| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Molecular Diagnostics and Genetics |
1 Molecular Medicine, Department of Medical Sciences, Uppsala University, Uppsala, Sweden.
aAddress correspondence to this author at: Molecular Medicine, Department of Medical Sciences, Uppsala University, Uppsala Academic Hospital, Entrance 70, 3rd Floor, Research Department 2, 751 85 Uppsala, Sweden. Fax 46-18-553601; e-mail Ann-Christine.Syvanen{at}medsci.uu.se.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
80% of the 25 000 human genes undergo alternative splicing. Alternative splicing may be associated with human diseases, particularly with cancer, but the molecular disease mechanisms are poorly understood. Convenient, novel methods for multiplexed detection of alternatively spliced transcripts are needed. Methods: We devised a new approach for detecting splice variants based on a tag-microarray minisequencing system, originally developed for genotyping single-nucleotide polymorphisms. We established the system for multiplexed detection of 61 alternatively spliced transcripts in a panel of 19 cancer-related genes and used it to dissect the splicing patterns in cancer and endothelial cells.
Results: Our microarray system detected 82% of the splice variants screened for, including both simple and complex splice variants, in at least 1 of the leukemia cell types analyzed. The intraassay CV values for our method ranged from 0.01 to 0.34 (mean, 0.13) for 5 replicate measurements. Our system allowed semiquantitative comparison of the splicing patterns between the cell lines. Similar, but not identical, patterns of alternative splicing were observed among the leukemia cell lines. Size analysis of the PCR products subjected to the tag-array minisequencing system and real-time PCR with exon-junction probes verified the results from the microarray system.
Conclusions: The microarray-based method is a robust and easily accessible tool for parallel detection of alternatively spliced transcripts of multiple genes. It can be used for studying alternative splicing in cancer progression and for following up drug treatment, and it may be a useful tool in clinical diagnostics for cancer and other disorders.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The most common forms of alternative splicing of mRNAs are the exclusion or inclusion of a complete exon, mutually exclusive splicing by selection of 1 exon variant, and the use of alternative 5'- or 3'-splice sites (8). Results obtained with bioinformatics methods indicate that the majority of the alternative splicing events may lead to changes in the protein product (4)(5). On the protein level, alternatively spliced mRNAs may lead to the use of alternative translation initiation sites, altered translation termination attributable to frame shifts, or the creation or removal of a stop codon, thus causing truncation or extension of the protein product (9).
For most genes, the functional significance of alternative splicing is poorly understood. Alternative splicing may be most important in complex systems in which the genetic information must be processed differently at different times, for example, in the immune system or during different developmental stages in which a very high degree of diversity is required, such as in the nervous system (5)(8)(10). Many genes that have multiple alternative splice variants, such as receptors and transcription factors, are involved in cell signaling and regulation of gene expression (9). An early study suggested that point mutations causing alternative splicing would be involved in 15% of the inherited diseases (11), and more recently, alternative splicing has been associated with a variety of diseases (12)(13). The most well-known forms of disease-related splicing defects are point mutations in genomic splice sites, exemplified by a mutation causing missplicing of exon 18 in the breast cancer gene, BRCA1 (14). Alternative splicing occurs frequently in human cancer cells (15)(16), and alternatively spliced tumor-specific transcripts might serve as diagnostic markers for cancer (17).
To increase our understanding of the relationship between alternative splicing and human disease, methods are required that allow detection of alternatively spliced mRNAs in a large number of clinically relevant tissue samples and in multiple genes. Highly multiplexed analyses using oligonucleotide microarrays are useful for genomic-scale surveys of alternative splicing, but the costs involved in these methods prohibit their application to large quantities of samples. The most commonly used method for detecting alternatively spliced transcripts in large sample sets are reverse transcription-PCR, followed by size analysis of the PCR products, or real-time PCR using PCR primers that span the splice junctions of the genes of interest (18)(19). A limitation of the PCR-based methods is that they are difficult to multiplex for parallel detection of alternative transcripts from multiple genes.
