Effects of ced-9 dsRNA on Caenorhabditis elegans and Meloidogyne incognita

Problem Statement: In metazoans Programmed Cell Death (PCD) is essenti al for proper development. Suppression of PCD is needed to guaran tee cell survival and in the nematode Caenorhabditis elegans the regulation of PCD is accomplished by the functi on of the ced-9 gene. Approach: In this work the use of double stranded RNA (dsRNA) to knock-down ced-9 gene function was tested as means to induce PCD. Results: Our results indicate that dsRNA targeting the cell d ath protection gene ced-9 is effective at decreasing the fecundity of C. elegans by up to 21%. The decreased fecundity correlated with an increased pr sence of cell corpses in developing embryos. Endogenous ced-9 transcript levels were reduced in progeny of ced-3 mutant nematodes fed bacteria expressing ced-9 dsRNA. These data suggest that nematode fecundity c an be reduced by ingestion and exposure to dsRNAs targeting regulation of the cell d ath pathway. In an attempt to determine if plant parasitic nematodes are susceptible to the targetin g of the PCD regulatory pathway we exposed Meloidogyne incognita, a plant parasitic nematode, to ced-9 sRNA; here we show that this exposure results in decreased gall formation in the tobacco plants. Conclusion/Recommendations: Our results provide the first steps toward using RNAi technolog ies to attempt nematode control by targeting cell death pathways. Ongoing research with transgenic pl ants designed to express dsRNA for ced-9-like sequences will further test the feasibility of gene rating plants with RNAi-based resistance to parasit ic nematodes.


INTRODUCTION
Nematode parasitism of animals and plants remains a significant global problem (Chitwood, 2003;De Silva et al., 2003). Research conducted using the free-living nematode Caenorhabditis elegans has contributed tremendously to our understanding of other nematode species and as such, C. elegans provides highly suitable model for testing potential methods of parasitic nematode control, particularly as older ineffective, or environmentally detrimental methods are phased out or become ineffective (Britton and Murray, 2006). In particular, C. elegans has contributed to our understanding of Programmed Cell Death (PCD). PCD in C. elegans eliminates a total of 131 cells (Kinchen and Hengartner, 2005). The genes egl-1, ced-4 and ced-3 are required for PCD in somatic cells of C. elegans and a loss-of-function (lf) mutation in any of them results in increased survival rates in cells that would otherwise be eliminated (Metzstein et al., 1998). The gene ced-9, responsible for inhibiting cell death, is epistatic to ced-3 and ced-4 and shares significant similarity in structure and function with the mammalian proto-oncogene bcl-2 (Chinnaiyan et al., 1997;Hengartner and Horvitz, 1994). A gain-offunction (gf) mutation in ced-9 prevents normally occurring cell deaths, while lf mutants lead to increased cell deaths, maternal-effect lethality and sterility (Hengartner et al., 1992).
RNA interference (RNAi) has been used extensively in recent years for the experimental targeting and silencing of specific genes in many species (Bakhetia et al., 2005;Geldhof et al., 2007;Wang and Barr, 2005). In this study we tested the effectiveness of Double Stranded (DS) RNAs for targeting and silencing the cell death protection gene ced-9 in C. elegans. We hypothesized that feeding or soaking C. elegans with double stranded (ds) ced-9 RNA would silence endogenous ced-9 expression and phenocopy a loss-of-function ced-9 mutant. Our data suggest that fecundity is significantly reduced in C. elegans treated with double stranded ced-9 RNA. We also tested the possibility that ds ced-9 RNA might be used to control reproduction in parasitic nematodes. Our work indicated that the plant root-knot nematode Meloidogyne incognita was also sensitive to ced-9 dsRNAs, demonstrated by reduced gall formation in tobacco plants after treating infective J2 M. incognita nematodes with ced-9 dsRNAs. These data suggest that targeting the cell death pathway of nematodes by RNAi may be an effective method for their control.

