MCB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Branicky, R.
Right arrow Articles by Hekimi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Branicky, R.
Right arrow Articles by Hekimi, S.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, May 2006, p. 3976-3985, Vol. 26, No. 10
0270-7306/06/$08.00+0     doi:10.1128/MCB.26.10.3976-3985.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Uncoupling the Pleiotropic Phenotypes of clk-1 with tRNA Missense Suppressors in Caenorhabditis elegans

Robyn Branicky, Phuong Anh Thi Nguyen, and Siegfried Hekimi*

Department of Biology, McGill University, 1205 Avenue Docteur Penfield, Montreal, Quebec, Canada H3A 1B1

Received 13 September 2005/ Returned for modification 30 November 2005/ Accepted 7 February 2006


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
clk-1 encodes a demethoxyubiquinone (DMQ) hydroxylase that is necessary for ubiquinone biosynthesis. When Caenorhabditis elegans clk-1 mutants are grown on bacteria that synthesize ubiquinone (UQ), they are viable but have a pleiotropic phenotype that includes slowed development, behaviors, and aging. However, when grown on UQ-deficient bacteria, the mutants arrest development transiently before growing up to become sterile adults. We identified nine suppressors of the missense mutation clk-1(e2519), which harbors a Glu-to-Lys substitution. All suppress the mutant phenotypes on both UQ-replete and UQ-deficient bacteria. However, each mutant suppresses a different subset of phenotypes, indicating that most phenotypes can be uncoupled from each other. In addition, all suppressors restore the ability to synthesize exceedingly small amounts of UQ, although they still accumulate the precursor DMQ, suggesting that the presence of DMQ is not responsible for the Clk-1 phenotypes. We cloned six of the suppressors, and all encode tRNAGlu genes whose anticodons are altered to read the substituted Lys codon of clk-1(e2519). To our knowledge, these suppressors represent the first missense suppressors identified in any metazoan. The pattern of suppression we observe suggests that the individual members of the tRNAGlu family are expressed in different tissues and at different levels.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Suppressor analysis has been routinely used in Caenorhabditis elegans for the study of gene function and the identification of genetic pathway components. Suppressor analysis of nonsense alleles has also led to the discovery of informational suppressors encoding components of the translational machinery, including suppressor tRNAs (for example, references 1, 21, 22, and 40). In fact, C. elegans is the only metazoan from which nonsense tRNA suppressors have been recovered in classic forward genetic screens.

We have used a suppressor approach to understand the pleiotropic phenotypes of the C. elegans clk-1 mutants. Mutations in clk-1 are highly pleiotropic, affecting the rates of many physiological processes over a wide range of time scales (41). These mutations result in an average lengthening of the cell cycle of early embryos, of embryonic and postembryonic development, and of the defecation, swimming, and pharyngeal pumping cycles of adults. clk-1 mutations also affect reproductive features, such as the rates of germ line development and egg production (35, 41), and lead to an increased life span (23).

clk-1 encodes a highly conserved hydroxylase (9, 18, 29, 36) that is required for the hydroxylation of 5-demethoxyubiquinone to 5-hydroxyubiquinone, which is converted by another enzyme (COQ-3) into ubiquinone (UQ; also called coenzyme Q and CoQ). In the absence of CLK-1, worms are devoid of UQ9 (the subscript refers to the number of isoprene units in the side chain and is a species-specific trait) (16, 30) and instead accumulate the precursor demethoxyubiquinone (DMQ9) (30). Several observations suggest that DMQ can partially substitute for UQ. In particular, mitochondrial respiration is only slightly affected in clk-1 mutants (2, 3, 10, 30). This is in contrast to the coq-3 mutants, which are also defective in UQ biosynthesis but do not make DMQ, and are inviable (12). Although DMQ may not be functional in Saccharomyces cerevisiae (33), it has been shown in both isolated membranes and in vitro studies that DMQ can function as an electron transporter in the mitochondrial electron transport chain (30, 39). Furthermore, mouse mclk1–/– embryonic stem cells that do not contain measurable amounts of UQ are still capable of high levels of mitochondrial respiration (25). However, despite the fact that DMQ appears capable of functioning as an electron transporter, it cannot entirely substitute for UQ in worms, as clk-1 mutants require UQ from their bacterial food source in order to proceed through development and become fertile adults (12, 16). When the mutants are fed Escherichia coli mutants that are defective in the biosynthesis of UQ8 (ubi mutants), they arrest development transiently and eventually grow up to become sterile adults (5).

