was labeled with [
-33P]ATP using the T4 polynucleotide kinase kit (Forward Labeling Reaction, Invitrogen Life Technologies, Inc., Groningen, The Netherlands) following the manufacturer’s protocol. The labeled oligonucleotide was then purified on MicroSpin G-25 columns (Amersham Biosciences).
The membranes were prehybridized for 4 h at 42 C in 10 ml 5x standard saline citrate (Eurobio, Les Ulis, France), 5x Denhardt’s, 0.5% sodium dodecyl sulfate, and 100 µg/ml sonicated salmon sperm DNA (Roche) buffer. After the addition of 400,000 cpm of the labeled plasmid oligonucleotide to the buffer, hybridization was performed for an additional 2 h at 42 C. After a washing step in a 2x standard saline citrate/0.1% sodium dodecyl sulfate buffer, the membranes were dried and exposed on a Super Resolution Cyclone Storage Phosphor Screen (Packard-PerkinElmer Life Sciences, Boston, MA; Fig. 2A). The signals were analyzed on a Cyclone Storage Phosphor System (Packard-Perkin-Elmer Life Sciences) and quantified with the Quantarray software (Packard-PerkinElmer Life Sciences).
|
-33P]deoxy (d)-CTP-labeled total cDNA of the thyroid tumors and the normal thyroid samples. Each sample was hybridized on an individual array. Five micrograms of total RNA were used for the RT. After addition of 8 µg deoxythymidine25 oligonucleotide (Promega Corp., Madison, WI), the mix was incubated 8 min at 68 C, and for 30 min while decreasing the temperature from 68 to 42 C. After the addition of 40 U RNasin ribonuclease inhibitor (Promega Corp.), 5x buffer, 6.7 mM dithiothreitol (Superscript II kit, Invitrogen Life Technologies, Inc.), 0.4 mM dATP, dCTP, dGTP, 2.4 µM dCTP, 30 µCi dCTP ([-33P]-labeled; PerkinElmer Life Sciences), and 200 U SuperScript II, the incubation was restarted for 1 h at 42 C. At this step, 200 U of the enzyme were added, and the incubation was prolonged for 1 h. The reaction was stopped by the addition of 0.33% sodium dodecyl sulfate, 13 mM EDTA, and 0.3 M NaOH and incubation for 30 min at 68 C, followed by 15 min at room temperature. Before purification on an AutoSeq G-50 column (Amersham Biosciences), 0.25 M Tris and 0.15 mM HCl were added. The purified labeled cDNA was then mixed with 2 µg dA80 oligonucleotide, denatured for 5 min at 100 C, and incubated for 2 h 30 min at 65 C in 1 ml hybridization buffer.
The stripped membranes were prehybridized in 10 ml prehybridization solution (see above) for 6 h at 68 C. The total cDNA-labeled probe was then added, and the incubation was performed for 48 h at 68 C. After the washing steps, the membranes were exposed, and the signals were detected as described above (Fig. 2
, B and C).
The total cDNA probe signal measurements were first corrected by means of the amount of spotted DNA measured by hybridization of the plasmid oligonucleotide as previously described (16). For each spot, the final measurement corresponded to the ratio of the sample signal to the oligonucleotide signal. Spots exhibiting weak plasmid hybridization (small amounts of spotted DNA) were considered as missing values. Spots with expression similar to the background were withdrawn from analysis. Normalization between arrays was obtained by dividing each gene measurement by the median expression value of the array. For each gene, we then calculated the O/T ratio, defined as the ratio between gene expression in the tumor vs. that in normal thyroid tissue. Only the genes for which the mean cut-off differential expression value for the four oncocytomas vs. normal thyroid samples was confirmed (i.e. O/T 
2) were retained for additional analysis.
