What’s new?
The genetics of Drosophila transgenics Gregg Roman
Summary In Drosophila, the genetic approach is still the method of choice for answering fundamental questions on cell biology, signal transduction, development, physiology and behavior. In this approach, a gene’s function is ascertained by altering either the amount or quality of the gene product, and then observing the consequences. The genetic approach is itself polymorphous, encompassing new and more complex techniques that typically employ the growing collections of transgenes. The keystone of these modern Drosophila transgenic techniques has been the Gal4 binary system. Recently, several new techniques have modified this binary system to offer greater control over the timing, tissue specificity and magnitude of gene expression. Additionally, the advances in post-transcriptional gene silencing, or RNAi, have greatly expanded the ability to knockdown almost any gene’s function. Regardless of the growing experimental intricacy, the application of these advances to modify gene activity still obeys the fundamental principles of genetic analysis. Several of these transgenic techniques, which offer more precise control over a gene’s activity, will be reviewed here with a discussion on how they may be used for determining a gene’s function. BioEssays 26:1243–1253, 2004. ß 2004 Wiley Periodicals, Inc. Introduction Genetics is a question that man poses to nature, obliging her to answer. The question is simple: what effects will be brought by
Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, TX 77030. E-mail:
[email protected] DOI 10.1002/bies.20120 Published online in Wiley InterScience (www.interscience.wiley.com).
Abbreviations: DES, Diethylstilbestrol; dsRNA, double-stranded ribonucleic acid; EIR, expressed inverted repeats; GFP, green fluorescent protein; PTGS, post-transcriptional gene silencing; RISC, RNA interference silencing complex; siRNAs, small interfering RNAs; TARGET, temporal and regional gene expression targeting; TSE, tissue-specific enhancer; UAS, Gal4-responsive Upstream Activation Sequences.
BioEssays 26:1243–1253, ß 2004 Wiley Periodicals, Inc.
altering a gene’s function? Animals harboring a mutation are analyzed for instructive phenotypes (e.g. changes in morphology, behavior, cellular identity, gene expression, etc.). One can begin with a phenotype and search for mutations that have the effect—forward genetics. Alternatively, one can start with a gene of interest, and see what phenotypes are produced from mutations within this gene—reverse genetics. Yet for any phenotype to be instructive, it is necessary to understand what effect the mutation has on the wild-type gene function. A description of these affects was codified by Hermann Muller in an astonishingly insightful treatise, over seven decades ago.(1) Mutations that are informative will result in either in a lossof-function or a gain of-function effect. The loss-of-function mutations include amorphs, hypomorphs and antimorphs (also known as dominant negatives), which all act to reduce or eliminate the normal gene function.(1) The gain-of-function mutations include the hypermorphs (resulting in increased gene function) and the neomorphs (resulting in a new gene function(1)). A hypermorphic phenotype would occur if the gene was expressed at levels exceeding a tolerable cellular range or if the gene product was constitutively activated, in many cases resulting in a phenotype that is opposed to the amorph phenotype. In contrast, the neomorphic phenotype might result if the gene is ectopically expressed at the wrong time or in a wrong cell type allowing for promiscuous gene activation, aberrant phosphorylation, or other abnormal cellular events. The phenotype of an amorph or null mutation (lacking any gene function) is the easiest to interpret as it directly indicates the requirement for the gene. The logic is that, if the absence of a gene (e.g. the eyeless gene of Drosophila) results in the absence of a function (e.g. eye development), then the gene is required for that function. Since gain-of-function mutations affect gene activity in different ways, deducing the wild-type gene function from the resulting phenotypes is necessary more complex. This is especially true for neomorphic mutations, which can result in changes that lie outside of the normal gene function; interpreting these phenotypes should be approached with resolute caution. Even so, the characterization of neomorphic mutations can frequently reveal exciting glimpses into the function of regulatory genes.(2–4) Knowing which of these classes of mutations you have is not always easy. The hypermorph and the antimorph are
BioEssays 26.11
1243
What’s new?
typically dominant. The neomorph can also be dominant but is more frequently semi-dominant.(1) Hypomorph and amorph mutations are generally recessive.(5) These latter two classes can be distinguished by the phenotypes when placed over a deficiency for the locus. The phenotype of the hypomorph will become more severe, and there will be no change in the phenotype of an amorph.(1) The nature of a dominant mutation can also be distinguished by titrating the amount of the wildtype gene. As you increase the copy number of the wild-type gene relative to the mutation, the antimorphic phenotype becomes less severe, the hypermorphic phenotype becomes more severe, and there is generally no change in the neomorphic phenotype.(1) In more recent years, Drosophila geneticists have increasingly turned to transgenes to modify gene activity. In principle, changes in a gene’s activity corresponding to any of Muller’s classes of mutations may be achieved through transgenesis. With P element vectors, it is possible to increase the copy number of either wild-type or mutant genes in almost any genetic background.(6) These vectors are most commonly used to either complement loss-of-function mutations or to generate hypermorphic conditions through the overexpression of a gene of interest. In 1993, the Gal4 binary system was introduced into Drosophila, giving rise to a quantum leap in the experimenter’s control of gene activity.(7) In this system, the yeast Gal4 transcription factor is expressed in specific tissues either through enhancer detector P-elements or through the use of defined promoter or enhancer sequences.(7) A second construct, the UAS responder, is also required; in this vector, a gene of interest is cloned behind a Gal4 responsive promoter.(7–9) Gal4 can then drive the transcription of this gene in the desired tissue. A modicum of control over the levels of gene expression can be obtained by either altering the copy number of either or both P-elements, or by changing the incubation temperature of the flies; Gal4 is more active at 298C than at 198C.(7) Recently, two new techniques have emerged that advance the Gal4 binary vector method by combining the existing spatial control of gene expression with experimenter control over the timing of gene expression. These techniques, the hormone-inducible Gal4 systems and the temperature-sensitive Gal80 TARGET system, also provide for improved control over the level of transcription from UAS responder transgenes. The ability to further delimit the expression in both space and time with the newly developed technologies will increase the specificity of transgenic approaches to gene activity modification, and provide for within genotype controls. Furthermore, the promising techniques for post-transcriptional gene silencing (PTGS) may vastly expand the ability to reduce gene activity. Our usage of these techniques is firmly based upon the genetic principles learned over the last century. In this review, I seek to illustrate, with relevant examples, these modern transgenic techniques and how they may be used for reduction-of-function and gain-of-function experiments.