Nematode and bacterial strains:
The strains of C. elegans (obtained from the Caenorhabditis Genetic Center, University of Minnesota) used in this work were: wild-type N2; ced-3 mutant strain (n717) IV; ced-4 mutant strain (n1162)III and gfp transgenic line (PD4251). Nematodes were cultured and synchronized according to standard protocols (Sulston and Hodgkin, 1998). E. coli strain OP50 was used as a food source for maintenance of the different free-living nematodes. The plant-parasitic nematode species Meloidogyne incognita (root-knot nematode) was kindly provided by Dr. Valerie Williamson, UC-Davis, CA.

Molecular biology techniques:
All reagents unless otherwise indicated were purchased from Fisher Scientific Company. All molecular biology techniques were conducted using standard protocols (Ausubel, 2003;Sambrook and Russell David, 2001) with the following modifications: cloning was initially carried out in E. coli strains DH5α or Top10 (Invitrogen) and plasmids were later transformed into E coli strain HT115(DE3) for C. elegans dsRNA feeding. The following primers were designed using ced-9 cDNA sequence (L26545) and were synthesized by Operon (Qiagen): P1 5'-ATAAGAATGCGGCCGCAGATGACACGCTGC ACGGCG-3'; P2 5'-GAAGGCCTTTACTTCAAGCTGAACATCATC-3'; P3 5'-CGAGATGAAGGAGTTTCTGG-3'; P4 5'-GCCAATCATCGACCACCGTC-3'. A cDNA of the ced-9 gene was cloned (908 bp) and its identity confirmed by DNA sequencing. The ced-9 gene was subcloned into the pLitmus28i vector (New England Biolabs). Constructs pLitmus28i::ced-9 (Fig. 1A) and pLitmus::Mal (positive control provided in New England Biolabs RNAi Synthesis Kit) were tested for their ability to produce dsRNAs under in vitro and in vivo conditions; transcripts of the correct size were obtained using both in vitro (Fig. 1B) and in vivo (Fig. 1C) conditions.
RNAi of the C. elegans ced-9 gene: In vitro dsRNA synthesis was conducted as follows: pLitmus28i::ced-9, pLitmus28i::gfp and pLitmus::Mal plasmid constructs were used as templates in PCR reactions with a T7 primer (5'-TAATACGACTCACTATAGG-3'), which flanks the insert in the pLitmus28i vector (New England BioLabs). pLitmus::Mal shares no homology with the C. elegans genome and was used to generate non-specific dsRNAs. 100 ng of purified plasmid DNA was added to PCR reactions with Pfu-turbo polymerase (Stratagene). PCR products were purified and 1 µg of DNA template was used in dsRNA synthesis reactions according to recommended protocols (New England BioLabs). dsRNA reactions were treated with DNase1 (Qiagen) and S1 nuclease (Promega) and purified dsRNA was resuspended in nematode soaking buffer (10.9 mM Na 2 HPO 4 , 5.5 mM KH 2 PO 4 , 2.1 mM NaCl, 4.7 mM NH 4 Cl, 3 mM spermidine and 0.05% gelatin in diH2O).
For in vivo dsRNA synthesis the plasmids pLitmus28i, pLitmus28i::ced-9, pLitmus28i::gfp and pLitmus::Mal, were transformed into E. coli strain HT115(DE3) and cultured as described (Timmons et al. 2001). Single colony-derived cultures were grown to an OD 600 of ~0.4 and dsRNA expression was induced with 1 mM IPTG for 4 h at 37°C. RNA was extracted from an aliquot of each sample, treated with DNase I and S1 Nuclease and resolved in 1% agarose to identify bacterial clones producing dsRNA bands of the correct molecular weight.