In silico analysis
The FASTA format of the insert sequences was compared with the databanks GenBank, dbEST (www.ncbi.nlm.nih.gov/), and UCSC Genome Browser (www.genome.ucsc.edu/cgi-bin/, July 2003 version) for identification. The best matching sequences were considered to be identical (e-value: 0–7 E-36; mean e-value, 3.04 E-37; median, 1E-64). For each unidentified confirmed sequence, we checked a putative gene prediction with a BLAT alignment of the sequence with the UCSC Genome Browser databank sequences. For each of the predicted genes, the corresponding protein sequence was analyzed with the Mitoprot package (http://ihg.gsf.de/ihg/) and the TargetP package (www.cbs.dtu.dk/services/TargetP) to predict a putative N-terminal mitochondrial targeting signal (see Table 2).
|
The expression level for the AF1q gene was explored by quantitative real-time PCR analysis using the actin gene as a reference, as previously described (10). The AF1q standard was obtained using the forward primer 5'-CCATCTTTGGAACACGCCAG-3' and the reverse primer 5'-TTTCTCCTGGTCTGCTGCAG-3'. The actin standard was obtained using the forward primer 5'-CGACATGGAGAAAATCTGGC-3' and the reverse primer 5'-AGGTCCAGACGCAGGATGG-3'.
Immunohistochemistry
The 11 formalin-fixed, paraffin-embedded, thyroid tumor samples (five oncocytic follicular carcinomas and six oncocytic follicular adenomas) and their normal counterparts were used for tissue array construction, as described by Kononen et al. (17). A triplet of tumor tissue and a triplet of the normal counterpart were arrayed for each sample. Immunostaining was performed using the standard avidin-biotin peroxidase technique. The primary antibodies were anticathepsin B (human liver; dilution, 1:400; Calbiochem, San Diego, CA) and antihuman CD51/61 (integrin Vß3-vitronectin receptor (ITGAV); clone 23C6; dilution, 1:10; Leinco Technologies, St. Louis, MO). Diaminobenzidine was used as the chromogen, and hematoxylin was used as the nuclear counterstain. For negative controls, the primary antibody was either omitted or replaced by a suitable concentration of normal IgG of the same species.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Because the differential display technique is known to produce a high rate of false-positive reactions (up to 80%) (13), we designed a nylon membrane macroarray assay to confirm the differential expression of clones among the 460 bands identified by differential display analysis of the two tumors. We spotted the ANT2 and UCP2 genes to act as positive controls on the membranes. The microarray measurement of mRNA expression was made in four other pairs of thyroid oncocytic adenomas and their normal paired controls (Fig. 2). We selected the clones differentially expressed in all four thyroid oncocytic adenomas. This highly selective, two-step approach enabled us to confirm 12 underexpressed and 49 overexpressed clones in the tumor among the 460 differentially expressed clones identified. Among these differentially expressed clones, the O/T ratio ranged from 0.71–6.52. The two control genes spotted on the macroarrays, i.e. the ANT2 and UCP2 genes, were confirmed as being overexpressed, with overexpression factors of 1.98 and 1.77, respectively, according to our previous observations (5, 6). For additional analysis, we chose to focus on the clones that showed at least a 2-fold differential expression compared with controls, although some of the excluded genes could be of interest (for example, the pendrin gene, which was down-regulated by a factor of 0.79). The cut-off value of an O/T of 2 or greater highlighted 30 overexpressed genes.
The 30 clones selected had factors of overexpression in the tumor ranging from 2.06–6.52 compared with normal thyroid tissue. Table 1 recapitulates the identification information for each of these sequences. Twenty-five of the 30 sequences (83%) corresponded to well-known genes, whereas five did not match any previously identified genes. Twelve of the 30 sequences (40%) were mtDNA-encoded genes (11 of 12) or a mtDNA sequence (one of 12). Ten of these sequences corresponded to genes coding for subunits of the four respiratory complexes. As shown in Table 1, many of these genes are among the most up-regulated genes in our study. However, the overexpression factor of the mtDNA-encoded genes in oncocytoma ranges from 2.15–6.52 (mean, 3.15). One of the remaining two clones corresponded to the 16S rRNA gene. The last sequence was identical to part of the RNA primer sequence, located in the mtDNA control region, which is necessary for initiation of mtDNA replication.
