- Email: [email protected]
Use of the giant multinucleate plasmodium of Physarum polycephalum to study RNA interference in the myxomycete
ANALYTICAL BIOCHEMISTRY Analytical Biochemistry 342 (2005) 194–199 www.elsevier.com/locate/yabio
Use of the giant multinucleate plasmodium of Physarum polycephalum to study RNA interference in the myxomycete Markus Haindl, Eggehard Holler ¤ Institute of Biophysics and Physical Biochemistry, University of Regensburg, D-93053 Regensburg, Germany Received 14 December 2004 Available online 12 April 2005
Abstract The plasmodium of Physarum polycephalum harbors billions of synchronized nuclei in a single cell of complex structure. Due to its synchrony and extreme size, it is used as a model to study events on a single cell level, such as cell cycle and diVerentiation. We show here for the Wrst time that this model, despite its enormous size and structural complexity, is accessible to RNA interference by simple injection of dsRNA or siRNA. The targeted gene is that of polymalatase, an intracellular adapter of poly(-L-malate) involved in the maintenance of the synchrony and functioning as an extracellular hydrolase of this polymer. Real-time reverse transcriptase polymerase chain reaction analysis revealed that the speciWc mRNA was knocked down to about 10% of the original level. The suppression of a single injection lasted for approximately 14 cell cycles (144 h) and could be prolonged for any time by repeated dsRNA injections. Western blots indicated that the knockdown of RNA was paralleled by a strong reduction in polymalatase synthesis. However, a change in the phenotype of the plasmodium could not be clearly observed. In principle, the plasmodium oVers an easy system for studying gene knockdown by RNA interference. 2005 Elsevier Inc. All rights reserved. Keywords: RNA interference; Polymalatase; Polynucleate cells; Plasmodium; Microinjection
Physarum polycephalum is one of the plasmodiumforming slime molds [1] of the class Myxomyceteae, a member of the monophylum Mycetozoa, and is more closely related to animals and fungi than to plants [2]. The plasmodium is a unique organism, representing a giant vegetative single cell in the life cycle of this slime mold. Its nuclei divide every 7–8 h in high synchrony, without dissolution of their nuclear envelope and in the absence of cytokinesis, thus giving rise to a giant multinucleate cell. The macroplasmodium (routinely in sizes of 100 cm2 or larger) grows easily in the laboratory as a Xat, disk-like cell on the surface of a solid support. *
Corresponding author. Fax: +49 941 943 2813. E-mail address: [email protected] (E. Holler). 0003-2697/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2005.03.031
Because of its large size, the plasmodium has evolved devices to allow the intracellular equilibration of cellular consitituents: (1) a complex macroscopic network of veins consisting of a rhythmically moving endoplasm surrounded by a resting ectoplasm, which displays a complicated structure of Wbrils and invaginations of the plasma membrane [3], and (2) a (putative) molecular transporter consisting of polymalate and an adapter [4 and references therein]. By adopting a structural similarity to the backbone of nucleic acids, polymalate binds nuclear proteins, and the adapter guides the polymalate– cargo complex to the various nuclei. In the extracellular matrix, secreted adapter functions as hydrolase (polymalatase) of extracellular polymalate. The plasmodium has been found suitable for studying various cellular and organismic aspects such as cell cycle,
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199
195
motility, and diVerentiation. Timely questions are the molecular basis of the synchrony [5] or the complex intracellular network recruiting commitment and development of sporulation [6,7]. The plasmodium is superior to any other known model organism by virtue of its enormous cellular size suitable for injections of materials such as nucleic acids or other materials [8], the high synchrony of cellular events [9], the fusibility of individual plasmodia allowing the transfer of molecular information [6,10], and the availability of both haploid and diploid species [10] to study gene dominance. However, attempts to knockout or overexpress genes by conventional techniques were mostly unsuccessful, rendering P. polycephalum an unsuccessful model system. To overcome this problem, the goals of our investigation were (1) to demonstrate that, in contrast to its large size and complex structure, the plasmodium was accessible to RNA interference by simple injection, (2) to assess whether repeated injections give rise to enhanced suppression eVects, (3) to measure the duration of the achieved RNAi eVect, (4) to test whether the eVect continued after replating of plasmodia, (5) to examine whether the eVect was reversible, and (6) to examine whether, in the present case of the knockdown of polymalatase, a change in phenotype was detectable.
Tris–acetate–EDTA-buVer, pH 8.0, before aliquots of 10 l were stored at ¡80 °C. For control injections, nonspeciWc dsRNA was generated by the same method using a PCR-derived fragment with 592-bp (nucleotides 142– 734) from the vector pGEM(R)-5zf(+) (Technical Services, Promega, USA). To synthesize siRNA, a motif based on the polymalatase gene (Accession No. AJ543320) was designed according to the outlines proposed by Elbashir et al. [12] employing the advanced RNAi software (oligoengine internet homepage). The templates 5⬘-TCG AGT AAG TAC TAG AGC TCC TAT AGT GAG TCG TAT TAG T-3⬘ (containing the sense segment) and 5⬘-AAG GAG CTC TAG TAC TTA CTC TAT AGT GAG TCG TAT TAG T-3⬘ (containing the antisense segment) were annealed with the T7-promotor segment 5⬘-ACT AAT ACG ACT CAC TAT AG-3⬘ for RNA synthesis in the presence of T7-RNA polymerase according to the method of Donze and Picard [11]. The resulting 19-bp RNA duplex contained 2 nt overhangs essential for the function of siRNA containing UU in the sense strand [13]. The template for transcription of nonspeciWc control siRNA had the sequences 5⬘-AAG GCG GTA ATA CGG TTA TCC-3⬘ as sense and 5⬘-GTG GAT AAC CGT ATT ACC GCC-3⬘ as antisense strands. All DNA synthetic work was carried out by MWG, Germany.
