post-embryonic development

Background pha-4 encodes a forkhead box (FOX) A transcription factor serving as the C. elegans pharynx organ identity factor during embryogenesis. Using Serial Analysis of Gene Expression (SAGE), comparison of gene expression profiles between growing stages animals and long-lived, developmentally diapaused dauer larvae revealed that pha-4 transcription is increased in the dauer stage. Results Knocking down pha-4 expression by RNAi during post-embryonic development showed that PHA-4 is essential for dauer recovery, gonad and vulva development. daf-16, which encodes a FOXO transcription factor regulated by insulin/IGF-1 signaling, shows overlapping expression patterns and a loss-of-function post-embryonic phenotype similar to that of pha-4 during dauer recovery. pha-4 RNAi and daf-16 mutations have additive effects on dauer recovery, suggesting these two regulators may function in parallel pathways. Gene expression studies using RT-PCR and GFP reporters showed that pha-4 transcription is elevated under starvation, and a conserved forkhead transcription factor binding site in the second intron of pha-4 is important for the neuronal expression. The vulval transcription of lag-2, which encodes a ligand for the LIN-12/Notch lateral signaling pathway, is inhibited by pha-4 RNAi, indicating that LAG-2 functions downstream of PHA-4 in vulva development. Conclusion Analysis of PHA-4 during post-embryonic development revealed previously unsuspected functions for this important transcriptional regulator in dauer recovery, and may help explain the network of transcriptional control integrating organogenesis with the decision between growth and developmental arrest at the dauer entry and exit stages.


Background
At the second larval molt, C. elegans may arrest development at the dauer stage in response to starvation and overcrowding, but can resume development to the adult when an environment favoring growth is encountered [1]. Entry into and exit from the dauer stage are determined by environmental cues, such as temperature, food supply and a constitutively released, dauer-inducing pheromone [2][3][4]. Genes involved in dauer formation are called daf genes. Dauer-constitutive (Daf-c) mutants form dauer larvae in an environment with abundant food, whereas dauerdefective (Daf-d) mutants fail to enter the dauer stage when they are starved or overcrowded. Most Daf-c mutants are temperature-sensitive, revealing the natural temperature dependence of dauer formation [5].
As a step toward understanding the genetic basis of diapause and longevity, Jones et al. [22] compared gene expression profiles of dauer larvae and mixed-stage, growing populations by SAGE. Transcripts that were enriched either in dauer larvae or in mixed stages were identified [22]. SAGE tags corresponding to pha-4, which encodes a FOXA transcription factor homolog, were detected in the dauer stage (8 tags), but not in mixed stages (no tags).
PHA-4 is regarded as the organ identity factor for the C. elegans pharynx [23][24][25]. In pha-4 mutants, pharyngeal cells are transformed into ectoderm and the development of mutant animals is arrested after hatching [25]. pha-4 mRNA is highly enriched in both the pharyngeal and intestinal primordia of the embryo, and low levels of pha-4 transcripts can be detected in the L3/L4 larval somatic gonad [26]. Candidate PHA-4 target genes in pharyngeal development have been identified [27], and analysis of target promoter sequences revealed several cis-regulatory elements. These elements are targets of unknown transcription factors, which function coordinately with PHA-4 to modulate gene expression in different pharyngeal cell types and at different developmental stages [28,29]. Ao et al. [28] found that the DAF-12 nuclear hormone receptor is one of these transcription factors. DAF-12 and PHA-4 function together to either activate or inhibit myo-2, the pharyngeal muscle gene, in response to environmental and developmental cues [28]. Thus, PHA-4 is an important regulator that is expressed in several cell types and controls a wide range of gene expression.
To assess PHA-4 functions in dauer larvae, we treated Dafc mutants with pha-4 RNAi. We found that PHA-4 is essential for dauer recovery, gonad and vulva development. The hermaphrodite somatic gonad includes distal tip cells, gonadal sheath cells, spermathecae and uterus. Distal tip cells regulate gonadal arm elongation and the switch from mitosis to meiosis in germline cells through the GLP-1/ Notch signaling pathway [30]. Gonadal sheath cells have physical contacts with the germline; the muscle components convey contractile properties required for ovulation. Moreover, the sheath cells, spermathecal and uterine precursor cells play regulatory roles in germline development, such as promoting germline proliferation, exit from pachytene, and/or gametogenesis [31]. Identification of PHA-4 targets in the somatic gonad should help explain the PHA-4 gonadal phenotype. The vulval transcription of lag-2, which encodes a ligand for the LIN-12/Notch lateral signaling pathway, is inhibited by pha-4 RNAi, indicating that LAG-2 functions downstream of PHA-4 in vulva development.

