"These lines are available only to scientists at nonprofit institutions.

"These lines are available only to scientists at nonprofit institutions.

being developed. These methods allow the expression patterns of most genes to be determined at the cellular level. However, if the resolution and sensitivity of these methods are not appropriate for the task in hand, consider screening either gene-trap or enhancer-trap lines. The general principle behind this approach is to integrate a reporter gene that either lacks a promoter (gene trap) or carries only a minimal promoter (enhancer trap) at various sites in the wild-type genome. If the vector happens to land in a gene, or if the minimal promoter is brought next to an endogenous enhancer element, the reporter gene, e.g., P-glu-curonidase (GUS) or green fluorescent protein (GFP), will be expressed in a specific fashion. Because the constraints on precise insertion are fewer for enhancer traps, these lines produce a higher frequency of expression patterns compared to gene-trap lines. A disadvantage, however, is that it may be difficult to identify which gene's enhancers cause the specific expression pattern.

For the latest availability of public collections, check the Web Site of the Arabidopsis stock center. As discussed above, generating a new collection using a transposon system is very time consuming (for details, see Bancroft et al. 1992; Aarts et al. 1995; Sundaresan et al. 1995; Wisman et al. 1998). Simple T-DNA vectors for enhancer trapping have been developed by the Jack and Bowman labs (Campisi et al. 1999; Eshed et al. 1999); a more sophisticated two-component system has been developed by Haseloff ( Home.html) (Berger et al. 1998). In the Haseloff system, the actual enhancer-trap vector contains a minimal promoter driving an artificial iraws-activator, GAL4:VP16. The activity of GAL4:VP16 in these lines is indirectly revealed through a GFP reporter under the control of GAL4-binding sites. An advantage of this system is that the enhancer-trapped GAL4:VP16 can be used to drive any other gene of interest that is also under the control of GAL4-binding sites (see Chapter 7). Disadvantages are that it is more complex than simple enhancer traps and that GAL4-binding sites can become inactivated by methylation (Galweiler et al. 2000).

Activation Tags

Screening loss-of-function mutants can be of limited use because they normally fail to identify genes that act redundantly or at multiple stages dur ing development. Genes that are not absolutely required for a certain pathway can still be identified through dominant mutant alleles if such genes are sufficient to activate that pathway in an ectopic manner. Conversely, inactivation of a certain pathway in loss-of-function mutants might kill a plant, but inappropriate activation of the same pathway in gain-of-function mutants might still be compatible with survival of the plant.

Gain-of-function mutations can be caused by mutations in the protein-coding sequence, in mRNA regulatory elements, or in transcriptional regulatory elements. There is currently no good way in which mutations of the first two types can be preferentially induced, but several approaches to alter normal transcriptional regulation of genes have been described.

The first approach described makes use of multimeric copies of an enhancer element from the constitutively active promoter of the CaMV 35S gene, which are inserted adjacent to the right border of a T-DNA vector. Although it is currently unclear whether the original implementation of this system in plant protoplasts was indeed successful (Hayashi et al. 1992), the method has subsequently been used with tissue explants of Arabidopsis (Kakimoto 1996) and with whole-plant transformation by vacuum infiltration (Weigel et al. 2000). A similar approach makes use of a complete CaMV 35S promoter pointing outward from a trans-posable Ds element (Wilson et al. 1996). Although screens using either enhancers on a transposon or a promoter pointing outward from a T-DNA have not yet been described in the literature, it is likely that these combinations will also work.

Genomic insertion of the multimeric enhancers or of the 35S promoter can lead to transcriptional activation of nearby genes, thus resulting in dominant gain-of-function phenotypes. Either scheme (insertion of enhancers or of the promoter) can be viewed as a variant of activation tagging, although the term "activation tagging" most often refers to the use of enhancers. To be precise, the investigator should distinguish between "enhancer activation tagging" and "promoter activation tagging." An advantage of enhancer activation tagging is that enhancers can insert both upstream and downstream from a gene to cause activation. A possible advantage of promoter activation tagging is that it has the potential to induce loss-of-function phenotypes via antisense RNA if the promoter inserts downstream from a gene. However such a case has not yet been described in the literature. A possible disadvantage of enhancer activation tagging is that the spectrum of genes responsive to 35S enhancers may be limited (Weigel et al. 2000).

