Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 May 10.
Published in final edited form as: Methods Mol Biol. 2007;372:207–217. doi: 10.1007/978-1-59745-365-3_15

Functional Analysis by Inducible RNA Interference in Drosophila melanogaster

Yuichi Matsushima, Cristina Adán, Rafael Garesse, Laurie S Kaguni
PMCID: PMC4862613  NIHMSID: NIHMS783761  PMID: 18314728

Summary

Ribonucleic acid (RNA) interference triggered by double-stranded RNA has become a powerful tool for generating loss-of-function phenotypes. It is used to inactivate genes of interest and represents an elegant approach to genome functional analysis by reverse genetics. In Drosophila, RNA interference has been used in both cell culture and animals. We have adopted this approach to reveal the physiological roles of a number of proteins involved in mitochondrial deoxyribonucleic acid metabolism, and present here experimental schemes to induce the stable expression of double-stranded RNA in Schneider cells and in transgenic Drosophila.

Keywords: Drosophila, RNA interference, RNAi, Schneider cells, transgenesis, UAS-GAL4 system

1. Introduction

Ribonucleic acid interference (RNAi) triggered by double-stranded RNA (dsRNA) was originally found in plants (posttranscriptional gene silencing; 1), fungi (“quelling”; 2) and Caenorhabditis elegans (3), but it has been shown to occur in virtually every organism examined, from protozoa to animals (4). The phenomenon of RNAi is triggered by dsRNA molecules that are cleaved into smaller RNA duplexes, 21–27 nucleotides long, by the ribonuclease (RNAse) III-type endonuclease Dicer. In Drosophila melanogaster, there are two Dicer isoforms: Dicer-1 (DCR-1) processes microRNAs (miRNAs), and Dicer-2 (DCR-2) is required for long dsRNA cleavage. The small dsRNA molecules, such as short interfering RNAs and miRNAs, are subsequently unwound and rearranged into effector complexes: RNA-induced silencing complex, RNA-induced transcriptional silencing or miRNA ribonucleoprotein particles. The RNA-induced silencing complex mediates the posttranscriptional degradation of homologous messenger RNAs, whereas RNA-induced transcriptional silencing promotes the condensation of heterochromatin, and miRNA ribonucleoprotein particles guide translational repression of messenger RNA targets (5). Although the basic mechanisms by which gene expression is suppressed are not completely understood, RNAi has become a powerful tool for generating loss-of-function phenotypes.

In Drosophila, RNAi has been used in both cell culture and animals. The transfection of dsRNA into Drosophila Schneider cells was found to have a gene-specific silencing function (6), but the transfected dsRNA only works for a short period. However, if the dsRNA is produced from a vector integrated into the genome (i.e., in an established RNAi cell line), then the RNAi effect can be monitored continuously. In living organisms, RNAi can be induced by injecting, feeding, or expressing dsRNA. Injection of dsRNA into Drosophila embryos disrupts gene activity efficiently (7), but its effect is transient, not inherited in the next generation, and genes expressed in later stages of development cannot be inactivated. To overcome these limitations, several strategies, mainly the use of the upstream activator sequences (UAS)-GAL4 system to induce controlled expression of the RNAi, have been developed to express dsRNA stably in transgenic Drosophila.

We are using a systematic RNAi approach to unravel the function of several essential factors that are constituents of the mitochondrial deoxyribonucleic acid (DNA) replication and transcription machinery in Drosophila. In this chapter, we describe protocols to generate RNAi stably in Schneider cell lines using inducible vectors. We also describe the general strategy to induce RNAi in D. melanogaster using the UAS-GAL4 system.

2. Materials

2.1. RNAi in Schneider Cells

2.1.1. Construction of the RNAi Vector

  1. pMt/Hy DNA (8): this vector can be obtained from the authors, or a variety of metallothionein promoter-based vectors can be obtained from the Drosophila Genomics Resource Center (http://dgrc.cgb.indiana.edu/news.html).

  2. Restriction enzymes.

  3. GeneElute agarose spin column (Sigma).

  4. Platinum Pfx DNA polymerase (Invitrogen).

  5. QIAquick polymerase chain reaction (PCR) purification kit (Qiagen).

  6. T4 DNA ligase.

  7. E. coli SURE cells (Stratagene).

  8. Electroporation device such as the E. coli pulser (Bio-Rad).

  9. Plasmid Midi Kit (Qiagen).

  10. 3 M sodium acetate, pH 5.2.

