| Nucleic Acids Research | Article |
Serial analysis of gene expression: rapid RT-PCR analysis of unknown SAGE tags
Introduction
Materials And Methods
Tag-sequences
RAST-PCR
Results
Discussion
References
Serial analysis of gene expression: rapid RT-PCR analysis of unknown SAGE tags
Received as resubmission June 21, 1999; Revised and Accepted July 19, 1999
ABSTRACT In a pilot study on SAGE on Reed-Sternberg cells we have sequenced 1055 tags representing 701 genes. Screening of the GenBank database resulted in the identification of a corresponding gene or EST for 490 of them. For 211 of the tags no homology could be detected. A major problem of the serial analysis of gene expression (SAGE) approach is how to further analyse the unknown tags. We have developed an RT-PCR-based method, rapid analysis of unknown SAGE tags (RAST-PCR), to analyse the expression of the corresponding genes. This approach can be used as a screening method to investigate whether or not the gene is differentially expressed between several cell types of interest.
INTRODUCTION
The serial analysis of gene expression (SAGE) technique allows the construction of a comprehensive expression profile in which each mRNA is defined by a specific 14mer (1-4). In contrast to subtraction and differential display techniques, SAGE also results in the quantification of expression levels of the corresponding genes. The effectiveness of the SAGE procedure was further demonstrated by the finding of 14 tags with an increased expression in p53 expressing colorectal cancer cells (5), an increased expression of MnSOD in a nitric oxide-resistant pheochromocytoma cell line (6), and high expression of the CC-chemokine TARC in Reed-Sternberg cells of Hodgkin's lymphoma (7). Recently, new improvements of the SAGE procedure have been reported which allow the use of limited amounts of input RNA (8), and enhance cloning efficiency via the cloning of longer concatemers (9,10). The human genome contains approximately 100 000 genes. At this moment, the sequences present in the GenBank files (known genes and EST) together comprise ~50% of all genes. Taking into account that not all sequences as present in the GenBank contain the most 3[prime] gene sequences, <50% of all the tags that can be identified with the SAGE technique will correspond to a known sequence or gene. Zhang et al. (4) detected homologies to known genes or EST sequences for 53-63% of the tags. Further analysis of the remaining unknown tags needs another approach; different to that used for the tags corresponding to known genes, for which it is relatively easy to select gene-specific primers or to obtain probes. Velculescu et al. (1) showed that a 13-nt primer could be used as a probe for the screening of a cDNA library, but this approach is labour intensive and requires a cDNA library of a specific cell type expressing the genes of interest. In this paper we describe an RT-PCR-based method which will allow a rapid analysis of unknown SAGE tags (RAST-PCR). The resulting PCR product can be used to obtain more gene-specific sequence data and in addition for the isolation of the full-length cDNA clone. The resulting, longer probes can be used as probes in a northern blot analysis or for an RNA in situ hybridisation analysis to determine the expression of the corresponding genes in various tissues.
MATERIALS AND METHODS
Tag-sequences
In a pilot experiment we applied the SAGE technique on the Hodgkin derived cell line L428 (7). We sequenced 1055 tags, representing 701 independent genes. A homology screening to the known human sequences (release 107.0 of the GenBank) and to EST sequences resulted in the identification of the corresponding sequence for 490 of the tags. The SAGE tag sequences identified in this study were used to set-up an RT-PCR procedure that will allow a rapid screening of the unknown SAGE tags. We selected six tags corresponding to known genes (Table 1) and 14 tags without homology to the known sequences.