For enhancer activation tagging, two vectors have proven to be most popular, pSKI074, which confers kanamycin resistance and is most useful for screens on MS media, and pSKI015, which confers resistance to the herbicide glufosinate (BASTA) and is most useful for selection of transgenic plants on soil (Weigel et al. 2000).

Because primary T-DNA transformants obtained from vacuum infiltration are not chimeric, T1 plants can be screened for dominant pheno-types. Screening the T1 generation has the advantage that any phenotype observed is likely to be dominant, since T1 plants are normally hemizy-gous for a T-DNA insertion (Ye et al. 1999) (see Vacuum Infiltration of Arabidopsis in Chapter 5). The dominant behavior of a mutation should be confirmed in subsequent generations. As with any other insertional mutant, it must be confirmed that the dominant mutation is linked genetically to a T-DNA insertion. Adjacent sequences can then be recovered either by TAIL-PCR (thermal asymmetric interlaced-polymerase chain reaction) (see Chapter 6) or by plasmid rescue. In most, if not all, cases, the activated gene is the one that is immediately next to the enhancers, which are inside the right border, and any gene found next to the right border should be tested for overexpression in the dominant mutant. A caveat to this experiment is that the 35S enhancers may not necessarily cause constitutive expression of the activated gene, but rather amplify the endogenous expression pattern. Thus, even if expression of this gene cannot be detected in either the wild type or the dominant mutant, it is still possible that it is the cause of the mutant phenotype.

When an overexpressed gene has been identified, it too must be confirmed as the cause of the dominant phenotype. The preferred method is to transform a wild-type plant with a piece of genomic DNA from the mutant that contains both the 35S enhancers and the activated gene. If the transformants display a phenotype similar to that of the original mutant, this confirms that the overexpressed gene is the cause of the dominant mutation. The activation-tagging vectors are constructed in such a way that it is often possible to recover such a piece of genomic DNA by plasmid rescue. Alternatively, a piece of wild-type genomic DNA that spans the gene of interest can be inserted into vectors that contain multimeric 35S enhancers, such as the pMN19 and pMN20 vectors (Weigel et al. 2000). Finally, if a cDNA of the overexpressed gene is available, it can be inserted into a T-DNA vector with a 35S promoter, and plants can be transformed with the 35S::cDNA construct.

As with any gain-of-function mutation, it is, of course, important to use other means to confirm that the activated gene has a bona fide role in the process of interest and that it is not merely mimicking the activity of a related factor. Examples include knocking out the overexpressed gene or studying the genetic interaction of the overexpressed gene with other genes in the same pathway.


Now that the Arabidopsis genome sequence is complete, many investigators are interested in studying the functions of genes whose sequences or expression patterns suggest roles in particular biological processes. Other reasons for selecting mutations in a specific gene may be because only partial loss-of-function mutations of a gene have been previously identified, or because only gain-of-function mutations have been isolated. Note also that finding a single-gene knockout may not always be sufficient to identify the true loss-of-function situation because of redundancy. This is a particular concern in the case of closely linked genes, for which a larger deficiency must be identified or completely different approaches such as those discussed in Chapter 8 must be used.

A powerful way to approach this problem is to identify a mutation in the gene of interest (Your Favorite Gene, YFG) and to test mutant plants for phenotypes that are predicted to result from loss of function of that gene. Currently, the way to do this is by PCR screening of pools of insertion lines, using one primer corresponding to YFG and one primer corresponding to the end of the insertion element. The synthesis of a product indicates the presence of an insertion in YFG. The pool is then repeatedly subdivided until a single plant carrying the desired insertion is identified. It is anticipated that this will become even easier in the future, as the genomic sequence of many insertion sites will be published on the Web (for an early effort, see Parinov et al. 1999).