PHA-4 is required for dauer recovery, gonad and vulva development
SAGE revealed that pha-4 transcription is elevated in dauer larvae [22], suggesting that, in addition to pharyngeal organogenesis, PHA-4 may function in dauer formation or recovery. We used semi-quantitative RT-PCR to compare pha-4 transcript levels in mixed-stages, dauer, and at different times during dauer recovery. pha-4 transcripts were much more abundant in the dauer stage than in mixed stages, and were also detected shortly after dauer larvae started to recover. pha-4 transcripts decreased during resumption of development (Fig. 1). Figure 1 pha-4 transcripts are enriched in dauer larvae. Semiquantitative RT-PCR experiments were performed to determine levels of pha-4 and rpl-21 mRNA. rpl-21, which encodes a ribosomal protein, serves as the internal control for equal loading. Experiments were performed three times with consistent results using three independent RNA preparations. Lane 1, mixed stages; lane 2, dauer; lane 3, 4 and 5, 2 hours, 24 hours and 36 hours after placement in food, respectively. Dauer larvae begin to feed approximately three hours after exposure to food.

pha-4 transcripts are enriched in dauer larvae
To determine PHA-4 function in the dauer stage, we knocked-down pha-4 expression by feeding Daf-c mutant animals E. coli expressing pha-4 double-stranded RNA [32] from the time of hatching from the egg. As a control, we fed the same strains with E. coli that carries the RNAi vector (L4440) without any insert. Semi-quantitative RT-PCR showed that pha-4 transcripts were efficiently reduced by the RNAi treatment ( Fig. 2A).
Dauer larvae with pha-4 RNAi treatment showed normal dauer characteristics, including SDS resistance and a constricted pharynx (not shown), suggesting that PHA-4 is not required for dauer formation. However, when the dauer larvae were transferred to an environment favoring growth (fresh food and low temperature), they exhibited decreased recovery compared to the control, 2 -32% vs. 96 -99% (Fig. 2B, Table 1). pha-4 RNAi-treated animals that resumed development to the adult had abnormal oocytes and embryos, and protruding vulvae (Fig. 2D, 2F). Adults fed the RNAi control under the same conditions were normal (Fig. 2C, 2E). Thus, PHA-4 plays essential roles in dauer recovery and gonad/vulva development.
To assess the possible role of PHA-4 in gonad and vulva development, a pha-4::gfp translational fusion carrying the 1.5 kb promoter (the first two exons and introns, and part of the third exon of pha-4 fused in-frame with the gfp coding region) was injected into the germline of daf-2(e1370). GFP was present in distal tip cells (DTCs), intestinal cells (Fig. 3A) and neurons (not shown) in dauer larvae, and in the spermathecae and uteri of L4 larvae (Fig.  3B).
PHA-4 is required for dauer recovery, gonad and vulva development Dauer larvae formed constitutively at 25°C in the presence of pha-4 RNAi from hatching showed significantly decreased recovery upon downshift to 15°C compared to control daf-7(e1372), daf-1(m40), daf-2(e1370), and age-1(m865) mutants. Percentages of dauer recovery, numbers of animals scored, and p values for t-tests were shown in Table 1. The entire experiment was performed twice with triplicates for each treatment.

DAF-16 is also important for dauer recovery, gonad and vulva development
Previous studies showed that DAF-16B, one of the daf-16 gene products, is expressed in the pharynx, somatic gonad and tail neurons [33]. The overlapping expression patterns of daf-16b and pha-4 and the essential roles of DAF-16 during dauer development suggest the possibility that DAF-16 may be involved in dauer recovery. Since daf-16 is a Daf-d mutant, we used a daf-16(mgDf47); daf-7(e1372) strain to examine dauer recovery and adults recovered from the dauer stage. The daf-16(mgDf47) null mutant carries the daf-16 mutation lacking the coding sequence for DNA binding domains [33]. The daf-7 Daf-c mutant affects the TGF-β pathway, and is not suppressed by daf-16 mutations [34]. The daf-16; daf-7 dauer larvae formed constitutively at 25°C showed significantly decreased recovery upon downshift to 15°C compared to the daf-7(e1372) dauer larvae (Fig. 4A, Table 1). Animals, which resumed development to the adult, showed defects in the gonad and vulva similar to pha-4 RNAi-treated animals ( Fig. 4B, 4C).