Materials and methods
Macroplasmodia and injection of dsRNA
dsRNA and siRNA
Microplasmodia of P. polycephalum strain M3CVII ATCC 204388 (American Type Culture Collection) were cultured by the method of Daniel and Baldwin [14] for 48 h at 24 °C. To start a macroplasmodium, 100 mg of wet microplasmodia were allowed to fuse on 2% agar (13.5 cm petri dish) containing the growth medium and cultured 22 h at 24 °C in the dark. The macroplasmodium (5 § 0.5 cm in diameter) was injected into a prominent vein with 10 l of a solution [8] containing 1 g of dsRNA, siRNA, or control RNAs and allowed to grow for 24 h before the plasmodium was scraped from the agar surface into liquid nitrogen and stored at ¡80 °C for analysis. In long-term experiments, macroplasmodia were grown and injected the same way, but 24 h after injection a 4-cm2 section of the macroplasmodium together with the agar was cut out and placed as an inoculum upside down on a fresh agar plate (Wrst generation). The remainder of the plasmodium was scraped from the agar plate and analyzed. The inoculum was allowed to grow for 60 h, receiving fresh injections of dsRNA after 24 and 48 h. Then a section was transferred for replating, the remainder of the plasmodium was prepared for analysis (after 60 h from replating) as above, and so on until three generations had been injected and prepared for analysis. In a second series, the plasmodium was injected only in the parental plasmodium, and the following generations
The DNA template to be transcribed into dsRNA was synthesized by PCR using the cDNA derived from P. polycephalum M3CVII plasmodia. Desalted primers speciWc for the polymalatase–cDNA (Accession No. AJ543320) with the sequences 5⬘-GTG TAA TAC GAC TCA CTA TAG GGA AAA GGA GGT TCT GAT CCT AGT-3⬘ (forward primer) and 5⬘-CAC TAA TAC GAC TCA CTA TAG GGA TCA CGA TGT CAT CAG CAA AAC-3⬘ (reverse primer), both containing the T7-polymerase promoter at their 5⬘ termini, were custom-made by MWG-Biotech, Germany. The resulting 589-bp DNA spanned the nucleotides 1083–1671 of the gene downstream of the origin of transcription and was used as template for in vitro dsRNA synthesis as described by Donze and Picard [11]. The 50-l reaction mixture was incubated for 15 h and then treated for 30 min with RNAse-free DNase (Qiagen, Germany) at 37 °C. The mixture was Wnally heated to 95 °C for 5 min and cooled to room temperature over a period of 7 h. Double-stranded RNA was precipitated with 2.5 volumes of ethanol/0.2 M sodium acetate, pH 4.9, pelleted at 20,000g for 20 min at 4 °C, washed with 70% ethanol, air-dried, and resuspended in 60 l of diethylpyrocarbonate-treated RNase-free water. The integrity of the dsRNA was validated by agarose gel electrophoresis in
196
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199
were analyzed without repeating injections. Control plasmodia were incubated in parallel and treated exactly as above. One received 10-l injections of control dsRNA, and the other did not receive injections. The method of injection described by Willibald et al. [8] employed a manually driven microinjector from Leitz (Wetzlar, Germany), a simple instrument consisting of a microsyringe and a microdosimeter, and the Binocular M5 from Wild (Heerbrugg, Switzerland).
Portions of 10 g protein were analyzed by Western blotting on a 10% SDS polyacrylamide gel/Immobilon-P PVDF membrane (Millipore, Germany) with the antibodies against polymalatase and actin. Immuno-reactive bands were visualized by chemiluminescence employing peroxidase-coupled second antibody and the reagents of NOWA (Mobitec, Germany).
Results Knockdown analysis by real-time RT-PCR RNA was extracted from plasmodia using the RNeasy-kit (Qiagen, Germany). One microgram RNA per sample was copied to cDNA using RevertAid H Minus M-MuLV Reverse Transcriptase (Fermentas, Germany) and oligo(dT)18 primer following the standard protocol of the provider. Realtime RT-PCR was performed using the Roche-LightCycler (Roche, Germany) with the Quantitect SYBR-Green PCR Kit (Qiagen) including HotStar-Taq and SYBR Green as Xuorescent dye for real-time detection of PCR products. For the analysis by real-time RT-PCR, the primer pairs were used as given below. Cycling conditions were 95 °C for 15 min, followed by 40 cycles of 95 °C for 15 s, 58 °C for 20 s, and 72 °C for 20 s. For quantiWcation, the target gene was normalized to the internal standard gene actin of P. polycephalum employing the untreated and the mock-injected samples as controls. As default settings of the Lightcycler Software Version 3.5.3 were used, measurement of probes in the linear range of the PCR was ensured. Desalted primer pairs for actin Ppa35 (Accession No. M21500) and Polymalatase (Accession No. AJ543320) were synthesized by MWG-Biotech, Germany: (i) actin forward primer, 5⬘-CAT GTG CAA GGC TGG ATT TGC TG-3⬘, (ii) actin reverse primer, 5⬘-ACC GAC GTA TGA GTC CTT TTG-3⬘, (iii) polymalatase forward primer, 5⬘-CAA AGG GAT TAT GAG ACA GCA G-3⬘, and (iv) polymalatase reverse primer, 5⬘-ACT GTG CCA TCC GCC TTC-3⬘.
Double-stranded RNA and siRNA induced speciWc decreases in gene expression at the levels of both mRNA (Fig. 1A) and protein synthesis (Fig. 1B). However, the control siRNA and control dsRNA containing unspeciWc sequences were without eVects (Figs. 1A and B). dsRNA suppressed 89% of the amount of polymalatasespeciWc mRNA seen for the uninjected control plasmodium. The eVect of siRNA was an 82% reduction. Injection of dsRNA in the range 0.2–1.0g resulted in a constant level of knockdown eYciency (data not shown). Addition of speciWc dsRNA (0.4 g/ml) to the culture medium of
Analysis by Western blotting Western blot analysis of polymalatase was performed using a previously described preparation of a speciWc antibody [15]. The actin-speciWc antibody was purchased from Sigma (Germany). Portions of the same samples analyzed by real time RT-PCR were homogenized in icecold buVer containing 20 mM Na–morpholinosulfonate, pH 7, 0.5 mM CaCl2, 15 mM MgCl2, 5 mM EGTA, 0.5 M hexyleneglycol, 19%(v/v) dextrane, 14 mM 2-mercaptoethanol, and protease inhibitor cocktail (Boehringer–Mannheim Germany). After pelleting at 600g, the supernatant was centrifuged 10 min at 10,000g and represented the cytoplasmic fraction.
