| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
1 Department of Chemistry and Biochemistry and the Molecular Biology Institute, University of California, Los Angeles.
2 Department of Cellular and Structural Biology, University of Texas Health Science Center at San Antonio.
Address correspondence to Pamela L. Larsen, PhD, Department of Cellular and Structural Biology, University of Texas Science Center at San Antonio, San Antonio TX 78229. E-mail: larsenp{at}uthscsa.edu
 |
A |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Key Words: Adult and dauer life span • Dauer formation • Protein L-isoaspartyl methyltransferase • daf-2 • daf-7
The repair methyltransferase has been conserved through evolution, and enzymes from prokaryotes and eukaryotes share a high degree of sequence similarity (9). Genes coding for this enzyme have been found in nearly all of the organisms examined. In Escherichia coli, loss of the pcm methyltransferase can result in reduced stationary-phase survival (10). In mice, animals lacking the pcmt-1 enzyme accumulate proteins containing modified aspartyl residues and start to die of uncontrolled seizures right after weaning, with only 50% remaining alive just before they reach sexual maturity (11–14).
In the nematode Caenorhabditis elegans, the repair methyltransferase is encoded by the pcm-1 gene (15). To elucidate the physiological role PCM-1 plays in C. elegans, a large portion of the gene was deleted using Tc1-transposon-mediated mutagenesis and excision in an him-8(e1439) background (16). The him-8; pcm-1 double mutant is similar to the control him-8 animals in morphology, fertility, and adult life span (16). Interestingly, the absence of the enzyme does not appear to cause a marked accumulation of damaged proteins (17). However, him-8; pcm-1 animals did show two pcm-1-specific phenotypes: pcm-1-null animals are selected against in long-term, competitive population studies, and they exhibit a shorter dauer life span when compared to him-8 dauer larvae (16).
The dauer stage is an alternate third larval stage that C. elegans will enter under conditions of restricted food supply, elevated temperature, or high pheromone concentration due to a dense population (18,19). This stage is characterized by several changes in the animal: They are thinner and longer than L3 larvae, their cuticle is especially thick, and their intestinal cells appear dark (20,21). Dauer larvae do not feed or defecate due to a constricted pharynx and sealed buccal and anal cavities. These features can contribute to their ability to resist environmental and chemical stresses such as desiccation and sodium dodecyl sulfate (SDS) treatment (20). Dauer larvae survive longer than any other wild-type stage (22). Many mutants have been identified that influence 
uer larva 
ormation (21). The phenotype of abnormal dauer larva formation is called Daf. The abnormality can either cause 
onstitutive (Daf-c) entry into the dauer larval stage under conditions where wild-type larvae would grow to reproductive adults or cause defects in uer larva ormation (Daf-d) resulting in morphologically abnormal, fewer, or no dauer larvae formed under inducing conditions.
To adequately understand the phenotypes of repair methyltransferase deletion mutants in the absence of secondary mutations, the pcm-1-null mutation was placed in a wild-type (N2) background. We found the significant phenotypic differences of these mutants in dauer larvae formation as well as longevity. We also found modulation of the pcm-1 dauer defect by several daf-c mutations in the three well-characterized branches of the dauer larva formation pathway (23–27). Mutations in the daf-7 or daf-2 genes each increased dauer longevity, whereas only the latter increases adult longevity 25°C, which suggests that some life-limiting processes in dauer larvae and adults are distinct. In addition, deletion of the pcm-1 gene has no effect on adult life span in either a wild-type or daf-2 mutant background. Finally, we show that overexpression of the pcm-1 gene significantly extends adult life span. These results point to a complex interplay of PCM-1 activity with signaling pathways leading to dauer formation and maintenance.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The mutations used in this study were: Linkage Group (LG) I: daf-16(m26); LG III: daf-2(m41, m596, and e1370), daf-7(m62 and e1372); LG IV: him-8(e1439); LG V: pcm-1(qa201 and tm363), daf-11(m47). To generate a single pcm-1 mutant, the him-8; pcm-1 strain was backcrossed to N2 five times and polymerase chain reaction (PCR) amplification was used to screen progeny for the pcm-1 deletion (16,29). The loss of the him-8 mutation was verified by placing virgin hermaphrodites singly on plates and observing no male progeny from any of the individuals. Double mutants were constructed by mating pcm-1 males with daf-c hermaphrodites (27). F2 Daf-c dauer larvae were cloned, and their progeny were screened for the pcm-1 deletion by PCR. After their progeny were collected, the adults were screened for the pcm-1 mutation by PCR. Several homozygous pcm-1 single- and double-mutant strains were isolated and behaved similarly.