1244
BioEssays 26.11
Hormone-inducible Gal4 systems The hormone-inducible Gal4 fusion proteins contain a ligandbinding domain of a mammalian nuclear hormone receptor, the Gal4 DNA-binding domain, and a transcriptional activation domain; these hybrids can activate transcription from a Gal4 promoter in a ligand-inducible manner.(10,11) Two such fusion proteins have been moved into Drosophila for use in binary transgenic systems.(12–14) The Gal4–estrogen receptor fusion protein (Gal4ER) can activate transcription from UAS responders after flies are fed b-estradiol,(12) while the Gal4progesterone receptor fusion protein (Gene-Switch) will activate transcription in the presence of the anti-progestin RU486 (Fig. 1(13,14)). The levels of RU486 used to obtain maximum induction appear, so far, to have no inherent effect on Drosophila development, behavior or aging, thereby averting a possible confound.(13–17) The critical features of these new hybrid Gal4 drivers to be considered here are the availability of lines with defined spatial expression patterns, and the control over the level of gene activation. A major advantage to the hormone-inducible Gal4 systems is that they will work with the thousands of extant UAS responder lines, but one drawback is that new driver lines will have to be produced. Enhancer detector P-elements have been generated for both the Gal4ER and Gene-Switch fusion proteins.(12,14) These P-elements utilize transcriptional enhancers located in the flanking genomic DNA to drive the expression of the hybrid Gal4s. The different enhancer detectors can be further mobilized to generate new lines with distinct insertion sites and expression patterns. Small collections of these lines have already been generated, from which many independent expression patterns have been found.(12,14) A much larger screen of the P{Switch2} enhancer detector element has recently taken place in the laboratories of Ronald Davis and Haig Keshishian (personal communication; Baylor College of Medicine and Yale University respectively) increasing the number of independent lines. Another successful strategy for producing a desired spatial expression pattern is to drive the hybrid Gal4s with defined promoters or transcriptional enhancers. The torso-like (tsl) regulatory region was used to drive expression of Gal4ER within the border cells and a subset of posterior cells of the follicle.(12) Both pan-neuronal and pan-muscular Gene-Switch lines were generated using the elav and mhc promoters, respectively.(13) Furthermore, a series of ten vectors designed for the rapid cloning of regulatory regions 50 to Gene-Switch coding sequences have been constructed.(16) These vectors have been used to make eye-specific and mushroom-body-specific Gene-Switch lines.(16,17) The level of gene expression in the uninduced state is a crucial consideration for experiments that use the inducible techniques. In both the Gal4ER and Gene-Switch systems, the level of transcriptional activation in the absence of ligand appears in most lines to be generally quite low and frequently
What’s new?
Figure 1. The Gene-Switch fusion protein offers temporal, spatial and quantitative control of transgene expression. Gene-Switch is a hybrid transcription factor composed of the Gal4 DNA-binding domain, the human progesterone receptor ligand-binding domain and the human p65 transcriptional activation domain.(11) In this system, spatial expression can be achieved by driving transcription of Gene-Switch with a tissue-specific enhancer (TSE) or promoter. In The P{Switch} enhancer detector elements, the TSEs are located in the flanking genomic DNA, while the minimal promoter is from the P-transposase.(14) Since the Gene-Switch activation of a Gal4-responsive promoter (UAS) is strongly dependent of the presence of the inducer RU486, the experimenter can control the onset of gene expression. The expression level of the gene of interest (GOI) can be controlled by sampling at different times after induction or by using different levels of the inducer and sampling at the same time. The amount of RU486 required for saturation of Gene-Switch activation may depend on the level of Gene-Switch fusion protein or on the tissue being analyzed. The principles behind Gene-Switch are very similar to those of the Gal4-ER fusion protein, where the inducer is b-estradiol.(12)
undetectable. In several Gal4ER-expressing lines, both the detection of b-galactosidase reporter activity and the stimulation of cell death through the induction of an UAS-diphtheria toxin A responder were completely dependent on the presence of b-estradiol.(12) Similar experiments have also shown most Gene-Switch lines to be ‘‘off’’ in the uninduced state.(13,14,16) Exceptions to this low level of basal activity come from Gene-Switch lines that also have very high levels of induced expression.(13) In one elav-Gene-Switch line, a 51- to 60-fold increase in reporter gene levels was obtained after induction. Thus, it may be that RU486 increases Gene-Switchdependent transcriptional activation by greater than 50 fold and, in most lines, the basal uninduced level of activation remains either below the detection threshold or low enough to have little impact on cellular functions. A clear benefit to the hormone-inducible Gal4 systems is the ability to turn ‘‘on’’ a gene of interest. However, one of the beautiful, if not wholly appreciated, aspects of inducible gene expression system, is that the level of the gene of interest can be controlled by manipulating the ligand concentration or the time after induction. In both the Gal4ER and Gene-Switch systems, the induction of an UAS responder can reasonably be expected to proceed at a rate influenced by the concentration of ligand-activated hybrid Gal4 molecules. The amount of
gene product may then accumulate at this rate until equilibrium is reached. It should be possible, therefore, to examine different levels of the gene of interest by sampling different times after the start of induction. This has been shown with the induction of GFP by an elav-Gene-Switch driver and with lacZ induction by both adult glia and mushroom body Gene-Switch drivers.(13,14,17) Additionally, the level of induction can be modulated by controlling the amount of ligand fed to the flies. A titration of RU486 concentrations fed to an elav-Gene-Switch line demonstrated sensitive control over the level of GFP produced.(13) Similar results were obtained for an enhancer detector expressing Gene-Switch within the adult glia.(14) Thus, Gene-Switch can act as a transcriptional rheostat. The induction of gene activity by the hybrid Gal4 molecules may frequently take hours to more than one day to reach peak levels. In the tsl- Gal4ER line, b-galactosidase activity was first detected in follicles after 12 hours of continuous feeding with DES, and had achieved maximum levels only after 2.5 days of continuous feeding.(12) In a Gene-Switch enhancer detector line that drives expression in the adult glia, b-galactosidase was first detected after 3 hours, and achieved a maximum level of activity at approximately 24 hours; the kinetics for the induction of a GFP reporter by an elav-Gene-Switch line were somewhat faster, reaching a peak of expression by
BioEssays 26.11
1245
What’s new?