|
Figure 3 shows the results of the immunostaining with the anticathepsin B and the anti-CD51/61 (ITGAV protein) antibodies of control samples and thyroid oncocytomas (six adenomas and five carcinomas). All tumors studied presented an overexpression of the two proteins compared with the control tissues. For the cathepsin B protein, cellular localization of the signal varied from one case to another. However, in the adenomas as well as the carcinomas, the signal was cytoplasmic and frequently concentrated at the apical border in the form of droplets of different sizes. For the integrin Vß3 protein (CD51/61), a dense cytoplasmic signal was found in adenomas (Fig. 3) as well as in carcinomas. The presence of droplets and a lining at the apical border was rarely observed.
|
Finally, in silico analysis allowed a gene prediction for three of the five isolated sequences that did not correspond to any known genes. One of these three sequences, 2q33.3, showed 100% identity (with respect to 59 bp) with the 3' extremity of the alternative variant gDec03 of the NDUFS1 gene. (The protein NDUFS1 is a nuclear-encoded component of the mitochondrial respiratory chain complex I.) The second sequence, 7p21.1, showed a 30% identity (with respect to 75 amino acids) with the elongation factor 1 of Macrosiphoniella ludovicianae involved in cytoplasmic translation (eIF1; e-value, 1.6). The gene LOC115294 was found to be similar to the hypothetical protein FLJ0883 involved in posttranslational modification (protein L-isoaspartate O-methyltransferase). These three predicted proteins (2q33.3, 7p21.1, and LOC115294) displayed a high probability of an N-terminal mitochondrial targeting signal by analysis with the Mitoprot and TargetP targeting prediction packages (Table 2). We did not find any particular homology concerning the unknown sequences located in 2q12.1 and 14q13.2.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The 30 overexpressed sequences included 10 of the 13 mtDNA-encoded genes corresponding to 13 subunits of four of the five respiratory chain complexes. The mtDNA-encoded genes are known to be up-regulated in the context of oncocytomas, which are characterized by increased mitochondrial biogenesis (5, 6, 19). The up-regulation of ND5, ND2, and UCP2 mRNA has been demonstrated with the RT-PCR method in 22 thyroid oncocytomas compared with the paired control tissues (5). Our results showed that nearly all mtRNA genes are up-regulated. The transcription of mtDNA is polycistronic; each strand of the molecule is entirely transcribed before cleavage of the different rRNA, transfer RNA, and mRNA sequences. The surprising variation in the level of expression of the 10 mRNAs may be explained by differences in posttranscriptional processing and in the half-life of mitochondrial mRNA.
Because the differential display technique uses polyT primer (HT11V) at the RT step, the up-regulation of the mitochondrial 16S rRNA, which is also mtDNA-encoded, was rather unexpected. However, it has been shown that 16S rRNA is polyadenylated by up to 10 residues (20), which is sufficient to allow RT.
We also isolated a fragment that matches the RNA replication primer sequence of the mtDNA control region. The transcription of this region produces an RNA sequence that, after cleavage, provides an RNA primer involved in the initiation of mtDNA replication (21). The overexpression of this sequence is of particular interest, because it is probably a key regulator of the balance between the two tightly coupled processes of replication and transcription of the mitochondrial genome. It has been shown in thyroid oncocytomas that this balance differs from that in normal thyroid tissues; the mRNA level is particularly high considering the mtDNA content (10). Although the RT of this nonpolyadenylated sequence is singular, its effective and reproducible overexpression may be involved in the uncoupling of replication and transcription.
In addition to the mtDNA-encoded genes, 13 known nuclear-encoded genes were also up-regulated in oncocytomas. Seven of these genes are associated with the cellular protein turnover at various levels of protein metabolism. In particular, the cathepsin B gene is the most overexpressed nuclear gene in our differential expression study, and its overexpression was confirmed at the protein level. The cathepsin B gene codes for a lysosomal cysteine endopeptidase located in the apical membrane of thyroid epithelial cells (22). This protein is involved in processing of the thyroglobulin hormone in the normal thyroid and is well known to be up-regulated at the transcriptional, translational, and posttranslational levels in human thyroid tumors (6, 23). It has therefore been suggested that this protein may be a reliable marker of the progression from the premalignant to the malignant state in the development of thyroid tumors (24, 25). The major dislocation of the protein at the apical border of the cells in all the tumors explored suggests an abnormal lysosome location or abnormal protein targeting or processing in oncocytoma.