In this study, we demonstrate that the minisequencing-based tag-microarray system routinely used in our laboratory for multiplexed genotyping of single-nucleotide polymorphisms (20)(21) can be adapted to multiplex detection of alternatively spliced transcripts. We applied the system to screen for several different types of alternatively spliced transcripts in a panel of 19 cancer-related genes in human leukemia cell lines and in normal endothelial cell lines.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The cells were grown in RPMI 1640 containing 100 mL/L fetal bovine serum. HAEC and HUVEC (Cascade Biologics, Inc.) cells were grown in medium 200 with Low Serum Growth Supplement (LSGS Kit; Cascade Biologics, Inc.). All cell lines were grown at 37 °C in a humidified atmosphere of 5% CO2. Passages 3–5 of the cultures were harvested at 80% confluence according to the manufacturer’s instructions.
RNA isolation
Total RNA was isolated from the cells by lysis in guanidine isothiocyanate (TRIZOL® Reagent; Gibco BRL), and the RNA samples were preserved at –70 °C until use. The purity and integrity of the RNA preparations were assessed by measuring their ultraviolet absorbance at 260 nm and 280 nm and by assaying the RNA on the Agilent 2100 Bioanalyzer (Agilent Technologies). High-quality RNA with an A260/A280 ratio above 1.9 and intact ribosomal 28S and 18S RNA were used for the microarray experiments.
assay design
We included 19 genes, with altogether 61 known alternatively spliced transcripts, in the assay panel (see Table S1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol52/issue2). The genes and their splice variants were identified in the database Atlas of Genetics and Cytogenetics in Oncology and Hematology (http://www.infobiogen.fr/services/chromcancer). Genes were selected from those with more than 1 transcript, according to Ensembl Human Build release, v.24.34e.1 and v.25.34e.1 (http://www.ensembl.org) or according to the NCBI Entrez/GenBank (http://www.ncbi.nlm.nih.gov/entrez). The selected genes are expressed in bone marrow or peripheral blood, according to the GeneCards resource (http://www.dkfz-heidelberg.de/GeneCards/cgi-bin) or according to the literature.
The PCR primers were designed using the Primer3 software (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi), with the goal to use as few PCR primers per gene as possible. Altogether, 42 PCR primer pairs were required for amplification of the 61 splice variants of all genes. The PCR products spanning the fragment of interest were analyzed with the BLASTn program (http://www.ncbi.nlm.nih.gov/BLAST) to ensure a specific hit to the human genome sequence. Table S1 in the online Data Supplement shows the expected sizes of the PCR products.
We designed 43 minisequencing primers that annealed immediately adjacent to nucleotides that define the 61 splice variants of the 19 selected genes (See Table S2 in the online Data Supplement). The 5' end of each minisequencing primer contained a 20-nucleotide tag sequence from the Affymetrix GeneChip Tag Collection. Absence of hairpin loop formation in the minisequencing primers was confirmed by use of Cybergene software (http://www.cybergene.se/primer.html). The oligonucleotides were synthesized by Integrated DNA Technologies. The PCR and minisequencing primer sequences are available from the authors on request.
first-strand CDNA synthesis and PCR
The RNA samples were treated with 1 U of RQ1 RNase-free DNase (Promega Corporation) per microgram of total RNA before reverse transcription. An aliquot of 3 µg of total RNA was used for first-strand cDNA synthesis in a total reaction volume of 20 µL, using the Superscript II reverse transcription assay (Gibco BRL) according to the manufacturer’s protocol. The quality of the cDNA was evaluated on a 1% agarose gel. We used 1 µL of a cDNA synthesis reaction mixture as template in PCR with 0.2 µM of each amplification primer and 1 µL of AccuPrimeTM Taq DNA polymerase in 25 µL of AccuPrime PCR Buffer 1 (Invitrogen Corporation). The PCR reactions were performed in a PTC225 Thermal Cycler (MJ Research), starting with a 3-min activation of the enzyme at 94 °C, followed by 40 amplification cycles of 94 °C for 30 s, 55 °C for 30 s, and 68 °C for 45 s, with a final extension at 72 °C for 7 min. Each alternatively spliced transcript was amplified in an individual PCR reaction. All PCR products from the same cell line sample were pooled and concentrated by use of Microcon® YM-30 Centrifugal Filter Devices (Millipore Corporation). The correct sizes of the PCR products were verified by use of an Agilent 2100 Bioanalyzer and the DNA 1000 LabChips (Agilent Technologies).