Fig. 1. Suppression of mRNA and protein synthesis after the injection of polymalatase (pma)-speciWc dsRNA or siRNA into macroplasmodia. Plasmodia were injected 22 h after inoculation and then grown for 24 h before analysis. (A) mRNA quantiWcation by real-time RT-PCR with reference to actin mRNA. Messenger RNA of the noninjected control plasmodia was set to 100%. Black column, noninjected control plasmodia; dark gray columns, control (nonspeciWc) siRNA and speciWc siRNA; light gray columns, control (nonspeciWc) dsRNA and speciWc dsRNA. The mRNA suppression was highly signiWcant (p > 0.001). Standard errors of three independent measurements are indicated. (B) Western blot of cytoplasmic samples of the plasmodia in (A). The band at 68 kDa refers to polymalatase and shows a decrease in intensity. The band at 42 kDa reXects actin as control.
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199
microplasmodia did not eVect suppression of mRNA synthesis, indicating that extrinsic RNA could not penetrate into the cytoplasm (data not shown). The eVect on protein expression was measured by Western blotting of samples from the cytoplasm. Actin was used as reference in this case. An inspection of Fig. 1B indicates a suppression for the 68-kDa polymalatase band in parallel with the suppression of mRNA. Long-term experiments were carried out to (1) estimate the time of persistence of the RNAi eVect and (2) to enhance the knockdown level by repeated injections. After the Wrst injection and growth, 1/20 (by mass) of the macroplasmodium was replated. The new plasmodium was injected and replated again, and so on. As shown in Fig. 2, suppression of polymalatase-speciWc mRNA persisted for approximately 144 h after a single dsRNA injection, corresponding to the duration of 18 cell cycles and spanning two successive inoculations to fresh agar plates. In the second “generation” the expression of polymalatase-speciWc mRNA and protein already began to recover. However, multiple injections did not enhance the level of knockdown. The level of protein suppression did not decrease either, consistent with the absence of a long-term response such as developing a resistance against injected dsRNA. Plasmodia of the Wrst to third generations have been inspected for changes in the phenotype. No changes in
197
the overall appearance and growth rate were seen except that the latter appeared to be insigniWcantly retarded (p > 0.2).
Discussion Of Myxomycetes, such as P. polycephalum, the plasmodium represents a cell with an extremely large dimension and a high structural complexity. Nevertheless, injection of speciWc dsRNA into the veins proved to induce an approximately 90% suppression of the polymalatase gene. The knockdown was speciWc as control dsRNA and siRNA were totally ineVective. Because the knockdown was not complete, this result suggested at least two explanations. (1) The artiWcially introduced RNA spread all over the plasmodium except in a few compartments, which were not accessible and thus retained their normal mRNA synthesis. However, this possibility seemed unlikely in face of the results of repeated injections and maintenance of the very same eVect for more than 200 h. (2) In a survey of more than 2850 cases of various siRNA “gene silencers” it has been reported that suppression was frequently incomplete [16]. This may suggest another possibility, namely that rates of mRNA synthesis and the RNAi-induced cleavage were of similar magnitudes, resulting in a steady
Fig. 2. Suppression of mRNA and protein synthesis after the injection of polymalatase (pma)-speciWc dsRNA into macroplasmodia. Plasmodia were injected with dsRNA 22 h after inoculation and then grown for 24 h before analysis. The Wrst and following generations were prepared by inoculation with a small amount of each precursor plasmodium on a fresh agar plate and grown for 60 h as described under Materials and methods. Black columns, noninjected control plasmodia; dark gray columns, water-injected control plasmodia; light gray columns, serial experiment with only the parent plasmodium injected; white columns, serial experiment with injections of the parent plasmodium and of the Wrst to third generations each at 24 and 48 h after replating. (A) mRNA quantifcation by real-time RT-PCR with reference to actin mRNA. Messenger RNA of the noninjected control plasmodia was set to 100%. Standard errors are indicated for three independent measurements. (B) Western blot analysis of the samples in (A). The band at 68 kDa refers to polymalatase and shows a decrease in intensity correlating with the suppression of polymalate-speciWc mRNA in (A). The band at 42 kDa reXects actin as control.
198
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199
state survival of mRNA. A variety of reasons for an incomplete break down has been reported, such as absence of optimal base-pairing between siRNA and target mRNA [17], limited accessibility of target mRNA due to stabile secondary structure [17–19], protective binding of proteins, and speciWc subcellular location of mRNA [19]. Low-abundance transcripts were less susceptible to siRNA-mediated degradation than mediumand high-abundance transcripts [19]. Long-term experiments revealed that a single injection was quite eYcient, the knockdown lasting through 144 h into the second generation (Fig. 2A). For comparison, the injected 1 g dsRNA (1.5 £ 1011 molecules) corresponded to 3000 molecules dsRNA per nucleus [5 £ 108 nuclei (g plasmodia)¡1] immediately after the injection and would have been 7.5 molecules of dsRNA at the time of analysis in the second generation. The sustentation of the knockdown eVect is in agreement with the proposed mechanism of RNAi silencing [20,21] and indicated a high eYciency. Repeated injection of dsRNA did not induce further decline in mRNA, and the mRNA returned to its original level after recovery during the third generation following single injection. Apparently, the regulation of mRNA production was not aVected by ongoing RNA interference. The phenotype of the plasmodium of the injected parent plasmodium through the third generation was found to resemble that of the control plasmodium, except for a slight, however, insigniWcant (p > 0.2) growth retardation. This was referred to the residual synthesis of polymalatase revealed by Western blotting. The persisting low level of polymalatase was apparently suYcient for the maintenance of the physiological status of the plasmodium. We have recently shown that an increase in polymalate content is related to an increase in growth rate of the plasmodium [22]. Similarly, we had expected a decrease in growth rate if the polymalatase (adapter)–polymalate complex were the functional unit and if the level of polymalatase was reduced by RNAi interference. A knockout of the protein would have paralyzed polymalate function as a molecular carrier of nuclear proteins. Protein concentrations in diVerent regions of the plasmodium would have fallen out of phase, resulting in loss of synchrony and consequently loss of the normal phenotype of the plasmodium. Although we could occasionally observe that growth behavior was slightly diVerent in dsRNA-treated cells and indeed some regions seemed to have fallen out of synchrony indicated by retarded growth, the observation was not statistically signiWcant. A complete knockout of the gene expression seems to be required to registrate a clear change in the phenotype. The method that we have introduced here for plasmodia is very simple and eVective. Because of the high synchronization of plasmodial events and the precision in timing of an injection, the method oVers the possibil-
ity to knockdown the temporary expression of genes in a time-resolved fashion. Moreover, the injection can be precisely placed at particular coordinates of the giant cell body (plasmodium) to study gene suppression over intracellular distances. The RNAi technique in conjunction with the unique properties of the plasmodium of P. polycephalum should allow with relatively little experimental expenditure study of versatile cellular phenomena that have hitherto not been accessible.