12 hours.(13,14) Hence, different levels of gene activity are achievable by sampling different times after induction. Interestingly, feeding starved adults a solution of 500 mM RU486 in 2% sucrose for as little as 1 hour provided for maximum levels of induction at 24 hours.(14) Thus, for saturating concentrations of RU486, the rate-limiting step of Gene-Switch transcriptional activation appears not to be the uptake and distribution of this ligand. Target The TARGET (Temporal And Regional Gene Expression Targeting) system has the potential of working with any existing GAL4 driver and UAS responder (Fig. 2). TARGET utilizes a temperature-sensitive Gal80 molecule to repress Gal4 activity.(18) In yeast, Gal80 will repress Gal4 activity in the absence of galactose by binding and inhibiting its transcriptional activation domain. Gal80 was first introduced into Drosophila to facilitate mosaic analysis; the loss of Gal80 from a daughter cell following mitotic recombination allows Gal4 to specifically mark patches of homozygous mutant cells in an otherwise heterozygous organism.(19) The Gal80ts protein was imported into Drosophila and shown to maximally repress Gal4 activity at 198C, while being permissive for Gal4 activity at 328C.(18) This temperature preference is the converse of Gal4’s own temperature sensitivity. Thus Gal80ts and Gal4 act as opposing forces, a sort of yin–yang of transcriptional activation, where lower temperatures inhibit transcription and elevated temperatures push transcription levels of the UAS responder higher. One strong benefit for TARGET is that temperature, as an inducer, is both rapid and
easily controlled, but one potential pitfall for TARGET is that elevated temperatures are not always neutral and can have undesired effects on the induced flies. For example, culture temperatures of 308C result in significantly more axonal branches and varicosities in the larval neuromuscular junction than at 258C.(20) Furthermore, Drosophila melanogaster show reduced mating, pupariation and oviposition when cultured above 328C, and males are typically infertile above 308C.(21) The utility of the trinary TARGET system comes in part from the ubiquitous, high level of expression of Gal80ts, such that Gal4 can be conditionally inhibited globally, and throughout development. In the TARGET system, the Gal80ts gene is driven to high levels by the Tubulin 1a-promoter.(18) TARGET offers great flexibility since two or more independent Gal80ts transgenes can also be added to the final genotype to boost the expression of Gal80ts. Additional Gal80ts transgenes may more fully repress transcriptional activation in lines that have particularly high levels of Gal4 or if the levels of the responder need to be extremely low in the uninduced state. Yet, the ability of a single Gal80ts to repress Gal4 at low temperatures is quite profound. One Gal80ts transgene appears to be able to completely repress a GFP reporter driven by an elav-Gal4 transgene. Additionally, a Gal80ts transgene was capable of fully suppressing the lethality produced when the GMRGal4 is combined with the UAS-head involution defect pro-apoptotic responder.(18) The kinetics of TARGET induction have been determined at the level of GFP mRNA levels, and are therefore not directly comparable to the kinetics of the hormone-inducible Gal4 systems for which protein accumulation or reporter activity has
Figure 2. The TARGET trinary system offers temporal, spatial and quantitative control of transgene expression. In this system, spatial expression can be achieved by driving transcription of Gal4 with a tissue-specific enhancer (TSE) or promoter. In theory, this technique can work with any Gal4 line. The ubiquitous expression of the temperature-sensitive Gal80ts gene is controlled by the Tubulin-1a promoter(18). At 198C, Gal80ts effectively inhibits Gal4 from activating transcription form a UAS promoter. The timing of transcriptional activation of a gene of interest (GOI) can be control by shifting the incubation temperature of the flies from 198C to higher temperatures(18). The rate of induction appears to be maximal at 328C but, at lower temperatures, intermediate levels of activation can be found.
1246
BioEssays 26.11
What’s new?
been measured. Nevertheless, GFP reporter mRNA was detected after three hours at 328C, and reached maximal levels after 6 hours at 328C. This induction resulted in a 42- to 48-fold increase in GFP message levels.(18) These data suggest that the high temperature induction of TARGET may be slightly faster than the Gene-Switch induction by feeding RU486. Like Gene-Switch, the TARGET system offers a rheostat capability. The level of induction can also be controlled either by the time from the period of induction or by the temperature used for inactivating the Gal80ts suppressor (Fig. 2(18)). Reduction-of-function transgenics In these experiments, a gene’s function is eliminated or inhibited by dominant transgenes. The need of a gene’s activity may then be inferred through the resulting phenotypes. There are several available methods that can disrupt a gene’s activity including dominant negative approaches, through the expression of selective peptide toxins, or through the use of post-transcriptional gene silencing. A gene is often required at several times and places during development. One challenge to these techniques is to limit the effect of the reduction-offunction to a given developmental time point or restricted to a specific tissue or cell type. In a classic paper, Ira Herskowitz(22) discussed methods of achieving the disruption of gene activity through the expression of dominant negative (antimorphic) mutations. This approach has been quite productive in Drosophila and is still in wide use (e.g.(23–25)). Dominant negative alleles will typically contain mutations that allow for the assembly of the altered protein into multimeric complexes, but leave the protein functionally inactive, thereby poisoning the entire complex.(22) As a consequence, this technique is limited to those genes whose products form such complexes. One such gene is shibire (shi), which encodes the GTPase dynamin.(26) Dynamin selfassembles into large macromolecular spirals that form collars around membranes pockets and regulates endocytosis.(27,28) In the shibirets (shits) mutation, synaptic vesicle recycling fails at restrictive temperatures, which leads to a rapid depletion of synaptic vesicles and the inhibition of synaptic signaling.(29) Toshihiro Kitamoto(30) generated a UAS responder that contains this temperature-sensitive allele of shibire. The targeted expression of the UASshits transgene has the extremely useful property of dominantly inhibiting synaptic signaling at temperatures above 298C, while having little effect on neuronal physiology at the lower permissive temperatures.(30) These properties have been exploited by several groups to dissect the neural and temporal elements of complex behaviors.(30–33) The limited expression of bacterial peptide toxins has also been used to interfere with discrete cellular functions. In most cases, these toxins will ablate the cells in which they are expressed; similar results can also be obtained through the
targeted expression of the pro-apoptotic genes reaper, head involution defect, or grim.(34) Early developmental expression of these genes frequently leads to the death of the organism, interfering with the analysis of later stages of development. To overcome this problem, conditional alleles of the ricin A chain and the diphtheria A toxin were identified and corresponding UAS responder lines made.(35,36) Both of these toxins disrupt cellular functions by interfering with protein translation. Although the conditional alleles proved useful, cellular ablation can still occur at permissive temperature when crossed with most Gal4 lines, presumably due to residual activity at permissive temperatures combined with high levels of expression. As stated earlier, when used in combination with the-inducible Gal4 systems however, the effect of the uninduced toxin is often minimal to non-existent.(12–14,16,18) The transgenic expression of the tetanus toxin light chain (TNT) has been an extremely useful technique for disrupting neural functions.(37) TNT specifically disrupts neurotransmission by cleaving synaptobrevin, a protein required for fusion of synaptic vesicle to the plasma membrane.(38) A novel conditional allele of TNT was recently generated by Keller et al.(39) to disrupt neural functions in adult flies. The approach of Keller et al.(39) was to insert a heterologous gene with a stop codon between the UAS promoter and the TNT coding sequences. The heterologous gene was also flanked by FRT-site-specific recombination sites. The transcriptional stop located between the UAS promoter and the TNT gene was then removed in vivo by heat-shock-inducible FLP recombinase, resulting in an intact UASTNT transgene.(39) Importantly, the FLP catalyzed activation of TNT could be induced in some, but not all, postmitotic cells. These conditional TNT alleles also proved to be somewhat leaky, perhaps limiting the general usefulness of this technique. It is possible, however, that using a different inducible system to control the recombinase expression may lower background activity. It may also be possible that Cre recombinase could be active in cells that are refractory to FLP recombinase. A possible solution to both these problems may have come from Heidmann and Lehner.(40) These authors have developed an inducible Cre recombinase system by fusing the human estrogen receptor ligand-binding domain to the Cre recombinase protein. In lines expressing this hybrid protein, recombination is dependent on the addition of estrogen. Post-transcriptional gene silencing (PTGS) is rapidly becoming one of the most powerful techniques for reducing gene activity. PTGS, also known as RNAi, is a conserved cellular process that exists in plants, nematodes, flies and mammals in which double-stranded RNA (dsRNA) induces the degradation, and thus silencing of homologous mRNAs.(41–44) PTGS is mechanistically linked to the protection against virus proliferation and to the production of the microRNA translational regulators.(45,46) In this process, the dsRNA is recognized by the RNase III-like enzyme dicer and cleaved into
BioEssays 26.11
1247
What’s new?