preparation of microarrays
The complementary tag oligonucleotides were covalently coupled to CodeLinkTM Activated Slides (Amersham Biosciences) by the mediation of an NH2 group at their 3' end. The oligonucleotides were printed in duplicate on the slides at a concentration of 25 µmol/L in 150 mmol/L sodium phosphate (pH 8.5), using a ProSys 5510A instrument (Cartesian Technologies Inc.) with 4 Stealth Micro Spotting Pins (TeleChem International Inc.) as described previously (21). The oligonucleotides were printed on the microscope slides in an "array of arrays" format that allowed the parallel analysis of up to 80 individual samples on each microscope slide (22). In each "subarray", a fluorophore-labeled oligonucleotide was included as a printing control. A reference oligonucleotide complementary to a synthetic template included in the primer extension reaction mixtures to monitor the difference in incorporation efficiency between the 4 nucleotides by the DNA polymerase was also included in each subarray. Finally, an oligonucleotide designed not to hybridize to any of the oligonucleotides present in the reaction mixture was included in each subarray to be used for background corrections. After printing, the slides were treated with ethanolamine and stored desiccated, in the dark, until use (20).
minisequencing
A 7-µL aliquot of the concentrated PCR products was treated with 5 U of Exonuclease I and 1 U of shrimp alkaline phosphatase (USB Corporation) to remove excess of PCR primers and deoxynucleoside triphosphates (dNTPs). The cyclic minisequencing reactions contained the 43 tagged minisequencing primers at 10 nmol/L, 0.1 µmol/L Texas Red-ddATP, Tamra-ddCTP, and R110-ddGTP, 0.2 µmol/L Cy5-ddUTP (Perkin-Elmer Life Sciences), and 0.064 U/µL of KlenThermaseTM DNA polymerase (GeneCraft). The minisequencing primer extension reactions were performed for 55 cycles of 95 °C and 55 °C, for 20 s each. The arrayed slides carrying the captured complementary tag oligonucleotides were preheated to 42 °C in a custom-made aluminum reaction rack with a silicon rubber grid forming 80 separate reaction wells on each slide (22). The minisequencing reaction products were added to each well on the slide and allowed to hybridize to the complementary tag oligonucleotides at 42 °C for 2.0 h to 2.5 h, followed by washing of the slide.
data analysis
The fluorescence signals were measured from the microarray slides by use of a ScanArray® Express instrument (Perkin-Elmer Life Sciences), and the signal intensities were determined with the QuantArray® analysis 3.1 software (Perkin-Elmer Life). The signals from the 4 different fluorophores were normalized by adjusting the laser power of the array scanner to give equal signals for the reference oligonucleotides in a subarray. The splice variants of each gene were detected based on the signal-to-noise ratios between the mean value of the fluorescence signals from each nucleotide expected to be incorporated, divided by the mean value of the background fluorescence of the respective nucleotide from the spot containing the oligonucleotide, to which no extension primer was expected to hybridize.
real-time PCR
PCR primers and dual-labeled TaqMan hybridization probes were designed for 9 of the genes by use of the Primer3 software (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi). Where possible, the TaqMan probes were designed to flank the exon junctions of the transcripts, but in some cases exon-specific probes were designed. The 5' ends of the probes for the 2 transcripts of a gene were labeled with either the fluorophore 6-carboxyfluorescein (6-FAM) or HEX, and the 3' ends were labeled with Tamra, as a quenching dye. A total of 28 PCR primers were purchased from Integrated DNA Technologies, and 19 labeled TaqMan probes were purchased from Thermo Electron Corporation. Real-time PCR was run using 0.2 µM of each amplification primer and labeled probe and 1 µL of the 20-µL cDNA synthesis reaction product in 25 µL of TaqMan Universal PCR Mastermix (Applied Biosystems) containing ROX as a reference dye. The PCR conditions were initial activation of the enzyme at 95 °C for 10 min, followed by 45 cycles of 95 °C for 15 s, 54 °C for 19 s, and 60 °C for 1 min in a Stratagene Mx3000P instrument. The signals for the 6-FAM and HEX fluorophores were corrected for the fluorescence of the ROX-labeled reference fluorophore, and threshold cycle (Ct) values, which correspond to the first PCR cycle at which a fluorescent signal is detectable, were recorded during PCR. Transcripts that yielded Ct values <44 were considered to be present in a cell line.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and the alternatively spliced transcripts screened for are detailed in Table S1 in the online Data Supplement. The majority of the selected genes are involved in hematopoiesis. Several of them are oncogenes, such as the BCL2, MYB, EVI1, TCL1B, and RUNX1genes. Chromosomal translocations involving RUNX1 are well documented and have been associated with several types of leukemia (23). Another of the genes, WT1, encodes a transcription factor that can activate the protooncogene MYC. Genes for kinases and kinase inhibitors, such as the CDKN2B, DOK1, KITLG, MAP4K4, NPM1, and PTK2Bgenes are also included in the panel. Using the tag-array minisequencing system, we screened for 61 possible alternatively spliced transcripts of these 19 genes in 5 leukemia cell lines and in 2 endothelial cell lines.
