| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Genomic Medicine Institute (F.W., R.E.T., C.E.), and Lerner Research Institute and Taussig Cancer Center (C.E.), Cleveland Clinic Foundation, Cleveland, Ohio 44195; Department of General Surgery and Transplantation (F.W., C.E.B., A.F.), University Hospital of Essen, 45122 Essen, Germany; and Department of Genetics (C.E.) and CASE Comprehensive Cancer Center (C.E.), Case Western Reserve University School of Medicine, Cleveland, Ohio 44106
Address all correspondence and requests for reprints to: Charis Eng, M.D., Ph.D., Genomic Medicine Institute, Cleveland Clinic Lerner Research Institute, 9500 Euclid Avenue, NE-50, Cleveland, Ohio 44195. E-mail: engc{at}ccf.org.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to identify deregulated miRNAs in FTC.
Design: We used two high-density expression arrays to identify miRNAs and their target genes that are differentially expressed between FTC and FA. Validation was done by quantitative RT-PCR. We further functionally characterized the effect of deregulated miRNAs in vitro using HEK293T, FTC133, and K5 cell lines.
Patients: In total, 45 primary thyroid samples (23 FTC, 20 FA, four normal control thyroid) were analyzed.
Results: Two specific miRNAs, miR-197 and miR-346, were significantly overexpressed in FTC. In vitro overexpression of either miRNA induced proliferation, whereas inhibition led to growth arrest. Overexpression of miR-197 and miR-346 repressed the expression of their predicted target genes in vitro and in vivo.
Conclusions: Our observations show that miR-197 and miR-346 contribute to FTC carcinogenesis. Both miRNAs and their target genes might potentially provide for novel molecular markers and act as novel targets for treatment by interference, which could potentially normalize the deregulated profile of many downstream target genes.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thyroid cancer derived from the follicular epithelial cells account for the great majority of all thyroid malignancies. Of these, follicular thyroid carcinoma (FTC) accounts for about 10–15%. However, in iodine-deficient areas, the incidence can be twice as high (10, 11). In the clinical setting, FTC poses a special diagnostic challenge due to the morphological and molecular similarities to the benign follicular adenoma (FA) (12). Different molecular profiles have been proposed to improve preoperative diagnosis (13, 14, 15, 16, 17). However, the accurate preoperative diagnosis of FTC, especially minimally invasive FTC, continues to be a challenge. In addition, whereas thyroid cancer in general has a favorable prognosis, FTC, when diagnosed at an advanced stage, is incurable, with 10-yr survival rates less than 40% (18). Therefore, the challenge is not only to identify molecular markers for highly accurate diagnostic tests but also to find new targets for treatment of locally advanced or metastatic thyroid cancer.
Despite much progress over recent years, there is a continued limited understanding of the molecular and biological relationship of the different benign thyroid neoplasias to each other and to thyroid carcinomas, in particular FTC (19, 20). In contrast to papillary thyroid carcinoma (PTC), the major underlying genetic alterations leading to follicular thyroid carcinogenesis remain heterogeneous, even obscure (19, 20, 21, 22, 23, 24, 25).
We hypothesized that the uniform deregulation of a specific set of miRNAs could induce down-regulation of a cascade of target tumor suppressor genes. It is believed that identifying such key molecular differences between FA, which are benign follicular neoplasias, and FTC, which are malignant follicular thyroid neoplasias, will result in discovering genes and events associated with FTC initiation. Therefore, we sought to elucidate the differences in global miRNA expression between FA and FTC to dissect out deregulated human miRNAs that eventually could aid much needed improvement in preoperative diagnosis of FTC vs. FA and treatment of this cancer.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In total, 47 thyroid samples (23 FTC, 20 FA, and four normal control thyroid) have been analyzed in this study (for detailed histologies see Supplemental Table 1, published as supplemental data on The Endocrine Society’s Journals Online web site at http://jcem.endojournals.org). No oncocytic or hypercellular adenomas have been analyzed in this study. A set of eight FAs and 12 FTCs were used for the miRNA-chip array, and a set comprising 12 FTCs and 12 FAs was analyzed on the GeneChip array. Six FTCs and six FAs overlapped in these two studies. Additional validation of the differentially expressed miRNAs was performed in an independent set of nine follicular neoplasias (five FTCs and four FAs) and four normal control thyroid, not used on the miRNA-chip. Gene expression validation was done in a set of 14 FTCs and nine FAs by quantitative RT-PCR. The study, which used anonymized unlinked samples, was approved by the participating Institutional Review Boards for Human Subjects’ Protection.