Several observations indicate a complex relationship between the Clk-1 phenotypes and UQ. First, clk-1 mutants exhibit a profound mutant phenotype when grown on UQ-replete bacteria, although their mitochondria contain bacterially derived UQ8 (17). In addition, there is no correlation between the absence of endogenous UQ9 and the severity of the overall phenotype. Indeed, no UQ can be detected in any of the three clk-1 mutants (e2519, qm30, and qm51), and all three accumulate the same amount of DMQ (30), yet for most, if not all, the phenotypes, the qm30 and qm51 null mutants, which produce no CLK-1 protein, are more severely affected than the missense e2519 mutant, which produces wild-type amounts of mutant CLK-1 protein (13). These observations suggest several possibilities. (i) Some of the mutant phenotypes could be due to the presence of DMQ rather than to the lack of UQ. Indeed, DMQ is generally not present in the wild type and likely has different redox properties than UQ does (31). (ii) UQ8 might not be functionally equivalent to UQ9. (iii) Endogenously synthesized UQ may not be functionally equivalent to exogenously supplied UQ, regardless of the chain length. (iv) CLK-1 could have other functions, in addition to the biosynthesis of ubiquinone.

Here we describe our analysis of mutants that were isolated in screens for growth suppressors of clk-1. We identified nine suppressors that are specific to the missense mutation clk-1(e2519) (Glu148Lys). All suppressors are dominant and suppress some, or, in some cases, all the Clk phenotypes on both UQ-replete (UQ+) and UQ-deficient (UQ–) bacteria. The suppressors still accumulate a large amount of DMQ9 but are able to synthesize a very small amount of UQ9. We found that the clk-1 mutants cannot be suppressed when this same amount of UQ9 is obtained exogenously (from bacteria). This suggests that viability can be supported by exogenous UQ but that wild-type rates of development and behavior require endogenous UQ biosynthesis. In addition, the large amount of DMQ9 retained in the suppressor mutants indicates that DMQ9 is unlikely to contribute to the Clk phenotypes.

We have cloned six of the suppressors, and all encode tRNAGlu genes whose anticodons are altered to read the substituted Lys codon of clk-1(e2519). To our knowledge, these suppressors represent the first missense suppressors identified in any metazoan. The pattern of suppressed phenotypes we observe among the suppressor mutants suggests that the individual members of the tRNAGlu family might be expressed in different tissues, at different times during development, and at different levels. It also indicates that most Clk phenotypes can be uncoupled from each other, an observation that is particularly relevant to understanding the long-life-span phenotype.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
General culture methods and strains. C. elegans worms were cultured under standard conditions at 20°C (37). For the most part, worms were fed either the standard E. coli strain OP50, which produces UQ8, or GD1 (14) or DM123 (28) (ubiG::Kan and yigR::Kan, respectively), which are UQ8-deficient E. coli strains. These strains were cultured in 2YT growth medium supplemented with 0.5% glucose and 50 µg/ml kanamycin and then on NGM plates supplemented with 50 µg/ml kanamycin to prevent contamination from other bacterial strains. For some experiments, worms were fed ispB mutant bacteria (ispB::Cmr) engineered to produce UQ8 or UQ9 (E. coli KO229 carrying pSN18 and KO229 carrying pKA3, respectively) (32). These bacteria were grown in 2YT growth medium and then on NGM plates containing 50 µg/ml of spectinomycin and ampicillin, respectively.

A two-step procedure was used to transfer worms from strain OP50 to the various test bacteria. First, the worms were transferred without bacteria to an intermediate plate seeded with the test bacteria. The worms were left to crawl around on these plates for ~2 h to rid them of any trace amount of OP50. The worms were then transferred from the intermediate plates to the experiment plates containing the same test bacteria.