To obtain transgenic rescue strains of the pcm-1 gene, the overlapping pcm-1 and C10F3.4 genes and an approximately 3 kb region upstream of each gene was PCR amplified from C. elegans genomic DNA using the Expand Long Template PCR System (Roche, Indianapolis, IN). The primers used were: forward, 5' TGATCAGTTGGGACCCACCATGAG 3'; reverse, 5' CTGCAGGAGTGCGGTGCTAATTTC 3'. A 10,853 bp PCR fragment was produced and was cloned into the pCR-XL-TOPO vector (Invitrogen, Carlsbad, CA) following the manufacturer's instructions. Transgenic lines were obtained by injection of the construct into the gonad (30) of pcm-1(qa201) adults along with the transformation marker plasmid containing myo-3::GFP. Mutations were introduced into the pcm-1/C10F3.4 rescue construct using the QuikChange II XL site-directed mutagenesis kit (Stratagene, La Jolla, CA). To mutate the methyltransferase active site, the sequence ALDVGS
S was changed to ALDVGS
S. The primers used were: forward, 5' GCTCTTGACGTTGGCTCAGTAAGTGTATATTTGACAGTTTGTATGGC 3'; reverse, 5' GCCATACAAACTGTCAAATATACACTTACTGAGCCAACGTCAAGAGC 3'. The DNA sequence of the mutagenized clones was confirmed. Transgenic pcm-1 nematodes containing the mutated clone demonstrated no PCM-1 enzyme activity.
Dauer Life-Span Assay
Synchronous entry into the dauer stage was initiated by crude pheromone in liquid culture with limiting amounts of bacteria. Eggs were isolated from gravid adults by hypochlorite treatment and allowed to hatch overnight at 25°C in the absence of food. These synchronized L1 larvae were counted, and the density was adjusted to 4000 larvae/mL for N2 and pcm-1 and 2000 larvae/mL for daf-2, daf-7, and daf-2; pcm-1. Induction of dauer formation for N2 and pcm-1 was by high temperature, high pheromone concentration, and low food concentration, whereas for the daf-c strains it was by high temperature and a mutation that signals inappropriate dauer entry in the presence of food alone. N2 and pcm-1 L1 larvae were placed in S Medium that was a mixture of 25% vol/vol fresh and 75% expended from liquid cultures that had contained growing populations for at least 2 weeks and filter sterilized. Each day, for 3 days, the N2 and pcm-1 cultures were given a small amount of food such that the cultures would consume the food within a day. The formation of dauer larvae was verified by visual inspection, and these conditions forced nearly 100% of the N2 larvae to enter the dauer stage. The daf-c larvae were cultured in fresh S Medium with food and were grown at 25°C.
Nematodes were harvested by sucrose flotation (28). Subsequently, dauer larvae were purified by 1% SDS treatment for 45 minutes at room temperature and washed four times in ice-cold M9. For the dauer life-span assay, animals were maintained in M9 at a density of 2000 dauer larvae/mL with ampicillin at 50 µg/mL and nystatin at 100 µg/mL, and were incubated at either 20°C or 25°C. The cultures were aerated on a platform rotary shaker (150 rpm) and sampled every 5 days. The rate of evaporation was measured by weight loss and was slow, such that nearly all loss was due to sampling. To determine the number alive, three separate samples were removed into sterile glass tubes using glass pipettes because we noted significant adhesion of animals to plasticware. To score at least 30 live larvae per plate, an appropriate volume was spotted on plates using glass micropipettes. The total number of dauer larvae was noted, then the number alive was counted based on movement when touched. At later time points when the samples were concentrated, the total number was based on the average value for the early time points, which were fairly stable. No larvae resumed development and grew in the cultures of aging dauer larvae.