Feeding and soaking of C. elegans: The feeding of C. elegans to monitor fecundity was conducted as follows: bacterially seeded NGM plates (24-well Falcon plates with NGM containing IPTG and 100 µg mL −1 ampicillin when appropriate) were incubated at 37°C overnight. Replica wells were seeded with 5 µL of HT115 (DE3) bacteria containing pLitmus28i empty vector, pLitmus::Mal, or pLitmus28i::ced-9. Two or 3 synchronized L3 or L4 stage hermaphrodites were placed in each well. After ~72 h, 1 treated nematode was selected at random from each well and placed into a well of a 96-well round-bottom plate containing 30 µL of liquid nematode media (3% yeast extract, 3% peptone and 1% dextrose supplemented with 0.5 mg mL −1 of hemoglobin and 40 µg mL −1 of cholesterol).
Nematodes were transferred to fresh media every 48 hours and total progeny counted at 120 h. Each randomly selected nematode was considered a biological replicate, a minimum of 12 biological replicates were analyzed per trial and at least 2 independent trials were carried out on each genotype. plasmid; similar empty vector, or vector containing the nonspecific Mal gene were also used. For in vitro generated transcripts (B) PCR amplified genes or entire plasmids were used as template in transcription reactions and dsRNA corresponding to the expected sizes were obtained: ~1 kb dsRNAs for ced-9 and ~0.9 kb for Mal). Lanes a, b and c represent transcripts from independent clones. Transcripts were also generated in vivo (C) in E. coli strain HT115(DE3). Independent clones (lanes a and b) expressing ced-9 and Mal revealed dsRNAs of the expected sizes (~0.9 kb and ~0.8 kb respectively), while bacterial samples containing empty pLitmus28i plasmids did not.
For each genotype the two negative controls were each used in at least one trial (pLitmus28i empty vector and pLitmus::Mal) and had similar effects on treated nematodes. Trials showed similar trends and replicates were pooled for final analysis (for wt, n=136 and n=132 for control and test treatment, respectively; for ced4, n=122 and n=120 for control and test treatments; for ced3 worms n=59 and n=60 for control and test treatments). PD4251 nematodes were fed HT115(DE3)::pLitmus28i::gfp or HT115(DE3)::pLitmus28i for a period of 72 hours and were mounted for observation by epi-fluorescence microscopy. The total number of GFP positive muscle and vulval cell nuclei were counted (n=10 and n=38 for control and test treatments, respectively).
For soaking experiments, 2-3 wild-type, ced-4(lf), or ced-3(lf) C. elegans L3 stage nematodes were added to 1 µL droplets (in closed 200 µL PCR tubes) of each treatment. Treatments were composed of soak buffer containing ced-9 or Mal dsRNA at a concentration of ~1 µg/µL, or soak buffer with no dsRNA. Nematodes were soaked for 24 h and then transferred to NGM plates seeded with OP50 for a two-day recovery. As hermaphrodites matured and began to produce eggs (~72 h), they were placed in 50 µL of liquid nematode media (in wells of 96-well flat-bottom plate), transferred to fresh media after 48 hours and total progeny counted at 96 hours. Each treated hermaphrodite was a biological replicate, a minimum of 10 biological replicates were analyzed per trial and at least two independent trials were conducted on each genotype. Replicates from all trials were pooled for final statistics as described above (for wt, n = 40 and n = 34 for control and test treatment, respectively; for ced4, n = 48 and n = 43 for control and test treatments; for ced3 worms n = 45 and n = 46 for control and test treatments). Similarly, PD4251 nematodes were treated with double stranded gfp dsRNA or soak buffer (n >30 for both treatments). Two sample T-tests were conducted using the Microsoft Excel data analysis function.
