A novel method for tissue-specific RNAi rescue Drosophila · 2017-05-29 · To reach unique restriction sites for further cloning, the two flanking primers were extended over the

Published online 29 May 2009 Nucleic Acids Research, 2009, Vol. 37, No. 13 e93doi:10.1093/nar/gkp450

A novel method for tissue-specific RNAi rescuein DrosophilaJoachim G. Schulz1,2,*, Guido David1 and Bassem A. Hassan2

1Laboratory of Glycobiology and Developmental Genetics and 2Laboratory of Neurogenetics, FlandersInstitute for Biotechnology (VIB) and Department of Human Genetics, Herestraat 49 bus 602, KatholiekeUniversiteit Leuven, 3000 Leuven, Belgium

Received April 8, 2009; Revised May 11, 2009; Accepted May 12, 2009

ABSTRACT

Targeted gene silencing by RNA interference allowsthe study of gene function in plants and animals.In cell culture and small animal models, geneticscreens can be performed—even tissue-specificallyin Drosophila—with genome-wide RNAi libraries.However, a major problem with the use of RNAiapproaches is the unavoidable false-positiveerror caused by off-target effects. Until now, this isminimized by computational RNAi design, compar-ing RNAi to the mutant phenotype if known, andrescue with a presumed ortholog. The ultimateproof of specificity would be to restore expressionof the same gene product in vivo. Here, we present asimple and efficient method to rescue the RNAi-mediated knockdown of two independent genesin Drosophila. By exploiting the degenerate geneticcode, we generated Drosophila RNAi EscapeStrategy Construct (RESC) rescue proteins contain-ing frequent silent mismatches in the completeRNAi target sequence. RESC products were nolonger efficiently silenced by RNAi in cell cultureand in vivo. As a proof of principle, we rescue theRNAi-induced loss of function phenotype of the eyecolor gene white and tracheal defects caused by theknockdown of the heparan sulfate proteoglycansyndecan. Our data suggest that RESC is widelyapplicable to rescue and validate ubiquitousor tissue-specific RNAi and to perform proteinstructure–function analysis.

INTRODUCTION

In model organisms, such as yeast, worm, fly and mouse,significant progress has been made to create tools for themanipulation of individual genes, such as random muta-genesis and gene targeting altering the genomic sequence,or RNA interference (RNAi) targeting of mRNA for

degradation. Lethality and pleiotropy compromise loss-of-function mutant studies for many genes, renderingidentification of late or tissue-specific phenotypes difficult.Gene knockdown by RNAi, when combined with thetransgenic Gal4/UAS system in Drosophila (1), is cell-autonomous and inducible in space and time. It is alsonontransitive (2,3), allowing isoform-specific knockdown.Drosophila genome-wide RNAi libraries have been gener-ated recently, making large-scale tissue-specific reversegenetic screens possible [(4) and http://www.shigen.nig.ac.jp/fly/nigfly].A major problem of RNAi, however, is off-target effects

(5–7). Even for a fully annotated genome, tolerance formismatches makes the design of unique RNAi targetsequences very difficult. This is particularly compoundedin Drosophila, where long double-stranded RNA of sev-eral hundred base pairs is usually required for efficientknockdown (8), thus increasing the risk of hitting unre-lated genes because of tandem trinucleotide repeats orsequences homologous to the RNAi target (5,6,9,10).Apart from that, activation of the innate immune system(11) and competition with the endogenous RNAi machin-ery (12) have been reported in mammals.Clearly, improved RNAi reliability by experimental val-

idation is highly desirable, ideally through rescue ofthe RNAi phenotype by the targeted gene itself. We setout to test if alteration of the cDNA at the nucleotide levelwhile maintaining the encoding of the native proteinwould circumvent RNAi-mediated silencing and thuspermit rescue. We then used this approach to unravel anovel function for the Drosophila proteoglycan syndecan(sdc) in tracheal development.

MATERIALS AND METHODS

RNAi design and synthesis

The Sdc RNAi target region was selected to be larger than300 bp, to target all existing splice forms, and not to con-tain sequences longer than 17 bp that are identical in otherloci. UAS::SdcdsRNA was designed as a genomic-cDNAhybrid construct targeting 357 bp of sdc exons 6 and 7.