Table 1. Overview of all tags analysed with the RAST-PCR
| Tag no. | Tag sequence | Corresponding gene | PCR productpredicted size(bp) | PCR productobtained size(bp) | Presence of NlaIII site |
| 125661 | actgggtcta | NM23-H2 | 204 | 200 | no |
| 148949 | agcacctcca | EF-2 | 112 | 110 | no |
| 208032 | atagtagctt | Fascin | 104 | 100 | no |
| 278636 | cacaaacggt | rib. protein S27 | 214 | 210 | no |
| 672779 | ggcacaaagg | TARC | 82 | 80 | no |
| 673954 | ggcagaggac | NM23-H1 | 186 | 190 | no |
| 70211 | acacagcaag | - | - | 600 | yes |
| 177315 | aggtcaggag | - | - | 260 | no |
| 291282 | cactactcac | - | - | 250 | no |
| 332055 | ccacacaccg | - | - | 160 | no |
| 335560 | ccactgtact | - | - | 120 | no |
| 355689 | cccgtccgga | - | - | 250 | no |
| 388150 | cctgtaatct | - | - | 320 | yes |
| 588196 | gattgcggat | - | - | 280 | no |
| 592194 | gcaagccaac | - | - | 100 | no |
| 618679 | gcctaagtcg | - | - | 250 | no |
| 633339 | gcgggcttgg | - | - | 200 | no |
| 860352 | tcagaagttt | - | - | 130 | no |
| 873906 | tcccccgtac | - | - | 110 | no |
| 1036353 | tttcaacaaa | - | - | 120 | yes |
RAST-PCR
Total RNA from different Hodgkin and non-Hodgkin cell lines was isolated with Trizol (Gibco/BRL, Paisley, UK), following the manufacturer's protocol. RNA samples were treated with DNase I (Roche Diagnostics, Almere, The Netherlands). First strand cDNA synthesis was primed with an oligo(dT)24 primer with a 5[prime] M13 tail [5[prime]-ctagttgtaaaacgacggccag-(t)24-3[prime]] according to the manufacturer's instructions using M-MLV Reverse Transcriptase (Gibco/BRL, Paisley, UK). The tag-specific primer consisted of 10 nt identified in the SAGE analysis with a 5[prime] NlaIII restriction site (catg) and 5 inosine nucleotides to increase the annealing temperature of the primers (5[prime]-IIIII-catg-tag sequence-3[prime]). For the PCR we used this 19-base tag-specific primer and a 20-nt primer (20bM13) corresponding to the 5[prime] tail of the oligo(dT) primer (5[prime]-agttgtaaaacgacggccag-3[prime]). The PCR was performed using standard conditions. The PCR program consisted of 5 min at 94°C, followed by 33 cycles of 30 s at 94°C, 30 s at 55°C, 30 s at 72°C. The final extension step consisted of 7 min at 72°C. Optimisation of the PCR was performed by testing different PCR buffers of which the MgCl2 concentration, the KCl concentration and the pH varied (PCR optimisation kit, Stratagene, La Jolla, CA, USA). For each PCR product, we verified whether the tag was present at the predicted location (downstream of the most 3[prime] NlaIII site in the full-length cDNA) within the analysed transcripts by digesting the PCR product with NlaIII. In case the tag sequence is indeed present at the predicted location, no NlaIII restriction site will be present in the obtained PCR product.
RESULTS
We tested the RAST-PCR procedure on six tags corresponding to known genes and on 14 tags corresponding to unknown genes all identified in the SAGE analysis of the L428 Hodgkin cell line (7). Total RNA isolated from different Hodgkin and non-Hodgkin lymphoma cell lines was used to test the RAST-PCR procedure. Application of the RAST-PCR protocol, using the standard PCR buffer, resulted in a PCR product in 17 of the 20 tag primers tested (for some examples see Fig. 1). A further improvement of the PCR for the remaining tags was achieved by using the PCR optimisation kit (Stratagene, La Jolla, CA, USA). PCR reactions were performed using all 12 different PCR buffers as present in the kit resulting in a PCR product for all of the three remaining tags. For two of them a good PCR product was obtained using buffer 8 containing 100 mM Tris-HCl (pH 8.8), 35 mM MgCl2, 750 mM KCl instead of 100 mM Tris-HCl (pH 9.0), 15 mM MgCl2 and 500 mM KCl in the normal PCR buffer. The sizes of the PCR products obtained for the known SAGE tags all corresponded to the size predicted from the reported sequences (Table 1). Restriction enzyme digestion of all 20 PCR products revealed the presence of an NlaIII site in three cases. No NlaIII restriction site was present in the remaining PCR products, indicating that the PCR product indeed represents a sequence downstream of the most 3[prime] NlaIII restriction site (Fig. 1). For these tags a rapid screening of a larger number of cell lines or tissues can be done to analyse whether or not the corresponding gene indeed is differentially expressed (for some examples see Fig. 2). In total we were able to obtain good PCR products for all of the 20 tags tested (e.g. 100%) using the RAST-PCR protocol. Restriction enzyme digestion with NlaIII indicated that 17 of the 20 tags (e.g. 85%) indeed represented true SAGE tags.
Figure 1. RAST-PCR analysis of unknown SAGE tags. PCR products and the corresponding NlaIII digests can be seen in the lanes. Lane 1, tag 177315 (catgaggtcaggag); lane 2, tag 177315 NlaIII digest; lane 3, 332055 (catgccacacaccg); lane 4, 332055 NlaIII digest; lane 5, 125661 (catgactgggtcta); lane 6, 125661 NlaIII digest; lane 7, 335560 (catgccactgtact); lane 8, 335560 NlaIII digest; lane 9, 355689 (catgcccgtccgga); lane 10, 355689 NlaIII digest; lane 11, 592194 (catggcaagccaac); lane 12, 592194 NlaIII digest; lane 13, 70211 (catgacacagcaag); lane 14, 70211 NlaIII digest. RAST-PCR was performed as described in the Results; primers used for the amplification were a 19-nt tag-specific primer (5[prime]-IIIII-catg-tag sequence-3[prime]) and a 20-nt primer corresponding to the 5[prime] tail of the oligo(dT) primer used for the cDNA synthesis reaction.
