Morphogenesis of Strongyloides stercoralis Infective Larvae Requires the DAF-16 Ortholog FKTF-1

Based on metabolic and morphological similarities between infective third-stage larvae of parasitic nematodes and dauer larvae of Caenorhabditis elegans, it is hypothesized that similar genetic mechanisms control the development of these forms. In the parasite Strongyloides stercoralis, FKTF-1 is an ortholog of DAF-16, a forkhead transcription factor that regulates dauer larval development in C. elegans. Using transgenesis, we investigated the role of FKTF-1 in S. stercoralis' infective larval development. In first-stage larvae, GFP-tagged recombinant FKTF-1b localizes to the pharynx and hypodermis, tissues remodeled in infective larvae. Activating and inactivating mutations at predicted AKT phosphorylation sites on FKTF-1b give constitutive cytoplasmic and nuclear localization of the protein, respectively, indicating that its post-translational regulation is similar to other FOXO-class transcription factors. Mutant constructs designed to interfere with endogenous FKTF-1b function altered the intestinal and pharyngeal development of the larvae and resulted in some transgenic larvae failing to arrest in the infective stage. Our findings indicate that FKTF-1b is required for proper morphogenesis of S. stercoralis infective larvae and support the overall hypothesis of similar regulation of dauer development in C. elegans and the formation of infective larvae in parasitic nematodes.


Introduction
Parasitism among nematodes appears to have arisen multiple times throughout evolution [1]. However, the exact mechanism by which nematodes developed parasitic life histories is unknown. Altering gene regulation through variation in conserved signaling systems, is a potential mechanism by which a free-living species might develop characteristics required for parasitism [2]. Insulinlike signaling regulates metabolism and lifespan in a variety of organisms including nematodes, insects and mammals [3,4]. In Caenorhabditis elegans, this signaling pathway mediates entry into the dauer larvae diapause by negatively regulating DAF-16, a forkhead transcription factor type O (FOXO) [5]. Biological requirements of C. elegans dauer larvae include increased resistance to stress and a metabolism altered to allow the animal to persist, potentially for months, in unfavorable environments [6]. Infective larvae of parasitic nematodes, such as S. stercoralis, have similar requirements for survival prior to host finding. The 'dauer hypothesis' recognizes the common physiological characteristics of dauer larvae and parasitic infective larvae, and proposes that the same molecular genetic mechanisms control the morphogenesis of both forms [7].
The life cycles of Strongyloides and Parastrongyloides spp., unusual among the parasitic nematodes, alternate between free-living and parasitic generations [8,9]. First-stage larval progeny of parasitic S. stercoralis females typically develop into free-living adults unless triggered by genetic, environmental or host-associated conditions to develop directly into infective third-stage larvae (L3i) [10].
Progeny of the free-living generation of S. stercoralis are uniformly fated to become L3i that invade the host and develop into parasitic females. Previous work identified the FOXO encoding gene fktf-1 (forkhead transcription factor-1) as the ortholog of C. elegans daf-16 in S. stercoralis [11]. In heterologous rescue experiments, a transgene construct designed to express FKTF-1b (isoform b) partially restored DAF-16 function to C. elegans daf-2;daf-16 double mutants [12] rescuing the dauer development phenotype. These data indicate that fktf-1b encodes a working forkhead transcription factor that can function in insulin-like signaling to regulate L3 development in C. elegans.
The more relevant question of whether FKTF-1b regulates infective larval development in S. stercoralis itself can now be addressed using new methods for transgenesis in this parasite [13,14]. In the present study, we transformed free-living adult female S. stercoralis with constructs encoding a 2.6 kb fktf-1b promoter controlling expression of GFP::FKTF-1b fusion proteins. We then examined first-stage larval progeny of these female worms for anatomical and intra-cellular localization of GFP-linked proteins and for phenotypes associated with expression of transgenes encoding mutant forms of FKTF-1b.