Dauer Formation Bioassay
Isolation of crude pheromone and the dauer formation assay were performed as previously described (31,32), with two exceptions: Streptomycin at 50 µg/mL was added to the plates rather than the bacterial slurry, and only 20 µL of OP50 in M9 (0.05 g cells/mL) was spotted on each plate. Three preparations of pheromone were used in the experiments presented here. Each pheromone preparation was titered using N2. This same wild-type strain was included in each experiment as a control. Eggs laid within 2 hours were cultured at 20°C for 48 hours at which time the developing larvae were scored as predominately L2d and L4 larvae. L4 larvae were removed from the plate. After 24 hours, the L2d larvae that had been left on the plate were scored for formation of dauer larvae. This same protocol was used for epistasis analysis except that the culture temperature was 25°C and no pheromone was added to the standard plate medium.
SDS Sensitivity
Larvae exhibiting dauer morphology were assayed for SDS resistance 96 hours after egg fertilization. Larvae were incubated in a 45 µL drop of 1% SDS in M9 for 45 minutes at room temperature on the lid of a sterile plate (20). The animals were then pipetted to a seeded plate. The number of larvae in the spot was scored, and the animals were allowed to recover at 15°C or 20°C, depending on the strain. Two to four days later, the number and larval stage of the live animals were scored. The number dead was also determined. Dead animals were the size of dauer larvae, did not move from the spot, and did not move when touched with a platinum wire pick.
Brood Size and Adult Life-Span Assays
The reproductive capacity was determined at 20°C. Individual adults were moved every 12 hours during the fertile period to a new plate with a thin bacterial lawn to aid in egg counts (27). Eggs were counted on each plate and then allowed to develop and counted again as L4 larvae. The life-span assays were conducted at 25°C, and Day 1 is the first day of adulthood (27). The animals were scored daily, excluding animals that had internal hatchlings, herniated gonads, or were lost.
|
RESULTS |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
igh 
ncidence of 
ale progeny from adult hermaphrodites. This him-8(e1439); pcm-1(qa201) strain showed a reduced brood size compared to N2 wild-type animals but a similar brood size to that of the him-8(e1439) control (16). Subsequent large-scale RNA interference (RNAi) experiments suggested that interfering with pcm-1 expression may cause 14% embryonic lethality (33). To dissect pcm-1 phenotypes from him-8 phenotypes and to detect the possible low levels of embryonic lethality, we backcrossed the him-8; pcm-1 strain to the N2 wild-type strain and generated a pcm-1(qa201) single mutant. We found the brood size and hatching efficiency of pcm-1(qa201) to be 274 ± 73 (n = 16) and 96% ± 2 (n = 8), respectively. This is not significantly different from the brood size of 286 ± 47 (n = 16) and hatching efficiency of 94% ± 5 (n = 12) for N2 at 20°C. Thus, deletion of pcm-1 does not reduce the number or viability of the eggs produced.
Reduced Life Span of pcm-1 and Increased Life Span of daf-7 and daf-2 Dauer Larvae
Previous work noted a reduced median survival of 24.5 days for him-8; pcm-1 dauer larvae compared to 27 days for him-8 dauer larvae (16). These median survival times are substantially shorter than the 45 days first reported for N2 wild type (22). To dissect the pcm-1 phenotypes from him-8 phenotypes, we examined the single mutant pcm-1(qa201) and compared it to N2. In addition, this allowed for comparisons to the previously published data on wild-type dauer longevity (22). In both of the previous experiments, animals were allowed to form dauer larvae as populations increased and food was depleted over a 10- to 14-day period (16,22). Here, survival may have been artificially increased if any of the dauer larvae resumed development and produced progeny. In our experiments, entry into the dauer stage was tightly controlled by pheromone treatment of synchronized L1 larvae, and all larvae entered the dauer stage within a 12-hour period. Complete absence of food during the dauer survival assay prohibited recovery and growth to reproductive adults thereby maintaining age synchrony.