  11. Phenol/chloroform/isoamyl alcohol (25:24:1, v/v/v).

  12. TE buffer: 10 mM Tris-HCl, pH 8.0, 1 mM ethylenediaminetetraacetic acid (EDTA).

2.1.2. Establishment of the RNAi Cell Line

  1. Schneider S2 cells.

  2. Schneider’s Drosophila medium (Gibco) supplemented with 10% (v/v) fetal bovine serum (Gibco).

  3. Effectene transfection reagent (Qiagen).

  4. Hygromycin B (50 mg/mL) (Invitrogen).

  5. 60-mm tissue culture dish (Corning).

  6. 25-cm2 flask (Corning).

  7. 75-cm2 flask (Corning).

  8. 125-cm2 flask (Corning).

2.1.3. Induction of dsRNA Expression

  1. 100 mM CuSO4.

  2. Phosphate-buffered saline: 135 mM NaCl, 10 mM Na2HPO4, 2 mM KCl, 2 mM KH2PO4.

  3. Lysis buffer: 10 mM Tris-HCl, pH 8.0, 5 mM EDTA, 1% (w/v) sodium dodecyl sulfate.

  4. BCA protein assay kit (Pierce).

2.2. RNAi in Flies

2.2.1. Construction of the RNAi Vector for Flies

For construction of the RNAi vector for flies, use pUAST DNA (9). This vector can be obtained from the Drosophila Genomics Resource Center (http://dgrc.cgb.indiana.edu/news.html).

2.2.2. Generation of UAS-IR (Inverted Repeat) Lines

There are no specific materials for generation of UAS-IR lines. General Drosophila lab supplies can be obtained from LabScientific.

2.2.3. Setting the UAS-IR × GAL4 Cross and RNAi Analysis

  1. Plastic vials (75 × 25 mm diameter) (LabScientific).

  2. Plastic fly bottles (100 × 25 mm diameter) (LabScientific).

  3. Nonabsorbent cotton plugs for fly bottles and vials.

  4. Thin brushes and forceps.

  5. Anesthetic device (carbon dioxide and related products).

  6. Dissecting microscope and halogen lamp.

  7. Fly food: 12 g/L agar; 100 g/L yeast (available from bakers’ suppliers); 100 g/L sugar; 35 g/L maize meal; and 3 mL/L propionic acid.

2.2.4. Phenotypic Analysis

There are no specific materials for phenotypic analysis. General Drosophila lab supplies can be obtained from LabScientific.

3. Methods

3.1. RNAi in Schneider Cells

3.1.1. Construction of the RNAi Vector

We describe the construction of an RNAi vector targeting the Drosophila mtTFB1 gene as an example. To construct RNAi vectors targeting other genes, use PCR primers appropriate for those genes (see Note 1). A schematic diagram depicting the structure of a dsRNA expression plasmid targeting the Drosophila mtTFB1 gene in Schneider cells is shown in Fig. 1.

  1. Digest 1 µg pMt/Hy plasmid DNA with XhoI and SpeI in a reaction volume of 20 µL.

  2. Electrophorese the reaction mixture in a 0.7% agarose gel.

  3. Purify the vector from the agarose gel using a standard method (e.g., using a GeneElute agarose spin column).

  4. Amplify the sense fragment using the mtTFB1 complementary DNA, platinum Pfx DNA polymerase and the pair of primers 5′-CGCctcgagactagt ACGGACAAGATAGTCAAGTCG-3′ and 5′-CGCcaattcGGGatcgatTAGCTTCTCAGCAACCTCCTC-3, and the antisense fragments with 5′-CGCctcgagactagtACGGACAAGATAGTCAAGTCG-3′ and 5′-CGCgaattcAAAaagcttTAGCTTCTCAGCAACCTCCTC-3′.