Results/Discussion
Anatomical expression patterns of fktf-1b First, we asked whether the localization of fktf-1b expression in S. stercoralis mimics that of C. elegans daf-16b, which is expressed primarily in the pharynx and body neurons [5]. First-stage S. stercoralis larvae expressed fktf-1b::gfp::fktf-1b ( Figure S1) predominantly in the procorpus of the pharynx ( Figure 1A, and Figure 1B, arrow) and the hypodermis ( Figure 1C and 1D). These expression patterns continued into the L3i ( Figure 1E and 1F). In S. stercoralis, remodeling of the short, trilobed rhabditiform pharynx of the L1 into the long, cylindrical filariform pharynx of the L3i is a hallmark of the transition to infectivity [15]. The rhabditiform pharynx, found in all free-living stages, has three main components: the procorpus, the isthmus and the terminal bulb. The procorpus of the pharynx is the muscular region anterior to the narrow isthmus and is primarily responsible for food intake [15]. The pharynx of the non-feeding L3i, is not contractile and has no readily identifiable lobes [15]. Interestingly, expression of the fktf-1b reporter construct in the filariform pharynx was restricted to a band ( Figure 1F arrow) analogous to the procorpus of a rhabditiform pharynx. The hypodermal cell layer is responsible for secretion of the cuticle in a stage specific manner [16]. The infective larval cuticle must not only protect the L3i, it must also allow the L3i to sense the presence of a host and secrete molecules facilitating invasion. Although the expression patterns of the fusion protein in L1 varied somewhat ( Figure 1G), the fact that the predominant sites of expression were the pharyngeal procorpus and the hypodermis bolsters confidence that the endogenous fktf-1b promoter is active in these tissues in wild-type larvae. This pattern of expression is consistent with a role for FKTF-1b in the development of structures characteristic of infective larvae.

FKTF-1b is regulated via phosphorylation
Insulin-like signaling negatively regulates the function of forkhead transcription factors, including DAF-16, via phosphorylation of serines or threonines at specific sites by Akt/PKB kinases [17]. To ascertain similar post-translational regulation of FKTF-1b, we transformed S. stercoralis with vectors encoding mutant versions of GFP::FKTF-1b that were predicted to behave as either constitutively phosphorylated or non-phosphorylated forms of the protein. Substitution of charged residues, either aspartic or glutamic acids, for serines at Akt/PKB phosphorylation sites in the forkhead domain of human FOXOs is sufficient for disruption of DNA binding by these proteins and for their export from the nucleus [18,19]. Homologous 'phospho-mimetic' mutations in predicted Akt/PKB sites of FKTF-1b also resulted in constitutive export of the fusion protein GFP::FKTF-1b(S238E/T240E) (encoded by pPV244, Figure S1) from nuclei of hypodermal cells in transgenic S. stercoralis L1 (Figure 2A and 2B and 2G). Likewise, disruption of all four predicted Akt/PKB sites in FKTF-1b by substitution of the neutral amino acid alanine for critical serine or threonine residues (see

Author Summary
Parasitic nematodes are an important threat to public health in much of the world. Understanding how these worms find and invade their hosts may lead to improved therapies. The infectious forms of many parasitic nematodes developmentally arrest as infective third-stage larvae that require hosts to reactivate. Development of these larvae has been compared to that of the diapausing dauer larvae of Caenorhabditis elegans. Our lab studies the development of the human nematode parasite Strongyloides stercoralis. We identified S. stercoralis' FKTF-1 as an ortholog of DAF-16, a forkhead transcription factor controlling dauer larval development in C. elegans. Transgenes were introduced into S. stercoralis to investigate the possibility that FKTF-1 regulates development of its infective larvae. We discovered that recombinant FKTF-1b tagged with GFP localizes to specific tissues remodeled in infective larvae. Furthermore, mutant forms of FKTF-1b designed to interfere with endogenous FKTF-1b function resulted in incomplete development of the infective larval structures and prevented some transgenic larvae from arresting in the infective stage. Indicating that FKTF-1b is required for the proper development of Strongyloides stercoralis infective larvae, our findings support the hypothesis of similar controls over parasitic and free-living nematode development and pave the way for further comparative studies. pPV243, Figure S1) resulted in strongly enhanced nuclear localization of the 'phospho-null' fusion protein GFP::FKTF-1b(4A) ( Figure 2E-2G). These data indicate that FKTF-1b's intracellular localization, and thereby its access to genomic response elements, is regulated by phosphorylation in a similar manner to DAF-16 and other FOXO-class transcription factors.

