Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (115K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Liu, W.
Right arrow Articles by Roberts, T. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Liu, W.
Right arrow Articles by Roberts, T. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 3616-3618  

Discovery of estrogen-responsive genes using an improved method which combines subtractive hybridization and PCR
Acknowledgements
References

Discovery of estrogen-responsive genes using an improved method which combines subtractive hybridization and PCR

Discovery of estrogen-responsive genes using an improved method which combines subtractive hybridization and PCR

Wei Liu, Wen Su*, Thomas M. Roberts

Division of Cellular and Molecular Biology, Dana-Farber Cancer Institute and Department of Pathology,Harvard Medical School, Boston, MA 02115, USA

Received March 9, 1998; Revised and Accepted June 11, 1998

ABSTRACT

Here we describe a reliable method for isolating genes that are differentially expressed in two cell populations. The method is a combination of subtractive hybridization and PCR. Among many improvements to previously described methods is the incorporation of a new technology into the procedure which sterilizes(inactivates) PCR amplicons, and thereby overcomes the limitation of similar procedures. To test this improved method, we conducted a search for estrogen-responsive genes. Estrogen-regulated genes dominated the subtracted libraries after four rounds of subtractive hybridizations. Four estrogen-regulated genes were identified from the initial screening.

Subtractive hybridization is often used to isolate genes that are expressed at different levels in two cell populations (1,2). The incorporation of PCR technology makes it possible to carry out unlimited cycles of subtractive hybridizations, since the subtracted products are amplified. Therefore, theoretically, it allows for the isolation of all differentially expressed genes (3-6). To make use of the PCR technology, it is critical that the driver cDNAs do not co-amplify with tracer cDNAs. We have identified that the limiting factor of previous PCR-subtractive hybridization procedures was the co-amplification of driver cDNA amplicons with tracer cDNA amplicons. This conclusion was derived from the observation that driver cDNAs appeared in tracer cDNA populations after the subtractive hybridization procedures. To overcome this limitation, we have introduced a DNA sterilization method into the PCR-subtractive hybridization procedure. By employing a novel chemical (isopsoralen IP-10) that sterilizes PCR amplicons without interfering with their hybridization properties, the driver cDNAs were subjected to a 106-108-fold sterilization prior to hybridization reactions, which is a 104-105-fold increase over previous methods. The artefacts of the PCR-subtractive hybridization procedure were greatly reduced by isopsoralen incorporation.

We applied this improved method to the search for estrogen-regulated genes. Identification of target genes of estrogen is a crucial step toward understanding the diverse functions of estrogen. To date, only a limited number of such genes have been identified; among these are pS2, the progesterone receptor, c-fos and c-myc (7-9). In the current study, we have identified four genes from initial screenings. Three of them were novel estrogen-responsive genes.

The improved subtractive hybridization method was developed based on the `gene expression screen', which involves multiple rounds of subtractive hybridization and PCR (5,6). The flow diagram to isolate up-regulated genes is presented in Figure 1. By using the positive control pS2 gene, we identified that the major obstacle in the `gene expression screen' procedure was that a significant fraction of the driver cDNAs was not extracted by streptavidin and, therefore, co-amplified with tracer cDNAs. Streptavidin extraction of driver cDNA in the hybridization mixture depends on an efficient biotinylation reaction. The probability to crosslink k number of biotins to one DNA molecule can be calculated by the equation

where [lambda] is the average number of biotin per DNA molecule (Poisson distribution). Even the best photobiotinylation reagents could only add one biotin per 50-100 bases of DNA. Assuming there are four biotins per DNA molecule on average, 2% of the DNA in the driver cDNA population will be without biotin ([lambda] = 4, k = 0 -> P = 0.02). Streptavidin extraction will not be able to remove these unbiotinylated DNAs. These DNAs must be sterilized to prevent them from co-amplifying with and contaminating tracer cDNAs. The restriction enzyme removal of the PCR linkers on driver cDNAs could only sterilize these driver cDNA amplicons up to 99-99.9% (102-103-fold sterilization). In our initial work, we detected that a significant amount of driver cDNAs co-amplified with tracer cDNAs, which severely interfered with the subtractive hybridizations. In an effort to further sterilize driver cDNAs, we introduced isopsoralen IP-10 incorporation into the procedure, which resulted in a 104-105-fold increase in sterilization efficiency (10,11). The total sterilization efficiency reached 106-108-fold. The driver cDNA contamination problem was therefore resolved by this amplicon sterilization technology.