*To whom correspondence should be addressed. Tel: +32 16 330 525; Fax: +32 16 330 522; Email: [email protected]

� 2009 The Author(s)This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

The forward genomic part consisting of sdc exon 6, intron6 and exon 7 was cloned from gDNA into pUAST,followed by the 3-kb sdc intron 5 to improve genomicstability. Then, exons 6 and 7 cDNA were cloned inreverse orientation and transformed into Sure2 repair-deficient bacteria (Stratagene) to avoid hairpin repair(Figure 4) (13). The UAS::wdsRNA line was obtainedfrom D. Smith (13).

RESC design

The boundaries of the RESC sequence were defined by theRNAi target boundaries. In total, 235 silent mutationswere introduced into the 687-bp w RNAi target regionand 72, 33 and 17 mutations to obtain 6-mer, 12-mer or24-mer changes into the 357-bp Sdc RNAi target region(Supplementary Figure 1). GC content, restriction sites,splice sites, repetitive nucleotide sequences and unavoid-able gaps (Methionine, Tryptophan) were taken intoaccount, leaving stretches of maximally five nucleotidesunchanged. Preferably, amino acids with different codonusage were swapped. Finally, the RESC sequence was ver-ified at amino acid level to be wild type (align p).

wRESC synthesis

wRESC was synthesized with 16 desalted oligonucleotideprimers (Invitrogen) in alternating directions, each 70nucleotides in length (+/–1 or 2 bp, G or C at 30 end)and with 20-bp overlap each. To reach unique restrictionsites for further cloning, the two flanking primers wereextended over the target region to contain wild-typesequence at their 50 end. Primer pairs (first round) or pur-ified PCR fragments together with the two outside primers(from round 2 on) were subjected to standard PCR elon-gation (Figure 1b). Finally, a 795-bp BspEI–SnaBI frag-ment was obtained from four rounds standard PCR andcloned to exchange the wild type w BspEI–SnaBI fragmentin pUAST.

sdcRESC synthesis

For sdcRESC every 6th, 12th or 24th nucleotide within theexon 6/7 RNAi target region was mutated (SupplementaryFigure 1). sdcRESC was synthesized with eight primers inalternating directions, 72–76 bp in length (+/–1 or 2 bp, Gor C at the 30 end) and with 22–25-bp overlap. To reachthe unique KspI restriction site in exon 5 the RESCsequence was extended by wild-type sequence. TheRESC fragment was obtained from three rounds standardPCR (Figure 1b) and cloned to exchange the sdc wild-typesequence in pUAST-sdc. All inserts were verified bysequencing.

Cell culture

Two Mio Drosophila S2 cells per 1ml Drosophila serum-free medium (Invitrogen) were transiently transfected with5 ml Cellfectin (Invitrogen) and 3 mg total DNA (pUAST-sdc with or without RESC modification, pUAST-SdcdsRNA or empty vector, and pMT-Gal4, 1:1:1 ratio),induced with 700 mM CuSO4 5 h after transfection, andharvested 48 h later.

Western blot

S2 cells or whole anesthetized flies were homogenized inlysis buffer (1% Triton X-100, 140mMNaCl, 10mM Tris)with complete protease inhibitor cocktail (Roche), boiledin reducing SDS sample buffer and loaded on 4–15%Biorad Ready Tris–HCl gels, and transferred to positivelycharged Zeta-probe blotting membrane (Biorad). Themembrane was blocked with 0.5% Casein (Merck) in0.6M NaCl, incubated with a-Sdc (14) or a-tubulin anti-body, and detected with alkaline phosphatase-conjugatedsecondary antibody (Promega) and CSPG (Tropix) onAmersham Hyperfilm.

Fly lines

Fly stocks were kept on standard fly food. Transgenic flies(sdcRESC, wRESC) were generated by standard P-element-mediated germline transformation. Transgenic lines weregenerated from sdcRESC with 6-mer changes. UAS::wdsRNA

flies were obtained from D. Smith, btl-Gal4 from the JapanNational Institute of Genetics, all other lines from theBloomington Stock Center. sdc cDNA was obtainedfrom J. Lincecum, sdc23 stock from G. Vorbruggen.

Phenotype analysis

The tracheal morphology of third instar larvae was ana-lyzed in a filet preparation by stereomicroscopy. Larvaewere cold anesthetized in 48C PBS and dissected with irisspring scissors (Fine Science Tools) from the ventral side.Although the lumen of the tracheae is easily visible withbright field microscopy, this does not reliably reflectmorphology because air filling is not always complete.Hence, reporter lines were generated that express mem-brane-bound CD8-GFP fusion protein in the tracheawith btl::Gal4. Eye colors of 5–7-day-old adult flies wereanalyzed under white light with a stereomicroscope(Leica).