miRNA-chip expression analysis
The miR chip analysis followed the design and protocols as described previously by Liu et al. (26) except that the human and mouse miRNA 11K version 2 chip was used. In brief, after biotin end labeling, the small RNAs were hybridized on a custom miRNA array chip that contains 460 mature miRNA probes (235 Homo sapiens, 222 Mus musculus, and three Arabidopsis manual). The 235 human miRNA are derived from a total of 319 (73.7%) unique, mature miRNAs known today. For each miRNA, 40-mer 5' amine modified C6 oligos were printed in quadruplicate on Amersham CodeLink-activated slides (Amersham, Piscataway, NJ). Quantification of biotin-containing transcripts was achieved after chip washing, processing, and incubation with streptavidin-Alexa647 using the Axon 4000B scanner and GENEPIX Pro 6.0 software package (Molecular Devices, Sunnyvale, CA). A detailed description of sequence selection, chip construction, and array protocols can be found on EMBL-EBI, Array Express no. E-TABM-68.
Microarray expression analysis
Total RNA extraction was performed under standard protocol using the TRIzol reagent (Invitrogen, Carlsbad, CA) and purified with the RNeasy kit (QIAGEN, Valencia, CA). The sample preparation, hybridization, and analysis were performed as described previously in detail (17, 22, 27). Chip data can be obtained from EMBL-EBI, Array Express no. E-MEXP-97.
miRNA and gene expression validation
The mirVana miRNA isolation kit was used for isolation and enrichment of small RNA fractions (Ambion, Austin, TX). miRNA expression analysis was done for miR-197, miR-328, and miR-346 by quantitative RT-PCR, according to the manufacturer’s protocols (Ambion). Optimized primers for the reverse transcription and PCR are commercially available (Ambion).
End point PCR was done with HotStar Taq polymerase (QIAGEN) and primers as follows: ACVR1, 5'-TTCCTCACTGAGCATCAACG and 5'-TAATGAGGCCAACCTCCAAG; TSPAN3, 5'-AGCCCTGCTTTTCATCATTG and 5'-TTCTGAATGCTGCGATCAAC; EFEMP2, 5'-GCCCAAACCTGTGTCAACTT and 5'-ATGAAGGCTGCTCTCGACAT; CFLAR 5'-TTTCTTTGCCTCCATCTTGG and 5'-GAAGCTCACAAGGGTCTTGC; and GAPDH, 5'-GGGCTGCTTTTAACTCTGGTAA and 5'-ATGGGTGGAATCATATTGGAAC.
Cell lines and culture conditions
The HEK293T, human embryonic kidney cells, two human follicular thyroid cancer cell lines (FTC133 and K5), and one human papillary thyroid cancer cell lines (NPA87) were cultured in DMEM supplemented with 10% fetal bovine serum and 100 U/ml penicillin and streptomycin (Life Technologies, Invitrogen). For cell growth assay, equal numbers (90,000) of cells were plated in 12-well plates. After 8, 12, 24, and 48 h, the medium was removed and the cells were washed and harvested. After trypsinization, viable cells (excluding trypan blue) were counted.
Transient overexpression of miRNAs
Precursor miRNAs (prec-miR-197 and prec-miR-346) (Ambion) were transiently transfected into HEK293T cells with the siPORT NeoFx transfection reagent (Ambion). For mock transfection conditions, prec-miR was substituted with random oligonucleotides at equal concentration. Optimal transfection efficiency was empirically determined at 3 µl siPORT NeoFx, 10 nM small RNA for 90,000 cells. All experiments were done in triplicate.
Suppression of endogenous miRNA function
We transfected commercially available anti-miR miRNA inhibitors (Ambion) directed against each of the mature sequences of miR-197 and miR-346 into two human thyroid carcinoma cell lines (FTC133 and K5) as well as NPA87 (human papillary thyroid carcinoma) cell line to study the effect on growth potential. Twenty to 80 nM of anti-miR oligonucleotides (Ambion) were transfected with the siPORT NeoFX transfection agent (3 µl) into the respective cells (90,000 cells/well of a 12-well plate).
