PCR fusion-based approach to create reporter gene constructs for expression analysis in transgenic C. elegans.

In the nematode Caenorhabditis elegans, gene expression patterns are most often determined by generating a fusion of the promoter of the gene under study to a reporter gene such as lacZ or gfp (3,4,7,8,11). Of the almost 20 000 genes in the worm, the spatial expression pattern of only a few hundred genes has been determined (http:// www.wormbase.org). If the reporter gene is fused to the full coding sequence of the gene under study, one can also obtain useful hints about the subcellular localization of the protein. Apart from revealing potential clues about the gene under study, reporter gene fusions serve as invaluable markers to assess the fate of individual cells in defined mutant backgrounds (1,13). The computational comparison of the promoter sequences of large numbers of co-expressed genes will also lead to a better understanding of the underlying logic of transcriptional control (14). Hence, the availability of a large number of reporter genes that provide gene expression information with temporal and spatial resolution is of significant interest in the post-genome era. A method to rapidly create reporter gene fusions on a large scale would thus be a desirable tool to have at hand. All current methods for generating reporter gene fusions encompass timeconsuming DNA subcloning and DNA purification protocols (3,4,8,11). Here a protocol is described to create gfp fusion constructs that are ready for injection into the C. elegans gonad within one day, with no need for subcloning procedures. The protocol is a modified version of previously described PCRbased fusions of overlapping DNA fragments (9,12) and is schematically outlined in Figure 1. The protocol entails a reaction in which two primary PCR products (Figure 1, product nos. 1 and 2) are fused by PCR with a set of nested primers. In PCR no. 1, the promoter (or the complete gene) under study is amplified from worm genomic DNA that was prepared by digestion with 60 μg/mL proteinase K (1 h lysis at 65oC; 15 min at 95oC inactivation; reaction buffer: 10 mM Tris-HCl, pH 8.2, 50 mM KCl, 2.5 mM MgCl2, 0.45% Nonidet P-40, 0.45% Tween 20, 0.01% gelatin) or, alternatively and often more efficiently, from a preparation of cosmid DNA. In the parallel PCR no. 2, the gfp coding sequence plus the generically used 3′ untranslated region (UTR) from the unc-54 gene are amplified from the standard Fire vector pPD95.75 (http://www.ciwemb.edu/pages/firelab.html) (4). The 3′ primer for the promoter/gene (Figure 1, termed “B”) has a 24-nucleotide overhang to the gfp vector pPD95.75. If product no. 1 contains the coding region of the gene of interest, then it is important to ensure with primer B that the gfp is fused in frame to the gene of interest. Primer “C” does not need to have an overlap to the specific promoter/gene to be amplified, thus making C a generic primer that can be used for every fusion reaction. Gel purification of the two primary products was often found to inhibit the fusion PCR for unknown reasons. The two primary products are thus used with no further purification for the fusion PCR, which further speeds up the whole process. The concentration of the product # 1 and # 2 is roughly estimated by agarose gel electrophoresis, and an aliquot of the reaction is then diluted with water to 10–50 ng/μL each product. In case the yield of the PCR product is less than 10 ng/μL, it can also be used undiluted; we have even encountered cases in which the first PCR product was invisible on a gel and, nevertheless, obtained a fusion product. After this estimation and dilution step, 1 μL each diluted (or undiluted) reaction is used in the fusion PCR. For this reaction, it is obligatory to use nested primers (Figure 1, A* and D*). Although in most cases one will get a single band from the PCR fusion, another band can occasionally be seen, possibly a gfp dimer; sometimes this additional band may be much stronger than the desired PCR fusion product. This band can be ignored and considered as some sort of “carrier DNA” for the ensuing microinjection into the worms. The concentration of product no. 3 is estimated by agarose gel electrophoresis, and the reaction is diluted Benchmarks