Feeding of mutant ced-3 C. elegans with ced-9 or Mal dsRNA and competitive PCR (cPCR) assay: Mutant ced-3 nematodes were plated in NGM plates containing OP50 and incubated for 72 h. Matured hermaphrodites were washed with M9 media (48 mM Na 2 HPO 4 , 22 mM KH2PO4, 9 mM NaCl, 19 mM NH4CL), incubated in a hypochlorite solution (final concentration of 5.25% Na-O-Cl and 250 mM KOH) for 10 min and the recovered embryos washed three times with M9. Embryos were transferred to NGM plates and incubated overnight with ced-9 or Mal dsRNA producing bacteria (NGM plates were supplemented with 1mM IPTG, 50 µg mL −1 ampicillin and 25 µg mL −1 tetracycline) and allowed to develop and feed for 72 hours. Embryos were extracted from these fed nematodes and divided in two fractions; the first fraction was used for RNA extraction; the second fraction was plated on NGM plates seeded with OP50 bacteria and incubated for an additional 24 hours for RNA extraction from developed nematodes. Embryo fractions were treated with hypochlorite solution and washed with M9 prior to RNA extraction to eliminate bacteria and avoid dsRNA cross-contamination. Total RNA was isolated from ced-3 mutant C. elegans nematodes or embryos using Trizol reagent protocol as recommended by the manufacturer (Invitrogen). Total RNA was treated with DNase I (1 unit µg −1 ) and purified by phenol chloroform extraction. Two micrograms of RNA was combined with 1 µg of ced-9 P4 primer for first-strand cDNA synthesis with Moloney Murine Leukemia Virus reverse transcriptase (M-MLV) according to manufacturer instructions (Promega). The cDNA was used as template for PCR with ced-9 primers (P3 and P4) in the presence of known quantities of the ced-9∆ competitor. The competitor was derived by fill-in and blunt-end ligation of pLitmus28i::ced-9 plasmid after digestion with Bpu10I and SapI restriction enzymes. The resulting plasmid (pLitmus28i::ced-9∆) contains a deletion of 88 base pairs in the ced-9 gene sequence. To quantify cDNA production during ced-9 cDNA synthesis we used the C. elegans gamma-tubulin gene, tbg-1 (F58A4.8), since gamma-tubulin is a single copy gene and is constitutively expressed in C. elegans (Consortium, 1998;Strome et al., 2001). The gammatubulin primers used were: upstream primer=5'ATGTCCGGTACGGGTGCCTTGA3' and downstream primer=5'TGGAGGCCTTCTGATTGCGAAATC3'. The PCR amplification was performed for 30 cycles as follows: hot start at 95°C for 5 min; denaturation at 95 o C for 30 sec.; annealing at 49 o C for 30 sec and extension at 72°C for 1 min. At the end of the 30 cycles products were extended at 72°C for 7 min. PCR products were visualized by ethidium bromide staining on 1% agarose gels and digitized using the 1D Image Analysis Software v3.5 (Eastman Kodak Company). The results shown are representative of three independent experiments.
Nomarski microscopy of C. elegans: Nomarski (Differential Interference Contrast) microscopy was used to study the morphology of C. elegans embryos derived from hermaphrodites fed E. coli expressing ced-9 or Mal dsRNA. Cell corpses were identified by their increased refractivity as described (Hengartner et al., 1992;Sulston et al., 1983). Embryos were mounted on agar-pad slides and visualized using an Olympus IX-70 research microscope. Images were recorded using MagnaFire camera and software (Optronics). Observations were conducted in multiple focal planes on the lateral sides of embryos ranging in age from ~350-420 min (n = 51 and n = 56 for test and control treatments, respectively). A two sample T-test was conducted as described above for testing the null hypothesis of equal means.

Maintenance and soaking of M. incognita:
Susceptible tomato plants (Rutger's Select, Tomato, Augusta, GA) were used to maintain M. incognita as follows: five to ten days old sand-germinated tomato plants (10 cm shoot and approximately 4-5 cm long roots) were infected with pre-parasitic J2s. Five to eight week old heavily galled plant roots were cut into small pieces using a razor blade. The chopped plant roots were vigorously shaken for 5 min with 10% bleach and subsequently poured onto a 250 micron mesh screen. Eggs were collected from the flow-through on a 25 micron mesh screen and further purified by centrifugation in 35% sucrose at 500 g for 10 min. Purified eggs were subjected to two additional treatments in 10% bleach (10 min each) followed by centrifugation at 500 g for 5 min and several rinses in sterile water. Sterilized eggs were hatched at 28°C. During this incubation period, juveniles were allowed to crawl through a sterile mesh strainer into a water container. Freshly hatched juveniles M. incognita J2 were re-suspended in soaking buffer for treatment with dsced-9 or dsMAL RNA.