Figure 3. Expressed inverted repeats result in post-transcriptional gene silencing through the production of siRNAs. In Drosophila, an inverted repeat is cloned behind a Gal4 responsive promoter. These two repeats are separated by an intron. The inverted repeats are transcribed in cells that are also expressing Gal4. The expressed inverted repeats (EIR) are shown in blue and purple, separated by an intron shown in white. The transcribed intron engages the splicesome complex, which increases the export of the expressed inverted repeat from the nucleus. The EIR folds back on itself to form a long stretch of double-stranded RNA. This dsRNA is a substrate for the Dicer RNase, which is a component of the RNA interference silencing complex (RISC), located in the cytoplasm. The Dicer RNase cleaves the exported dsRNA into smaller, 21 nucleotide small inhibitory RNAs (siRNAs). The siRNAs are used by the RISC complex as a template for destroying homologous messages. Through this mechanism, specific messages can be targeted for knockdown in a cell-autonomous manner.
21 nucleotide fragments known as small interfering RNAs (siRNAs; Fig. 3(47)). The siRNAs are used as guides by the RNA interference silencing complex (RISC) to recognize and cleave homologous sequences, reducing target transcripts.(48,49) The dicer RNase physically associates with the RISC complex.(50) Members of the RISC complex, including dicer are cytoplasmically localized in both vertebrates and Drosophila.(51,52) The separate dicer and RISC enzymatic activities are also localized in the cytoplasm.(46,53) Therefore the nuclear export of the dsRNA substrate is an essential feature of PTGS and nuclear RNAs appear immune. Recently many laboratories have generated dsRNA producing transgenes to reduce or eliminate gene function (e.g.(54–57)). In this technique, mRNAs that contain inverted repeats are induced from UAS transgenes (Fig. 3). The expressed inverted repeats (EIR) form dsRNA that becomes a template for siRNA formation and PTGS. There are several features of PTGS in Drosophila that make this an incredibly powerful tool for reduction-of-function studies. The sequence targeted can be highly specific, the effect appears to be cell autonomous, and can be easily modulated experimentally. The rules for producing effective expressed inverted repeat
1248
BioEssays 26.11
constructs are becoming clearer and, in most cases, are quicker and require less work than generating mutations de novo in the gene of interest. In plants and C. elegans, PTGS produces a general systemic effect where the knockdown of targeted messages can occur very far away from the initial site of siRNA induction, although neurons of C. elegans and the guard cells of plants appears to escape systemic PTGS response.(41,58,59) In C. elegans, the transmembrane protein product of sid-1 is required for the spreading of PTGS,(60) a Drosophila homolog of this protein has not been found. The systemic PTGS response appears to be absent in both Drosophila and mouse. The evidence supporting cell autonomy for PTGS in Drosophila came initially from a failure to see spreading of the phenotypic effects outside of regions defined by Gal4 expression (e.g.(56,61–63) Van Roessel et al.(64)) found, by examining the PTGS of GFP variants using laser scanning confocal microscopy, that the effect could be strictly limited to engrailed defined embryonic segments. An important point from this study was that GAP junctions present between epidermal and ectodermal cells would not support PTGS spreading.(64) Similarly, Roignant et al.(65) using immunohistochemistry, failed
What’s new?
to find any spreading of EIR-induced PTGS of the batman, EcRB1 and Pcaf genes into adjacent cells. In this study, batman expressed inverted repeat clones were made within third instar tissues, and again PTGS spreading went undetected in the surrounding cells. Thus it appears that EIR-induced PTGS is generally a cell-autonomous process in Drosophila. In plants and C. elegans, there is also a transitive process in which secondary siRNAs appear that are homologous to sequences on the mRNAs that are either 50 or 30 to the original targeted sequence.(66,67) Thus PTGS can ‘‘spread’’ to include targets upstream or downstream of the original siRNA. This can limit the effectiveness of EIR-induced PTGS by eliminating entire classes of mRNAs that share sequence with the target siRNAs. In both plants and C. elegans, the transitive attribute of PTGS requires an RNA-dependent RNA polymerase(68–70) Drosophila, however does not appear to have any sequences homologous to this RNA-dependent RNA polymerase gene family.(71) The question of transitivity in Drosophila was addressed experimentally by Roignant et al.(65) These authors found that PTGS could effectively work against an individual splice form of the ecdysone receptors by targeting exons unique to a specific isoform. The failure to directly identify siRNAs from common exons and the failure to silence mRNAs of the non-targeted splice variants indicates that transitive ‘‘spreading’’ did not occur. This point was again demonstrated by looking for transitivity in larvae expressing a batman–GFP fusion and an EIR–GFP construct under the daughterlessGal4 driver.(65) In this experiment, GFP but not batman siRNAs were detected by RNAse protection. Additionally, there was no suppression of the native batman gene in these animals.(65) Taken together, these data suggest that extreme specificity of EIR-induced PTGS can be achieved in Drosophila through the appropriate selection of target sequences. The effectiveness of EIR-induced PTGS as a research tool appears to be increased by the ability to modulate the response experimentally. The severity of the PTGS response is sensitive to the level of EIR.(56,57,72) Control of EIR levels can be relatively easily achieved by using independent transgenes with different position effects, or by increasing the copy number of EIR transgenes present in the animal. Furthermore, this dose-dependent response suggests that PTGS should work quite well with inducible expression systems. Lam and Thummel(73) found that hsp70Bb promoter provided significant control over the timing of EIR-induced PTGS of the EcR and bFTZ-F1 genes. In another experiment, hsp-Gal4 was used to induce yellow EIR.(72) Simply by varying the number heat-shock treatments, the response would range from very weak to very severe PTGS of yellow.(72) The Tet-on-inducible system has also been used to knockdown the levels of phosphoglucomutase.(74) In these experiments, induction of the pgm EIR with doxycycline resulted in a 1.3- to 24-fold reduction in mRNA levels, while a significantly more modest reduction in enzymatic levels were found.