Figure 1. Generation of RESC DNA from oligonucleotides. (a)Illustration of primer positions and PCR reactions required to generatea RESC DNA fragment in four rounds of PCR starting from 16 oligonu-cleotides. In each round, the two corresponding PCR products and out-side primers from the previous round are used for amplification. Arrowsindicate direction of elongation. (b) Agarose gel showing wRESC PCRproducts and their size obtained in four PCR rounds as illustrated in (a).

e93 Nucleic Acids Research, 2009, Vol. 37, No. 13 PAGE 2 OF 8

Primers

sdc RESC, 6-mer changes.For 1Atcgatggtccgcggatcggtggcaacgatggagatattacagagcgcggaccgggtgctggtggtagcaatgtg

For 2Tcaatagtcagccgtccgatacaaaaggcattgatcataggccgaacggaaacgaggtggtgataatgagtgag

For 3Tcccaaccgggaatactggcagctgtgattggaggtgcggtcgtcggcctcctgtgcgcgattctcgtcgtgatg

For 4Gacgaaggatcgtatgccctggatgagcctaagcgttcgccagccaataattcctatgccaagaacgcgaacaac

Rev 1TTTGTATCGGACGGCTGACTATTGACGTTGGTGTTCGGGTCTAGTTCATGCACATTGCTACCACCAGCACCCGG

Rev 2CAGCTGCCAGTATTCCCGGTTGGGAGAAGAAACTGGACGTCCGATCATCCTCACTCATTATCACCACCTCGTTTC

Rev 3CATCCAGGGCATACGATCCTTCGTCTTTCTTTCGCATTCGGTAGACGATAAACATCACGACGAGAATCGCGCACAG

Rev 4GAGGTACCCTCGAGCCGGAGCTCGCATATTCTCACGCATAGAATTCTCTGTTGTTCGCGTTCTTGGCATAGG

sdc RESC, 12-mer changes.For 1Atcgatggtccgcggatcggtggcaacgatggagatattacagagcgcggaccgggtgctggtggcagcaatgtg

For 2Tgaatagtcagccctccgatacaaagggcattgatcacaggccgaacggcaacgaggtggtcataatgagcgag

For 3Tcccagcccggaatactggctgctgtgattggcggtgcggtcgttggcctcctctgcgccattctcgtggtcatg

For 4Gacgagggatcgtatgcgctggatgagccaaagcgttcgccggccaataattcgtatgccaaaaatgcgaacaac

Rev 1TTTGTATCGGAGGGCTGACTATTCACGTTGGTGTTGGGGTCTAATTCGTGCACATTGCTGCCACCAGCACCCGG

Rev 2CAGCAGCCAGTATTCCGGGCTGGGAGAAGAAGCTGGACGTGCGATCATCCTCGCTCATTATGACCACCTCGTTGC

Rev 3CATCCAGCGCATACGATCCCTCGTCTTTCTTCCTCATTCGGTACACGATAAACATGACCACGAGAATGGCGCAGAG

Rev 4AGAGGTACCCTCGAGCGCGAGTCGCATTATCTCACGCGTAGAATTCGCGGTTGTTCGCATTTTTGGCATACG

sdc RESC, 24-mer changes.For 1Atcgatggtccgcggatcggtggcaacgatggagatattacagagcgcggaccgggtgctggtggcagcaacgtg

For 2Tgaatagtcagccctccgacacaaagggcattgatcacaggcccaacggcaacgaggtggtcatcatgagcgag

For 3Tcccagcccggaattctggctgctgtgattggcggtgccgtcgttggcctcctctgcgccatactcgtggtcatg

For 4Gacgagggatcctatgcgctggatgagccaaagagatcgccggccaataattcgtatgcgaaaaatgcgaacaac

Rev 1TTTGTGTCGGAGGGCTGACTATTCACGTTCGTGTTGGGGTCTAATTCGTGCACGTTGCTGCCACCAGCACCCGGRev 2CAGCAGCCAGAATTCCGGGCTGGGAGAAGAAGCTCGACGTGCGATCATCCTCGCTCATGATGACCACCTCGTTGC

Rev 3CATCCAGCGCATAGGATCCCTCGTCTTTCTTCCTCATGCGGTACACGATAAACATGACCACGAGTATGGCGCAGAG

Rev 4AGAGGTACCCTCGAGCGCGAGTGCCATTATCTCACGCGTAGAACTCGCGGTTGTTCGCATTTTTCGCATACG

w RESC primers.For 1Ggctccggatggcggcagctggtcaaccggacacgcggactattctgcaacgagcgacacatcccagctc