Dominant interfering transgenes cause altered intestinal morphology in L1s
Its anatomical localization and intra-cellular trafficking support the hypothesis that FKTF-1b is an ortholog of DAF-16 and that through it, insulin-like signaling regulates S. stercoralis' larval development. More conclusive testing of this hypothesis requires experimental manipulation of gene function and evaluation of phenotypic outcomes. Thus far, S. stercoralis, like many other parasitic nematodes, has proven insensitive to targeted gene silencing via RNAi [20]. Therefore, we opted for an approach based on transgenesis in which we express altered forms of FKTF-1b designed to interfere with the function of the endogenous transcription factor. Two such mutant proteins, encoded by plasmids pPV251 and pPV298, respectively ( Figure S1), are tagged with GFP and carry the four 'phospho-null' mutations described above, causing them to be sequestered in the nucleus where they presumably out-compete native FKTF-1b for response elements in the genome. In addition, both mutant proteins are truncated within the C-terminal domain immediately downstream of the fourth regulatory phosphorylation site, ablating key transactivator binding motifs. In one of the dominant interfering proteins, encoded by pPV251 and dubbed GFP::FKTF-1b(dominant-repressor), the truncated C-terminal domain is fused to the repressor domain of Ce-PIE-1, a protein responsible for the transcriptional repression characterizing the germline precursor of C. elegans [21]. In the other mutant protein, encoded by pPV298 and dubbed GFP::FKTF-1b(dominant-negative), the truncated Cterminal domain is not linked to Ce-PIE-1.
Upon hatching, larvae expressing either of the dominant interfering constructs were shorter but virtually identical in form to larvae expressing GFP-tagged wild-type FKTF-1b at similar levels ( Figure S2), indicating that the S. stercoralis transcription factor, like DAF-16 in C. elegans [22], does not play a significant role in embryonic development. By contrast, at 24 hours, S. stercoralis L1 expressing either of the dominant interfering mutants of FKTF-1b exhibited phenotypic changes in the form and apparent function of their intestinal cells, with these being most evident in larvae expressing the dominant-repressor construct. These phenotypes ranged from flattening of the normally apically rounded intestinal cells and a decrease in the number of cytoplasmic storage granules in the presence of GFP::FKTF-1b(dominant-negative) (compare Figure 3J to Figure 3H and Figure 3I) to an almost complete loss of intestinal cell architecture and of cytoplasmic storage granules in the presence of the GFP::FKTF-1b(dominant-repressor). Perhaps due to compromised intestinal cell function, S. stercoralis L1 expressing GFP::FKTF-1b(dominant-repressor) exhibited significant (P,0.01) growth retardation at 24 hours ( Figure 3N). Owing to the severity of the associated phenotypes, none of the larvae expressing GFP::FKTF-1b(dominant-repressor) survived beyond the L1. The fact that S. stercoralis L1 expressing comparable levels of wild-type FKTF-1 tagged with GFP ( Figure 3B, 3E, and 3I) were morphologically similar to untransformed larvae ( Figure 3A and 3H) argues against the observed phenotypes being due to nonspecific effects of recombinant protein expression. Therefore, it is clear from these findings that FKTF-1 is necessary for normal development of intestinal cells in pre-infective larvae of S. stercoralis and specifically for accumulation of storage granules, which may contain reserves necessary for survival of the L3i.