In our synchronized-entry dauer life-span assay, the survival of N2 dauer larvae at 20°C was similar to that previously observed with 50% survival at 40 days (compared to 45 days) and with a maximal survival of 75 days (compared to 80 days) (Figure 1A) (22). As expected, pcm-1(qa201) dauer larvae showed reduced survival, but they displayed even shorter life spans than those described previously (Figure 1A) (16). Dauer larvae with a deletion in the pcm-1 gene had a median life span only one quarter of the N2 dauer life span at 20°C. At the beginning of the assay, the larvae were treated with SDS to select for only the dauer larvae, because dauer-like or younger larvae are killed by this treatment. All pcm-1 dauer larvae were dead by day 40, whereas 50% of N2 dauer larvae remained alive at that time. These results indicate that PCM-1 plays a major role in survival of dauer larvae.
|
As poikilotherms, many processes in C. elegans, such as growth and adult longevity, are affected by temperature (35,36). When we assayed dauer life span at a higher temperature, we found that the metabolically depressed dauer larvae do respond to temperature. There is a substantial decrease in life span for all genotypes at 25°C (Figure 1B). The median survival for wild-type dauer larvae was only 27 days at 25°C compared to 40 days at 20°C. The median survival for daf-7 at 25°C is 33 days, which is also reduced from that observed at 20°C. The median survival of 35 days for the daf-2 and daf-2(m41); pcm-1 dauer larvae show that they are still long-lived at 25°C, compared to N2. The maximum life span for pcm-1 mutant dauer larvae was reduced by half. This short dauer life-span phenotype is suppressed by a mutation in the daf-2 gene. The dramatic reduction in dauer life span observed with a 5° increase in cultivation temperature is surprising, considering the perception that the dauer stage is an unchanging stage that is resistant to environmental insults.
Deletion of the pcm-1 Gene Inhibits Dauer Formation in Response to Pheromone
To understand the reduced survival of pcm-1 mutant dauer larvae, we first tested for a possible defect in dauer formation. Wild-type animals that sense pheromone and have a limited food supply will enter into the dauer stage (19). We found that pcm-1 mutants formed fewer dauer larvae in response to pheromone relative to N2 larvae (Table 1). With 20 µL of pheromone per plate at 20°C, pcm-1 formed only 24% of the dauer larvae formed by N2. We saw no dauer larva formed by Daf-d daf-16(m26) larvae (negative control) and observed defects in dauer formation at different concentrations of pheromone (Table 1, Figure 2) (23). In addition, we analyzed the him-8; pcm-1 strain for the Daf-d phenotype using 60 µL of crude pheromone. The him-8 mutation does not appear to affect dauer formation because the result for the single mutant is indistinguishable from N2 in response to pheromone. However, the double mutant him-8; pcm-1 formed only 62% of the dauer larvae formed by the single mutant him-8 at this high pheromone concentration (Table 1). We were also able to confirm the effect of the pcm-1 mutation on dauer formation using a second deletion allele, pcm-1(tm363) (Table 2). Finally, we performed a rescue experiment by preparing transgenic worms with a wild-type clone of the pcm-1 locus in the pcm-1 deletion strain. We also prepared transgenic worms with a mutated clone of the pcm-1 gene where two glycine residues, which are essential for enzymatic activity, were changed to valine residues. No methyltransferase activity could be detected in the mutant transgenic worms where activity was present in the wild-type transgenic worms (data not shown). Transgenic larvae were then tested in a pheromone-induced dauer formation assay. Comparison of the control transgenic strain, PL50, and a strain carrying a wild-type clone of the pcm-1 locus, PL51, shows substantially more dauer larvae in the latter (Table 2). We conclude that this clone rescues the dauer formation defect of pcm-1(qa201). This rescue depended on the transgene having the wild-type pcm-1 sequence; animals with the mutant transgene (PL54 and PL55 strains) formed fewer dauer larvae than did animals with the wild-type transgene (PL51 strain) (Table 2). These results clearly demonstrate the importance of an active PCM-1 methyltransferase in dauer formation.