DAF-16 and PHA-4 function in parallel pathways to regulate dauer recovery
Since pha-4 transcripts are elevated in dauer larvae, transcription factors in dauer signaling pathways are candidates for promoting pha-4 expression in dauer larvae. The overlapping expression patterns and similar loss-of-function phenotypes of daf-16 and pha-4 suggest the possibility that DAF-16 may be a regulator of pha-4 transcription.
pha-4 is transcribed into three major transcripts: pha-4a, b and c [26]. pha-4 b and c are trans-spliced with the SL1 leader mRNA at the beginning of exons 2 and 3, respectively (Fig. 5A). Introns 1 and 2 are large (2247 bp and 1329 bp, respectively), considering that the median size of confirmed introns in the C. elegans genome is 65 bp pha-4 is expressed in the somatic gonad  We mutated the element in the pha-4::gfp construct, but did not observe altered gfp expression in transgenic animals (data not shown). It is possible that pha-4a expression is not affected by the mutation. To test whether pha-4b and pha-4c are regulated through this binding site, we cloned a 3.8-kb DNA fragment of pha-4 (which carries the first two introns, the second exon and part of the third exon) in frame with the gfp coding sequence. We then mutated the potential DAF-16 binding site, and injected these two constructs into daf-2 animals. Animals carrying the construct with the normal DAF-16 binding site showed gfp expression only in head neurons (Fig. 5C); whereas animals carrying the mutated DAF-16 binding site showed very weak gfp expression in the same cells (Fig. 5D). This suggests that the potential DAF-16 binding site in the second intron of pha-4 has regulatory activity in vivo.
DAF-16 is activated and localized to the nucleus shortly after food deprivation [36]. We compared pha-4 mRNA levels between well-fed and starved L3 larvae, the alternative third stage to dauer larvae. Starved wild-type L3 larvae, which have high DAF-16 activity, had higher levels of pha-4 transcripts than well-fed wild-type L3 larvae, which have low DAF-16 activity (Fig. 5E). However, starvation also increases pha-4 transcription in the daf-16(mgDf47) null mutant background, although both well-fed and starved daf-16 animals showed slightly lower pha-4 transcription compared to wild-type N2 animals (Fig. 5E).
These results suggest that there may be more than one transcriptional regulator promoting pha-4 transcription in response to starvation.
To test the genetic interaction between DAF-16 and PHA-4, we treated the daf-16; daf-7 double mutant, which is Daf-c, with the control or pha-4 RNAi in dauer recovery assays. pha-4 RNAi treatment further reduced recovery of daf-16; daf-7 dauer larvae (Fig. 5F, Table 1). This result suggests that DAF-16 and PHA-4 may function in parallel pathways to regulate dauer recovery.