Dominant interfering transgenes cause aberrant morphogenesis in S. stercoralis L3
With regard to the dauer hypothesis and the role of insulin signaling in infective larval development, the most significant results in the present study were the morphogenetic changes seen in L3 expressing the GFP::FKTF-1b(dominant-negative) transgene. Under the null hypothesis, all of our transgenic larvae should develop to L3i. While L3i expressing the wild-type GFP::FKTF-1b fusion protein were morphologically identical to their nontransgenic counterparts ( Figure 4A and 4B, compare to Figure  S3A), L3 expressing the dominant-negative transgene ( Figure 4C and 4D) exhibited some indications of bypassing developmental arrest and failing to undergo the pharyngeal and intestinal remodeling characteristic of L3i. Three of the 11 transgenic L3 appeared to initiate an aberrant molt to the fourth stage as evidenced by the existence of a pointed tail inside a notched L3i cuticle cast ( Figure 4E and 4F compared to wild-type Figure 4G). The notched tail is characteristic of infective larvae and is created by pairs of 'L3i-specific' alae [15]. Another L3 expressing the dominant-negative construct exhibited an elongated rhabditiform pharynx complete with a grinder-like structure ( Figure 4H and 4I) instead of the expected filariform pharynx ( Figure S3B and Figure  S3C). Incomplete remodeling of the rhabditiform pharynx and initiation of a supernumerary molt in culture are consistent with expression of the interfering FKTF-1b transgenes in the pharyngeal procorpus and the hypodermis. Initiation of ecdysis by L3 in combination with retention of some rhabditiform pharyngeal characteristics as we observed suggests that worms expressing GFP::FKTF-1b(dominant-negative) were developing in the direction of a second-generation free-living L4. While such a form occurs in some strongyloidoid species (e.g. Strongyloides planiceps, Parastrongyloides trichosuri), it does not exist in the natural life cycle of S. stercoralis [8,23].
Five of the 11 transgenic L3 expressing the dominant-negative construct exhibited changes consistent with a failure to remodel the free-living intestine into the darkened, radially constricted intestine of the L3i. In some cases, the L3 intestine retained bacteria ( Figure 4E) and in others, it failed to constrict and close ( Figure 4I to the left of the black triangle). The incompletely remodeled intestine seen in the transgenic L3 is consistent with the defects in intestinal structure seen in the L1. Together, these data indicate that FKTF-1b is required for the proper remodeling of the pharynx and the intestine of a free-living larva into structures characteristic of the infective larva.
Our findings support the 'dauer hypothesis' [7] by showing that the forkhead transcription factor FKTF-1b, presumably under the control of insulin-like signaling, regulates infective larval development in Strongyloides stercoralis in a manner similar to the dauer regulatory functions of DAF-16 in Caenorhabditis elegans. Furthermore, in this study, we have demonstrated the utility of transgenesis in S. stercoralis for investigation not only of temporal and spatial patterns of gene expression, but also of endogenous gene function. This work opens new avenues of inquiry into the genes involved in the shift between free-living and parasitic states in Strongyloides stercoralis and ultimately into the evolution of parasitism in nematodes generally.

Parasite maintenance and culture
The UPD strain of Strongyloides stercoralis was maintained in immuno-suppressed dogs and cultured as described [24]. Freeliving adult S. stercoralis were isolated from two-day-old coprocultures via Baermann funnels. The worms were washed twice with sterile deionized water to reduce carryover of fecal bacteria and plated on Nematode Growth Medium (NGM agar) plates seeded with Escherichia coli OP50. All cultures of S. stercoralis were incubated at 22uC unless otherwise noted.
Transgene construction gfp reporters. In C. elegans, the promoter for splice form daf-16b exists in the large intron between exons 4 and 5 [25]. The fktf-1 genomic sequence also includes a large, approximately 6 kb, intron between the end of the fktf-1a-specific exons (exons 1-3) and the beginning of the fktf-1b-specific exon 4 [11]. We used 2.6 kb sequence directly 59 of the fktf-1b-specific ATG codon plus the first 90 bp of the coding sequence as the fktf-1b promoter element. Using the primers SsHindBprF1 and SsBprPstR2 (Table S1), we amplified the putative fktf-1b promoter from plasmid pPV57. The PCR product was cloned into pAJ01, a promoter-less vector containing the gfp coding sequence fused to the Strongyloides era-1 39 UTR, with HindIII and PstI restriction enzymes using standard techniques. The resulting fktf-1b::gfp::era-1 reporter construct was designated pPV232 ( Figure S1).  To create the fusion protein GFP::FKTF-1b, the original construct pPV232 was altered via site-directed mutagenesis (QuikChange Site-Directed Mutagenesis, Stratagene, La Jolla, California, USA), to remove the terminator for the gfp coding region and to introduce the unique restriction site BspEI using the primers GFPnoTER-BspEF and R (Table S1). The clone with the appropriate mutation and the new restriction site was designated pPV233 (fktf-1b::gfp(no ter)::era-1). The cDNA sequence for fktf-1b had been previously cloned into a plasmid vector pPV207. The coding region of fktf-1b was amplified from pPV207 using the primers: BspE-FkBF (Table S1) and FkB-AvrIIR (Table S1) and inserted into pPV233 using the restriction enzymes BspEI and AvrII via standard techniques. The clone containing the fulllength fktf-1b promoter with gfp fused N-terminal to and in frame with the fktf-1b coding region was designated pPV234 (fktf-1b::gfp::fktf-1b::era-1) ( Figure S1).
Phosphorylation mutant constructs. Phosphorylation-null and phosphorylation-mimetic constructs were created from pPV234 using site-directed mutagenesis (QuikChange Site-Directed Mutagenesis, Stratagene, La Jolla, California, USA). All phosphorylation-null mutagenesis reactions were carried out on the cDNA sequence of fktf-1b in the pCR-BluntII-TOPO cloning vector (Invitrogen, Carlsbad, California, USA) to improve the efficiency of the reactions. The phosphorylation-null construct was created in three stages, each stage mutating the key regulatory residues in one of the three functional domains. Primers FK-1bPsite1F2 (Table S1) and FK-1bPsite1R (Table S1) mutated threonine 22 to alanine in the N-terminal domain. Primers SsD16Psite2F and SsD16Psite2R (Table S1) mutated both serine 238 and threonine 240 to alanine. Primers FkCPsite3KOF2 and FkCPsite3KOR2 (Table S1) mutated the phosphorylation site serine 317 to alanine in the C-terminal transactivation domain. Once all four sites were mutated, the wild-type fktf-1b coding region was removed from the pPV234 vector, and the phosphonull fktf-1b(4A) coding region was cloned in using BspEI and AvrII sites using standard techniques. The phospho-null construct was designated pPV243 (fktf-1b::gfp::fktf-1b(4A)::era-1) ( Figure S1).