A second method to modulate EIR levels is through temperature. Fortier and Belote(56) found that the EIR-induced PTGS of tranformer-2 was temperature dependent using three independent Gal4 drivers—more severe at 298C and practically absent at 228C. One possibility for this temperature dependence is that Gal4 is more active at 298C than at lower temperatures,(7) so the EIR is expressed at higher levels resulting in greater PTGS. However, results from PTGS in plants suggest that it may be a more general phenomenon. The pathological effects of virus infection can be masked by higher temperature.(75,76) The temperature-masking effect can be attributed to the loss of virus-specific siRNA formation at temperatures below 228C, whereas at 248C siRNA formation was significantly enhanced.(77) Therefore, the temperature-sensitive event occurs at or before the dicer processing step and, in plants, higher temperatures lead to substantially more PTGS than lower temperatures. There are now several ‘‘rules’’ or suggestions that may lead to an efficient design of EIR transgenes. The actual orientation of the inverted repeats does not seem to matter since efficient PTGS has been achieved with either sense–antisense or antisense–sense constructs.(56,72) The length of the repeat region may be a critical factor, since Riechhart et al.(62) demonstrated that EIRs containing 200 nucleotide repeats provided poor PTGS as compared to repeats of greater than 500 nucleotides. The ability to clone the inverted repeat in E. coli is markedly increased by the additional of a spacer of at least 20–50 base pairs between the repeats.(72,78) Several studies have successfully used intronic spacers to separate the inverted repeats.(61,62,79) Kalidas and Smith(61) generated lush, white and Gaq EIRs by cloning into UAS vectors, genomic DNAs containing introns followed by an inverted cDNA sequence immediately after an intron. All three constructs demonstrated significantly stronger and more uniform PTGS than previous reports.(56,57,72) Reichhart et al.(62) examined the PTGS strength of necrotic with EIR constructs containing either a stem loop intervening spacer, or with an intronic spacer between the inverted repeats. In the resulting transgenic flies, significantly stronger PTGS was seen in the lines containing the intronic spacer.(62) What is not clear from these studies is whether the intron needs to be the spacer, or whether the increase potency of these constructs is simply because of the presence of an intron within the transcribed region, regardless of its position. In mammalian systems, the expression of recombinant proteins frequently requires the presence of an intron.(80–82) In Xenopus oocytes, spliced mRNAs are exported from the nucleus more rapidly and to much higher levels than are the identical mRNAs that did not contain the intron.(83) In the C6/36 Aedes cultured cells, adding an intron to luciferase increased the levels of expression from 12- to 60-fold, depending on the promoter.(84) Thus, it is possible the intron simply promotes the nuclear export of the EIR for more efficient processing by the dicer RNase and
BioEssays 26.11
1249
What’s new?
greater PTGS. Interestingly, when Svoboda et al.(78) varied the position of the SV40 intron within their mos EIR construct, they found that the highest PTGS in mouse oocytes occurred when the intron was 50 of the inverted repeat; placement of the intron between the inverted repeats actually inhibited mos PTGS relative to the absence of an intron. These data suggest that the position of an intron within the EIR may be very important, and should be further analyzed in Drosophila. In summary, when building an EIR construct for PTGS, the length of the repeat should be greater than 200 nucleotides, the repeats may be separated by a spacer for more efficient cloning, and the construct should contain an intron. However, care should be taken in choosing which intron to add to your EIR construct. Several Drosophila introns have been described that contain complex transcriptional regulatory sequences.(85–88) The placement of such an intron within the EIR construct could drive expression of the EIR in an unintended tissue or cell type. There are a few general provisos to keep in mind since EIR expression is still a young technique in Drosophila. It remains possible that some constructs may suppress translation of more than one gene. Whenever possible, phenotypes induced by EIR transgenes should be compared with those of the loss-of-function mutations. It is also possible that some tissues may be refractory to this technique. In C. elegans, for example, RNAi works very poorly in gonads and in the nervous system.(89) Lastly, it may even be possible that some postdevelopmental cells may have suppressors of PTGS, or may lack the machinery necessary for PTGS. Gain-of-function transgenics The use of UAS responders in gain-of-function studies frequently involves the expression of a wild-type cDNA in an otherwise mutant background. These experiments can determine the cells or tissues responsible for a given loss-of-function phenotype. For example, Zars et al.(90) found that the expression of a wild-type rutabaga (rut) cDNA in the mushroom bodies of a rut mutant was sufficient to rescue the olfactory learning phenotype of this rut mutation, thereby suggesting rut function within this brain region is important for learning and memory. Two important experimental points on gain-of-function experiments were made in this study.(90) First, the authors drew the comparison of the rut rescue with Gal4 lines to mosaic analysis. All of the Gal4 lines used in the Zars, et al. study(90) also express in cells outside of the mushroom bodies. The positioning of the rut function within mushroom bodies, therefore, was only possible because mushroom body neurons were the only cells in common among all the tested Gal4 lines capable of rescuing the phenotype. The second point is that, in these lines, Gal4 is expressed throughout development. Thus, one could not determine when rut was required for olfactory learning and memory.
1250
BioEssays 26.11
The inducible Gal4 systems have been utilized recently to address this specific question of a developmental versus an acute physiological role for rutabaga in olfactory learning and memory. McGuire et al.(18) using the TARGET system found that rut expression induced within the mushroom bodies posteclosion for 6 days rescued the olfactory learning and memory phenotype of a severe rut mutation. These authors also found that expression of rut during development, but not during adulthood was insufficient for rescuing the rut phenotype. In a similar series of experiments, Mao et al.(17) used a mushroom-body-specific Gene-Switch line to control the timing of rut expression within mushroom bodies. These authors found that induction of rut expression for as little as 12 hours before training was enough to rescue the olfactory learning and memory phenotype of a severe rut allele, while expression during development, but not during adulthood, failed to improve the learning scores of the rut mutant. These results indicate that the time constraint for rut expression is very close to the time of training and, therefore, rut is most likely to be involved in the physiology of learning.(17,18,90) Other gain-of-function studies utilize either ectopic or overexpression of non-mutant genes in a wild-type fly to examine that gene’s function. While these studies mostly give rise to neomorphic phenotypes, they also are frequently quite informative: an accurate and developed interpretation of these experiments will require information on the wild-type gene function. This information may come from: (1) a phenotype from an amorphic mutation or (2) biochemical information on gene function, which could possibly be presumed through sequence similarity to a known gene product. The phenotypes arising from ectopic expression of the eyeless gene is an excellent example of the former. Loss-of-function mutations in eyeless lead to a deletion of the eyes.(91) Yet when ectopically expressed in imaginal discs, eyeless can drive eye formation, indicating that the expression of this gene is sufficient for eye formation in some contexts.(92,93) In another classic example of this gain-of-function experimental approach, Sto¨rtkhul and Kettler(94) utilized the Gal4 system to increase the number of olfactory receptor neurons that expressed the or43a gene. Since this gene encoded a putative G protein-coupled receptor and was expressed in a small subset of olfactory receptor neurons, it was assumed to be an odorant receptor.(95,96) The flies in which or43a expression was increased gained an increased olfactory sensitivity to the odorant benzaldehyde as measured by electroantennagrams, suggesting that OR43a is activated by this odorant.(94) A companion paper confirmed that, when expressed in Xenopus oocytes, OR43a can activate G proteins in response to benzaldehyde.(97) There are significant advantages to these gain-of-function experiments. Information can be obtained about the function of a gene in the absence of a mutation, such as with or43a.(94) Gain-of-functions studies of this kind may also provide a phenotype when, due to functional
What’s new?