M. incognita susceptibility assay:
Tobacco plants were germinated on half-strength Murashige and Skoog (MS) medium (Murashige and Skoog, 1962) and transferred to square (10x10mm) plates containing MS medium and 1% Agargel (Invitrogen). The tobacco seedlings were placed on the surface of media in square plates and allowed to grow in the incubator for 2 weeks at 28°C and at a photoperiod of 16/8h (light/dark). 30 tobacco plants were used per treatment; 5 plants/plate and 2 roots per plant were inoculated at the root apex with approximately with 20 M. incognita J2 nematodes. Prior to infection, M. incognita J2s were soaked 24 hours in soaking buffer containing dsced-9 or dsMAL RNA at a concentration of 1 µg µL −1 . Positive infections (root thickenings caused by growth of adult females) were scored every 48 hours for 10 days. Each root thickening developed into a gall after about 20 days (not shown). Sodium-hypochlorite-acid-fuchsin staining methods (Byrd et al. 1983) were used to confirm nematode infections.
Wild-type C. elegans soaked in ced-9 dsRNA produced significantly fewer offspring than control treatments (Fig. 2B), although the effect was not as strong as that observed in feeding assays. An average decrease of ~16% in fecundity was observed (p<0.05). Conversely, ced-4(lf) and ced-3(lf) C. elegans mutants showed no difference among treatment groups (Fig. 2B).
Gfp silencing in C. elegans by RNAi confirmed the effectiveness of dsRNA delivery methods: the effectiveness of our RNAi delivery methods was confirmed by silencing the expression of Green Fluorescent Protein (GFP) in the gfp transgenic C. elegans strain PD4251 (Fig. 2C). In wild-type C. elegans, ced-9 was more efficiently silenced by dsRNA feeding; however in C. elegans strain PD4251, the gfp transgene was more effectively silenced by dsRNA soaking (Fig. 2C).
Endogenous levels of ced-9 mRNA are reduced following treatment with ced-9 dsRNA: we quantified the expression level of the endogenous ced-9 mRNA in progeny of treated nematodes using a cPCR assay (Brutnell and Dellaporta, 1994;Death et al., 1999). Since lf alleles of ced-9 lead to sterility and maternaleffect lethality and ced-9; ced-3 double mutant nematodes are viable (Hengartner et al., 1992), we tested the specificity of RNAi directed at ced-9 expression in a ced-3 background. The endogenous concentration of ced-9 mRNA in wild-type nematodes was approximately 0.5 picograms per 5 µL of cDNA template used for competitive PCR (not shown). We then quantified the concentration of ced-9 and gamma-tubulin, as an internal control, in progeny of mutant ced-3 C. elegans that were fed bacteria expressing ced-9 or Mal dsRNA. The expression levels of gamma-tubulin were similar in both ced-9 and Mal dsRNA fed individual nematodes (Fig. 3A  lanes 3 and 4) and embryos (Fig. 3A lanes 7 and 8). However, endogenous ced-9 mRNA was lower in both adult nematodes (compare lanes 1 and 2 of Fig.  3A) and embryos (compare lanes 5 and 6 of Fig. 3A) progeny of the ced-9 dsRNA fed nematodes.