Dominant-interfering
constructs. The phospho-null construct, GFP::FKTF-1b(4A), which sequesters in the nucleus, still contains functional forkhead (DNA binding) domain and Cterminal transactivation domains and thus should out compete endogenous FKTF-1b for DNA binding sites. Truncation of the C-terminal domain in the phospho-null GFP::FKTF-1b, should create a protein constitutively bound to DNA but unable to activate transcription. Subsequently, replacing the endogenous transactivation domain of FKTF-1b with a repressor domain should create a protein actively repressing transcription specific to FKTF-1b's DNA binding domain.
The transactivation domain was truncated using an introduced ClaI site immediately downstream of the regulatory phosphorylation site S317. The phospho-null fktf-1b(4A) in the pCR-BluntII-TOPO vector (Invitrogen, Carlsbad, California, USA) was again used for mutagenesis using the primers FKTFtrcMutCla1F and FkMutTrcCla R2 (Table S1) to introduce the ClaI site in the desired location. The plasmid was cut using BspEI and ClaI to release the fktf-1b(4A, truncated) coding region (BspEI-fktf-1b(4a, trunc)-ClaI).
The coding region for the C. elegans gene pie-1 was amplified from RNA prepared from a pool of C. elegans N2 worms using the primers Ce-pie-1ATGF and Ce-pie-1StopR (Table S1) and cloned into a pCR-BluntII-TOPO cloning vector (Invitrogen, Carlsbad, California, USA). The active repressor domain of PIE-1 consists of 81 amino acids at the C-terminus (CTD) [21]. Using the primers ClaI-pieCTDF and Ce-pie-1teravrR (Table S1) the repressor domain alone was PCR amplified using PfuTurbo (Stratagene, La Jolla, California, USA). The resulting PCR product was digested with ClaI and AvrII for cloning, (ClaI-pieCTD-AvrII).

Transformation of S. stercoralis
Adult female S. stercoralis were transformed with transgene encoding plasmids via intra-gonadal microinjection using standard protocols [24,26]. Coding plasmids were injected at a concentration of 10-100 ng/ml with non-coding plasmids being used as necessary to make up the total DNA concentration to 100 ng/ml. Following injection, worms were transferred to clean NGM OP50 plates with an excess of males and incubated at 22uC.
For general expression patterns, plates were scored at 24 hour intervals for adult survival and frequency of transgene expression among F1 progeny. For specific time points, adults were transferred to clean NGM OP50 plates at three to five hour intervals to obtain egg cohorts. All plates with eggs were checked at hourly intervals for the presence of transgenic progeny. When the time of hatch was known, transgenic larvae were transferred to clean plates marked with the time point and examined after the  Table S2 for counts). (N) All larvae were measured from the tip of the buccal cavity to the tip of the tail using the ImageJ program [27]. A minimum of 10 larvae per category per timepoint, with the exception of GFP::FKTF-1b(wild-type) L3i where n = 8, were examined and measured (Table S2). At 1 hour post-hatch, larvae expressing either of the dominant-interfering transgenes were slightly shorter than the wild-type larvae (P,0.05). At 24 hours post-hatch, only the GFP::FKTF-1b(dominant-repressor) transgene-expressing larvae were significantly shorter than the wild-type (P,0.01). The transgenic L3i expressing the GFP::FKTF-1b(dom-neg) transgene were not significantly shorter than the wild-type L3is (P.0.05). doi:10.1371/journal.ppat.1000370.g003 appropriate interval. Transgenic progeny for which the time of hatch was not known were used for analysis of L3i development. The low transformation rate, ,5%, of S. stercoralis larvae made it impractical to accumulate sufficient numbers of individuals for both phenotypic study (Table S2 for phenotype counts) and confirmation of a full-length gfp::fktf-1b transcript. However, we have confirmed that C. elegans transformed with the same gfp::fktf-1b coding sequence under the control of the daf-16 promoter exhibit GFP fluorescence and express a full length transcript encoding the fusion protein (data not shown). Current methods only allow us to observe transgene expression in F1 generation following transformation.