Figure 1. The flow diagram to isolate estrogen up-regulated genes. `+' and `-' indicate with and without estrogen treatment, respectively. To isolate estrogen down-regulated genes, the procedure was performed in parallel, with the `+' and `-' sign exchanged. `BD' represents biotinylated driver DNA. The hybridization was carried out for 20 and 2 h as indicated. The -2 cDNA and -1 cDNA pools in the drivers of 3rd and 4th subtraction help suppress low and high abundant common genes, respectively. Double-stranded cDNA was synthesized from 2 µg of mRNA extracted from MCF-7 cells using the `Copy kit' (Invitrogen). The cDNA was divided into two aliquots, digested with the 4 bp restriction enzymes MspI or MspI plus TaqI, respectively. Digested cDNA fragments (both driver and tracer) were combined and ligated with the ds-phosphorylated linker: 5[prime]-CGACGGCCAGGAAGCTTTTA-3[prime] and 3[prime]-GCTGCCGGTCCTTCGAAAATGC-Pi-5[prime]. The linker-ligated cDNA fragments in the size range of 0.2-1.2 kb were excised from an agarose gel and subjected to PCR amplification (94°C for 2 min; 94°C for 1 min; 58°C for 1 min; 72°C for 1 min, 25 cycles; 72°C for 10 min). Driver cDNAs were prepared by digesting 5 µg of amplified products with 250 U of HindIII, treating with isopsoralen IP-10 (HRI Associate, CA) and biotinylating with Photoprobe Biotin (Vector Laboratories) following the manufacturer's instructions. The tracer cDNA (0.5 µg) and driver cDNA (5 µg) were hybridized in a buffer containing 750 mM NaCl, 5 mM EDTA, 0.1% SDS and 25 mM HEPES pH 7.5. The hybridizations were performed in a volume of 10 µl at 68°C. The removal of biotinylated DNA was carried out by extraction with streptavidin beads (UltraLink, PIERCE). The streptavidin beads (100 µl of 50% slurry in 2 M LiCl, 10 mM HEPES-KOH, 1 mM EDTA, pH 7.5) were added into the hybridization mixture at the end of every hybridization and the tubes were incubated at room temperature with rocking for 2 h. The subtracted cDNA (in the liquid) was separated from the streptavidin beads using Millipore Ultrafree-MC 0.45 µm filter units (Millipore).