The Bristol strain N2 was used as the wild type. The clk-1 alleles used in this study were e2519 and qm30. The mutants used for linkage and mapping are as follows: linkage group I (LGI), unc-11(e47), dpy-5(e61), unc-75(e950), unc-101(m1), unc-54(e190); LGII, lin-31(n301), unc-85(e1414), clr-1(e1745), dpy-10(e128), unc-4(e120), unc-53(e404); LGIII, dpy-17(e164); LGIV, unc-24(e138), LGV, dpy-11(e224); LGX, unc-1(e719), lon-2(e678), unc-6(e78), dpy-6(e14), unc-115(mn481), unc-9(e101), unc-84(e1410), dsc-1(qm133), unc-7(e5).

Isolation of suppressor mutations. clk-1(e2519) worms were mutagenized with ethyl methanesulfonate essentially as described previously (37). Briefly, worms at the fourth larval stage (L4) were incubated with 50 mM ethyl methanesulfonate for 4 h at 20°C, with agitation every 30 min. The mutagenized P0 worms were transferred to 90-mm plates, 20 worms/plate and left to produce progeny. The resulting F1 worms were bleached as gravid adults (using a solution of 20:3:2 of H2O:10% NaOCl:5 M KOH), and the resulting F2 worms were left to hatch in M9 buffer for ~48 h, under which condition they will arrest as L1 stage larvae. For the growth rate suppressor screen, the L1 worms were spotted onto 90-mm plates seeded with strain OP50 and were screened 2 or 3 days later for the presence of fast-growing worms. For the growth arrest/sterility suppressor screen, the L1 worms were spotted onto 90-mm plates seeded with strain DM123 and were screened 3 or 4 days later for the presence of nonarrested worms. The plates were also rescreened a few days later for the presence of F3 eggs laid by a fertile F2 worm that was missed in the initial screening, but no such eggs were found.

The number of haploid genomes screened was estimated from the number of F2 worms scored and the number of F1 worms from which they originated (assuming that each F1 worm carries two independently mutagenized haploid genomes and that 4 F2 worms need to be scored from each F1 to be able to score the two haploid genomes). On the basis of this, we estimate that we scored 200,000 haploid genomes in the growth rate suppressor screen (from which we isolated two suppressors, qm194 and qm195) and 100,000 haploid genomes in the growth arrest/sterility suppressor screen (from which we isolated seven suppressors, qm196, qm197, qm198, qm199, qm210, qm211, and qm213), which suggests that the growth arrest/sterility suppressor screen is a much more sensitive screen. It should also be noted that because all the suppressors that we recovered were dominant, the actual numbers of haploid genomes screened are likely greater than our estimates. We also screened approximately 100,000 haploid genomes in the clk-1(qm30) background in both screens, from which we did not recover any suppressors.

The mutants described here in detail (qm199, qm210, qm211, and qm213 mutants) have been backcrossed to the clk-1(e2519) strain at least three times.

Mapping and cloning of suppressors. All linkage and mapping were carried out using strains that were homozygous for clk-1(e2519). For linkage analyses, homozygous suppressor mutants were crossed with mutants homozygous for a recessive visible mutation. F2 animals homozygous for the visible mutation were transferred to strain DM123 plates as young adults. The entire F3 brood was scored as either suppressed (when 75 or 100% of the F3 worms were suppressed) or not suppressed (when 0% were suppressed). Linkage is indicated by <75% of the F3 plates being suppressed. The recombination distance (p) was calculated by considering the recombination frequency (f; the number of suppressed plates/total number of plates) to be equal to 2p/(1 + p). f was also used to calculate the 95% confidence interval from the binomial distribution.

For three-point mapping experiments, hermaphrodites homozygous for two closely linked markers were crossed with homozygous suppressor males. Recombinant F2 worms were singled as adults. To assay for the presence of the suppressor mutation, four F3 worms with the recombinant phenotype (either homozygous or heterozygous for the recombinant chromosome) from each recombinant F2 were singled out as adults onto strain DM123 plates. Their entire F4 brood was scored as either suppressed (75% or 100% suppressed) or not suppressed (0% suppressed). If any of the F4 plates were suppressed, the recombinant F2 was considered to have carried the suppressor mutation. The 95% confidence intervals were calculated from the binomial distribution.