For 2Gagtcgcgtaccctggagagttgctggcagtcatgggtagctcaggcgcgggcaagaccacgctcctgaac

For 3Ccaggtatccccgtcgggcatgaggctcctgaacgggcagccagtcgatgcaaaagaaatgcaagcgcgg

For 4Attgggtcgctcaccgcgcgagagcatctcatctttcaagcgatggtccgaatgccgcggcacctcacttac

For 5Tcatacaagaactcagcttgtcgaagtgccaacataccattataggcgtcccgggacgggtcaagggactaag

PAGE 3 OF 8 Nucleic Acids Research, 2009, Vol. 37, No. 13 e93

For 6Tgcatcggaagcgctgacggaccctccactgcttatatgtgacgaacctacgagtggcctagatagtttc

For 7Aaactcagccaaaagggaaagacggtgattctcactatccaccaacctagctctgaactattcgaactgttc

For 8Gagtcgcctttctaggaaccccttcggaggcagtagatttcttctcatacgtgggtgcccagtgtcctac

Rev 1CTCTCCAGGGTACGCGACTCCACAAACATTCTTCAGTAAGTGCTTTCTAGGAGCTGGGATGTGTCGCTCG

Rev 2TGCCCGACGGGGATACCTGGATCCCTTGTGGGGACCGGAAAGCCAGCGCGTTCAGGAGCGTGGTCTTGCC

Rev 3CGCGCGGTGAGCGACCCAATGAATAGATCGTCTTGTTGCACGTACGCACACCGCGCTTGCATTTCTTTTG

Rev 4CAAGCTGAGTTCTTGTATGACTTGGTCGACTCGTGCGACGCGTTGGCGGTAAGTGAGGTGCCGCGGCATTC

Rev 5CCGTCAGCGCTTCCGATGCAAAGGCTAGCCTTTTCCGCTCGCCTCCGCTTAGTCCCTTGACCCGTCCCGG

Rev 6TTTCCCTTTTGGCTGAGTTTCTTAAGGACTTGCACCACGGAATGAGCAGTGAAACTATCTAGGCCACTCG

Rev 7GGTTCCTAGAAAGGCGACTCGCCCTTCAGCCATAAGCAGAATCTTATCGAACAGTTCGAATAGTTCAGAG

Rev 8CCTACGTAAAAGTCCGCCGGATTGTAGTTGGTAGGACACTGGGCACCCAC.

RESULTS

Rescue of RNAi-mediated w loss-of-function

RNAi specificity is ideally demonstrated by re-expressionof the knocked-down gene. However, a transgene contain-ing the original RNAi target sequence is subject to RNAi aswell, and deletion of the target sequence while maintaininggene function is possible only if noncoding sequence istargeted. A possible way to circumvent this problem maybe to introduce silent mismatches into the rescue constructsuch that the mRNA sequence is altered while maintainingthe protein coding sequence. The degenerate nature of thegenetic code permits sequence alterations for 18 out of the

20 amino acids and the stop codon (exceptions areMethionine and Tryptophan), allowing approximatelyone change every third nucleotide of a cDNA.We reasonedthat even when considering the tolerance of the RNAi forsome mismatches, this degree of alteration would be suffi-cient to escape the RNAi machinery.

To test this idea in vivo and in vitro, we chose to rescuetwo genes in Drosophila, one with a known and one with anovel phenotype. The white (w) gene is required for redeye color in Drosophila and was previously shown to beeffectively knocked down by RNAi in vivo (13,15)(Figure 2a). We decided to rescue w with a rescue con-struct containing 235 silent mutations in the 687-bp targetsequence and named it wRESC (RNAi Escape StrategyConstruct). Since RNAi in Drosophila is not transitive,only the w RNAi target sequence but not the flankingsequences were mismatched. To introduce this highnumber of mutations, site-directed mutagenesis by whichonly a few new mutations can be generated per round ofPCR is not an efficient approach. In addition, a naturalRESC template that could be used for amplification doesnot exist. Hence, we decided to create the new sequencecompletely from custom-synthesized oligonucleotides.We divided the target fragment including overlappingwild-type sequence on both ends for further cloning into16 oligonucleotides, 70–75-nt long and with an overlap of20–25 nt. By successive elongation of the oligonucleotidesand purification of intermediate fragments the final wRESC

fragment was obtained after four rounds of PCR(Figure 1a and b). The wRESC fragment was cloned toreplace the corresponding part of the wild-type sequenceof w in a standard vector used for P-element-mediatedgermline transformation, where w routinely serves as amarker gene expressed under its own promoter. The orig-inal vector in which the w sequence was unaltered servedas a control. Several lines of both w and wRESC transgenicflies were obtained which restore eye color to w nullmutant flies, showing that the wRESC transgene is func-tional (Figure 2b). However, whereas w RNAi efficientlytargeted combinations of up to three w transgenes, result-ing in complete loss of eye color (Figure 2a and c), wRESC

was not targeted, as judged by the fact that wRESC and wRNAi + wRESC flies show identical eye color (Figure 2b).