Figure 2. RAST-PCR analysis of unknown SAGE tags on a larger number of Hodgkin and non-Hodgkin cell lines. (A) RT-PCR analysis for the primer corresponding to SAGE tag 208032 (catgatagtagctt). (B) RT-PCR analysis for the primer corresponding to SAGE tag 278636 (catgcacaaacggt). (C) RT-PCR for GAPDH. The analysed samples are Pop (EBV transformed B-cell); Ray (EBV transformed B-cell); L428 (Hodgkin); L591 (Hodgkin); L1236 (Hodgkin); DEV (L&H Hodgkin); Rose (large cell B-cell lymphoma); Raji (Burkitt's lymphoma); Ver (B-cell non-Hodgkin lymphoma); Jurkat (T-cell leukemia). RT-PCR analysis was performed as described in the Results. Primer 208032 is an example of a tag corresponding to the fascin gene and shows differential expression in the cell lines analysed, whereas primer 278636 is an example of an unknown tag corresponding to a gene which is expressed in all the cell lines.
DISCUSSION
A major problem of the SAGE approach is how to further analyse the unknown SAGE tags. Therefore, we have developed an RT-PCR based method, which can be used to determine whether or not a tag indeed represents a differentially expressed gene. In addition, the resulting tag-PCR product can be used as a probe in a northern blot analysis to obtain information about the level of expression and the size of the corresponding transcript. Moreover, the tag-PCR product can be sequenced to obtain more gene-specific sequence information, to isolate full-length cDNA clones and to analyse gene expression in various tissues using an RNA in situ hybridisation. This approach allows a rapid analysis of the expression of unknown tags in tissue or cell populations of interest.
In total we were able to obtain a good PCR product for 100% of the cases. In three cases the presence of an NlaIII restriction site was demonstrated indicating that those PCR products were not derived from the gene corresponding to the SAGE tag. Remarkably, the size of one of these PCR products was relatively long (600 bp) as compared to the other PCR products (with an average length of 177 bp) and to the theoretical fragment size (256 bp) that will be obtained after digestion with NlaIII. In none of the other PCR products was an NlaIII restriction site present, indicating that these PCR products (e.g. 85% of the analysed tags) indeed represent the genes corresponding to the SAGE tags. Application of the RAST-PCR protocol as described above will allow quick screening of unknown SAGE tags, thereby making it possible to distinguish the specifically over- or under-represented tags from those expressed at levels comparable to the level in control tissue.
REFERENCES
*To whom correspondence should be addressed. Tel: +31 50 361 4811; Fax: +31 50 363 2510; Email: s.poppema{at}path.azg.nl
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. P. So, R. F. B. Turner, and C. A. Haynes Increasing the efficiency of SAGE adaptor ligation bydirected ligation chemistry Nucleic Acids Res., July 6, 2004; 32(12): e96 - e96. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Fizames, S. Munos, C. Cazettes, P. Nacry, J. Boucherez, F. Gaymard, D. Piquemal, V. Delorme, T. Commes, P. Doumas, et al. The Arabidopsis Root Transcriptome by Serial Analysis of Gene Expression. Gene Identification Using the Genome Sequence Plant Physiology, January 1, 2004; 134(1): 67 - 80. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Matsumura, S. Reich, A. Ito, H. Saitoh, S. Kamoun, P. Winter, G. Kahl, M. Reuter, D. H. Kruger, and R. Terauchi Gene expression analysis of plant host-pathogen interactions by SuperSAGE PNAS, December 23, 2003; 100(26): 15718 - 15723. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Hauser, Y.-J. Li, S. Takeuchi, R. Walters, M. Noureddine, M. Maready, T. Darden, C. Hulette, E. Martin, E. Hauser, et al. Genomic convergence: identifying candidate genes for Parkinson's disease by combining serial analysis of gene expression and genetic linkage Hum. Mol. Genet., March 15, 2003; 12(6): 671 - 677. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. D. Patino, O. Y. Mian, and P. M. Hwang Serial Analysis of Gene Expression: Technical Considerations and Applications to Cardiovascular Biology Circ. Res., October 4, 2002; 91(7): 565 - 569. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Kreeft, C. J.A. Moen, M. H. Hofker, R. R. Frants, E. Vreugdenhil, M. J.J. Gijbels, L. M. Havekes, and N. A. Datson Identification of Differentially Regulated Genes in Mildly Hyperlipidemic ApoE3-Leiden Mice by Use of Serial Analysis of Gene Expression Arterioscler Thromb Vasc Biol, December 1, 2001; 21(12): 1984 - 1990. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-J. Chen, J. D. Rowley, and S. M. Wang Generation of longer cDNA fragments from serial analysis of gene expression tags for gene identification PNAS, January 4, 2000; 97(1): 349 - 353. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||