The approach used by the TRiP is to generate transgenic animals with an RNAi hairpin under UAS-Gal4 control. The hairpin-containing transgenes are inserted via site-specific recombination into genomic loci known to be optimal for expression. Specific information follows.

The TRiP wanted to avoid the variability associated with some P-element-based approaches, such as random integration into transcriptionally inactive regions. To do this, we developed a set of vectors (VALIUM series) based on the phiC31 targeted integration method (Groth et al., 2004). To identify attP landing sites that permit optimal transgene regulation and expression, we generated a set of over 20 attP landing sites randomly distributed across the genome and used an integrated UAS::luciferase reporter to measure basal and Gal4-mediated expression at these sites (Markstein et al., 2008). To determine which, if any, of the remaining landing site loci would permit high levels of inducible expression, we measured the levels of luciferase activity induced by different ubiquitous or tissue-specific Gal4 drivers. Based on these analyses we selected attP40 and attP2 as sites for the integration of hairpin constructs on the second and third chromosomes, respectively.

The TRiP generated a series of vectors referred to as the "VALIUM series" (Vermilion-AttB-Loxp-Intron-UAS-MCS) (Ni et al., 2008; Ni et al., in preparation). Here we only describe VALIUM1 and VALIUM10 as all TRiP stocks are constructed in one of these vectors.

Both of these vectors contain vermilion as a selectable marker and an attB sequence to allow for phiC31 targeted integration at genomic attP landing sites. We chose vermilion rather than mini-white as the proper gene dosage of white has been found to be important in behavioral studies, a high-priority target application in the vector design. Further, based on our finding that UAS sites promote transcription in an additive fashion, we designed the VALIUM vectors with two pentamers of UAS, one of which can be removed using the Cre/loxP system. Other features of VALIUM vectors are the hsp70 TATA promoter, which has been shown to be an effective promoter in UAS vectors.

VALIUM1, our first vector, contains a multiple cloning site (MCS) that allows a single PCR product to be cloned in both orientations to generate the hairpin construct. Additionally, VALIUM1 contains two introns: the white intron, located between the inverted DNA repeats, which has been shown to reduce toxicity in bacteria; and the ftz intron, followed by the SV40 polyA tail to facilitate hairpin-RNA processing and export from the nucleus. This vector proved effective for transgenic RNAi (Ni et al., 2008) and a number of TRiP stocks have been generated in VALIUM1. The TRiP continued to optimize the vectors and is currently using VALIUM10 which has been found to be more effective for RNAi knockdown. VALIUM10 differs from VALIUM1 in a number of ways: 1. it contains insulator sequences that increase significantly the level of expression of the hairpins (Ni et al., in preparation); 2. instead of the MCS sites of VALIUM1, VALIUM10 contains a recombination system that facilitates the cloning of the hairpins, and 3. VALIUM10 contains two ftz introns.

Primers for the TRiP hairpins are designed using the DRSC's amplicon design tool "Snapdragon". The PCR product from the hairpin is designed to be 400 - 600bp long. Portions of gene transcript sequences are used that are common to all splice-forms of the gene of interest. Ideally regions that are free of 19bp matches to other genes are used. If there are no such regions of sufficient length, then regions with less than five 21bp matches to other genes are used. Primer3 (Rozen and Skaletsky, 2000) is used to cull the primers once an appropriate sequence is chosen. Any primer pairs whose reverse primer has the sequence CCAC at the 5' end are rejected. An extension of CACC is added to the 5' end of the forward primer.

The targeted method to generate hairpin lines has many practical advantages over P-element based methods: 1. the frequency of recovering transformants using the integrase method, either following co-injection of integrase mRNA or by injecting into the nanos-integrase strain, is almost five-fold higher than in conventional P-element transformation; 2. establishment of the lines is greatly facilitated as no mapping of the transformants to a specific chromosome is needed; and 3. unlike P-element-based methods, insertions into the attP landing site are homozygous viable.

Further, since the efficacy of the transgenic RNAi technology depends upon the level of expression of the UAS-driven constructs, the VALIUM constructs with their modular number of UAS copies allow the generation of a phenotypic series. From the original 10XUAS construct, a 5XUAS derivative can be recovered, and because the attP-containing chromosome can be homozygosed, it is also possible to generate both 15XUAS and 20XUAS combinations (Ni et al., 2008). The ability to generate a phenotypic series from 5XUAS to 20XUAS may prove useful, in particular, when variation of the gene expression dosage is important for phenotypic studies of pleiotropic genes. To generate different levels of RNAi knockdown, the features described above can be used together with other means to vary expression (different Gal4 lines of different strength, temperature) or processing (with coexpression of UAS-Dcr2, Dietzl et al., 2007) of the hairpin construct.

Ni JQ, Liu LP, Binari R, Hardy R, Shim HS, Cavallaro A, Booker M, Pfeiffer B, Markstein M, Wang H, Villalta C, Laverty T, Perkins L, Perrimon N. A Drosophila Resource of Transgenic RNAi Lines for Neurogenetics. Genetics. 2009; published ahead of print on June 1, 2009 as doi: 10.1534/genetics.109.103630

Dietzl, G. et al. (2007) A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature 448, 151-156.

Groth, A.C., Fish, M., Nusse, R. & Calos, M.P. (2004) Construction of transgenic Drosophila by using the site-specific integrase from phage phiC31. Genetics 166, 1775-1782.

Markstein, M., Pitsouli, C., Villalta, C., Celniker, S. and Perrimon, N. (2008) Exploiting position effects and the gypsy retrovirus insulator to engineer precisely expressed transgenes. Nature Genetics. 135, 1439-1449.

Ni, J-Q., Markstein, M., Binari, R., Pfeiffer, B., Liu, L-P., Villalta, C., Booker, M., Perkins, L. A., and Perrimon, N. (2008) Vector and Parameters for Targeted Transgenic RNAi in Drosophila melanogaster. Nature Methods 5, 49-51.

Rozen, S. & Skaletsky, H. (2000) Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 132, 365-86.