  5. Purify the PCR products with a PCR purification kit (e.g., QIAquick PCR purification kit).

  6. Digest the sense PCR fragment with XhoI and EcoRI and digest the antisense PCR fragment with SpeI and EcoRI.

  7. Electrophorese the reaction mixtures in a 1% agarose gel.

  8. Purify the fragments from the agarose gel using an appropriate method as above.

  9. Ligate the vector DNA with the sense and antisense fragments using T4 DNA ligase at 16°C overnight. Use a molecular ratio of vector/sense fragment/antisense fragment of 1:3:3.

  10. Transform E. coli host cells with 1 µL of ligation mixture using the E. coli pulser and plate on an Luria-Bertani (LB) plate containing ampicillin at 100 µg/mL. We use E. coli SURE cells as the host.

  11. Recover plasmids from the colonies and check their identity and integrity by restriction endonuclease digestion.

  12. Select positive colonies and purify the plasmid DNA (e.g., using a Qiagen plasmid Midi Kit).

  13. Dissolve the DNA in 400 µL TE buffer.

  14. Add 400 µL phenol/chloroform/isoamyl alcohol (25:24:1, v/v/v), mix well, and then separate the phases by centrifugation at 12,000g for 5 min.

  15. Transfer the aqueous phase to a new tube, add 40 µL 3 M sodium acetate, pH 5.2, and 800 µL ethanol, mix, and incubate at room temperature for 10 min.

  16. Centrifuge at 12,000g for 10 min and discard the supernatant. Rinse the pellet with 600 µL 70% ethanol. Centrifuge at 12,000g for 5 min and discard the supernatant.

  17. Dry the pellet at room temperature for 10 min and dissolve the pellet in 50 µL of TE buffer.

Fig. 1.

Fig. 1

Open in a new tab

Schematic representation of a dsRNA expression plasmid targeted to the Drosophila mtTFB1 gene. Gray arrows show sense and antisense sequences. The open box indicates the loop region. The loop region is 24 nt and contains three restriction enzyme sites, ClaI–CCC–EcoRI–AAA–HindIII.

3.1.2. Establishment of the RNAi Cell Line

  1. Culture Drosophila Schneider S2 cells at 25°C in Drosophila Schneider medium supplemented with 10% (v/v) fetal bovine serum. Subculture cells to 3 to 5 × 106 cells/mL every third to fifth day.

  2. Place 4 mL cells at a density of 3 to 5 × 106 cells/mL into a 60-mm dish 24 h prior to transfection.

  3. Transfect the cells using Effecten (Qiagen) according to the manufacturer’s instructions.

  4. Incubate for 24–48 h at 25°C.

  5. Transfer the cells from a 60-mm dish into a 25-cm2 flask containing 4 mL fresh Drosophila Schneider medium supplemented with 10% (v/v) fetal bovine serum and 200 µg/mL hygromycin.

  6. Incubate at 25°C until cells reach a density of 10 to 15 × 106 cells/mL or for 7–10 d.

  7. Transfer the cells into a 75-cm2 flask containing 10 mL fresh Drosophila Schneider medium supplemented with 10% (v/v) fetal bovine serum and 200 µg/mL hygromycin.

  8. Incubate at 25°C until cells reach a density of 15 to 20 × 106 cells/mL or for 5–7 d.

  9. Transfer the cells into a 125-cm2 flask containing 20 mL fresh Drosophila Schneider medium supplemented with 10% (v/v) fetal bovine serum and 200 µg/mL hygromycin.

  10. Incubate at 25°C until cells reach a density of 15 to 20 × 106 cells/mL or for 5–7 d.

  11. Transfer 7 mL of the culture into a 125-cm2 flask containing 20 mL fresh Drosophila Schneider medium supplemented with 10% (v/v) fetal bovine serum and 200 µg/mL hygromycin.

  12. Incubate at 25°C until cells reach a density of 15 to 20 × 106 cells/mL or for 5–7 d.

  13. Culture the selected Schneider S2 cells at 25°C in Drosophila Schneider medium supplemented with 10% (v/v) fetal bovine serum, subculturing to 3 to 5 × 106 cells/mL every third to fifth day.