References [1] T.G. Burland, K.L. Solnica, J. Bailey, D.B. Cunningham, W.F. Dove, Patterns of inheritance, development and the mitotic cycle in the protist Physarum polycephalum, Adv. Microb. Physiol. 35 (1993) 1–69. [2] S.L. Baldauf, A.J. Roger, I. Wenk-Siefert, W.F. Doolittle, A kingdom-level phylogeny of eukaryotes based on combined protein data, Science 290 (2004) 972–977. [3] D. Kessler, Plasmodial structure and motility, in: H.C. Aldrich, J.W. Daniel (Eds.), Cell Biology of Physarum and Didymium, vol. 1, Academic Press, Paris, 1982, pp. 145–208. [4] B.-S. Lee, M. Vert, E. Holler, Water-soluble aliphatic polyesters: Poly(malic acid)s, in: Y. Doi, A. Steinbüchel (Eds.), Biopolymers, Polyesters, vol. 3a, Wiley-VCH, Weinheim, 2002, pp. 75–103. [5] S. Doerhoefer, C. Windisch, B. Angerer, O.I. Lavrik, B.-S. Lee, E. Holler, The DNA-polymerase inhibiting activity of poly(-L-malic acid) in nuclear extract during the cell cycle of Physarum polycephalum, Eur. J. Biochem. 269 (2002) 1253–1258. [6] W. Marwan, Detecting functional interactions in a gene and signaling network by time-resolved somatic complementation analysis, BioEssays 25 (2003) 950–960. [7] G. Golderer, E.R. Werner, S. Leitner, P. Gröbner, G. WernerFelmayer, Nitric oxide synthase is induced in sporulation of Physarum polycephalum, Genes Dev. 15 (2001) 1299–1309. [8] B. Willibald, W. Bildl, B.-S. Lee, E. Holler, Is -poly(L-malate) synthesis catalysed by a combination of -L-malyl-AMP-ligase and poly(L-malate) polymerase? Eur. J. Biochem. 269 (1999) 1085– 1090. [9] J.J. Tyson, Periodic phenomena in Physarum, in: H.C. Aldrich, J.W. Daniel (Eds.), Cell Biology of Physarum and Didymium, vol. 1, Academic Press, New York, 1982, pp. 61–110. [10] J. Dee, Genetics of Physarum polycephalum, in: H.C. Aldrich, J.W. Daniel (Eds.), Cell Biology of Physarum and Didymium, vol. 1, Academic Press, New York, 1982, pp. 212–251. [11] O. Donze, D. Picard, RNA interference in mammalian cells using siRNAs synthesized with T7 RNA polymerase, Nucleic Acids Res. 30 (e46) (2002) 1–4. [12] S.M. Elbashir, J.P.A. Martinez, W. Lendeckel, T. Tuschl, Functional anatomy of siRNAs for mediating eYcient RNAi in Drosophila melanogaster embryo lysate, EMBO J. 20 (2001) 6877– 6888. [13] S.M. Elbashir, J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, T. Tuschl, Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells, Nature 411 (2001) 494–498. [14] F.W. Daniel, H.H. Baldwin, Methods of culture for plasmodial myxomycetes, Methods Cell Physiol. 1 (1964) 9–14. [15] M. Karl, E. Holler, Multiple polypeptides immunologically related to -poly(L-malate) hydrolase (polymalatase) in the plasmodium of the slime mold Physarum polycephalum, Eur. J. Biochem. 251 (1998) 405–412. [16] siRNA-Catalog, Santa Cruz Biotechnology, Inc., Santa Cruz, 2004.
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199 [17] K. Yoshinari, M. Miyagishi, K. Taira, EVects on RNAi of the tight structure, sequence and position of the targeted region, Nucleic Acids Res. 32 (2004) 691–699. [18] T. Holen, M. Amarzguioui, M.T. Wiiger, E. Babaie, H. Prydz, Positional eVects of short interfering RNAs targeting the human coagulation trigger tissue factor, Nucleic Acids Res. 30 (2002) 1757–1766. [19] X. Hu, S. Hipolito, R. Lynn, V. Abraham, S. Ramos, F. WongStaal, Relative gene-silencing eYciencies of small interfering RNAs targeting sense and antisense transcripts from the same genetic locus, Nucleic Acids Res. 32 (2004) 4609–4617.
199
[20] C. Arenz, U. Schepers, RNA interference: from an ancient mechanism to a state of the art therapeutic application?, Naturwissenschaften 90 (2003) 345–359. [21] T. Sijen, J. Fleenor, F. Simmer, K.L. Thijssen, S. Parrish, L. Timmons, R.H. Plasterk, A. Fire, On the role of RNA ampliWcation in dsRNA-triggered gene silencing, Cell 107 (2001) 465–476. [22] M. Karl, R. Anderson, E. Holler, Injection of poly(-L-malate) into the plasmodium of Physarum polycephalum shortens the cell cycle and increases growth rate, Eur. J. Biochem. 271 (2004) 3805– 3811.