The key mapping data for qm199, qm210, qm211, and qm213 have been included in Table 1. The approximate map positions of qm195 and qm197 were determined by linkage analyses and are also given in Table 1. On the basis of the map positions of these six mutations, we identified the corresponding genes using a candidate approach. Genomic DNA was extracted from the mutants, and the candidate genes were sequenced. Primer sequences and PCR conditions are available upon request. qm195 was found to contain the same mutation in the same gene as qm199, although they were isolated independently in different screens. qm194, qm196, and qm198 have all been linked to various regions of LGX; however, we have not determined to which genes they correspond.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Genetic positions and molecular identities of suppressor genes

 
Phenotypic analyses. (i) Brood size. The total broods of individual animals were counted. For brood size measurements on bacterial strains DM123 and GD1, the P0 worms were transferred to the UQ– bacteria as young adults, and the broods produced by their F1 progeny were counted.

(ii) Embryonic development. Two- to four-cell embryos were dissected from gravid hermaphrodites and monitored every hour until hatching.

(iii) Postembryonic development. Eggs were selected and put on plates and examined 3 h later. Larvae that had hatched during this time were considered to be 1.5 h old at the end of the interval. Larvae were examined every hour until they molted into adults.

(iv) Life span. On day 0, eggs were selected and put on plates and left to hatch for 6 h. Eggs that had hatched during this interval were used for the experiment. The worms were monitored daily until they died. Worms that were believed to have died from either internal hatching or gut extrusion were not included in the study and were replaced by other worms.

(v) Defecation. Worms were scored for three to five consecutive defecation cycles as described previously (4), and the mean cycle length was calculated for each worm. For experiments on E. coli KO229(pKA3) and KO229(pSN18), N2 worms were grown on the test bacteria for three generations and the clk-1 mutant strains for four generations.

(vi) Pharyngeal pumping. Worms were scored for 3 to 5 individual minutes, and the mean number of pumps per minute was calculated for each worm.

(vii) Statistical analyses. The Student t test was used for all phenotypes, except the life span phenotype, for which the log rank test was used. All tests were one tailed. Differences between the samples were considered to be significant when P was <0.05. In the text, we consider a significant difference from clk-1(e2519) but not N2 to indicate full suppression; significant differences from both N2 and clk-1(e2519) to indicate partial suppression; and a significant difference from N2 but not clk-1(e2519) to indicate no suppression.

High-performance liquid chromatography (HPLC) analyses. Worms near starvation were collected from plates and were washed several times with M9 buffer. The worms grown on E. coli KO229(pSN18) were cleaned by sucrose flotation. Approximately 400 µl of worms was used for each run. The extraction was performed essentially as described previously (34). Briefly, worm pellets were put in 1 ml of water and homogenized in the presence of 50 µl of 2,6-di-tert-butyl-4-methyl-phenol (BHT) (10 mg/ml in ethanol) (Sigma). One hundred microliters of homogenate was taken at this step to measure protein concentration, using the Bio-Rad protein assay. One milliliter of 0.1 M sodium dodecyl sulfate was added, and the samples were further homogenized. Two milliliters of ethanol was added, and the samples were vortexed. Two milliliters of hexane was added, and the samples were vortexed and then centrifuged at 1,000 x g for 5 min at 4°C. The upper organic layer was collected, and the hexane extraction was repeated. The samples were evaporated using a cooled speed vacuum and were kept at –80°C. The Beckman Coulter System Gold hardware and the 32 Karat software were used to analyze the samples. Shortly after reconstitution with 750 µl mobile phase (70% methanol and 30% ethanol), the samples were loaded on a reverse-phase column (Inertsil ODS-3 C18 column [5 µm; 4.6 x 250 mm; GL Science]), and elution was monitored by a UV detector at 275 nm.