Genes associated with other functions, such as transcription regulation, cell adhesion, and signal transduction, were also well represented. Among these genes, several have been shown to be involved in tumorigenesis or the cell cycle, e.g. TCF8 (transcription factor 8), ZMYND12 (zinc finger MYND domain containing 12 proteins), and ITGAV (integrin V). ITGAV encodes the integrin -chain V protein involved in cell adhesion (cell to cell junctions) and signal transduction. Integrins are known to be involved in various cancer processes, such as invasion, angiogenesis, and metastasis (26). The overexpression of genes involved in cell adhesion and proteolysis has been described in thyroid oncocytomas at a similar level of overexpression (6).
Of particular interest was the overexpression of the nuclear-encoded gene AF1q, which codes for a 9-kDa protein of unknown function with no similarity to any other protein. The AF1q gene was previously implicated in a translocation t(1;11)(q21;q23) responsible for cases of leukemia (27). The up-regulation of this gene has been shown to be greater in leukemic cell lines than in normal hemopoietic tissues (28). Moreover, the expression of this gene in pediatric acute myeloid leukemia may correlate with the clinical outcome of the patients (29). Our results, displaying no differences in the overexpression of the gene between adenoma and carcinoma oncocytic thyroid tumors, suggest that overexpression of the AF1q oncogene may be implicated in thyroid tumorigenesis. Finally, we revealed three putative genes encoding potential mitochondrial proteins, in which the functions have to be explored.
The set of 30 genes, overexpressed by a factor of at least 2, detected in thyroid oncocytic tumors reveals a great up-regulation of mtDNA genes and of the genes implicated in protein metabolism (seven known nuclear-encoded genes, two in silico predicted mitochondrial nuclear-encoded genes, and the mtDNA-encoded 16S RNA). These results show that mitochondrial proliferation is clearly associated with intense protein turnover and raise an important question. Is this increased protein turnover related to the oncocytoma tumorigenesis? We could postulate that increased protein turnover is necessary for the intense mitochondrial proliferation or that saturation of the cytoplasm by mitochondria could disturb cell trafficking and processing of the proteins, as suggested by relocation of the cathepsin B protein in tumors.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online December 28, 2004
Abbreviations: mtDNA, Mutations in mitochondrial DNA; UCP2, uncoupling protein 2.
Received July 9, 2004.
Accepted December 20, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. N. Co, W. P. Tsang, T. W.L. Wong, H. H. Cheung, T. Y. Tsang, S. K. Kong, and T. T. Kwok Oncogene AF1q enhances doxorubicin-induced apoptosis through BAD-mediated mitochondrial apoptotic pathway Mol. Cancer Ther., October 1, 2008; 7(10): 3160 - 3168. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Krause, M. Eszlinger, O. Gimm, S. Karger, C. Engelhardt, H. Dralle, and D. Fuhrer TFF3-Based Candidate Gene Discrimination of Benign and Malignant Thyroid Tumors in a Region with Borderline Iodine Deficiency J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1390 - 1393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Bawa-Khalfe, J. Cheng, Z. Wang, and E. T. H. Yeh Induction of the SUMO-specific Protease 1 Transcription by the Androgen Receptor in Prostate Cancer Cells J. Biol. Chem., December 28, 2007; 282(52): 37341 - 37349. [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]
|
|
|
|
 |
|
 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. Kozma, J. J. Keusch, B. Hegemann, K. B. Luther, D. Klein, D. Hess, R. S. Haltiwanger, and J. Hofsteenge Identification and Characterization of abeta1,3-Glucosyltransferase That Synthesizes the Glc-beta1,3-Fuc Disaccharide on Thrombospondin Type 1 Repeats J. Biol. Chem., December 1, 2006; 281(48): 36742 - 36751. [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 |