|
|
|
We then analyzed the interaction of the pcm-1 and the daf-2 genes with pheromone induction of dauer formation because pcm-1 larvae respond poorly and some daf-c mutations result in animals that are more sensitive to pheromone than N2 (32). We found that daf-2(m41) formed transient dauer larvae with no pheromone at 20°C, and formed nearly 100% dauer larvae with only 5 µL of pheromone per plate at 20°C (data not shown). In the double mutant, we found that the daf-2 mutation enables pcm-1-null mutants to form more dauer larvae than the single pcm-1 mutant in response to pheromone (Table1). However, the pcm-1 mutation reduces the ability of daf-2 mutants to be sensitized to form dauer larvae by pheromone (Table 1). These results suggest that animals with the daf-2(m41) allele, like N2 animals, depend on PCM-1 function to reliably execute dauer formation in the presence of pheromone at 20°C.
SDS Sensitivity of pcm-1 Dauer Larvae
The formation of dauer-like or partial dauer larvae has been described for mutations in other daf-d genes; these larvae have abnormalities in specialized dauer structures including the alae, lateral hypodermal cells, and the pharynx, as well as sensitivity to 1% SDS (24,38). We examined the pcm-1 dauer larvae alae and lateral hypodermal bodies by differential interference contrast microscopy, and no obvious differences were observed. Measurements of the pharynx were made, and we detected no differences in the maximal width of the anterior bulb (metacarpus), the maximal width of the terminal bulb and isthmus, or the length from the buccal cavity to the grinder. Thus, there are no gross morphological differences in the pcm-1 dauer larvae.
N2 dauer larvae survive 1% SDS treatment and, after removal from SDS and upon feeding, resume development to reproductive adults (20). The pcm-1 larvae picked for SDS treatment were thin and long, did not forage, appeared not to pump, had dark intestinal cells, and displayed typical dauer posture. The majority of these apparent dauer larvae lacking the pcm-1 gene, however, are highly sensitive to SDS treatment (Table 3). In these experiments, we measured sensitivity 96 hours after egg fertilization; dauer larvae tested after 72 hours showed a reduced sensitivity (data not shown). In each experiment, the percentage of pcm-1 dauer larvae killed by SDS treatment is much higher than the percentage of N2 dauer larvae killed. This defect is also noted in the original pcm-1-null strain, him-8; pcm-1, which has a significantly higher percentage of larvae killed by SDS treatment than the him-8 mutant control (Table 3). Because we find no obvious defects in the alae, lateral hypodermal cells or the pharynx, the pcm-1 mutation does not appear to cause morphological defects that lead to SDS sensitivity.
|
Epistasis Analysis of pcm-1 for Dauer Larvae Formation
It is possible to order gene function into genetic pathways by analyzing the phenotype of double mutants whose single mutants have opposite phenotypes. Numerous such epistasis studies have been carried out with mutations that affect dauer formation (23–27). Here, we tested the functional relationships between pcm-1 and daf-2, daf-7, and daf-11. Mutations in the daf-2 gene lead to multiple temperature-sensitive phenotypes: Daf-c, embryonic arrest, morphological differences, increased adult life span, and late progeny production (23,34,39,40). The daf-2 gene encodes an insulin-like receptor tyrosine kinase (40). Mutations in the daf-7 gene, which shows homology to transforming growth factor-beta (TGF-β), cause temperature-sensitive constitutive dauer formation (23,37). Mutations in the daf-11 gene, encoding a guanylyl cyclase, also cause temperature-sensitive constitutive dauer formation (24,41). Finally, larvae with a mutation in the daf-12 gene are Daf-d, and these mutations suppress the Daf-c defect of daf-7 and interact with particular alleles of daf-2 (23–27). The daf-12 gene shows homology to nuclear hormone receptors (42,43). In wild-type larvae under growth conditions, the inferred order of gene function places the TGF-β-like daf-7 and the guanylyl cyclase daf-11 pathways both functioning to inactivate the daf-12 gene, thereby preventing dauer formation. The insulin-like signaling daf-2 pathway prevents dauer formation by inactivating the daf-16 gene.