Use of the giant multinucleate plasmodium of Physarum polycephalum to study RNA interference in the myxomycete Markus Haindl, Eggehard Holler ¤ Institute of Biophysics and Physical Biochemistry, University of Regensburg, D-93053 Regensburg, Germany Received 14 December 2004 Available online 12 April 2005
Abstract The plasmodium of Physarum polycephalum harbors billions of synchronized nuclei in a single cell of complex structure. Due to its synchrony and extreme size, it is used as a model to study events on a single cell level, such as cell cycle and diVerentiation. We show here for the Wrst time that this model, despite its enormous size and structural complexity, is accessible to RNA interference by simple injection of dsRNA or siRNA. The targeted gene is that of polymalatase, an intracellular adapter of poly(-L-malate) involved in the maintenance of the synchrony and functioning as an extracellular hydrolase of this polymer. Real-time reverse transcriptase polymerase chain reaction analysis revealed that the speciWc mRNA was knocked down to about 10% of the original level. The suppression of a single injection lasted for approximately 14 cell cycles (144 h) and could be prolonged for any time by repeated dsRNA injections. Western blots indicated that the knockdown of RNA was paralleled by a strong reduction in polymalatase synthesis. However, a change in the phenotype of the plasmodium could not be clearly observed. In principle, the plasmodium oVers an easy system for studying gene knockdown by RNA interference. 2005 Elsevier Inc. All rights reserved. Keywords: RNA interference; Polymalatase; Polynucleate cells; Plasmodium; Microinjection
Physarum polycephalum is one of the plasmodiumforming slime molds [1] of the class Myxomyceteae, a member of the monophylum Mycetozoa, and is more closely related to animals and fungi than to plants [2]. The plasmodium is a unique organism, representing a giant vegetative single cell in the life cycle of this slime mold. Its nuclei divide every 7–8 h in high synchrony, without dissolution of their nuclear envelope and in the absence of cytokinesis, thus giving rise to a giant multinucleate cell. The macroplasmodium (routinely in sizes of 100 cm2 or larger) grows easily in the laboratory as a Xat, disk-like cell on the surface of a solid support. *
Corresponding author. Fax: +49 941 943 2813. E-mail address: [email protected] (E. Holler). 0003-2697/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2005.03.031
Because of its large size, the plasmodium has evolved devices to allow the intracellular equilibration of cellular consitituents: (1) a complex macroscopic network of veins consisting of a rhythmically moving endoplasm surrounded by a resting ectoplasm, which displays a complicated structure of Wbrils and invaginations of the plasma membrane [3], and (2) a (putative) molecular transporter consisting of polymalate and an adapter [4 and references therein]. By adopting a structural similarity to the backbone of nucleic acids, polymalate binds nuclear proteins, and the adapter guides the polymalate– cargo complex to the various nuclei. In the extracellular matrix, secreted adapter functions as hydrolase (polymalatase) of extracellular polymalate. The plasmodium has been found suitable for studying various cellular and organismic aspects such as cell cycle,
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199
195
motility, and diVerentiation. Timely questions are the molecular basis of the synchrony [5] or the complex intracellular network recruiting commitment and development of sporulation [6,7]. The plasmodium is superior to any other known model organism by virtue of its enormous cellular size suitable for injections of materials such as nucleic acids or other materials [8], the high synchrony of cellular events [9], the fusibility of individual plasmodia allowing the transfer of molecular information [6,10], and the availability of both haploid and diploid species [10] to study gene dominance. However, attempts to knockout or overexpress genes by conventional techniques were mostly unsuccessful, rendering P. polycephalum an unsuccessful model system. To overcome this problem, the goals of our investigation were (1) to demonstrate that, in contrast to its large size and complex structure, the plasmodium was accessible to RNA interference by simple injection, (2) to assess whether repeated injections give rise to enhanced suppression eVects, (3) to measure the duration of the achieved RNAi eVect, (4) to test whether the eVect continued after replating of plasmodia, (5) to examine whether the eVect was reversible, and (6) to examine whether, in the present case of the knockdown of polymalatase, a change in phenotype was detectable.
Tris–acetate–EDTA-buVer, pH 8.0, before aliquots of 10 l were stored at ¡80 °C. For control injections, nonspeciWc dsRNA was generated by the same method using a PCR-derived fragment with 592-bp (nucleotides 142– 734) from the vector pGEM(R)-5zf(+) (Technical Services, Promega, USA). To synthesize siRNA, a motif based on the polymalatase gene (Accession No. AJ543320) was designed according to the outlines proposed by Elbashir et al. [12] employing the advanced RNAi software (oligoengine internet homepage). The templates 5⬘-TCG AGT AAG TAC TAG AGC TCC TAT AGT GAG TCG TAT TAG T-3⬘ (containing the sense segment) and 5⬘-AAG GAG CTC TAG TAC TTA CTC TAT AGT GAG TCG TAT TAG T-3⬘ (containing the antisense segment) were annealed with the T7-promotor segment 5⬘-ACT AAT ACG ACT CAC TAT AG-3⬘ for RNA synthesis in the presence of T7-RNA polymerase according to the method of Donze and Picard [11]. The resulting 19-bp RNA duplex contained 2 nt overhangs essential for the function of siRNA containing UU in the sense strand [13]. The template for transcription of nonspeciWc control siRNA had the sequences 5⬘-AAG GCG GTA ATA CGG TTA TCC-3⬘ as sense and 5⬘-GTG GAT AAC CGT ATT ACC GCC-3⬘ as antisense strands. All DNA synthetic work was carried out by MWG, Germany.