Sdc RESC in vitro

Available mutants of the single fly homolog of thevertebrate transmembrane heparan sulfate proteoglycansyndecan (sdc) are all derived from imprecise excisionscausing deletions that potentially affect a nearby gene onthe complementary strand called Smad anchor for receptoractivation (sara). sara shares the enhancer region with sdc,making sdc an ideal candidate to confirm the validity ofthe RESC strategy.

While a maximum number of nucleotides were mis-matched in wRESC, we now first tested to what extent thenucleotides in sdcRESC had to be exchanged in order toescape RNAi. We generated three UAS::sdcRESC vectorsfor rescue experiments, with mismatches every 6th, 12thor 24th bp in exons 6 and 7, giving rise to 72, 33 and17 mutations in the 357-bp target region, respectively

e93 Nucleic Acids Research, 2009, Vol. 37, No. 13 PAGE 4 OF 8

(Supplementary Figure 1). In Drosophila S2 cells in cul-ture, all sdc constructs were expressed at similar levels inthe absence of RNAi, but under sdc RNAi conditions theexpression levels decreased with increasing similaritybetween hairpin sequence and target region in the rescueconstruct. Expression of sdcRESC with 24-mer-mismatchspacing was not different from wild-type sdc, suggestingthat a mutation every 24th base pair is insufficient toescape silencing by RNAi. The 12-mer changes resultedin some level of escape; however, the strongest effect wasobserved with the 6-mer construct (Figure 3a).

Sdc RESC in vivo

To confirm the RESC effect in vivo, transgenic flies weregenerated that inducibly express sdcRESC with 6-mer mis-matches. In the absence of RNAi, both sdc and sdcRESC

were efficiently translated, while under RNAi conditionsmost wild-type sdc but very little, if any, sdcRESC wasdegraded (Figure 3b).

Both, sdc mutants and RNAi animals, show an almostidentical nonfusion phenotype in the dorsal branchesof the tracheal system (Schulz et al., submitted for pub-lication). When sdcRESC was expressed in the tracheae ofsdc RNAi animals, the dorsal branch phenotype was com-pletely reversed (Figure 3c), confirming that Sdc is neces-sary for tracheal development and that the RESCtechnique permits tissue-specific rescue of RNAi-mediatedphenotypes.

DISCUSSION

The discovery of RNAi and its exploitation for the versa-tile, rapid and efficient study of gene function hasthe potential to revolutionize genetics and genetic screensin small model organisms such as Drosophila andCaenorhabditis elegans. However, a major drawback ofRNAi-mediated gene targeting is off-target effects, espe-cially in Drosophila where long dsRNA give rise to manyshort dsRNA species. Thus far, off-target effects wereaddressed by in silico analysis of the RNAi sequence, useof more than one nonoverlapping RNAi per targetedgene, rescue by a putative ortholog or paralog, or com-parison to known mutant phenotypes (16,17). The latter isof particular importance for conclusions about biologicalfunction but limited by the fact that null alleles are onlyavailable for a minority of genes, and that even amongthose not all show easily discernable phenotypes. Themost powerful proof of RNAi specificity would berescue by the endogenous target gene itself. Here, we pres-ent a simple method that exploits the degenerate geneticcode to introduce frequent silent mismatches in the entiretarget region in an otherwise wild-type rescue construct.On average, exchange of every third base in a codingsequence is possible. Our data indicate that exchangingevery sixth nucleotide—equivalent to three to four mis-matches in a �21-bp dsRNA—only in the segment tar-geted by the RNAi is sufficient to rescue RNAi mediatedknockdown. In contrast to the in vivo experiments, the

Figure 2. Photographs (100�) from eyes of age-matched adult flies demonstrating eye color rescue with wRESC. w RNAi causes total loss of eyecolor, which is restored with wRESC but not w3. Genotypes are (a) w1118 (upper left); w1118;tub::Gal4,w+(1)/+ (upper right);w1118;;UAS::wdsRNA,w+(2)/+ (lower left); w1118;;tub::Gal4,w+(1)/UAS::wdsRNA,w+(2) (lower right). (b) w1118,w+(RESC)/w1118 (left); w1118,w+(RESC)/w1118;; tub::Gal4,w+(1)/UAS::wdsRNA,w+(2) (right) (c) w1118,w+(3)/w1118 (left); w1118,w+(3)/w1118;; tub::Gal4,w+(1)/UAS::wdsRNA,w+(2) (right).