Protein isolation and Western blot
Protein was isolated from tumor samples using radioimmunoprecipitation assay buffer [50 mM Tris (pH 8.0), 150 mM NaCl, 1% Triton, and 0.1% sodium dodecyl sulfate] containing proteases and subsequently sonication. Protein extracts (15 µg) were separated on a 10% SDS-PAGE gel and electrophoretically transferred onto nitrocellulose. After blocking for nonspecific binding, blots were then incubated with either ACVR1 (Abgent, San Diego, CA) or actin (Sigma, St. Louis, MO) primary antibody (1:1000 in 3% BSA). After incubation with an antirabbit secondary antibody (1:2500 dilution in 5% milk; Promega, Madison, WI), the protein bands were visualized using enhanced chemiluminescence as described by the manufacturer (Amersham Pharmacia Corp., Piscataway, NJ).
Statistical methods
For the miRNA-chip data, spots flagged as poor quality during image analysis were excluded from analysis. The average intensity over quadruplicate spots for each miRNA was computed, and a log base 2 transformation was then applied to the expression values. A median-centering array normalization procedure was then performed to allow for comparison across arrays. Our primary interest was comparing miRNA expression between FA and FTC patient samples. Because array samples were hybridized at two different times, we accounted for the possibility of a batch effect by using a two-way ANOVA with batch as a block variable. The two hybridization sets included both FA and FTC samples, with three FAs and five FTCs in the first and five FAs and seven FTCs in the second set. A nominal significance level of 0.001 was used in all statistical comparisons. BRB-ArrayTools version 3.3 (National Cancer Institute, Rockville, MD) was used for all analyses. GeneChip HG-U133A raw data were analyzed with the DNA-Chip Analyzer Software (dChip) developed by Li and Wong (www.dchip.org) as described by us previously in detail (17). We used linear diagonal discriminant analysis for class prediction in the gene expression data. The performance of the predictor was tested using leave-one-out cross-validation method based on 2000 random permutations. A two-tailed Student’s t test for independent samples, assuming equal variance, was used to determine difference between mean gene expressions in the validation analysis and cell growth assay. For analysis between groups, Fisher two-tailed exact test was used.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and Supplemental Table 2, published as supplemental data on The Endocrine Society’s Journals Online web site at http://jcem.endojournals.org). Two miRNAs (miR-192 and miR-197) have previously been experimentally validated in human (i.e. are truly human miR expressed in human tissues), whereas miR-328 and miR-346 are only predicted human homologs; we have now shown their expression in human tissue (Fig. 1) (28).
|
|
In an independent set of nine follicular thyroid neoplasias (five FTCs and four FAs) and four normal control thyroids, we could validate the differential expression of the mature miR-197 (overexpressed in FTCs vs. FAs by 2.00-fold, P = 0.0044) and miR-346 (1.37-fold expressed in FTCs over FAs, P = 0.049) using quantitative RT-PCR (Fig. 1
) (29). miR-192 was restricted to in silico analyses because specific reverse transcription and PCR primers for miR-192 could not be designed and tissue availability did not allow for analysis by Northern blot hybridization. However, for miR-328, even though the average expression was higher in FTCs, compared with FAs, this difference did not meet statistical significance in this validation set (P > 0.08; data not shown) and was not pursued further.
Functional effect of identified miRNAs
We determine the functional consequences of miRNA overexpression by transient transfection of two of the identified and most robustly validated miRNAs (miR-197 and miR-346) in a human nonneoplastic cell line (HEK293T). First, we confirmed transfection efficiency by detecting overexpression of miR-197 and miR-346 above endogenous levels (Fig. 2A). At 12 and 24 h after miR-197 or miR-346 transfection, significantly induced cell proliferation was noted with approximately 1.5-fold more viable cells than before transfection (P = 0.003–0.049; see Fig. 2B and legend). For both miR-197 and miR-346, expressional levels were seen to peak at 12 h after transfection and begin to return to basal levels by 24 h (Fig. 2A). The nonviable cell population increased by factors of 1.7- to 2.28-fold and thus mirrored the increase in the viable cell count observed in miR-197 and miR-346 transfected cells (Fig. 2C).