Materials and methods
Macroplasmodia and injection of dsRNA
dsRNA and siRNA
Microplasmodia of P. polycephalum strain M3CVII ATCC 204388 (American Type Culture Collection) were cultured by the method of Daniel and Baldwin [14] for 48 h at 24 °C. To start a macroplasmodium, 100 mg of wet microplasmodia were allowed to fuse on 2% agar (13.5 cm petri dish) containing the growth medium and cultured 22 h at 24 °C in the dark. The macroplasmodium (5 § 0.5 cm in diameter) was injected into a prominent vein with 10 l of a solution [8] containing 1 g of dsRNA, siRNA, or control RNAs and allowed to grow for 24 h before the plasmodium was scraped from the agar surface into liquid nitrogen and stored at ¡80 °C for analysis. In long-term experiments, macroplasmodia were grown and injected the same way, but 24 h after injection a 4-cm2 section of the macroplasmodium together with the agar was cut out and placed as an inoculum upside down on a fresh agar plate (Wrst generation). The remainder of the plasmodium was scraped from the agar plate and analyzed. The inoculum was allowed to grow for 60 h, receiving fresh injections of dsRNA after 24 and 48 h. Then a section was transferred for replating, the remainder of the plasmodium was prepared for analysis (after 60 h from replating) as above, and so on until three generations had been injected and prepared for analysis. In a second series, the plasmodium was injected only in the parental plasmodium, and the following generations
The DNA template to be transcribed into dsRNA was synthesized by PCR using the cDNA derived from P. polycephalum M3CVII plasmodia. Desalted primers speciWc for the polymalatase–cDNA (Accession No. AJ543320) with the sequences 5⬘-GTG TAA TAC GAC TCA CTA TAG GGA AAA GGA GGT TCT GAT CCT AGT-3⬘ (forward primer) and 5⬘-CAC TAA TAC GAC TCA CTA TAG GGA TCA CGA TGT CAT CAG CAA AAC-3⬘ (reverse primer), both containing the T7-polymerase promoter at their 5⬘ termini, were custom-made by MWG-Biotech, Germany. The resulting 589-bp DNA spanned the nucleotides 1083–1671 of the gene downstream of the origin of transcription and was used as template for in vitro dsRNA synthesis as described by Donze and Picard [11]. The 50-l reaction mixture was incubated for 15 h and then treated for 30 min with RNAse-free DNase (Qiagen, Germany) at 37 °C. The mixture was Wnally heated to 95 °C for 5 min and cooled to room temperature over a period of 7 h. Double-stranded RNA was precipitated with 2.5 volumes of ethanol/0.2 M sodium acetate, pH 4.9, pelleted at 20,000g for 20 min at 4 °C, washed with 70% ethanol, air-dried, and resuspended in 60 l of diethylpyrocarbonate-treated RNase-free water. The integrity of the dsRNA was validated by agarose gel electrophoresis in
196
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199
were analyzed without repeating injections. Control plasmodia were incubated in parallel and treated exactly as above. One received 10-l injections of control dsRNA, and the other did not receive injections. The method of injection described by Willibald et al. [8] employed a manually driven microinjector from Leitz (Wetzlar, Germany), a simple instrument consisting of a microsyringe and a microdosimeter, and the Binocular M5 from Wild (Heerbrugg, Switzerland).
Portions of 10 g protein were analyzed by Western blotting on a 10% SDS polyacrylamide gel/Immobilon-P PVDF membrane (Millipore, Germany) with the antibodies against polymalatase and actin. Immuno-reactive bands were visualized by chemiluminescence employing peroxidase-coupled second antibody and the reagents of NOWA (Mobitec, Germany).
Results Knockdown analysis by real-time RT-PCR RNA was extracted from plasmodia using the RNeasy-kit (Qiagen, Germany). One microgram RNA per sample was copied to cDNA using RevertAid H Minus M-MuLV Reverse Transcriptase (Fermentas, Germany) and oligo(dT)18 primer following the standard protocol of the provider. Realtime RT-PCR was performed using the Roche-LightCycler (Roche, Germany) with the Quantitect SYBR-Green PCR Kit (Qiagen) including HotStar-Taq and SYBR Green as Xuorescent dye for real-time detection of PCR products. For the analysis by real-time RT-PCR, the primer pairs were used as given below. Cycling conditions were 95 °C for 15 min, followed by 40 cycles of 95 °C for 15 s, 58 °C for 20 s, and 72 °C for 20 s. For quantiWcation, the target gene was normalized to the internal standard gene actin of P. polycephalum employing the untreated and the mock-injected samples as controls. As default settings of the Lightcycler Software Version 3.5.3 were used, measurement of probes in the linear range of the PCR was ensured. Desalted primer pairs for actin Ppa35 (Accession No. M21500) and Polymalatase (Accession No. AJ543320) were synthesized by MWG-Biotech, Germany: (i) actin forward primer, 5⬘-CAT GTG CAA GGC TGG ATT TGC TG-3⬘, (ii) actin reverse primer, 5⬘-ACC GAC GTA TGA GTC CTT TTG-3⬘, (iii) polymalatase forward primer, 5⬘-CAA AGG GAT TAT GAG ACA GCA G-3⬘, and (iv) polymalatase reverse primer, 5⬘-ACT GTG CCA TCC GCC TTC-3⬘.
Double-stranded RNA and siRNA induced speciWc decreases in gene expression at the levels of both mRNA (Fig. 1A) and protein synthesis (Fig. 1B). However, the control siRNA and control dsRNA containing unspeciWc sequences were without eVects (Figs. 1A and B). dsRNA suppressed 89% of the amount of polymalatasespeciWc mRNA seen for the uninjected control plasmodium. The eVect of siRNA was an 82% reduction. Injection of dsRNA in the range 0.2–1.0g resulted in a constant level of knockdown eYciency (data not shown). Addition of speciWc dsRNA (0.4 g/ml) to the culture medium of
Analysis by Western blotting Western blot analysis of polymalatase was performed using a previously described preparation of a speciWc antibody [15]. The actin-speciWc antibody was purchased from Sigma (Germany). Portions of the same samples analyzed by real time RT-PCR were homogenized in icecold buVer containing 20 mM Na–morpholinosulfonate, pH 7, 0.5 mM CaCl2, 15 mM MgCl2, 5 mM EGTA, 0.5 M hexyleneglycol, 19%(v/v) dextrane, 14 mM 2-mercaptoethanol, and protease inhibitor cocktail (Boehringer–Mannheim Germany). After pelleting at 600g, the supernatant was centrifuged 10 min at 10,000g and represented the cytoplasmic fraction.