Large collections of T-DNA insertion lines have been generously provided to the Arabidopsis stock centers at Ohio State University and Nottingham by the individuals who constructed them. The stock centers distribute pools of DNA representing these lines. When an investigator has identified a pool containing an insertion in YFG, subpools can be ordered from the stock center. When the subpool containing the desired insertion has been identified, seed of the population used to make the subpools of DNA can be obtained and used to identify a plant containing the insertion in YFG. At the time of writing, 175,000 T-DNA insertion lines of various types are available from the A8RC (see Table 2.1). DNA pools are only available for the 22,000 lines in the Feldmann and Jack collections, but pools representing other lines are expected to become available in the future. A National Science Foundation-supported service will, for a fee, screen for mutations in a particular gene (

Several ways exist to make the process of selecting insertions by PCR more efficient. DNA prepared from individual plants can be pooled in a multidimensional manner, such that a relatively small number of PCRs are sufficient to identify a single plant carrying the desired insertion. Furthermore, several groups are working to isolate and sequence DNA fragments flanking insertion sites and are depositing the sequences in databases. When these databases reach sufficient size, it will be possible to find an insertion in YFG simply by consulting the databases and ordering seed from the stock center.

A method for reverse genetic screens of EMS-mutagenized populations called "TILLING" (Targeting Induced Local Lesions IN Genomes) has been reported recently (McCallum et al. 2000). DNA was prepared from small pools of M2 plants, and 500-bp fragments of a gene of interest were amplified by PCR. Denaturing high-performance liquid chromatography (dHPLC) was then used to identify single-base mismatches corresponding to mutations in one of the plants in the pool. This method is potentially very powerful, because it allows isolation of types of mutations that cannot be obtained using insertion mutagenesis. For example, it is possible to obtain single-base mutations that cause partial loss of function in essential genes without killing the plant, as well as mutations that cause altered functions such as unregulated activity or altered substrate specificity. The National Science Foundation has funded efforts to make TILLING lines publicly available.

Screen rig DNA Pools for T-DNA Insertions in YFG

This protocol is suitable for screening a collection of 60,480 T-DNA insertions in ecotype Ws as described previously (Krysan et al. 1999). It can easily be adapted for screening other collections, taking care to ensure that the primers for the ends of the insertion element are compatible with the element used.


1. Obtain the following primers for the left and right ends of the T-DNA:


Better results are obtained if the primers are PAGE-purified. If a known insertion is present in the pools that are being screened, it is a good idea to try to detect that insertion before searching for an insertion in YFG. Such a positive control ensures that the PCR conditions are correct.

2. Design primers for YFG.

One primer reading in from the 5' end of the coding sequence and one primer reading in from the 3' end are needed. (When looking for insertions in the promoter region, place the 5' primer just upstream of the 5' end of the promoter.) Use the following guidelines: Select primers 29 bp long with a G+C fraction between 34% and 50%. Between positions 19 and 29, there should be fewer than ten G+C bases and no more than one of the bases at positions 28 and 29 should be G or C, These primers should also be PAGE-purified.

3. Test the gene-specific primers by performing three PCRs, using wild-type genomic DNA as the template. Carry out one reaction with the 5' YFG primer and the 3' YFG primer, one with the 5' YFG primer alone, and one with the 3' YFG primer alone.

The reaction with both primers should yield a product of the correct size, whereas the other two reactions should yield no product. If the 5' and 3' primers do not yield a product, they may be spaced too far apart. In this case, design some additional primers that are closer together and use them for testing the 5' and 3' primers. If this is not successful, it suggests that the original primers did not work well.

4. There are 30 superpools of DNA, with each superpool representing 2025 T-DNA insertion lines. For each superpool in the collection, perform four PCRs, one with the 5' YFG primer and the left border primer, one with the S" YFG primer and the right border primer, one with the 3' YFG primer and the left border primer, and one with the 3' YFG primer and the left border primer. PCR mix:

lOx PCR buffer

Taq* polymerase (5 units/jil)

Template DNA

Sterile water

5 |x1 1.5 JLU 5 |jJ 1^1 2.5 Lil 12 ng to a final volume of 50 jil

*Use a hot start or an enzyme mix that includes an and-Taq antibody for an automatic hot start, such as Platinum Taq (Life Technologies) or AdvanTAge Taq (CLONTECH).

CAUTION: MgCl2 (see Appendix 3)

PCR program:

Initial denaturation:


5 min

Amplification cycle:


30 sec

Was this article helpful?

0 0
Caring For Your Lawn

Caring For Your Lawn

The Secret of A Great Lawn Without Needing a Professional You Can Do It And I Can Show You How. A Great Looking Lawn Doesn't Have To Cost Hundreds Of Dollars Or Require The Use Of A Professional Lawn Care Service. All You Need Is This Incredible Book.

Get My Free Ebook

Post a comment