The hormone-starved MCF-7 cells (human breast cancer cells) provide a controlled system for examining the effects of estrogen, since addition of estrogen to these cells will induce estrogenic responses at both cellular and molecular levels. The subtraction procedure was examined using control genes pS2, GAPDH and [beta]-actin. Figure 2A shows the effect of estrogen on the expression of these control genes. The pS2 gene was activated by estrogen, as expected. To our surprise, the GAPDH gene, which is commonly regarded as a house keeping gene, was up-regulated ~2-fold by estrogen. The expression of [beta]-actin showed no detectable difference by estrogen addition and was used as a negative control. Figure 2B presents the results of the subtractive hybridization monitored by Southern blotting. As the number of rounds of subtractive hybridization were increased, the estrogen up-regulated gene pS2 was highly enriched and the non-regulated gene [beta]-actin was suppressed to an undetectable level. The GAPDH cDNA was suppressed to an undetectable level in the procedure to isolate down-regulated genes (Fig. 2B, the `-' lanes); but survived in the pool that is enriched for up-regulated genes (Fig. 2B, the `+4' cDNA). The success of the procedure was also shown by the cross-hybridization experiment between the +4 cDNA and -4 cDNA pools. Results revealed no detectable hybridization between +4 cDNA and -4 cDNA. This demonstrates that the cDNA fragments enriched in the +4 cDNA were greatly suppressed in the -4 cDNA pool and vice versa.A
Figure 2. The subtractive hybridization procedure monitored by control cDNA probes. (A) Northern blotting analysis of [beta]-actin, GAPDH and pS2 gene expression. `+' and `-' represent RNAs isolated from estrogen-responsive MCF-7 cells, which were treated with 10 nM of estrogen and without estrogen, respectively. Ten micrograms of total RNA were loaded in each lane. The membrane was probed with pS2, rehybridized with GAPDH and then actin. The blot was quantified using the PhosphorImager from Molecular Dynamics. (B) The subtractive hybridizations monitored by Southern blotting. The numbers on top indicate the number of subtractive hybridizations the cDNA has undergone. At the end of every subtractive hybridization, an aliquot of the subtracted cDNAs was amplified by PCR and 1 µg of each amplified cDNA pool was loaded on the gel. The left part of the gel (`- estrogen') represents the enrichment of down-regulated genes, the right part (`+ estrogen') represents the enrichment of up-regulated genes. The `-0' and `+0' cDNAs (starting cDNAs) were derived from the double-stranded cDNAs synthesized from mRNAs. The double-stranded cDNAs were digested by restriction enzymes, ligated to linkers, amplified by PCR and the PCR products were used as the starting `-0' and `+0' cDNAs.

To identify the genes that are regulated by estrogen, +4 and -4 cDNA libraries were constructed using the positive selection vector pZErO-1 (Invitrogen). The +4 cDNA library was screened against the +4 cDNA pool in colony hybridizations. Colonies that displayed strong hybridization signals should contain genes that are abundant in the +4 cDNA population, which are likely to be up-regulated by estrogen. Isolated cDNA fragments from these clones were analyzed by northern blotting to confirm their regulation by estrogen. The first clone identified contained a cDNA fragment of the highly up-regulated pS2 gene. This clone hybridized to ~80% of the colonies in the +4 cDNA library. To suppress the presence of pS2, the pS2 insert was amplified by PCR, sterilized, biotinylated and used as the driver DNA in further subtractive hybridizations. More up-regulated genes were isolated in the next round of screening. The next cDNA isolated represented a gene that is up-regulated 4-fold by estrogen (Fig. 3). Homology search reveals that the cDNA is 100% homologous to the human X-box binding protein-1 (hXBP-1), which is a basic region-leucine zipper transcription factor. It is involved in the regulation of human MHC class II gene transcription (12,13). The localization of hXBP-1 during embryo development also suggests a role for it in bone development (14). The -4 cDNA library was screened likewise against -4 cDNA pool. In addition to pS2 and hXBP-1, two other novel genes were identified and confirmed in northern blots from our initial screenings. One of those two genes is down-regulated by estrogen. Both have no homology to any GenBank sequences.


Figure 3. An example of northern hybridizations using a cDNA fragment isolated from subtracted libraries. The numbers to the right indicate the positions of the 0.24-9.5 kb RNA ladder (GIBCO BRL). The sequence of this cDNA fragment showed 100% homology to the human X-box binding protein-1 (hXBP-1).

In conclusion, the subtractive hybridization procedure was successful, as demonstrated by the genes discovered and the Southern blotting with control cDNA probes. It is also sensitive enough to enrich a gene (GAPDH) that is regulated only 2-fold by estrogen judged by northern blotting (Fig. 2A). The highly regulated or the moderately regulated but abundant genes will be isolated first, and by suppressing these genes, the moderately regulated or less abundant genes will be discovered. This procedure should, in theory, enable us to isolate all up- and down-regulated genes.