Fig. 1. Suppression of mRNA and protein synthesis after the injection of polymalatase (pma)-speciWc dsRNA or siRNA into macroplasmodia. Plasmodia were injected 22 h after inoculation and then grown for 24 h before analysis. (A) mRNA quantiWcation by real-time RT-PCR with reference to actin mRNA. Messenger RNA of the noninjected control plasmodia was set to 100%. Black column, noninjected control plasmodia; dark gray columns, control (nonspeciWc) siRNA and speciWc siRNA; light gray columns, control (nonspeciWc) dsRNA and speciWc dsRNA. The mRNA suppression was highly signiWcant (p > 0.001). Standard errors of three independent measurements are indicated. (B) Western blot of cytoplasmic samples of the plasmodia in (A). The band at 68 kDa refers to polymalatase and shows a decrease in intensity. The band at 42 kDa reXects actin as control.
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199
microplasmodia did not eVect suppression of mRNA synthesis, indicating that extrinsic RNA could not penetrate into the cytoplasm (data not shown). The eVect on protein expression was measured by Western blotting of samples from the cytoplasm. Actin was used as reference in this case. An inspection of Fig. 1B indicates a suppression for the 68-kDa polymalatase band in parallel with the suppression of mRNA. Long-term experiments were carried out to (1) estimate the time of persistence of the RNAi eVect and (2) to enhance the knockdown level by repeated injections. After the Wrst injection and growth, 1/20 (by mass) of the macroplasmodium was replated. The new plasmodium was injected and replated again, and so on. As shown in Fig. 2, suppression of polymalatase-speciWc mRNA persisted for approximately 144 h after a single dsRNA injection, corresponding to the duration of 18 cell cycles and spanning two successive inoculations to fresh agar plates. In the second “generation” the expression of polymalatase-speciWc mRNA and protein already began to recover. However, multiple injections did not enhance the level of knockdown. The level of protein suppression did not decrease either, consistent with the absence of a long-term response such as developing a resistance against injected dsRNA. Plasmodia of the Wrst to third generations have been inspected for changes in the phenotype. No changes in
197
the overall appearance and growth rate were seen except that the latter appeared to be insigniWcantly retarded (p > 0.2).
Discussion Of Myxomycetes, such as P. polycephalum, the plasmodium represents a cell with an extremely large dimension and a high structural complexity. Nevertheless, injection of speciWc dsRNA into the veins proved to induce an approximately 90% suppression of the polymalatase gene. The knockdown was speciWc as control dsRNA and siRNA were totally ineVective. Because the knockdown was not complete, this result suggested at least two explanations. (1) The artiWcially introduced RNA spread all over the plasmodium except in a few compartments, which were not accessible and thus retained their normal mRNA synthesis. However, this possibility seemed unlikely in face of the results of repeated injections and maintenance of the very same eVect for more than 200 h. (2) In a survey of more than 2850 cases of various siRNA “gene silencers” it has been reported that suppression was frequently incomplete [16]. This may suggest another possibility, namely that rates of mRNA synthesis and the RNAi-induced cleavage were of similar magnitudes, resulting in a steady
Fig. 2. Suppression of mRNA and protein synthesis after the injection of polymalatase (pma)-speciWc dsRNA into macroplasmodia. Plasmodia were injected with dsRNA 22 h after inoculation and then grown for 24 h before analysis. The Wrst and following generations were prepared by inoculation with a small amount of each precursor plasmodium on a fresh agar plate and grown for 60 h as described under Materials and methods. Black columns, noninjected control plasmodia; dark gray columns, water-injected control plasmodia; light gray columns, serial experiment with only the parent plasmodium injected; white columns, serial experiment with injections of the parent plasmodium and of the Wrst to third generations each at 24 and 48 h after replating. (A) mRNA quantifcation by real-time RT-PCR with reference to actin mRNA. Messenger RNA of the noninjected control plasmodia was set to 100%. Standard errors are indicated for three independent measurements. (B) Western blot analysis of the samples in (A). The band at 68 kDa refers to polymalatase and shows a decrease in intensity correlating with the suppression of polymalate-speciWc mRNA in (A). The band at 42 kDa reXects actin as control.
198
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199
state survival of mRNA. A variety of reasons for an incomplete break down has been reported, such as absence of optimal base-pairing between siRNA and target mRNA [17], limited accessibility of target mRNA due to stabile secondary structure [17–19], protective binding of proteins, and speciWc subcellular location of mRNA [19]. Low-abundance transcripts were less susceptible to siRNA-mediated degradation than mediumand high-abundance transcripts [19]. Long-term experiments revealed that a single injection was quite eYcient, the knockdown lasting through 144 h into the second generation (Fig. 2A). For comparison, the injected 1 g dsRNA (1.5 £ 1011 molecules) corresponded to 3000 molecules dsRNA per nucleus [5 £ 108 nuclei (g plasmodia)¡1] immediately after the injection and would have been 7.5 molecules of dsRNA at the time of analysis in the second generation. The sustentation of the knockdown eVect is in agreement with the proposed mechanism of RNAi silencing [20,21] and indicated a high eYciency. Repeated injection of dsRNA did not induce further decline in mRNA, and the mRNA returned to its original level after recovery during the third generation following single injection. Apparently, the regulation of mRNA production was not aVected by ongoing RNA interference. The phenotype of the plasmodium of the injected parent plasmodium through the third generation was found to resemble that of the control plasmodium, except for a slight, however, insigniWcant (p > 0.2) growth retardation. This was referred to the residual synthesis of polymalatase revealed by Western blotting. The persisting low level of polymalatase was apparently suYcient for the maintenance of the physiological status of the plasmodium. We have recently shown that an increase in polymalate content is related to an increase in growth rate of the plasmodium [22]. Similarly, we had expected a decrease in growth rate if the polymalatase (adapter)–polymalate complex were the functional unit and if the level of polymalatase was reduced by RNAi interference. A knockout of the protein would have paralyzed polymalate function as a molecular carrier of nuclear proteins. Protein concentrations in diVerent regions of the plasmodium would have fallen out of phase, resulting in loss of synchrony and consequently loss of the normal phenotype of the plasmodium. Although we could occasionally observe that growth behavior was slightly diVerent in dsRNA-treated cells and indeed some regions seemed to have fallen out of synchrony indicated by retarded growth, the observation was not statistically signiWcant. A complete knockout of the gene expression seems to be required to registrate a clear change in the phenotype. The method that we have introduced here for plasmodia is very simple and eVective. Because of the high synchronization of plasmodial events and the precision in timing of an injection, the method oVers the possibil-
ity to knockdown the temporary expression of genes in a time-resolved fashion. Moreover, the injection can be precisely placed at particular coordinates of the giant cell body (plasmodium) to study gene suppression over intracellular distances. The RNAi technique in conjunction with the unique properties of the plasmodium of P. polycephalum should allow with relatively little experimental expenditure study of versatile cellular phenomena that have hitherto not been accessible.