|
Commercially available miRNA inhibitors (Ambion) were used to suppress the functional effect of endogenous miRNA-197 and miR-346. FTC-133 cells under control conditions resulted in a 2.31-fold increase in cell number within 48 h (absolute cell count at 48 vs. 0 h; Fig. 3A). In contrast, transfection of anti-miR-197 and/or anti-miR-346 into FTC-133 cells resulted in a 2-fold growth suppression, i.e. a 1.11- to 1.5-fold increase in cell number instead of the control 2.31-fold was noted during the same time period (48 vs. 0 h). The effect of this miRNA inhibition on FTC-133 cell proliferation was significant (P = 0.0128, 0.0016, and 0.0026, respectively; Fig. 3A). A similar effect was seen in a second human FTC cell line (K5) (Fig. 3B), whereas neither inhibitor showed any effect in the NPA-87 cell line, which lacks endogenous miRNA-197 and miRNA-346 overexpression (data not shown). The number of nonviable cells did not differ between anti-miR oligonucleotide and control conditions (Fig. 3C).
|
We used the MicroCosm Web resource (version 2.0) maintained by the Sanger Institute (Cambridge, UK) to predict potential miRNA target sequences and reinterrogated the data from our previously published gene expression array [HG-U133A, 12 FTCs and 12 FAs] for these target genes (17). For miR-197, 57 of the 496 represented target genes showed significant underexpression in FTCs, compared with FA when using a cutoff value of –1.5-fold and a maximum P value of 0.05 (Supplemental Table 3, published as supplemental data on The Endocrine Society’s Journals Online web site at http://jcem.endojournals.org). Using the same criteria, 24 of the 278 target genes for miR-346 and 51 of 379 target genes predicted for miR-192 were significantly underexpressed in FTCs, compared with FAs (Supplemental Tables 4 and 5, published as supplemental data on The Endocrine Society’s Journals Online web site at http://jcem.endojournals.org).
To ensure specificity of our findings in the context of FTC, we repeated this analysis using the predicted target genes for miR-221, miR-222, and miR-146a, which are specific for papillary thyroid carcinogenesis (30). These analyses revealed that the PTC-miRs are not differentially regulated between FTC and FA. Between 418 and 566 target genes were present on the HG-U133A chip, but of those, only 20 (miR-146a, 4.8%) to 29 (miR-222, 5.1%) genes were significantly underexpressed in FTCs. This is significantly less than what we observed for the FTC-specific miR-192 (13.5%, P < 0.000004), miR-346 (8.6%, P < 0.018), and miR-197 (11.5%, P < 0.00011).
Validation of predicted target genes
To verify that in silco predicted miRNA target genes can be regulated by the respective miRNA in vitro, we selected two of 57 miR-197 targets (ACVR1, TSPAN3) and two of 24 miR-346 target genes (EFEMP2, CFLAR) for proof of principle (Supplemental Tables 3 and 4). We were able to successfully validate the two target genes (ACVR1 and TSPAN3) for miR-197 that were significantly underexpressed in FTCs, compared with FAs (1.9- and 1.5-fold, P = 0.00039 and P = 0.03) and normal thyroid control (Fig. 4). For ACVR1, differences in gene transcript expression are reflected by protein levels as well (Fig. 4, D and E). Similarly, the two miR-346 target genes EFEMP2 and CFLAR were underexpressed by 2.2-fold (P = 0.035) and 1.9-fold (P = 0.000014) in FTCs, compared with FAs (Fig. 4).