3.1.3. Induction of dsRNA Expression

  1. Dilute the cells to 3 to 5 × 106 cells/mL and add CuSO4 to a final concentration of 0.4 mM.

  2. Incubate at 25°C and subculture to 3 to 5 × 106 cells/mL every third day.

  3. After 10 d culture, harvest the cells by centrifugation at 2000g for 5 min.

  4. Wash with phosphate-buffered saline and centrifuge at 2000g for 5 min.

  5. Add lysis buffer, heat at 100°C for 5 min, and centrifuge at 12,000g for 10 min.

  6. Transfer the lysate to a fresh tube and assay the protein concentration (e.g., using the BCA protein assay kit).

  7. Check the suppression level of the target protein by immunoblot analysis (Fig. 2) (see Notes 2 and 3).

Fig. 2.

Fig. 2

Open in a new tab

Expression of d-mtTFB1-targeted RNAi in Schneider cells. Immunoblot analysis of mitochondrial extracts probed with affinity-purified rabbit antiserum against d-mtTFB1 (14). Schneider cells with no plasmid (control) or carrying pMt/Hy (vector) or RNAi vector (RNAiB1) were cultured for 10 d in the presence or absence of 0.4 mM CuSO4. Protein extracts (20 µg) were fractionated by 10.5% sodium dodecylsulfatepolyacrylamide gel electrophoresis, transferred to nitrocellulose filters, and probed with rabbit antiserum against d-mtTFB1 or d-mtTFB2 (15) as indicated (see Notes 2 and 3).

3.2. RNAi in Flies

One of the most valuable tools available for scientists working in Drosophila is the availability of the UAS-GAL4 system (9), which is shown schematically in Fig. 3. In this dual system, its two components (UAS and GAL4 transgenic lines) are maintained as independent parental stocks until needed. After crossing, the resulting F1 generation will express the gene (or RNAi) of interest in the pattern driven by GAL4. At present, numerous GAL4 drivers for constitutive or tissue-specific overexpression have been reported in the literature, and a list of them can be obtained from public stock centers (http://fly.bio.indiana. edu/gal4.htm). The flexibility of the UAS-GAL4 system has allowed its use for the analysis of many biological processes, including the induction of loss-of-function phenotypes through overexpression of RNAi constructs (10).

Fig. 3.

Fig. 3

Open in a new tab

Schematic representation of the UAS-GAL4 strategy. UAS-IR and GAL4 lines are maintained as independent stocks. GAL4 drivers express the transcriptional regulator GAL4 with a constitutive or tissue-specific pattern depending on the endogenous enhancer that is located in its proximity. There is a large collection of GAL4 lines available from public stock centers. UAS-IR lines are generated by P-element-mediated transformation. A construct containing in tandem several GAL4 binding sites upstream of the inverted repeat of the target gene is inserted randomly in the genome. Leaky expression is generally very low, showing no phenotype. After crossing the GAL4 drivers with the UAS-IR stocks, the F1 generation expresses the dsRNA directed by GAL4 and induces the RNAi. In this way, a knock-down of the gene of interest is obtained in the animal. The scheme is based on that published by Brand and Perrimon (9).

3.2.1. Construction of the RNAi Vector for Flies

Construct the vector as described in Subheading 3.1.1., substituting the pMt/Hy plasmid with the appropriately digested pUAST vector.

3.2.2. Generation of UAS-IR Lines

Transgenesis in Drosophila is based on the use of P-element transformation. There are several standard protocols to obtain transgenic animals that are based on the original method described by Spradling and Rubin (11). Current protocols are similar, with efficiencies ranging from 3 to 12%. We routinely use the protocol of Déjardin and Cavalli, which is available online with a detailed description (http://www.igh.cnrs.fr/equip/cavalli/link.labgoodies.html). It is highly efficient, yielding approx 10 independent transgenic animals from 100 injected embryos. Here, we offer detailed explanatory notes regarding important considerations that we have encountered in our laboratory experience (see Note 4).

3.2.3. Setting the UAS-IR × GAL4 Cross and RNAi Analysis

  1. Collect virgin females from the GAL4 driver stock. Depending on the number of crosses planned, start collecting virgins at the beginning of the week and set the crosses at the end of the week. At the same time, collect newly hatched UAS-IR males and plan to use both parental lines as controls.

  2. Set the RNAi crosses using the appropriate number of flies according to the size of the food vials selected. Place no less than 10 males and 20 females in a small vial. For a big vial, use 30 males and 50 females and scale it up to 150 flies if needed. When using bottles, use at least 60 males and 100 females.