|
Our strategy for detecting the splice variants expressed from the genes is based on the extension of gene-specific minisequencing primers with fluorescent dideoxynucleoside triphosphates (ddNTPs), each of which define a specific splice variant (see Table S2 in the online Data Supplement). For most genes, minisequencing primers that anneal immediately adjacent to a splice junction were used, followed by extension of the primer with fluorescently labeled nucleotides that define the alternatively spliced transcripts, as exemplified in Fig. 2A
for the NPM1 gene. In this gene, extension of the primer that anneals to the 3'-nucleotides of exon 1 with a fluorescently labeled ddA defines the presence of exon 2, and extension of the primer with a labeled ddG defines the presence of exon 3. The same design was used for splice variants caused by alternative 5'- or 3'-splice sites. In those genes in which the first nucleotide position after a splice junction was identical in 2 alternatively spliced variants, the minisequencing primer was designed to anneal 1 or more nucleotides beyond the splice junction, adjacent to the first nucleotide that differs between the splice variants. For the BCL2L11 gene, 3 splice variants were detected by use of a single minisequencing primer, but most genes with multiple alternatively spliced transcripts required more than 1 minisequencing primer. For example, the MAP4K gene, in which 4 alternatively spliced variants were identified, required 1 minisequencing primer to define the exon that supersedes exon 15 in 2 of the alternative transcripts. A second minisequencing primer defines the presence or absence of exon 17 in transcripts containing exon 16 (Fig. 2B
). WT1 is the only one of the selected genes with a splicing pattern that can be only partly defined by our system. In WT1, 2 primers define 2 exons, spaced far apart, that might be deleted in different transcripts. Extension of one of the primers with either ddA or ddC defines the presence or absence of exon 5, and the extension of a second primer with ddA or ddG depends on the presence or absence of an additional sequence at the end of exon 9. Although we can detect the presence or absence of different exons, we do not know which of the exons are present on the same transcript
|
Each minisequencing primer contains a unique tag sequence in its 5' end that allows the extension products to be captured at a specified position on the microarray surface (Fig. 2C
). Thus, a 4-color fluorescence signal pattern will be generated on the microarray to detect each of the 61 alternatively spliced variants. Fig. 2D
shows the 4-color fluorescence pattern from the extended primers in the K-562 cell line in one of the subarrays.
detection of alternatively spliced transcripts
We screened 5 leukemia cell lines (HL-60, BSM, NB-4, K-562, and SD-1) and 2 endothelial cell lines (HAEC and HUVEC) for the 61 detectable splice variants in 5 parallel reactions on the same microarray slide. Table S2 in the online Data Supplement exemplifies the numeric fluorescence and CV values measured from the microarray from the HL-60 cell line, and it illustrates the interpretation of the result based on the signal-to-noise (s/n) ratios obtained after scanning the microarray at 4 wavelengths and after careful normalization of the power of the 4 lasers, using the reaction control signals. The s/n ratios were calculated by dividing the mean value of the expected signal from a primer extension reaction that defines a splice variant by the mean background fluorescence value from the negative control spot for the appropriate fluorophore. We used a conservative detection limit based on a s/n ratio >10 to be confident that all of the detected splice variants were real and not the result of signals from a possible small sequence-dependent misincorporation of labeled ddNTPs by the DNA polymerase, which might occur for some variants. Table S3 in the online Data Supplement provides the s/n ratios measured for the splice variants in each of the 19 genes in the cell lines. According to the applied criterion of s/n ratios >10, 50 of the possible 61 splice variants (82%) were detected in at least 1 of the cancer cell lines. In the endothelial cell lines, the number of detected alternatively spliced transcripts was lower, 31 (51%), presumably because the cancer-related genes are not expressed or are expressed at lower concentrations in the endothelial cell lines.