redundancy, the loss-of-function will not produce one. For instance, if Drosophila perceives benzaldehyde through multiple receptors, which is almost certainly the case, then an or43a loss-of-function mutant would have a very subtle to imperceptible behavioral phenotype. The ability to perform gain-of-function studies on the genomic scale was made possible by the development of the specialized EP transposable element.(98) This P-element construct contains a Gal4 UAS promoter at the terminus of the P-element-facing toward the 30 flanking sequence; when inserted within a proximal promoter region, it brings the flanking gene under Gal4 control. A large collection of EP lines has spawned a cottage industry of gain-of-function genetic screens, by allowing the overexpression or misexpression of hundreds of genes in tissues defined by the extant Gal4 driver lines.(98) A phenotype arising from these screens would most likely be neomorphic. Still, the responsible gene would be identified by the site of the EP-element insertion, and the sequence of this gene may provide insights into its cellular function. Additionally, the EP element will generally be a hypomorphic mutation, allowing for the characterization of a loss-of-function phenotype if one was not already available. The use of an inducible Gal4 system in these gain-offunction screens would offer a few theoretical advantages over the usual Gal4 system. First, a critical time for the emergence of the phenotype could be directly determined by modifying the time of induction. By inducing gene expression from the EP element at defined times, the resulting phenotype would be less likely to be caused by earlier, undesired and potentially unknown sites of expression. The induction may also suggest whether the phenotype arises as the end point of a cascade of gene-activation events, which may take some time to develop, or whether the phenotype is the direct result of the ectopically produced gene. A second benefit is that the level of ectopic activation may be modulated. The use of the inducible system’s rheostat function allows for an allelic series within a single genotype. From this type of experiment, one may discern whether there are phenotypic thresholds and whether the induced phenotype is a linear response to levels of ectopic expression. The single genotype allelic series would be particularly instructive for genes that function through concentration gradients (e.g. the gap genes(99)), or for genes that operate as part of a counting mechanism (e.g. runt, vap-33A(100,101)). Conclusions It has now been 103 years since William Castle, at Harvard University, first used Drosophila melanogaster in biological experimentation.(102) During this past century, a fairly detailed understanding of this organism’s biology has developed. As a consequence of this understanding, the questions asked are also becoming progressively more refined, and more highly developed tools are required to answer these new questions.
Yet, there is also a universal benefit to controlling as many variables as possible in experimentation. When it comes to altering gene activity, the primary variables that should be controlled are how, when, where and to what degree. The recent technical breakthroughs of the inducible Gal4 systems and the PTGS techniques hold the promise of allowing us to control these variables to a much greater extent than was previously possible. As a result, a gene’s function may be determined faster, more precisely, and with significantly less effort than ever before. References 1. Muller HJ. 1932. Further studies on the nature and causes of gene mutations. In Jones DF, editor, Proceedings of the Fifth International Congress of Genetics. Austin: Genetics Society of America. pp 215– 255. 2. Brunner E, Brunner D, Fu W, Hafen E, Basler K. 1999. The dominant mutation Glazed is a gain-of-function allele of wingless that, similar to loss of APC, interferes with normal eye development. Dev Biol 206:178– 188. 3. Mozer BA. 2001. Dominant Drop mutants are gain-of-function alleles of the muscle segment homeobox gene (msh) whose overexpression leads to the arrest of eye development. Dev Biol 233:380–393. 4. Lai EC. 2003. Drosophila tufted is a gain-of-function allele of the proneural gene amos. Genetics 163:1413–1425. 5. Wright S. 1929. The evolution of dominance. Am Natural 63:556–561. 6. Rubin GM, Spradling AC. 1982. Genetic transformation of Drosophila with transposable element vectors. Science 218:348–353. 7. Brand AH, Perrimon N. 1993. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401–415. 8. Rørth P. 1998. Gal4 in the Drosophila female germline. Mech Dev 78: 113–118. 9. Roman G, He J, Davis RL. 1999. New series of Drosophila expression vectors suitable for behavioral rescue. Biotechniques 27:54–56. 10. Webster NJ, Green S, Jin JR, Chambon P. 1988. The hormone-binding domains of the estrogen and glucocorticoid receptors contain an inducible transcription activation function. Cell 54:199–207. 11. Wang Y, O’Malley BW Jr, Tsai SY, O’Malley BW. 1994. A regulatory system for use in gene transfer. Proc Natl Acad Sci USA 91:8180–8184. 12. Han DD, Stein D, Stevens LM. 2000. Investigating the function of follicular subpopulations during Drosophila oogenesis through hormone-dependent enhancer-targeted cell ablation. Development 127: 573–583. 13. Osterwalder T, Yoon KS, White BH, Keshishian H. 2001. A conditional tissue-specific transgene expression system using inducible GAL4. Proc Natl Acad Sci USA 98:12596–12601. 14. Roman G, Endo K, Zong L, Davis RL. 2001. P[Switch], a system for spatial and temporal control of gene expression in Drosophila melanogaster. Proc Natl Acad Sci USA 98:12602–12607. 15. Wang MC, Bohmann D, Jasper H. 2003. JNK signaling confers tolerance to oxidative stress and extends lifespan in Drosophila. Dev Cell 5:811–816. 16. Roman G, Davis RL. 2002. Conditional expression of UAS-transgenes in the adult eye with a new gene-switch vector system. Genesis 34:127–131. 17. Mao Z, Roman G, Zong L, Davis RL. 2004. Pharmacogenetic rescue in time and space of the rutabaga memory impairment by using GeneSwitch. Proc Natl Acad Sci USA 101:198–203. 18. McGuire SE, Le PT, Osborn AJ, Matsumoto K, Davis RL. 2003. Spatiotemporal rescue of memory dysfunction in Drosophila. Science 302:1765–1768. 19. Lee T, Luo L. 1999. Mosaic analysis with a repressible cell marker for studies of gene function in neuronal morphogenesis. Neuron 22:451–461. 20. Zhong Y, Wu CF. 2004. Neuronal activity and adenylyl cyclase in environment-dependent plasticity of axonal outgrowth in Drosophila. J Neurosci 24:1439–1445.
BioEssays 26.11
1251
What’s new?