References [1] T.G. Burland, K.L. Solnica, J. Bailey, D.B. Cunningham, W.F. Dove, Patterns of inheritance, development and the mitotic cycle in the protist Physarum polycephalum, Adv. Microb. Physiol. 35 (1993) 1–69. [2] S.L. Baldauf, A.J. Roger, I. Wenk-Siefert, W.F. Doolittle, A kingdom-level phylogeny of eukaryotes based on combined protein data, Science 290 (2004) 972–977. [3] D. Kessler, Plasmodial structure and motility, in: H.C. Aldrich, J.W. Daniel (Eds.), Cell Biology of Physarum and Didymium, vol. 1, Academic Press, Paris, 1982, pp. 145–208. [4] B.-S. Lee, M. Vert, E. Holler, Water-soluble aliphatic polyesters: Poly(malic acid)s, in: Y. Doi, A. Steinbüchel (Eds.), Biopolymers, Polyesters, vol. 3a, Wiley-VCH, Weinheim, 2002, pp. 75–103. [5] S. Doerhoefer, C. Windisch, B. Angerer, O.I. Lavrik, B.-S. Lee, E. Holler, The DNA-polymerase inhibiting activity of poly(-L-malic acid) in nuclear extract during the cell cycle of Physarum polycephalum, Eur. J. Biochem. 269 (2002) 1253–1258. [6] W. Marwan, Detecting functional interactions in a gene and signaling network by time-resolved somatic complementation analysis, BioEssays 25 (2003) 950–960. [7] G. Golderer, E.R. Werner, S. Leitner, P. Gröbner, G. WernerFelmayer, Nitric oxide synthase is induced in sporulation of Physarum polycephalum, Genes Dev. 15 (2001) 1299–1309. [8] B. Willibald, W. Bildl, B.-S. Lee, E. Holler, Is -poly(L-malate) synthesis catalysed by a combination of -L-malyl-AMP-ligase and poly(L-malate) polymerase? Eur. J. Biochem. 269 (1999) 1085– 1090. [9] J.J. Tyson, Periodic phenomena in Physarum, in: H.C. Aldrich, J.W. Daniel (Eds.), Cell Biology of Physarum and Didymium, vol. 1, Academic Press, New York, 1982, pp. 61–110. [10] J. Dee, Genetics of Physarum polycephalum, in: H.C. Aldrich, J.W. Daniel (Eds.), Cell Biology of Physarum and Didymium, vol. 1, Academic Press, New York, 1982, pp. 212–251. [11] O. Donze, D. Picard, RNA interference in mammalian cells using siRNAs synthesized with T7 RNA polymerase, Nucleic Acids Res. 30 (e46) (2002) 1–4. [12] S.M. Elbashir, J.P.A. Martinez, W. Lendeckel, T. Tuschl, Functional anatomy of siRNAs for mediating eYcient RNAi in Drosophila melanogaster embryo lysate, EMBO J. 20 (2001) 6877– 6888. [13] S.M. Elbashir, J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, T. Tuschl, Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells, Nature 411 (2001) 494–498. [14] F.W. Daniel, H.H. Baldwin, Methods of culture for plasmodial myxomycetes, Methods Cell Physiol. 1 (1964) 9–14. [15] M. Karl, E. Holler, Multiple polypeptides immunologically related to -poly(L-malate) hydrolase (polymalatase) in the plasmodium of the slime mold Physarum polycephalum, Eur. J. Biochem. 251 (1998) 405–412. [16] siRNA-Catalog, Santa Cruz Biotechnology, Inc., Santa Cruz, 2004.
RNA interference in plasmodia of myxomycetes / M. Haindl, E. Holler / Anal. Biochem. 342 (2005) 194–199 [17] K. Yoshinari, M. Miyagishi, K. Taira, EVects on RNAi of the tight structure, sequence and position of the targeted region, Nucleic Acids Res. 32 (2004) 691–699. [18] T. Holen, M. Amarzguioui, M.T. Wiiger, E. Babaie, H. Prydz, Positional eVects of short interfering RNAs targeting the human coagulation trigger tissue factor, Nucleic Acids Res. 30 (2002) 1757–1766. [19] X. Hu, S. Hipolito, R. Lynn, V. Abraham, S. Ramos, F. WongStaal, Relative gene-silencing eYciencies of small interfering RNAs targeting sense and antisense transcripts from the same genetic locus, Nucleic Acids Res. 32 (2004) 4609–4617.
199
[20] C. Arenz, U. Schepers, RNA interference: from an ancient mechanism to a state of the art therapeutic application?, Naturwissenschaften 90 (2003) 345–359. [21] T. Sijen, J. Fleenor, F. Simmer, K.L. Thijssen, S. Parrish, L. Timmons, R.H. Plasterk, A. Fire, On the role of RNA ampliWcation in dsRNA-triggered gene silencing, Cell 107 (2001) 465–476. [22] M. Karl, R. Anderson, E. Holler, Injection of poly(-L-malate) into the plasmodium of Physarum polycephalum shortens the cell cycle and increases growth rate, Eur. J. Biochem. 271 (2004) 3805– 3811.