|
). Interestingly, overexpression of miR-346 resulted in a continuous reduction of EFEMP2 mRNA levels at both 12 h (down 1.2-fold) and 24 h (down 1.89-fold) (Fig. 5B). In contrast, overexpression of miR-346 did not significantly influence the transcript levels of CFLAR in our HEK293T model (Fig. 5B). Neither miRNA had any effect on the gene transcription of nontarget genes (e.g. ACVR1, TSPAN3 for miR-346 and CFLAR, EFEMP2 for miR-197).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Functional effect of miR-197 and miR-346
Overexpression of our most robustly validated miRNAs (miR-197 and miR-346) induces marked proliferation in vitro. As proof of principle, we validated the functional link between miR-197 and miR-346 and the transcriptional suppression of three target genes. First, EFEMP2 (or fibulin 4) is involved in stabilization and organization of extracellular matrix structures (34). There is evidence that EFEMP2 harbors tumor-suppressor functions, which we show to be inhibited by miR-346 deregulation (34, 35). Second, as a functional consequence of deregulated miR-197 in FTC, ACVR1 as well as tetraspanin 3 (TSPAN3) becomes underexpressed. Activin A and TGF-B1 are ligands for the activin A receptors type 1 (ACVR1) and have been shown to be potent growth inhibitors in various human cells, including thyroid epithelium (36). Whereas no functional data exist on TSPAN3, there are such data for CD63, another member of the tetraspan superfamily with highest homology to TSPAN3 (37). Expression levels have been shown to be inversely correlated with the metastatic potential in melanoma (37, 38). Finally, our findings show the limitations of in silco analysis when identifying miRNA target genes. For one (CFLAR) of the four genes tested in this study, a functional link between the miRNA and the potential target gene could not be established in vitro despite in silico evidence.
Implications of deregulated miRNAs for the accurate preoperative diagnosis of FTC
The overexpression of a small set of miRNAs with subsequent cascading down-regulation of target tumor suppressor genes represents a powerful mechanism in which a small but significant (1.2- to 2-fold range) overexpression can lead to larger downstream perturbations that inactivate numerous genes potentially participating in FTC genesis. These miRNAs and their target genes might therefore provide novel molecular markers to accurately differentiate malignant (FTC) and benign thyroid neoplasia (FA). Based on the set of differentially expressed miRNAs (miR-192, miR-197, miR-328, and miR-346) in our miRNA-chip experiment, we were able to correctly predict class labels (FTC vs. FA) in 74% of all cases. However, we need to critically discuss the usefulness of miRNAs for diagnostic purposes because in follicular thyroid neoplasias, the diagnosis must rely on material obtained from fine needle aspiration biopsies, and it is our observation that needle washout material does not provide enough of the small RNA fraction for reproducible analysis (our unpublished observation). We therefore suggest that the target genes of these miRNAs might provide for better diagnostic markers. Using the common approach of diagonal linear discriminant analysis and leave-one-out cross-validation method (17, 31), our miRNA target gene classifier (ACVR1, TSPAN3, and EFEMP2) achieved an accuracy of more than 87% to differentiate between FTC and FA in two independent sample sets (see Results). Whereas the molecular markers presented here perform similarly well as other proposed models based on gene expression profiling such as reported by Cerutti et al. (14) (e.g. 83% accuracy) or Umbricht et al. (16) (e.g. 77% accuracy), it does not perform superiorly to our previously identified three-gene signature (96.7% accuracy) (14, 16, 17). Nonetheless, all minimally invasive FTCs (03E077, 03E191, and 03E192) were correctly identified as a malignancy using the miRNA target gene classifier (ACVR1, TSPAN3, and EFEMP2). Considering the advancement over the last years to identify and validate such molecular markers, one currently unanswered question will need to be addressed. That is, if indeed there is an adenoma-carcinoma sequence in follicular thyroid cancer, what will be the treatment of choice for those patients diagnosed with FA preoperatively?
Suppression of endogenous miRNA expression: clinical implications
In our human thyroid cancer cell line models, the introduction of synthetic chemically modified anti-miRNA oligonucleotides directed against miR-197 or miR-346 induced a significant growth arrest. This phenomenon was observed in both FTC-133 and K5 FTC cells, whereas the papillary thyroid cancer cell line (NPA87), lacking deregulation of these miRNAs, was not affected. Recently it has been discussed and tested that interference with miRNA function opens novel opportunities for therapeutic intervention (9, 39, 40, 41). Our current study provides in vitro evidence for the feasibility of this approach for FTC, something that clearly will need further in vivo validation. However, it is tempting to propose that the interference with the deregulated miRNA profile in FTC might allow reactivation of suppressed target genes and ultimately affect an array of downstream targets to reverse the malignant phenotype or at least cause growth arrest. In addition, our findings indicate that the interference with specific miRNA(s) is not only cancer-type specific but also could be subhistology-specific in a given type of cancer, in our case, specific for FTC. In contrast, we show here that miR-221 and miR-222, which are implicated in PTC carcinogenesis, do not play a role in follicular neoplasia development (30).