To order pcm-1 function relative to other genes in the dauer formation pathway, the phenotype of double mutants was analyzed. In this case, dauer formation was induced at 25°C by daf-c mutations in the absence of pheromone. Appropriately, N2 and the pcm-1-null mutant did not form dauer larvae at this temperature, whereas all of the daf-c mutants did form dauer larvae (Table 4). No daf-7; pcm-1 or daf-2; pcm-1 double-mutant larvae grew to L3, L4, or adult at the restrictive temperature similar to their daf-c single-mutant controls. Similarly, loss of the pcm-1 gene did not affect the ability of daf-11(m47) mutant animals to enter the dauer stage, resembling previous studies of this allele (24). We conclude that the pcm-1 deletion does not suppress the Daf-c phenotype of daf-2, daf-7, or daf-11. There is a statistically significant increase in the pre-dauer larval arrest for two of the daf-2; pcm-1 double mutants (Table 4). The dauer formation observed here for the daf-2(m41); pcm-1 larvae appears to be contrary to the results of the pheromone assay, but here the dauer-formation cues were temperature and the daf-2 reduction-of-function mutation. Lack of PCM-1 does not inhibit dauer formation of these daf-c mutations; therefore, PCM-1 is not required for these mutants to enter the dauer stage.
|
|
Deletion of the pcm-1 Gene Does Not Affect Adult Hermaphrodite Life Span in Either a Wild-Type or daf-2 Mutant Background
To test the hypothesis that the protein repair activity of the pcm-1 gene contributes to adult survival at the elevated temperature of 25°C where more protein damage would be expected to occur, we assayed the wild type and four single mutant isolates of pcm-1. Overall, we noted no statistically significant decrease in adult life span of pcm-1 deletion mutants compared to N2 (Figure 3, Table 6). In one experiment, a statistically significant increase in life span was computed for three of the four isolates tested. However, this trend was not observed when the experiment was repeated (Table 6). The failure to reproduce the slight, statistically significant life-span increase reinforces the practice of replicate testing with more than one isolate to confirm life-span differences.
|
|
Overexpression of the pcm-1 Gene Increases Adult Life Span
The transgenic strains created to establish that the pcm-1 deletion was responsible for the dauer defect contain many copies of the pcm-1 gene. By quantitative reverse transcription-PCR, a 20-fold increase in messenger RNA level was found in the transgenic strain PL51, which verified overexpression of pcm-1. To test whether overexpression of PCM-1 has an effect, we determined the adult life span of transgenic animals. We assayed adults from both strains carrying wild-type pcm-1 and strains carrying enzymatically defective pcm-1 transgenes. As before, we conducted the adult life-span experiment at the elevated temperature of 25°C, expecting more protein damage to occur. A substantial increase in adult life span was observed for adults overexpressing the pcm-1 gene (Figure 3B, Table 6). No statistically significant difference in adult life span was detected for transgenic adults carrying a defective pcm-1 compared to N2 (Figure 3B, Table 2). The similar longevity for these two genotypes demonstrates that the exposure to ultraviolet light as larvae, to select transgenic animals, did not affect adult life span. The observed longevity suggests that there is a benefit of excess PCM-1, even though there is no evidence for survival defects in adult animals lacking PCM-1.
|
DISCUSSION |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The effect of the loss of the PCM-1 methyltransferase on dauer larva, the stage specialized for stress resistance and long-term survival under unfavorable conditions, is consistent with results from other organisms. In bacteria, the loss of the pcm gene has no apparent effect in log-phase growth but significantly reduces the survival of stationary phase cells under environmental stresses (10). In mice, repair-methyltransferase deficient pups have seizures and die a premature death shortly after weaning (13,14,46,47). Interestingly, adult life span is extended in Drosophila that overexpress the Pcmt methyltransferase when cultured at 29°C, but not when the flies are cultured at 25°C, illustrating that the methyltransferase becomes important to the survival of the animal under stress (48). This appears to be the case in C. elegans as well.