  3. Pass the crossed RNAi lines daily after allowing flies to remain in the same vial for a couple of days. Keep the incubation temperature constant.

  4. Isolate total RNA and protein from embryos, larvae, or adults (13). Place the crosses in suitable cages to collect materials at regular time intervals. If the number of flies is small, then collect several egg lays and store them at 4°C to gather enough material. Collect larvae and proceed with fresh material if possible for best results, although they may be kept frozen at −70°C at the desired stage. Ten larvae from each cross generally provide sufficient amounts of RNA and protein for analysis. Anesthetize and freeze adults at −70°C until use.

  5. Check the RNAi-induced suppression level of the target protein by immunoblot analysis or the target RNA by quantitative reverse transcriptase PCR (14).

3.2.4. Phenotypic Analysis (see Notes 5–8)

  1. Pass the crossed RNAi lines daily for 1–2 wk.

  2. Inspect the vials daily and note relevant events and dates on which they occur, such as the beginning of the first larval instar, third larval instar wandering, prepuparium formation, pupariation, and eclosion.

  3. Perform a detailed progeny analysis on selected vials. Count the number of flies (males and females) daily. If the progeny are not expected to have the same genotype, then also count the number of flies of each possible genotype daily.

  4. Determine the rate of pupal lethality, life-span, and reproductive capacity in terms of egg laying or perform behavior assays as appropriate.

Acknowledgments

The work in our laboratories was supported by National Institutes of Health grant GM45295 to L. S. K. and Ministerio de Ciencia y Tecnología, Spain (grant BFU2004–04591) and Instituto de Salud Carlos III, Redes de centros RCMN (C03/08), and Temáticas (G03/011) to R. G.

Footnotes

1

The length of the stem region of a hairpin type of dsRNA is usually longer than 300 bp. Such constructs have been shown to cause more than 90% reduction of expression of the target gene. However, a shorter dsRNA expression vector with less than a 100-bp stem region is also effective (12).

2

Because of leaky expression from metallothionein promoter, the RNAi construct may repress the target transcript under uninduced conditions (see Fig. 2).

3

The RNAi cell lines are stable for at least 3 mo. Check the target protein level by immunoblot monthly. If sufficient suppression is not apparent, then it is better to establish a new cell line.

4

Although dsRNA can be obtained by several different strategies, in general two identical fragments of 0.5–1 kb that have sequence identity to the gene to be knocked down are cloned into the pUAST vector in a head-to-head or tail-to-tail orientation. By traditional P-element transformation, a transgenic fly that carries the inverted repeats under UAS control is obtained. Normally, the transgene will remain silent in the absence of GAL4, and only leaky expression of dsRNA will occur.

5

Always work with several UAS-IR independent transgenic lines. Even though dsRNA production will be driven by GAL4, sites of insertion will still have an effect on transgene expression, and some insertions may be lethal because of positional effects.

6

Depending on the driver, RNAi can be set off early in development or not and may be constitutive or tissue specific. Therefore, the use of reporter genes such as green fluorescent protein (GFP) or lacZ (using UAS-GFP or UAS-lacZ lines currently available) is recommended to obtain the most accurate spatiotemporal expression pattern given by each GAL4 driver.

7

When mating UAS-IR and GAL4 parental lines, do not underestimate the number of flies needed. Crosses should have at least 10–20 individuals. Setting small crosses can result in lethality not related to the RNAi phenotype. Remember to set the appropriate controls so that, by comparison, all F1 progeny represent the same gene dosage. Mate UAS-IR virgins to GAL4 driver males and the reverse. When possible, keep parental stocks balanced and maintain each homozygous for IR or GAL4 insertion. When homozygous stocks are used, 100% of the progeny will express dsRNA and show the RNAi phenotype.

8

The function of GAL4 is temperature dependent. Flies can be maintained at a temperature ranging from 16 to 29°C. At low temperature, GAL4 shows little ability to activate transcription; thus, RNAi will have a milder effect. At high temperature, there is increasing activation of transcription by GAL4 that will lead to a more powerful effect of RNAi. Apart from trying different GAL4 drivers, inducing RNAi at different temperatures should also be considered.