PAGE 5 OF 8 Nucleic Acids Research, 2009, Vol. 37, No. 13 e93

rescue appears to be incomplete in vitro. Yet, as opposedto the stable expression under in vivo conditions, theexpression of the transgenes in cell culture was achievedby triple transient transfections, and it is unlikely that allcells with knockdown also express the rescue construct.Constructs with mismatch of every 12th nucleotide orless, yielding �21-bp fragments with only two peripheralor a single central mismatch, were still largely degraded.This confirms that RNAi target recognition is extremelysequence specific and several mismatches per shortdsRNA still allows for some translation (15,18,19).When RNAi is performed with long hairpins of up to

1000 bp such as in Drosophila (8), site-directed mutagen-esis would be too time consuming to introduce the high

number of mutations required to modify the completetarget sequence of the rescue construct. Instead, we sug-gest introducing mutations by using long, overlappingoligonucleotides as templates for amplification and syn-thesis of the entire RESC fragment. Assuming oligonu-cleotides that are 80-nt long and overlap by 20 bp,synthesis of a target region of 500 bp or less requiresthree PCR rounds or seven reactions with 8 primers, fortarget regions between 500 and 1000 bp four PCR roundswith 16 primers suffice. Due to the transitive nature ofDrosophila RNAi, it is sufficient to mismatch only thedirect RNAi target sequence, thus the size of the RESCfragment is independent from the total gene product size.

To simplify future RESC design, we created a simpleand user-friendly web-based tool to generate RESC andthe necessary oligonucleotides sequences for any gene at:https://med.kuleuven.be/cme-mg/lng/RESC/.

To demonstrate the feasibility and functionalityof RESC, we rescued phenotypes associated with twoindependent genes in vitro and in vivo. The eye color phe-notype of the w gene was restored by wRESC containingmismatches in 212 of 229 codons of the target region, butnot in the nontargeted sequence. The targeting of sdcrepresents a common case where despite the availabilityof putative null alleles, their specificity cannot be comple-tely established. In this case the combination of RNAi andRESC reveals novel tissue-autonomous and gene-specificfunctions. Furthermore, once a RESC construct is gener-ated it can be exploited further in rescue-based structure–function analysis.

The combination of the transgenic RNAi and RESCrescue with the Gal4/UAS system and its variationsoffers several advantages. First, wild-type protein can beknocked down and exchanged for by wild-type or mutantprotein using the same promoter at any time during devel-opment, and in any chosen tissue. Second, tissue-specificexpression avoids ‘anatomical off-target effects’, in con-trast to systemic RNA injection. Third, without Gal4both UAS::dsRNA and UAS::RESC can be kept in acombined stock in off-modus, avoiding modifier accumu-lation over several generations of stock keeping. Fourth,the effects are dominant, making screens more feasible.

Figure 3. sdcRESC rescue in vitro and in vivo. (a) aSdc western blot fromDrosophila S2 cells transiently transfected with pMT-Gal4, with (+) orwithout (–) UAS::SdcdsRNA, and UAS::sdc with wild-type (wt), orRESC sequence (�24, 24-mer changes; �12, 12-mer changes; �6, 6-mer changes). (b) aSdc western blot from adult flies ubiquitouslyexpressing Sdc without (wt) or with (�6) RESC sequence, in the pres-ence (+) or absence (–) of ubiquitous Sdc RNAi. (c) Quantification ofunfused segments of tracheal dorsal branches per animal, in wild-type(sdc+/+), sdc23 homozygous (sdc�/�) mutants, tracheal Sdc RNAi alone,or with tracheal rescue sdcRESC. n� 10; error bars=CI; �P< 0.01.