We then conducted cPCR assay and found that ced-3 mutant C. elegans fed ced-9 dsRNA consistently demonstrated reduced (0.12 picograms; Fig. 3B) ced-9 levels compared to nematodes fed Mal dsRNA (0.5 picograms; Fig. 3C). Results were similar when the endogenous ced-9 mRNA was measured in progeny embryos (2 picograms in ced-9 dsRNA fed nematodes compared to 4 picograms in Mal dsRNA fed nematodes; Fig. 3B and C). There were no differences in endogenous ced-9 concentrations between wild-type or ced-3 mutant C. elegans fed Mal dsRNA (not shown). C. elegans embryos exposed to ced-9 dsRNA were observed by Nomarski microscopy. One focal plane of a treated embryo (A) and control embryo (B) at the comma stage. Arrows in (A) and (B) point to cell corpses; identified by their increased refractivity under Nomarski microscopy as described previously (Hengartner et al., 1992;Sulston et al., 1983). (C) The graph shows a comparison between ced-9 and Mal dsRNA treatments. Values are means +/-SEM. Both comma stage (350-380 min) embryo data and 1½ fold stage (~380-420 min) embryo data were combined and ced-9 dsRNA treatments were compared to Mal dsRNA. Scale bar is 10 µm. (t-test, p≤0.05*) .05* at 5days post-infestation and t-test, p≤0.001*** at 7 and 10 days post-infestation.

Decreased fecundity was associated with increased cell (extra space between increased and cell) death corpses in C. elegans embryos:
we detected a significant increase (19%) in ectopic cell deaths occurring in comma stage (350-380 min) embryos treated with ced-9 dsRNA and a 16% increase in the number of cell deaths occurring in the 1½-fold stage (400-420 min) embryos relative to controls. Overall, a 17% increase in cell deaths was detected in 350-420 min embryos derived from ced-9 dsRNA fed wild-type C. elegans relative to, (delete comma, space corrected between elegans and relative) Mal dsRNA fed nematodes ( Fig. 4; p<0.05).

M. incognita juveniles soaked in ced-9 dsRNA prior to plant inoculation demonstrate decreased efficiency of gall formation in infected tobacco plants:
Juveniles of the plant parasitic nematode M. incognita were soaked with ced-9 or Mal dsRNA prior to inoculation of tobacco roots. In two independent experiments, we observed that M. incognita J2s soaked with ced9 dsRNA prior to inoculation of tobacco plants, showed a reduction of 40% in the number of galls generated when compared to worms soaked with dsMal RNA (Fig. 5). This (edited) reduction in the number of galls was statistically significant between two treatments; t-test, p0.05* at 5 days post-infestation and t-test, p≤0.001*** at 7 and 10 days post-infestation, (SPSS 13.0 statistical analysis software for windows was used) values were means of two independent experiments and in each experiment 30 plants were used per treatment.

DISCUSSION
Plant-parasitic nematodes pose a serious worldwide threat to agriculture and as older or environmentally unfriendly methods for their control are phased out or become ineffective, new methods must be actively sought. Our research suggests that nematode numbers can be affected by exposure to dsRNAs encountered in their environment and food, and indicates that cell death regulatory genes may be useful targets for engineering nematode resistance in host crop species. Wild-type C. elegans was susceptible to dsced-9 RNA in both feeding and soaking experiments, with up to a 21% reduction in fecundity following treatment. The reduced fecundity correlated with an (delete) increased (d added) in (delete) ectopic cell deaths in treated embryos. A similar analysis in a (the instead of a?) plant parasitic nematode M. incognita also demonstrated susceptibility to ced-9 dsRNA.
The effectiveness of our system was confirmed by silencing the expression of green fluorescent protein (gfp) with greater than 90% efficiency in the gfp transgenic C. elegans strain PD4251 (data not shown). These results were comparable to those achieved in other laboratories (Timmons et al., 2001). When comparing the silencing of gfp and ced-9 in C. elegans, it appeared that silencing efficiency was dependent on the method of RNAi treatment employed and the gene targeted. Gene-specific differences in RNAi susceptibility have been reported between feeding and microinjection procedures (Kamath et al., 2001).