DIC and Fluorescent Microscopy
Transgenic larvae were identified based on GFP fluorescence using an Olympus SZX12 stereomicroscope with coaxial epifluorescence. For more detailed examination of particular tissues and individual cells, larvae were immobilized on 4% Agar Noble (Sigma, St. Louis, Missouri, USA) pads in 10 mM (L1) or 20 mM (L3i) levamisole and observed using an Olympus BX60 compound microscope equipped with Nomarski Differential Interference Contrast (DIC) optics and epifluorescence (Olympus America Inc., Center Valley, Pennsylvania, USA). Specimens were imaged with a Spot RT Color digital camera and images were processed using either the Spot Advanced image analysis software package (Diagnostic Instruments, Inc., Sterling Heights, Michigan, USA) or Adobe Photoshop 7.0. All image-processing algorithms (e.g. brightness and contrast adjustments) were applied in a linear fashion to the entire image.

Measurements
Worm lengths were measured using the ImageJ program available from the National Institutes of Health (http://rsb.info. nih.gov/ij/) [27]. Calibrations were done by determining the distance of 10 mm on a micrometer in pixels and then setting the scale in the program. All measurements were done in duplicate using the freehand line option, taking the average of the results for analysis.

Statistical Analysis
As categorical data, expression patterns, localization and phenotypes, were analyzed using x 2 tests. Analysis of the expression patterns of the fktf-1b promoter constructs was based on the null hypothesis that the expression patterns were not specific to the hypodermis and the pharynx. Categories of intracellular GFP localization were analyzed based on the null hypothesis that the phosphorylation status of the FKTF-1b protein had no effect on its localization. In order to compare the mean lengths of wild-type and transgenic larvae, we used the Mann-Whitney test, which makes no assumptions as to the population distribution of the observations. Figure S1 Diagrams of fktf-1b constructs used to transform S. stercoralis. pPV232 encoding the fktf-1b::gfp transcriptional reporter. pPV234, the GFP::FKTF-1b(wt) fusion protein expression vector. pPV243 (GFP::FKTF-1b(4A)) has all four canonical Akt/PKB phosphorylation sites mutated to alanine. pPV244 (GFP::FKTF-1b(S238E/T240E) has the phosphorylation sites in the forkhead domain changed to the phospho-mimetic glutamic acid. pPV251 (GFP::FKTF-1b(dom-rep)) and pPV298 (GFP::FKTF-1b(domneg)) both contain the four alanine mutations. pPV251 encodes a chimeric fktf-1b with the repressor domain from Ce-pie-1 replacing the endogenous transactivation domain. pPV298 is truncated just downstream of the fourth regulatory phosphorylation site and thus lacks either a transactivation or a repressor domain.     Wild-type and transgenic larvae examined for developmental abnormalities. Images of larvae at 1 hour, 24 hours, and L3i timepoints were examined for intestinal structure abnormalities, pharyngeal abnormalities, presence of storage granules in intestinal cells, and overall body integrity. Intestinal structure abnormalities were defined as atrophy of the intestine or structural defects. Pharyngeal abnormalities included loss of musculature, terminal bulb irregularities, and loss of synchronized contractions. The larva was positive for granules if the majority of intestinal cells near the primordial gonad contained granules. Overall body damage takes into account loss of cells outside of the intestine, damaged tail architecture, and any structural abnormality other than in the intestine and pharynx. Found at: doi:10.1371/journal.ppat.1000370.s005 (1.12 MB TIF)