Figure 4. Illustration of the RESC rescue principle. Small interferingnucleotide duplexes (siRNA) derived from cleavage of long, double-stranded RNAi by the RNase Dcr-2 guides the RNA-induced silencingcomplex guides (RISC) complex to complementary mRNA that is sub-sequently cleaved and degraded. RESC constructs are mismatched inthe RNAi target region and therefore escape RISC-mediateddegradation.

e93 Nucleic Acids Research, 2009, Vol. 37, No. 13 PAGE 6 OF 8

Our technique has implications beyond Drosophila.In mammals, RNAi is already successfully used toknock down dominant mutant human disease genes suchas Huntingtin (20,21), but in human gene therapy a prob-lem may arise from targeting of the endogenous wild-typecopy (20). A RESC-based approach rescue might be avaluable strategy to prevent this.

The main limitations of the Gal4-driven RNAi/RESCcombination are intrinsic to the Gal4/UAS system. First,RESC expression levels are not endogenous and, dueto position effects, subject to variation between thetransgenic lines. The latter may be overcome by using aP-element replacement strategy or fC31 integrase totarget a common locus. Overexpression is more difficultto avoid when using the Gal4/UAS system alone, sincelowering expression by decreasing the temperature willalso affect RNAi performance. To uncouple RNAi andrescue, two independent systems with different inductionmodes, such as the Gal4 and lexA systems may be usedsimultaneously (22). Second, the use of tissue-specificdriver lines can only cause knockdown in cells where thetarget mRNA is present, but it induces expression in allcells expressing the driver, which is not necessarily identi-cal. Here, the endogenous promoter of the gene of interestwould be ideal. Another limitation is that synonymouscodons are not necessarily expressed with the same accu-racy (23) and at the same level (24), but this is probably ofminor importance given the efficiency of the rescuewe observe. In our web-based tool we suggest the mostfrequently used alternative codon.

An alternative to the RESC approach might be the useof cDNA from other Drosophila species. However, findinga related species with the right balance between sufficientchange at the nucleotide level to avoid knockdown and nocritical change at the amino acid level in order to fullyrestore function is not an easy task.

In summary, the combination of RESC rescue withlarge-scale RNAi libraries are a powerful approach tostudy gene function in vitro and tissue-specifically in vivowith high reliability because off-target effects can beexcluded experimentally.

SUPPLEMENTARY DATA

Supplementary Data are available at NAR online.

ACKNOWLEDGEMENTS

The authors thank Helga Ceulemans for excellent techni-cal assistance, Stein Aerts (Leuven) for establishingthe online tool to generate RESC constructs, theBloomington Drosophila Stock Center, the NationalInstitute of Genetics (Shizuoka), Dean Smith (Dallas),and Gerd Vorbruggen (Gottingen) for providing flystocks, and John Lincecum (Boston) for sdc cDNA andantibody.

FUNDING

The National Fund for Scientific Research - Flanders(FWO) (grant G.0498.05 to G.D.) and grants

G.0542.08N and G.0543.08N (to B.A.H); FlandersInstitute for Biotechnology (VIB) (to G.D. and B.A.H.);Interuniversity Attraction Poles (IUAP VI-20) Program ofthe Belgian Federal Government (to G.D.); andK.U.Leuven (to B.A.H.; Impuls, CREA, GOA; and toG.D., GOA). Funding for open access charge: VIB.

Conflict of interest statement. None declared.

REFERENCES

1. Brand,A.H. and Perrimon,N. (1993) Targeted gene expression as ameans of altering cell fates and generating dominant phenotypes.Development, 118, 401–415.

2. Roignant,J.Y., Carre,C., Mugat,B., Szymczak,D., Lepesant,J.A.and Antoniewski,C. (2003) Absence of transitive and systemicpathways allows cell-specific and isoform-specific RNAi inDrosophila. RNA, 9, 299–308.

3. Celotto,A.M. and Graveley,B.R. (2002) Exon-specific RNAi: a toolfor dissecting the functional relevance of alternative splicing. RNA,8, 718–724.

4. Dietzl,G., Chen,D., Schnorrer,F., Su,K.C., Barinova,Y., Fellner,M.,Gasser,B., Kinsey,K., Oppel,S., Scheiblauer,S. et al. (2007) Agenome-wide transgenic RNAi library for conditional geneinactivation in Drosophila. Nature, 448, 151–156.

5. Kulkarni,M.M., Booker,M., Silver,S.J., Friedman,A., Hong,P.,Perrimon,N. and Mathey-Prevot,B. (2006) Evidence of off-targeteffects associated with long dsRNAs in Drosophila melanogastercell-based assays. Nat. Methods, 3, 833–838.

6. Ma,Y., Creanga,A., Lum,L. and Beachy,P.A. (2006) Prevalence ofoff-target effects in Drosophila RNA interference screens. Nature,443, 359–363.