The pcm-1-null mutants cannot remain as dauer larvae as well as wild type. This observation distinguishes pcm-1 Daf-d from che and osm mutants that have defects in chemosensory neurons and thereby fail to form dauer larva in response to pheromone (21,32,49). The pcm-1 response to increasing concentrations of pheromone resembles wild-type larvae lacking ASJ neurons, which results in fewer dauer larvae formed in response to pheromone (50). The ASI neuronal marker (daf-7::GFP) is appropriately regulated in response to pheromone in pcm-1-null mutants. PCM-1 may have a role in processing information in pheromone-sensing neurons or in formation of the non-ASI pheromone-sensing neurons. It is also possible that the integration of sensory cues in the absence of PCM-1 favors reproductive growth by default. In dauer larva formation, the decision whether to form a dauer larva or grow to an adult is not equally weighted, in that the decision to grow is irreversible and the decision to form a dauer larva is continually reassessed. In the pcm-1 mutant case, the sensory cues appear to be received and transduced to signal initiation of dauer morphogenesis. However, many of the pcm-1 mutants fail to remain committed to the dauer state.
Dauer formation of pcm-1 mutants at 20° in the presence of pheromone was modulated by daf-2. The daf-2(m41) mutation allows pcm-1-null mutants to form more dauer larvae in response to pheromone. Coincidently, the pcm-1 mutation inhibits the ability of daf-2(m41) to form as many dauer larvae as the daf-2 single mutant. Because both the daf-2 and pcm-1 phenotypes are observed in the double mutant, it may be inferred that these genes appear to function in parallel pathways. However, because the mutant daf-2 allele used is not a null allele and probably has residual activity, the observation is also consistent with daf-2 acting downstream of pcm-1, with pcm-1 negatively regulating daf-2.
The daf genes have been ordered by analysis of epistasis data into parallel TGF-β–like, insulin-like, and cyclic guanosine monophosphate signaling pathways that converge in the process of dauer formation (23–27). We performed epistasis analysis with pcm-1 to understand its function in the context of the classic dauer formation pathways. In our epistasis analysis with the daf-c; pcm-1 double mutants, we found that all the larvae attempt to form dauer larvae in response to the internally generated signal due to a daf-c mutation in any of the three signaling pathways. Therefore, the pcm-1 deletion does not suppress the Daf-c phenotype of daf-2, daf-7, or daf-11. Both mutant phenotypes are observed in the daf-c; pcm-1 double mutants because SDS-sensitive dauer-like larvae are constitutively formed. We found that mutations in daf-7 and daf-11 do not suppress the SDS sensitivity of pcm-1 mutant dauer larvae and actually increase the penetrance of dauer-like larvae formation. Similar genetic interactions with mutations in daf-2 and daf-7 have been previously noted for rop-1, leading to the conclusion that this gene has general effects on the dauer formation process (51). This could also be the case with pcm-1. Based on previous reports, the genetic interactions of pcm-1 are reminiscent of daf-9. One interpretation of our results, considering a classic Daf-d phenotype, is that pcm-1 functions in parallel or downstream of daf-7 and daf-11 and in parallel or upstream of daf-2. However, in the studies with transgenic strains we observed inappropriate exit from the dauer stage, which might be considered transient dauer formation. If inability to remain dauer is the primary defect, then another interpretation would be that pcm-1 does not function directly in the decision to initiate dauer formation, as do daf-2, daf-7, and daf-11, but rather in the decision to remain dauer.