In conclusion, our study shows that a small set of differentially regulated miRNAs are specifically deregulated in follicular thyroid cancer and potentially participate in the transformation from benign to malignant neoplasia. These small RNAs and their target genes might point us to new targets to improve preoperative diagnosis of follicular nodule and even therapy for a disease that continues to challenge us in the clinical setting.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure statement: The authors have nothing to disclose.
First Published Online July 5, 2006
Abbreviations: FA, Follicular adenoma; FTC, follicular thyroid carcinoma; miR and miRNA, micro-RNA; PTC, papillary thyroid carcinoma.
Received March 30, 2006.
Accepted June 28, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is frequently downregulated in a diversity of sporadic nonmedullary thyroid carcinomas. Oncogene 22:3412–3416[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  F.-Z. Wang, F. Weber, C. Croce, C.-G. Liu, X. Liao, and P. E. Pellett Human Cytomegalovirus Infection Alters the Expression of Cellular MicroRNA Species That Affect Its Replication J. Virol., September 15, 2008; 82(18): 9065 - 9074. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Eszlinger, K. Krohn, S. Hauptmann, H. Dralle, T. J. Giordano, and R. Paschke Perspectives for Improved and More Accurate Classification of Thyroid Epithelial Tumors J. Clin. Endocrinol. Metab., September 1, 2008; 93(9): 3286 - 3294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Tam The Emergent Role of MicroRNAs in Molecular Diagnostics of Cancer J. Mol. Diagn., September 1, 2008; 10(5): 411 - 414. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Bruchova, M. Merkerova, and J. T. Prchal Aberrant expression of microRNA in polycythemia vera Haematologica, July 1, 2008; 93(7): 1009 - 1016. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. N. Nikiforova, G. C. Tseng, D. Steward, D. Diorio, and Y. E. Nikiforov MicroRNA Expression Profiling of Thyroid Tumors: Biological Significance and Diagnostic Utility J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1600 - 1608. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sengupta, J. A. den Boon, I-H. Chen, M. A. Newton, S. A. Stanhope, Y.-J. Cheng, C.-J. Chen, A. Hildesheim, B. Sugden, and P. Ahlquist MicroRNA 29c is down-regulated in nasopharyngeal carcinomas, up-regulating mRNAs encoding extracellular matrix proteins PNAS, April 15, 2008; 105(15): 5874 - 5878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Weber and C. Eng Update on the Molecular Diagnosis of Endocrine Tumors: Toward -omics-Based Personalized Healthcare? J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1097 - 1104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Landi, F. Gemignani, A. Naccarati, B. Pardini, P. Vodicka, L. Vodickova, J. Novotny, A. Forsti, K. Hemminki, F. Canzian, et al. Polymorphisms within micro-RNA-binding sites and risk of sporadic colorectal cancer Carcinogenesis, March 1, 2008; 29(3): 579 - 584. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Pan, X. Luo, T. Toloubeydokhti, and N. Chegini The expression profile of micro-RNA in endometrium and endometriosis and the influence of ovarian steroids on their expression Mol. Hum. Reprod., November 1, 2007; 13(11): 797 - 806. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Tang, J. Gal, X. Zhuang, W. Wang, H. Zhu, and G. Tang A simple array platform for microRNA analysis and its application in mouse tissues RNA, October 1, 2007; 13(10): 1803 - 1822. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Zhang, J. E. Dahlberg, and W. Tam MicroRNAs in Tumorigenesis: A Primer Am. J. Pathol., September 1, 2007; 171(3): 728 - 738. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Negrini, M. Ferracin, S. Sabbioni, and C. M. Croce MicroRNAs in human cancer: from research to therapy J. Cell Sci., June 1, 2007; 120(11): 1833 - 1840. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Blenkiron and E. A. Miska miRNAs in cancer: approaches, aetiology, diagnostics and therapy Hum. Mol. Genet., April 15, 2007; 16(R1): R106 - R113. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