ACKNOWLEDGEMENTS

We thank Drs Brenda Williams, John Alberta and Charles Stiles for the advice about the manuscript. We thank Dr Myles Brown for helpful discussions. This work was supported by the Massachusetts Breast Cancer Research Grant (to W.L.) and by National Institutes of Health Grant CA30002 (to T.M.R.). W.S. was a recipient of a Fellowship from the Leukemia Society of America.

REFERENCES

1. Davis,R.L., Weintraub,H. and Lassar,A.B. (1987) Cell, 51, 987-1000. MEDLINE Abstract

2. Duguid,J.R., Rohwer,R.G. and Seed,B. (1988) Proc. Natl Acad. Sci. USA, 85, 5738-5742.

3. Duguid,J.R. and Dinauer,M.C. (1990) Nucleic Acids Res, 18, 2789-2792. MEDLINE Abstract

4. Wieland,I., Bolger,G., Asouline,G. and Wigler,M. (1990) Proc. Natl Acad. Sci. USA, 87, 2720-2724.

5. Wang,Z. and Brown,D.D. (1991) Proc. Natl Acad. Sci. USA, 88, 11505-11509.

6. Wang,Z. and Brown,D.D. (1993) J. Biol. Chem., 268, 16270-16278. MEDLINE Abstract

7. Brown,A.M., Jeltsch,J.M., Roberts,M. and Chambon,P. (1984)Proc Natl Acad. Sci USA, 81, 6344-6348.

8. Weisz,A. and Bresciani,F. (1993) Crit. Rev. Oncog., 4, 361-388.

9. Nardulli,A.M., Greene,G.L., O'Malley,B.W. and Katzenellenbogen,B.S. (1988) Endocrinology, 122, 935-944. MEDLINE Abstract

10. Cimino,G.D., Metchette,K.C., Tessman,J.W., Hearst,J.E. and Isaacs,S.T. (1991) Nucleic Acids Res, 19, 99-107. MEDLINE Abstract

11. Isaacs,S.T., Tessman,J.W., Metchette,K.C., Hearst,J.E. and Cimino,G.D. (1991) Nucleic Acids Res, 19, 109-116. MEDLINE Abstract

12. Liou,H.C., Boothby,M.R., Finn,P.W., Davidon,R., Nabavi,N., Zeleznik,L.N., Ting,J.P. and Glimcher,L.H. (1990) Science, 247, 1581-1584. MEDLINE Abstract

13. Ono,S.J., Liou,H.C., Davidon,R., Strominger,J.L. and Glimcher,L.H. (1991) Proc. Natl Acad. Sci. USA, 88, 4309-4312.

14. Clauss,I.M., Gravallese,E.M., Darling,J.M., Shapiro,F., Glimcher,M.J. and Glimcher,L.H. (1993) Dev. Dyn, 197, 146-156.

*To whom correspondence should be addressed at present address: Eli Lilly and Company, Drop Code 0444, Lilly Corporate Center, Indianapolis, IN 46285, USA. Tel: +1 317 277 7706; Fax: +1 317 277 2934; Email: su_eric_w@lilly.com

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 20 Jul 1998
Copyright©Oxford University Press, 1998.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Cancer Res.Home page
H. Qi, C. Fillion, Y. Labrie, J. Grenier, A. Fournier, L. Berger, M. El-Alfy, and C. Labrie
AIbZIP, a Novel bZIP Gene Located on Chromosome 1q21.3 That Is Highly Expressed in Prostate Tumors and of Which the Expression Is Up-Regulated by Androgens in LNCaP Human Prostate Cancer Cells
Cancer Res., February 1, 2002; 62(3): 721 - 733.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
D. W. Losordo and J. M. Isner
Estrogen and Angiogenesis : A Review
Arterioscler Thromb Vasc Biol, January 1, 2001; 21(1): 6 - 12.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (115K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Liu, W.
Right arrow Articles by Roberts, T. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Liu, W.
Right arrow Articles by Roberts, T. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?