21. David JR, Allemand R, van Herrewege J, Cohet Y. 1983. Ecophysiology: abiotic factors. In: Ashburner M, et al. editors. The genetics and biology of Drosophila Vol. 3d. London: Academic Press, pp 105–170. 22. Herskowitz I. 1987. Functional inactivation of genes by dominant negative mutations. Nature 329:219–222. 23. Cherbas L, Hu X, Zhimulev I, Belyaeva E, Cherbas P. 2003. EcR isoforms in Drosophila: testing tissue-specific requirements by targeted blockade and rescue. Development 130:271–284. 24. Niimi T, Clements J, Gehring WJ, Callaerts P. 2002. Dominant-negative form of the Pax6 homolog eyeless for tissue-specific loss-of-function studies in the developing eye and brain in Drosophila. Genesis 34: 74–75. 25. Belote JM, Fortier E. 2002. Targeted expression of dominant negative proteasome mutants in Drosophila melanogaster. Genesis 34:80–82. 26. van der Bliek AM, Meyerowitz EM. 1991. Dynamin-like protein encoded by the Drosophila shibire gene associated with vesicular traffic. Nature 351:411–414. 27. Hinshaw JE, Schmid SL. 1995. Dynamin self-assembles into rings suggesting a mechanism for coated vesicle budding. Nature 374:190–192. 28. Stowell MH, Marks B, Wigge P, McMahon HT. 1999. Nucleotidedependent conformational changes in dynamin: evidence for a mechanochemical molecular spring. Nat Cell Biol 1:27–32. 29. Kosaka T, Ikeda K. 1983. Possible temperature-dependent blockage of synaptic vesicle recycling induced by a single gene mutation in Drosophila. J Neurobiol 14:207–225. 30. Kitamoto T. 2001. Conditional modification of behavior in Drosophila by targeted expression of a temperature-sensitive shibire allele in defined neurons. J Neurobiol 47:81–92. 31. Waddell S, Armstrong JD, Kitamoto T, Kaiser K, Quinn WG. 2000. The amnesiac gene product is expressed in two neurons in the Drosophila brain that are critical for memory. Cell 103:805–813. 32. Dubnau J, Grady L, Kitamoto T, Tully T. 2001. Disruption of neurotransmission in Drosophila mushroom body blocks retrieval but not acquisition of memory. Nature 411:476–480. 33. McGuire SE, Le PT, Davis RL. 2001. The role of Drosophila mushroom body signaling in olfactory memory. Science 293:1330–1333. 34. Wing JP, Zhou L, Schwartz LM, Nambu JR. 2001. Distinct cell killing properties of the Drosophila reaper, head involution defective, and grim genes. Cell Death Diffn 5:930–939. 35. Moffat KG, Gould JH, Smith HK, O’Kane CJ. 1992. Inducible cell ablation in Drosophila by cold-sensitive ricin A chain. Development 114:681–687. 36. Bellen HJ, D’Evelyn D, Harvey M, Elledge SJ. 1992. Isolation of temperature-sensitive diphtheria toxins in yeast and thEIR effects on Drosophila cells. Development 114:787–796. 37. Sweeney ST, Broadie K, Keane J, Niemann H, O’Kane CJ. 1995. Targeted expression of tetanus toxin light chain in Drosophila specifically eliminates synaptic transmission and causes behavioral defects. Neuron 14:341–351. 38. Keller A, Sweeney ST, Zars T, O’Kane CJ, Heisenberg M. 2002. Targeted expression of tetanus neurotoxin interferes with behavioral responses to sensory input in Drosophila. J Neurobiol 50:221–233. 39. McMahon HT, Ushkaryov YA, Edelmann L, Link E, Binz T, et al. 1993. Cellubrevin is a ubiquitous tetanus-toxin substrate homologous to a putative synaptic vesicle fusion protein. Nature 364:346–349. 40. Heidmann D, Lehner CF. 2000. Reduction of Cre recombinase toxicity in proliferating Drosophila cells by estrogen-dependent activity regulation. Dev Genes Evol 211:458–465. 41. Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, et al. 1998. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391:806–811. 42. Hamilton AJ, Baulcombe DC. 1999. A species of small antisense RNA in posttranscriptional gene silencing in plants. Science 286:950–952. 43. Clemens JC, Worby CA, Simonson-Leff N, Muda M, Maehama T, et al. 2000. Use of double-stranded RNA interference in Drosophila cell lines to dissect signal transduction pathways. Proc Natl Acad Sci USA 97: 6499–6503. 44. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, et al. 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411:494–498.
1252
BioEssays 26.11
45. Voinnet O. 2001. RNA silencing as a plant immune system against viruses. Trends Genet 17:449–459. 46. Hutvagner G, Zamore PD. 2002. A microRNA in a multiple-turnover RNAi enzyme complex. Science 297:2056–2060. 47. Bernstein E, Caudy AA, Hammond SM, Hannon GJ. 2001. Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature 409:363–366. 48. Hammond SM, Bernstein E, Beach D, Hannon GJ. 2000. An RNAdirected nuclease mediates post-transcriptional gene silencing in Drosophila cells. Nature 404:293–296. 49. Elbashir SM, Lendeckel W, Tuschl T. 2001. RNA interference is mediated by 21- and 22-nucleotide RNAs. Genes Dev 15:188–200. 50. Hammond SM, Boettcher S, Caudy AA, Kobayashi R, Hannon GJ. 2001. Argonaute2, a link between genetic and biochemical analyses of RNAi. Science 293:1146–1150. 51. Billy E, Brondani V, Zhang H, Muller U, Filipowicz W. 2001. Specific interference with gene expression induced by long, double-stranded RNA in mouse embryonal teratocarcinoma cell lines. Proc Natl Acad Sci USA 98:14428–14433. 52. Findley SD, Tamanaha M, Clegg NJ, Ruohola-Baker H. 2003. Maelstrom, a Drosophila spindle-class gene, encodes a protein that colocalizes with Vasa and RDE1/AGO1 homolog, Aubergine, in nuage. Development 130:859–871. 53. Lee Y, Jeon K, Lee JT, Kim S, Kim VN. 2002. MicroRNA maturation: stepwise processing and subcellular localization. EMBO J 21:4663– 4670. 54. Chuang CF, Meyerowitz EM. 2000. Specific and heritable genetic interference by double-stranded RNA in Arabidopsis thaliana. Proc Natl Acad Sci USA 97:4985–4990. 55. Tavernarakis N, Wang SL, Dorovkov M, Ryazanov A, Driscoll M. 2000. Heritable and inducible genetic interference by double-stranded RNA encoded by transgenes. Nat Genet 24:180–183. 56. Fortier E, Belote JM. 2000. Temperature-dependent gene silencing by an expressed inverted repeat in Drosophila. Genesis 26:240–244. 57. Martinek S, Young MW. 2000. Specific genetic interference with behavioral rhythms in Drosophila by expression of inverted repeats. Genetics 156:1717–1725. 58. Palauqui JC, Elmayan T, Pollien JM, Vaucheret H. 1997. Systemic acquired silencing: transgene-specific post-transcriptional silencing is transmitted by grafting from silenced stocks to non-silenced scions. EMBO J 16:4738–4745. 59. Voinnet O, Vain P, Angell S, Baulcombe DC. 1998. Systemic spread of sequence-specific transgene RNA degradation in plants is initiated by localized introduction of ectopic promoterless DNA. Cell 95:177– 187. 60. Winston WM, Molodowitch C, Hunter CP. 2002. Systemic RNAi in C. elegans requires the putative transmembrane protein SID-1. Science 295:2456–2459. 61. Kalidas S, Smith DP. 2002. Novel genomic cDNA hybrids produce effective RNA interference in adult Drosophila. Neuron 33:177–184. 62. Reichhart JM, Ligoxygakis P, Naitza S, Woerfel G, Imler JL, et al. 2002. Splice-activated UAS hairpin vector gives complete RNAi knockout of single or double target transcripts in Drosophila melanogaster. Genesis 34:160–164. 63. Enerly E, Larsson J, Lambertsson A. 2002. Reverse genetics in Drosophila: from sequence to phenotype using UAS-RNAi transgenic flies. Genesis 34:152–155. 64. Van Roessel P, Hayward NM, Barros CS, Brand AH. 2002. Two-color GFP imaging demonstrates cell-autonomy of GAL4-driven RNA interference in Drosophila. Genesis 34:170–173. 65. Roignant JY, Carre C, Mugat B, Szymczak D, Lepesant JA, et al. 2003. Absence of transitive and systemic pathways allows cell-specific and isoform-specific RNAi in Drosophila. RNA 9:299–308. 66. Sijen T, Fleenor J, Simmer F, Thijssen KL, Parrish S, et al. 2001. On the role of RNA amplification in dsRNA-triggered gene silencing. Cell 107: 465–476. 67. Vaistij FE, Jones L, Baulcombe DC. 2002. Spreading of RNA targeting and DNA methylation in RNA silencing requires transcription of the target gene and a putative RNA-dependent RNA polymerase. Plant Cell 14:857–867.