The fact that ced-9 dsRNA treated C. elegans continued to produce offspring throughout their life indicates that the penetrance of the phenotype was incomplete. It is not uncommon that weaker phenotypes are achieved using RNAi, or that silencing occurs in only a subset of cells in comparison to genetic mutants (Tabara et al., 1998). We observed high expression of the gfp transgene in embryos within dsRNA treated hermaphrodites, supporting the idea that some embryonically expressed transcripts may be difficult to silence (Kamath et al., 2001). It has been shown that maternally-derived ced-9 is passed into the developing egg (Hengartner et al., 1992) and it is possible that this subset of RNAs was more efficiently degraded by RNAi of ced-9. Consequently, the techniques employed in this study may have been insufficient for silencing embryonically expressed transcripts, leading to increased survival. In C. elegans, stronger ced-9(lf) mutants have been characterized as having massive ectopic cell deaths in developing embryos, maternal effect lethality and partial sterility (Hengartner et al., 1992). On the other hand, weak loss-of-function alleles have also been described in which nematodes are viable (Riddle, 1997). Early high throughput RNAi assays did not report phenotypes associated with the ced-9 gene (Gonczy et al., 2000;Maeda et al., 2001), while other studies have reported increased apoptosis, embryo lethality, and other cell death phenotypes (Lettre et al., 2004;Sonnichsen et al., 2005). The results presented here support previous studies and indicate that a quantifiable reduction in fecundity followed ced-9 RNAi treatments.
It is possible that non-specific silencing of other genes occurred in C. elegans treated with ced-9 dsRNA. The ced-9 locus transcribes four mRNA species two of which include both ced-9 and cyt-1. Cyt-1 encodes a protein similar to cytochrome b560 and is involved in electron transport in mitochondria (Senoo-Matsuda et al., 2003). Mutants in cyt-1, such as mev-1, lead to increased sensitivity to oxidative stress and aging (Senoo-Matsuda et al., 2003). In C. elegans, cyt-1 mutants display structural abnormalities of the mitochondria, anion generation, loss of membrane potential, changes in ced-9 protein levels and ced-3 dependent increases in cell death occurring in 400-min embryos. Thus it is plausible that non-specific degradation of cyt-1 contributed to the phenotypes we observed. However, three lines of evidence suggest that the decrease in fecundity we observed in C. elegans was due to specific silencing of ced-9 and consequent up-regulation of the cell death pathway. First, decreased fecundity was only observed in wild-type C. elegans and not in ced-3 or ced-4 mutants, suggesting the effect was due to an up-regulation of the proteins CED-4 and CED-3 through reduced ced-9 levels. Second, endogenous levels of ced-9 RNA were reduced in ced-3 mutants treated with ced-9 dsRNA. Third, a significant increase in ectopically occurring cell deaths was observed in treated nematodes.
C. elegans shares common ancestry with plantparasitic nematodes (Riddle, 1997) and it has been demonstrated that much like C. elegans, Meloidogyne artiellia, Heterodera glycines and Globodera pallida are sensitive to RNAi (Fanelli et al., 2005;Urwin et al., 2002). This suggests that RNAi mechanisms are conserved among nematode species. Homology search across Nematoda phylum using the cyt-1/ced-9 fulllength transcripts (L26545) identified homology to cyt-1 and ced-9 sequences in some species of plant parasitic nematodes (not shown). PCR-Southern blot hybridizations to C. elegans ced-9 probe detected similar sized sequences in other nematode genera, including Pristionchus, Zeldia and Meloidigyne (our unpublished data). Our susceptibility assay on M. incognita suggests that this species may contain an orthologous ced-9 gene susceptible to RNAi; however, our inability to clone a ced-9-like gene using the ced-9 primers suggests sequence divergence among the two species. Consequently, it remains to be seen whether or not genes with similar sequences from unrelated biochemical pathways were silenced by double stranded ced-9 RNAs, contributing to reduced gall formation.

CONCLUSION
In conclusion, our results provide the first steps toward using RNAi technologies to attempt nematode control by targeting cell death pathways. Ongoing research with transgenic plants designed to express dsRNA for ced-9-like sequences will further test the feasibility of generating plants with an RNAi-based resistance to parasitic nematodes.