Knocking-down pha-4 by RNAi inhibits lag-2 transcription in vulval cells
lag-2 is a candidate PHA-4 target, since pha-4 and lag-2 are both expressed in DTCs in dauer larvae, and there are two potential PHA-4 binding sites in the 1 kb lag-2 promoter. lag-2 encodes a ligand for Notch signaling pathways in gonad and vulva development [37,38]. LAG-2 expression in DTCs regulates entry into mitosis versus meiosis through the GLP-1/Notch signaling pathway [38]. Dauer larvae carrying an integrated lag-2::gfp reporter exhibited lag-2 expression in IL1 neurons and DTCs (data not shown).
We crossed the integrated lag-2::gfp reporter into the daf-2(e1370) mutant to test the effect of pha-4 RNAi on lag-2 expression in DTCs and to assess possible post-dauer gonadal defects. We found that lag-2::gfp expression in DTCs and IL1 neurons is not affected by pha-4 RNAi treatment, nor did lag-2 RNAi-treated animals show postdauer gonad defects that were observed in animals treated with pha-4 RNAi (data not shown). GLD-1, a KH motifcontaining RNA binding protein, functions downstream of LAG-2 in the GLP-1/Notch signaling pathway in germ-  Table 1. The entire experiment was performed twice with triplicates for each treatment.
lag-2 is also expressed in P6.p and its descendants, the cells that adopt the 1° vulval fate [37,40,41]. LAG-2 is one of the ligands that function through LIN-12/Notch lateral signaling to promote P5.p and P7.p to adopt 2° vulval fate [37]. Previous studies showed that a lag-2 temperature-sensitive (ts) mutant has a protruding vulva (Pvl) when it is grown at the restrictive temperature [42]. The Pvl phenotype is similar to that of pha-4 RNAi-treated animals. Thus, it is possible that LAG-2 functions downstream of PHA-4 in vulva development. Using the same lag-2::gfp transgene as shown in previous studies [40,41], we observed lag-2::gfp expression in four P6.p descendant cells in L4 larvae. pha-4 RNAi greatly decreased lag-2::gfp expression in those P6.p descendant cells ( Figure 6). Therefore, PHA-4 is required for lag-2 expression in vulval cells in vivo. Although electrophoretic mobility shift assays (EMSAs) indicated that PHA-4 binds to the lag-2 promoter in vitro (data not shown), it is not clear whether PHA-4 directly promotes lag-2 transcription in vulval precursor cells. Taken together, the data show that PHA-4 is important for lag-2 transcription in vulva development, but does not affect lag-2 expression in distal tip cells for germline development.

PHA-4 plays essential roles in post-dauer development
PHA-4 is expressed in every pharyngeal cell during embryonic development, and it directly regulates expression of many pharyngeal genes [23,24,[27][28][29]. SAGE and RT-PCR data showed that pha-4 transcription is elevated in the dauer stage. Our RNAi analysis revealed that PHA-4 plays essential roles in dauer recovery, gonad and vulva development. Thus, PHA-4 serves as a key regulator to control different aspects of embryonic and larval development. pha-4 is expressed in the somatic gonad. Previous studies suggested that there is an unknown signal from gonadal sheath cells, uterus and spermathecae precursor cells that regulates germline development [31]. Identification of PHA-4 targets in gonad and vulva development may help characterize this unknown signal. binding site in the non-coding region of pha-4. We have shown that this potential DAF-16 binding site has regulatory function on pha-4 expression in head neurons. However, this regulation is unlikely to be important for the RNAi phenotypes we observed since gene expression in wild-type neurons is refractory to RNAi.

DAF-16 and PHA-4 function in parallel pathways during dauer recovery
pha-4 shows increased transcription under starvation, which also causes increased DAF-16 activities. This observation is consistent with the recent study that PHA-4 plays important roles in dietary restriction-mediated life span extension and pha-4 transcription is increased upon dietary restriction or in the eat-2 mutant, which serves as a genetic mimic of dietary restriction [43]. However, the increased pha-4 transcription upon starvation is independent of daf- 16

Other potential regulators of PHA-4
The regulation of pha-4 expression and activity is complex. pha-4 expression in the intestine is inhibited by the let-7 micro RNA, which is essential for the transition from the L4 larval stage to adult [44]. Interestingly, let-7 also inhibits DAF-12 expression in hypodermal seam cells [44]. Since micro RNAs usually regulate translation of the mRNA, increased mRNA levels do not guarantee elevated activity. It will be interesting to examine whether PHA-4 protein levels and activities are increased in dauer larvae.
There is a tract of 29 C residues in the second intron of pha-4 (Fig. 5A). Deletions in this sequence can be frequently detected in the dog-1 (DEAH helicase) mutant [45], which exhibits germline and somatic deletions in genes containing such poly G/C tracts. However, it is not yet known whether this poly C tract plays a regulatory role in pha-4 expression. Recently, Updike and Mango [46] have reported a screen for suppressors of pha-4 loss-offunction. One of the suppressors (px63) turned out to be an allele of pha-4, with elevated expression. The pha-4(px63) allele carries a 156 bp deletion in the second intron, and an insertion of ≥ 5.3 kb between the 3' end of intron 2 and the middle of exon 5 [46]. Interestingly the deletion overlaps with the polyC tract, and the insertion is likely to overlap with the potential DAF-16 binding site and flanking sequences. Updike and Mango proposed that the DNA sequence rearrangement in pha-4(px63) either disrupts a negative regulatory element or introduces a positive element for pha-4 expression [46]. It will be interesting to test which mutation is responsible for the elevated pha-4 expression and whether the polyC tract or the potential DAF-16 binding site is involved.