The predicted biochemical outcome of the absence of PCM-1 is increased accumulation of damaged protein. Such accumulation was not detected previously in aged pcm-1 dauer larvae, so degradation of L-isoaspartyl–containing proteins by active proteolytic pathways may occur, because no biochemical evidence suggests the presence of another L-isoaspartyl-O-methyltransferase in C. elegans (17). Alternatively, if the methyltransferase has a limited cellular distribution, it may be difficult to detect changes in whole animal lysates arising from repair in the PCM-1–expressing cells. However, aged dauer larvae have been shown to accumulate other types of damage that is reversed upon exit of dauer and resumption of development (52). It is possible that L-isoaspartyl damage accumulates with age in proteins responsible for the development and maintenance of dauer larvae and is at the root of the inappropriate decision to resume development in the presence of pheromone in the pcm-1 mutants. Another possibility is that the incomplete proteolysis of L-isoaspartyl–containing proteins leads to fragments that can compromise the function of these or other cells (14,53). PCM-1 may function to ensure the complete proteolytic digestion of interfering or unneeded proteins during dauer formation, considering the increased need due to autophagy-dependent tissue remodeling that is crucial for dauer formation (54,55). Finally, PCM-1 may have functions in dauer maintenance that are independent from a role in protein damage repair. We speculate that this role involves a neuroendocrine function that is shared between the various neurons that regulate dauer formation via different signal transduction pathways (50). This role may be the common aspect between nematodes and mice, because neuronal function is most affected in mice lacking PIMT (7).
Finally, the observation of the enhanced life span of mutant dauer larvae in this work is novel. The increased dauer life span of the two daf-c genotypes, daf-2 and daf-7, was similar. This result differs from the adult Age phenotype because daf-7 adults are not long-lived, at least at 25°C (27,56). At 20°C in the presence of fluorodeoxyuridine, it has recently been reported that there is an increase in the adult life span of daf-7 mutants (57). Enhanced longevity of daf-7 and daf-2 dauer larvae may result from the altered entry of the larvae into the dauer stage, physiological differences of the daf-c dauer larvae, or a combination of the two. Instead of entering dauer by sensing high pheromone and low food, temperature-sensitive daf-c mutations induce dauer larva formation while feeding and thus may have more energy stores initially. Alternatively, there may be differences in the gene expression and metabolism of daf-c dauer larvae from that of wild-type pheromone-induced dauer larvae. That there is overlap in the genes that regulate adult life span and dauer life span has been known for some time. However, as in the case of pcm-1 and daf-7, a novel observation is that there are also genes that have no effect on adult life span, but influence dauer life span. The pcm-1 dauer larvae may be impaired in long-term survival because they inappropriately resume development. In the dauer survival assay, there is no pheromone and no food, so even if a dauer larva attempted to resume development, lack of food would prevent growth. In essence, we suggest that the pcm-1 dauer larvae are short-lived because they initiate an energetically demanding process and starve to death. The results presented here demonstrate that longevity can depend on the proper cellular response to external signals.
|
A |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
We thank C. Clarke, T. Jonassen, J. D. Lowenson, and H. Yu for their critical review of the manuscript. We are grateful to B. Ellebracht for performing the GFP assays, for the pharyngeal measurements, and for isolation of transgenic lines. We thank J. D. Lowenson for performing methyltransferase activity assays and K. Hale and J. Wada for their assistance in the dauer life span experiments. We also want to express our appreciation to our colleagues for their gifts of plasmids and strains. D. R. Riddle provided strain DR1751. S. Mitani (Toyko Women's Medical University) provided the pcm-1(tm363) strain. A. van der Bliek provided the myo-3::mitoGFP plasmid, and he and his laboratory members gave us continuing helpful advice.
Kelley L. Banfield is now with the Laboratory of Molecular Technology, SAIC-Frederick, Inc., National Cancer Institute at Frederick, Frederick, MD.
Tara A. Gomez is now with the California Institute of Technology, Pasadena.
|
F |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Received September 22, 2007
Accepted April 25, 2008
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