What’s new?
68. Smardon A, Spoerke JM, Stacey SC, Klein ME, Mackin N, et al. 2000. EGO-1 is related to RNA-directed RNA polymerase and functions in germ-line development and RNA interference in C. elegans. Curr Biol 10:169–178. 69. Mourrain P, Beclin C, Elmayan T, Feuerbach F, Godon C, et al. 2000. Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell 101:533–542. 70. Dalmay T, Hamilton A, Rudd S, Angell S, Baulcombe DC. 2000. An RNA-dependent RNA polymerase gene in Arabidopsis is required for posttranscriptional gene silencing mediated by a transgene but not by a virus. Cell 101:543–553. 71. Stein P, Svoboda P, Anger M, Schultz RM. 2003. RNAi: mammalian oocytes do it without RNA-dependent RNA polymerase. RNA 9:187– 192. 72. Piccin A, Salameh A, Benna C, Sandrelli F, Mazzotta G, et al. 2001. Efficient and heritable functional knock-out of an adult phenotype in Drosophila using a GAL4-driven hairpin RNA incorporating a heterologous spacer. Nucleic Acids Res 29:E55–5. 73. Lam G, Thummel CS. 2000. Inducible expression of double-stranded RNA directs specific genetic interference in Drosophila. Curr Biol 10: 957–963. 74. Allikian MJ, Deckert-Cruz D, Rose MR, Landis GN, Tower J. 2002. Doxycycline-induced expression of sense and inverted-repeat constructs modulates phosphogluconate mutase (Pgm) gene expression in adult Drosophila melanogaster. Genome Biol 3: research0021.1– 0021.10. 75. Johnson J. 1922. The relation of air temperature to the mosaic disease of potatoes and other plants. Phytopathology 12:930–938. 76. Harrison BD. 1956. Studies on the effect of temperature on virus multiplication in inoculated leaves. Annu Appl Biol 44:215–226. 77. Szittya G, Silhavy D, Molnar A, Havelda Z, Lovas A, Lakatos L, Banfalvi Z, Burgyan J. 2003. Low temperature inhibits RNA silencing-mediated defence by the control of siRNA generation. EMBO J 22:633–640. 78. Svoboda P, Stein P, Schultz RM. 2001. RNAi in mouse oocytes and preimplantation embryos: effectiveness of hairpin dsRNA. Biochem Biophys Res Commun 287:1099–1104. 79. Lee YS, Carthew RW. 2003. Making a better RNAi vector for Drosophila: use of intron spacers. Methods 30:322–329. 80. Hamer DH, Leder P. 1979. Splicing and the formation of stable RNA. Cell 18:1299–1302. 81. Buchman AR, Berg P. 1988. Comparison of intron-dependent and intron-independent gene expression. Mol Cell Biol 8:4395–4405. 82. Chung S, Perry RP. 1989. Importance of introns for expression of mouse ribosomal protein gene rpL32. Mol Cell Biol 9:2075–2082. 83. Luo MJ, Reed R. 1999. Splicing is required for rapid and efficient mRNA export in metazoans. Proc Natl Acad Sci USA. 96:14937–14942. 84. Zieler H, Huynh CQ. 2002. Intron-dependent stimulation of marker gene expression in cultured insect cells. Insect Mol Biol 11:87–95. 85. Tourmente S, Chapel S, Dreau D, Drake ME, Bruhat A, et al. 1993. Enhancer and silencer elements within the first intron mediate the transcriptional regulation of the beta 3 tubulin gene by 20-hydroxyecdysone in Drosophila Kc cells. Insect Biochem Mol Biol 23:137–143.
86. Meredith J, Storti RV. 1993. Developmental regulation of the Drosophila tropomyosin II gene in different muscles is controlled by muscle-typespecific intron enhancer elements and distal and proximal promoter control elements. Dev Biol 59:500–512. 87. Keplinger BL, Guo X, Quine J, Feng Y, Cavener DR. 2001. Complex organization of promoter and enhancer elements regulate the tissueand developmental stage-specific expression of the Drosophila melanogaster Gld gene. Genetics 157:699–716. 88. Adachi Y, Hauck B, Clements J, Kawauchi H, Kurusu M, et al. 2003. Conserved cis-regulatory modules mediate complex neural expression patterns of the eyeless gene in the Drosophila brain. Mech Dev 120: 1113–1126. 89. Kennedy S, Wang D, Ruvkun G. 2004. A conserved siRNA-degrading RNase negatively regulates RNA interference in C. elegans. Nature 427:645–649. 90. Zars T, Fischer M, Schulz R, Heisenberg M. 2000. Localization of a short-term memory in Drosophila. Science 288:672–675. 91. Richards MH, Furrow EY. 1922. A preliminary report on the optic tract of eyeless flies. Proc Ok Acad Sci 2:41–45. 92. Halder G, Callaerts P, Gehring WJ. 1995. Induction of ectopic eyes by targeted expression of the eyeless gene in Drosophila. Science 267: 1788–1792. 93. Chen R, Halder G, Zhang Z, Mardon G. 1999. Signaling by the TGFbeta homolog decapentaplegic functions reiteratively within the network of genes controlling retinal cell fate determination in Drosophila. Development 126:935–943. 94. Stortkuhl KF, Kettler R. 2001. Functional analysis of an olfactory receptor in Drosophila melanogaster. Proc Natl Acad Sci USA 98: 9381–9385. 95. Clyne PJ, Warr CG, Freeman MR, Lessing D, Kim J, et al. 1999. A novel family of divergent seven-transmembrane proteins: candidate odorant receptors in Drosophila. Neuron 22:327–338. 96. Vosshall LB, Amrein H, Morozov PS, Rzhetsky A, Axel R. 1999. A spatial map of olfactory receptor expression in the Drosophila antenna. Cell 96:725–736. 97. Wetzel CH, Behrendt HJ, Gisselmann G, Stortkuhl KF, Hovemann B, et al. 2001. Functional expression and characterization of a Drosophila odorant receptor in a heterologous cell system. Proc Natl Acad Sci USA 98:9377–9380. 98. Rørth P. 1996. A modular misexpression screen in Drosophila detecting tissue-specific phenotypes. Proc Natl Acad Sci USA 93:12418–12422. 99. Struhl G. 1989. Differing strategies for organizing anterior and posterior body pattern in Drosophila embryos. Nature 338:741–744. 100. Duffy JB, Gergen JP. 1991. The Drosophila segmentation gene runt acts as a position-specific numerator element necessary for the uniform expression of the sex-determining gene Sex-lethal. Genes Dev 5:2176– 2187. 101. Pennetta G, Hiesinger P, Fabian-Fine R, Meinertzhagen I, Bellen H. 2002. Drosophila VAP-33A directs bouton formation at neuromuscular junctions in a dosage-dependent manner. Neuron 35:291–306. 102. Allen GE. 1975. The introduction of Drosophila into the study of heredity and evolution, 1900–1910. Isis 66:322–333.
BioEssays 26.11
1253