We also calculated a signal fraction (s/f) value that describes the fraction of the total signal contributed by each detectable splice variant to the total signal from all detected splice variants of a particular gene. It should be noted that the incorporation efficiency between the 4 labeled ddNTPs in the primer extension reaction may differ in a sequence-dependent fashion, despite normalization of the laser power. Therefore, the s/f values do not provide an accurate measure of the actual abundance of each splice variant, but they do facilitate comparison of the splicing patterns between the cell lines. For some of the genes, the s/f values revealed differences in the splicing patterns between the cell lines. For example, the only transcript of RUNX1 detected in the endothelial cell lines is the b-variant, and the leukemia cell lines express all of the transcripts except the c-variant. The RUNX1 c-variant is only detected in the BSM and SD-1 cells, both of which are ß-lymphoblastoid cell lines. Similarly, the splicing pattern of the RUNX1T1 gene in the BSM and SD-1 cells differs from that in the 3 other leukemia cell lines. The PTK2B c-variant was clearly expressed in the cancer cell lines and absent in the endothelial cells. The CDKN2B a-variant is the dominant transcript in the endothelial cell lines, whereas the cancer cell lines express only or mainly the b-variant.
To validate our results, we compared the splice variants detected in the microarray analysis from the 7 cell lines with the sizes of the PCR products originally subjected to minisequencing primer extension using the Bioanalyzer instrument (Table 1
). For 385 (90%) of the 427 alternatively spliced transcripts screened for by the minisequencing system, the size analysis verified our results, either positive or negative. A PCR product corresponding to the expected size was clearly detectable for 266 of the 291 (91%) splice variants detected by the microarray analysis, and 119 of the undetectable splice variants had no visible PCR product of the expected size. A PCR fragment of the expected size was observed as a very weak band for 17 of the possible splice variants that remained undetected by the microarray analysis, whereas the microarray system detected 25 additional alternative transcripts, most likely reflecting a higher sensitivity of detection of the microarray system.
|
Finally, to validate the results from the minisequencing analysis (Table 2
), we used real-time PCR using TaqMan probes for 9 randomly selected genes with simple splicing patterns. With one exception, all 95 of the splice variants that were detected by the TaqMan probes were also detected by the microarray system, and 84% of the transcripts detected by the minisequencing system were detected by real-time PCR. The transcripts that remained undetected by the TaqMan probes had low signals on the microarray, which implies that they were expressed at low concentrations in these cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A major difference between our minisequencing microarray strategy for detection of splice variants and hybridization using oligonucleotide microarrays is that our system is based on PCR. An obvious advantage of using PCR is the possibility of detecting splice variants that are present at very low abundance in an RNA sample. A limitation shared by all PCR-based methods, compared with well-designed hybridization microarrays (24), is that determination of the relative abundances of the splice variants of a gene that are present in a sample is only semiquantitative. The reason for this limitation is that the efficiency of the PCR amplification depends on the sequence of the PCR primers and on the sizes and sequences of the amplicons. For detecting alternative transcripts, the sequences and/or the sizes of the amplicons, or alternatively, the sequence of at least one of the PCR primers for each transcript, differ by necessity. Thus, in PCR-based methods, the measured signals reflect the relative amount of PCR product corresponding to each splice variant, rather than the relative abundance of the alternatively spliced transcripts from each gene in the original sample. Accurate quantification of the original amount of each splice variant would require a competitive PCR strategy with a calibrator that is highly similar to each alternative splice variant. Recently, a primer extension strategy similar to ours, for detecting alternative splice variants, was described (25). This system is also based on PCR, followed by primer extension, but uses detection of the primer extension products by mass spectrometry instead of fluorescence, as in our method. Analogous to our method, quantitative determination of the relative amount of alternatively spliced transcripts in the original sample is hampered by differences in PCR efficiency between amplicons of different sizes, although this limitation does not appear to be recognized in the study. However, as shown by the data in Table S3 in the online Data Supplement and Fig. 3
, our system allows comparison of the relative abundances of the splice variants from each gene between the cell samples. Thus, our PCR-based system may be used for sensitive monitoring of changes in the relative abundances of alternatively spliced transcripts—for example, in serially collected samples or between tissues.