7. Moffat,J., Reiling,J.H. and Sabatini,D.M. (2007) Off-target effectsassociated with long dsRNAs in Drosophila RNAi screens. TrendsPharmacol. Sci., 28, 149–151.

8. Tuschl,T., Zamore,P.D., Lehmann,R., Bartel,D.P. and Sharp,P.A.(1999) Targeted mRNA degradation by double-stranded RNAin vitro. Genes Dev., 13, 3191–3197.

9. Aleman,L.M., Doench,J. and Sharp,P.A. (2007) Comparison ofsiRNA-induced off-target RNA and protein effects. RNA, 13,385–395.

10. Echeverri,C.J., Beachy,P.A., Baum,B., Boutros,M., Buchholz,F.,Chanda,S.K., Downward,J., Ellenberg,J., Fraser,A.G., Hacohen,N.et al. (2006) Minimizing the risk of reporting false positives in large-scale RNAi screens. Nat. Methods, 3, 777–779.

11. Sledz,C.A., Holko,M., de Veer,M.J., Silverman,R.H. andWilliams,B.R. (2003) Activation of the interferon system byshort-interfering RNAs. Nat. Cell. Biol., 5, 834–839.

12. Grimm,D., Streetz,K.L., Jopling,C.L., Storm,T.A., Pandey,K.,Davis,C.R., Marion,P., Salazar,F. and Kay,M.A. (2006) Fatalityin mice due to oversaturation of cellular microRNA/short hairpinRNA pathways. Nature, 441, 537–541.

13. Kalidas,S. and Smith,D.P. (2002) Novel genomic cDNA hybridsproduce effective RNA interference in adult Drosophila. Neuron, 33,177–184.

14. Spring,J., Paine-Saunders,S.E., Hynes,R.O. and Bernfield,M. (1994)Drosophila syndecan: conservation of a cell-surface heparan sulfateproteoglycan. Proc. Natl Acad. Sci. USA, 91, 3334–3338.

15. Lee,Y.S., Nakahara,K., Pham,J.W., Kim,K., He,Z.,Sontheimer,E.J. and Carthew,R.W. (2004) Distinct roles forDrosophila Dicer-1 and Dicer-2 in the siRNA/miRNA silencingpathways. Cell, 117, 69–81.

16. Arziman,Z., Horn,T. and Boutros,M. (2005) E-RNAi: a webapplication to design optimized RNAi constructs. Nucleic AcidsRes., 33, W582–W588.

17. Hsu,Y.C., Chern,J.J., Cai,Y., Liu,M. and Choi,K.W. (2007)Drosophila TCTP is essential for growth and proliferation throughregulation of dRheb GTPase. Nature, 445, 785–788.

18. Zeng,Y., Yi,R. and Cullen,B.R. (2003) MicroRNAs and smallinterfering RNAs can inhibit mRNA expression by similarmechanisms. Proc. Natl Acad. Sci. USA, 100, 9779–9784.

PAGE 7 OF 8 Nucleic Acids Research, 2009, Vol. 37, No. 13 e93

19. Doench,J.G., Petersen,C.P. and Sharp,P.A. (2003) siRNAs canfunction as miRNAs. Genes Dev., 17, 438–442.

20. Harper,S.Q., Staber,P.D., He,X., Eliason,S.L., Martins,I.H.,Mao,Q., Yang,L., Kotin,R.M., Paulson,H.L. and Davidson,B.L.(2005) RNA interference improves motor and neuropathologicalabnormalities in a Huntington’s disease mouse model. Proc. NatlAcad. Sci. USA, 102, 5820–5825.

21. Xia,H., Mao,Q., Eliason,S.L., Harper,S.Q., Martins,I.H., Orr,H.T.,Paulson,H.L., Yang,L., Kotin,R.M. and Davidson,B.L. (2004)RNAi suppresses polyglutamine-induced neurodegeneration in amodel of spinocerebellar ataxia. Nat. Med., 10, 816–820.

22. Lai,S.L. and Lee,T. (2006) Genetic mosaic with dualbinary transcriptional systems in Drosophila. Nat. Neurosci., 9,703–709.

23. Akashi,H. (1994) Synonymous codon usage in Drosophilamelanogaster: natural selection and translational accuracy. Genetics,136, 927–935.

24. Carlini,D.B. and Stephan,W. (2003) In vivo introductionof unpreferred synonymous codons into the DrosophilaAdh gene results in reduced levels of ADH protein. Genetics, 163,239–243.

e93 Nucleic Acids Research, 2009, Vol. 37, No. 13 PAGE 8 OF 8