References

  • 1.Jorgensen RA. Sense cosuppression in plants: past, present and future. In: Hannon GJ, editor. RNAi : A Guide to Gene Silencing. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 2003. pp. 5–22. [Google Scholar]
  • 2.Cogoni C, Irelan JT, Schumacher M, Schmidhauser TJ, Selker EU, Macino G. Transgene silencing of the al-1 gene in vegetative cells of Neurospora is mediated by a cytoplasmic effector and does not depend on DNA–DNA interactions or DNA methylation. EMBO J. 1996;15:3153–3163. [PMC free article] [PubMed] [Google Scholar]
  • 3.Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature. 1998;391:806–811. doi: 10.1038/35888. [DOI] [PubMed] [Google Scholar]
  • 4.Cogoni C, Macino G. Post-transcriptional gene silencing across kingdoms. Curr. Opin. Genet. Dev. 2000;10:638–643. doi: 10.1016/s0959-437x(00)00134-9. [DOI] [PubMed] [Google Scholar]
  • 5.Tomari Y, Zamore PD. Perspective: machines for RNAi. Genes Dev. 2005;19:517–529. doi: 10.1101/gad.1284105. [DOI] [PubMed] [Google Scholar]
  • 6.Hammond SM, Bernstein E, Beach D, Hannon GJ. An RNA-directed nuclease mediates post-transcriptional gene silencing in Drosophila cells. Nature. 2000;404:293–296. doi: 10.1038/35005107. [DOI] [PubMed] [Google Scholar]
  • 7.Kennerdell JR, Carthew RW. Use of dsRNA-mediated genetic interference to demonstrate that frizzled and frizzled 2 act in the wingless pathway. Cell. 1998;95:1017–1026. doi: 10.1016/s0092-8674(00)81725-0. [DOI] [PubMed] [Google Scholar]
  • 8.Koelle MR, Talbot WS, Segraves WA, Bender MT, Cherbas P, Hogness DS. The Drosophila EcR gene encodes an ecdysone receptor, a new member of the steroid receptor superfamily. Cell. 1991;67:59–77. doi: 10.1016/0092-8674(91)90572-g. [DOI] [PubMed] [Google Scholar]
  • 9.Brand AH, Perrimon N. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development. 1993;118:401–415. doi: 10.1242/dev.118.2.401. [DOI] [PubMed] [Google Scholar]
  • 10.Duffy JB. GAL4 system in Drosophila : a fly geneticist’s Swiss army knife. Genesis. 2002;34:1–15. doi: 10.1002/gene.10150. [DOI] [PubMed] [Google Scholar]
  • 11.Spradling AC. P element-mediated transformation. In: Roberts DB, editor. Drosophila a Practical Approach. Oxford, UK: IRL Press; 1986. pp. 175–197. [Google Scholar]
  • 12.Farr CL, Matsushima Y, Lagina AT, 3rd, Luo N, Kaguni LS. Physiological and biochemical defects in functional interactions of mitochondrial DNA polymerase and DNA-binding mutants of single-stranded DNA-binding protein. J. Biol. Chem. 2004;279:17,047–17,053. doi: 10.1074/jbc.M400283200. [DOI] [PubMed] [Google Scholar]
  • 13.Lefai E, Calleja M, Ruiz de Mena I, Lagina AT, 3rd, Kaguni LS, Garesse R. Overexpression of the catalytic subunit of DNA polymerase gamma results in depletion of mitochondrial DNA in Drosophila melanogaster. Mol. Gen. Genet. 2000;264:37–46. doi: 10.1007/s004380000301. [DOI] [PubMed] [Google Scholar]
  • 14.Matsushima Y, Adan C, Garesse R, Kaguni LS. Drosophila mitochondrial transcription factor B1 modulates mitochondrial translation but not transcription or DNA copy number in Schneider cells. J. Biol. Chem. 2005;280:16,815–16,820. doi: 10.1074/jbc.M500569200. [DOI] [PubMed] [Google Scholar]
  • 15.Matsushima Y, Garesse R, Kaguni LS. Drosophila mitochondrial transcription factor B2 regulates mitochondrial DNA copy number and transcription in Schneider cells. J. Biol. Chem. 2004;279:26,900–26,905. doi: 10.1074/jbc.M401643200. [DOI] [PubMed] [Google Scholar]

RESOURCES