PHA-4 is required for vulval transcription of lag-2
lag-2 encodes the ligand for Notch signaling pathways in the gonad and vulva. Our results showed that PHA-4 is important for lag-2 transcription in vulval cells, but does not affect lag-2 expression in distal tip cells for germline development. Vulval transcription of lag-2 is regulated by the inductive EGFR/MAPK signaling pathway, and LIN-31 is the transcription factor functioning downstream of this pathway in vulva development [37]. Although we found that PHA-4 binds an element in the lag-2 promoter in vitro, we did not detect any PHA-4 expression in vulval cells where lag-2 is expressed using the partial PHA-4::GFP translational fusion reporter. More work will be required to determine whether PHA-4 directly promotes lag-2 transcription, or whether it functions through components of the EGFR/MAPK pathway.

Conclusion
DAF-16 and PHA-4 play essential roles during dauer recovery, and PHA-4 functions upstream of LIN-12/Notch signaling during vulval development. The dauer/nondauer decision is determined by food, population density and temperature [5]. The essential functions of PHA-4 in both pharyngeal and reproductive development in response to environmental factors suggest a link between nutrient in-take and reproduction. Hence, the analysis of PHA-4 functions in dauer larvae will help explain how the overall network of transcriptional control is formed, and how different aspects of C. elegans development are integrated.

Strains
Strains were maintained as described by Brenner [47]. Strain names and genotypes of animals used were: GR1329 daf- 16 RNAi A 1.3-kb coding region of the pha-4 gene was amplified from C. elegans genomic DNA using primers 5' CGG AAT TCG TTT TAC CAC TGG CAC CAC 3' and 5' CCC AAG CTT CTG GTA TAC TCC GTT GGT G 3'. The PCR products were cloned into the RNAi vector L4440 between the EcoR I and Hind III sites. The pha-4 RNAi construct was transformed into E. coli HT115(DE3). RNAi bacteria cultivation and double-stranded RNA induction were performed as described by Kamath et al. [32]. In all RNAi assays, E. coli HT115(DE3) carrying the empty RNAi vector L4440 was fed to the same strain as controls.

Dauer recovery assays
About ten Daf-c gravid adults were transferred to RNAi plates, and allowed to lay eggs for 4 to 6 hours before they were removed. The plates were incubated at 25°C for three days until dauer larvae formed constitutively. The dauer larvae were then transferred to fresh RNAi plates and incubated at 15°C to resume development. Animals that became L4 larvae or young adults after 3 days at 15°C were scored as recovered. Gonad and vulva morphologies of these adult animals were examined using a Zeiss Axioscope with Nomarski optics.

RT-PCR
Total RNA was extracted using the Trizol reagent (Invitrogen) following the manufacturer's instructions. Reverse transcription followed by PCR reactions were performed to determine gene transcription levels from different samples as previously described [48]. The following gene-specific primers were used to amplify transcripts of interest: pha-4, 5' GCG GAG CTC ATG AAC GCT CAG GAC TAT CTG 3' and 5' CGC AAG CTT TAG GTT GGC GGC CGA GTT C 3' ; rpl-21, 5' ATG ACT AAC TCC AAG GGT C 3' and 5' TCA CGC AAC AAT CTC GAA AC 3' pha-4::gfp reporters A 5.5-kb DNA fragment that contains the 1.5-kb promoter through part of the third exon of pha-4 was amplified from C. elegans genomic DNA using primers 5' CAA CGA GAG GGC ATG CTG TGA AC 3' and 5' CGG GAT CCT GAT ATG GTT GGT AGT TTA ACG 3'. The PCR products were cloned into the vector pPD95.67 (a gift from Dr. Andrew Fire) between the Sph I and BamH I sites. The pha-4::gfp clone (50 ng/μl) was injected into the ovaries of daf-2(e1370) young adults along with the pRF4 rol-6(su1006) dominant transformation marker (100 ng/μl).