|
Real-time PCR with hybridization probes targeted at the splice junctions is the most frequently used method for detection of alternatively spliced transcripts at present (19)(26)(27). In addition, in real-time PCR, accurate quantitative analysis requires a standard sequence to correct for sequence-dependent differences in PCR efficiency (18)(19)(28). Compared with real-time PCR, an advantage of our microarray-based system is that multiple splice variants can be analyzed in parallel to generate splicing patterns of multiple genes, whereas real-time PCR is difficult to multiplex. Another drawback in real-time PCR, as well as other hybridization-based methods (24), is that a unique probe is required for each splice variant under study, compared with a system based on primer extension, in which the same primer can be used to define alternative transcripts of the same gene. Yet another advantage of our microarray-based system is its array-of-arrays format, which facilitates detection of up to 200 primer extension products in 80 samples on each microarray slide (21), in contrast to conventional microarrays for expression profiling, which at present can analyze only a single sample per microarray.
A method based on invasive cleavage of RNA (the Invader Assay) avoids a PCR step and may therefore be potentially more accurate for determining the relative abundance of alternative transcripts than the PCR-based methods (29). However, sequence-dependent factors may affect the efficiency of the consecutive, linear, invasive cleavage reactions used for signal amplification in this method. An interesting method for multiplexed analysis of alternative transcripts is to allow 2 oligonucleotides that flank the splice junctions in each splice variant to hybridize to the RNA targets, followed by ligation of the probes, amplification with the aid of universal PCR primers sequences, capture of the products on bead arrays by use of sequence-specific oligonucleotide tags, and detection by a label on one of the PCR primers (30). Another interesting approach describes the use of polymerase colony technology to detect alternatively spliced transcripts by use of fluorescently labeled exon-specific hybridization probes (31), or by detecting exon-specific single nucleotides by minisequencing (32). The "polony" technology permits amplification of individual nucleic acid molecules embedded in an acrylamide matrix. As each polony originates from a single cDNA molecule, the relative amounts of each splice variant can be determined by counting the polonies corresponding to each splice variant. This method combines the high sensitivity of PCR with the possibility of quantitative analysis of the alternatively spliced transcripts. The novel technologies described above rely on equipment and reagents that are not generally available and thus have not yet gained widespread use.
Recent systematic studies using both bioinformatics and experimental approaches imply that many genes display different patterns of alternative splicing in cancer tissues and in normal tissues (8)(26). Thus, alternatively spliced transcripts of cancer-related genes have the potential to serve as markers of cancer progression and for follow-up of treatment with anticancer drugs (33). In this study, we used a panel of 19 genes, which may be involved in leukemia in different ways, as a model system for developing a microarray-based minisequencing system for detecting alternatively spliced transcripts. We show that the system can be designed for detecting both simple and complex patterns of alternative splicing and for monitoring changes in the splicing patterns of multiple genes. On the basis of our results, we conclude that our system is a robust and easily accessible alternative for studying the role of alternative splicing in cancer progression and for sensitive follow-up of the consequences of drug treatment. The format of our assay is also well suited for clinical diagnostics, in which typically a limited set of biomarkers are analyzed in many samples in each series.
This study was supported by grants from the Swedish Research Council, the K & A Wallenberg Foundation, and the Swedish Cancer Society.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Nonstandard abbreviations: HAEC, human aortic endothelial cell; HUVEC, human umbilical vein endothelial cell; dNTP, deoxynucleoside triphosphate; ddNTP, dideoxynucleoside triphosphate; Ct, threshold cycle; s/n, signal-to-noise ratio; and s/f, signal fraction.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit genes. Unique splicing patterns are observed in dorsal root ganglia. J Biol Chem 2004;279:46234-46241.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  R. Klinck, A. Bramard, L. Inkel, G. Dufresne-Martin, J. Gervais-Bird, R. Madden, E. R. Paquet, C. Koh, J. P. Venables, P. Prinos, et al. Multiple Alternative Splicing Markers for Ovarian Cancer Cancer Res., February 1, 2008; 68(3): 657 - 663. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Ben-Dov, B. Hartmann, J. Lundgren, and J. Valcarcel Genome-wide Analysis of Alternative Pre-mRNA Splicing J. Biol. Chem., January 18, 2008; 283(3): 1229 - 1233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Milani, M. Gupta, M. Andersen, S. Dhar, M. Fryknas, A. Isaksson, R. Larsson, and A.-C. Syvanen Allelic imbalance in gene expression as a guide to cis-acting regulatory single nucleotide polymorphisms in cancer cells Nucleic Acids Res., March 12, 2007; 35(5): e34 - e34. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