For peak identification, a "standard" containing commercial UQ9 (Sigma) mixed with lipid extract from clk-1(qm30) worms grown on E. coli OP50 (for DMQ9 and UQ8) was prepared. For quantification, the amount of UQ9 was determined by comparison to UQ9 standards of known concentrations, which were also used to estimate the amounts of DMQ9 and UQ8. The recovery rate was estimated by determining how much UQ6 (Sigma) was present in the sample after a known amount had been added to the samples before the lipid extraction was performed. The amount of quinone was normalized to the amount of protein.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Identification of suppressors of the growth phenotypes of clk-1(e2519). clk-1 mutants exhibit a highly pleiotropic phenotype: when the worms were grown on UQ-replete bacteria, the mutants grow slowly, have slowed behaviors, delayed reproduction, and an increased life span (23, 35, 41), whereas when they were grown on UQ-deficient bacteria, the mutants transiently arrest development at the second larval (L2) stage, eventually growing up to become sterile adults (5, 12). We carried out suppressor screens in an effort to better understand the bases of the growth phenotypes and also to further explore the mutants' requirement for dietary UQ. The first screen was aimed at identifying mutations that could suppress the low growth rate of clk-1 mutants on the standard UQ-replete bacterial strain OP50. We screened for suppressors in both the clk-1(qm30) and clk-1(e2519) backgrounds. We were not able to find any suppressors in the qm30 background but isolated two suppressors (qm194 and qm195) in the e2519 background. The second screen was aimed at finding suppressors of the growth arrest and/or sterility on the UQ-deficient DM123 bacteria. Again we screened on both the qm30 and e2519 backgrounds and found suppressors only in the e2519 background. This second screen was a much more efficient, or sensitive, screen, as we were able to isolate seven suppressors (qm196, qm197, qm198, qm199, qm210, qm211, and qm213), although we screened only half as many haploid genomes as had been screened in the first screen (see Materials and Methods). All the suppressors isolated in this screen suppressed both the growth arrest and sterility on the UQ-deficient bacteria.

As mentioned above, we were unable to find suppressors in the qm30 background, which suggested that the suppressors may be e2519 specific.

To test this directly, we transferred a number of the suppressors into the qm30 background and found that none could suppress any phenotypes on UQ+ or UQ– bacteria (data not shown), indicating that we had in fact isolated nine e2519-specific suppressors. Given the number of haploid genomes screened, it is likely that the growth phenotypes and dietary requirements of the qm30 mutants cannot be suppressed by any simple gene inactivation.

Growth suppressors suppress both UQ– and UQ+ phenotypes. In order to determine whether the mutants isolated in the two screens represent two different classes of suppressors, we tested both groups of suppressors for suppression of the Clk-1 phenotypes on UQ– and UQ+ bacteria. We found that all nine suppressors are able to suppress growth arrest and sterility on UQ– bacteria (Fig. 1 and data not shown). Indeed, all the suppressors are fully fertile on the UQ– bacteria. In some cases, the suppressors even have wild-type brood sizes.


Figure 1
View larger version (38K):
[in this window]
[in a new window]
  		FIG. 1. Brood sizes of Bristol strain N2, clk-1, and clk-1 suppressor strains grown on UQ-replete (OP50) and UQ-deficient (DM123 and GD1) bacteria. The bars represent the mean brood sizes ± standard deviations (error bars) produced by 12 to 17 animals. The two black arrows indicate the complete sterility of clk-1(e2519) worms on DM123 and GD1 bacteria.

 
We then tested whether the suppressors are able to suppress the growth and behavioral Clk phenotypes observed on UQ+ bacteria (Fig. 2). For these analyses, we focused on four of the mutants that we had repeatedly backcrossed and which we believe to be representative of the nine mutants isolated. We found that all mutants are able to partially suppress the low rate of embryonic development of clk-1(e2519), with qm211 having the weakest effect (Fig. 2A). All mutants are also able to partially suppress the slow postembryonic development, with the exception of qm211, which did not suppress this phenotype at all (Fig. 2B). Similarly, we found that, with the exception of qm211, the suppressors are able to fully suppress the increased life span of clk-1(e2519) (Fig. 2C). In contrast, we found that all the suppressors could fully suppress the low pharyngeal pumping rate of clk-1(e2519) (Fig. 2D). For the low defecation rate of clk-1(e2519), we found that qm213 suppresses fully and qm211 partially, while qm199 and qm210 do not suppress this phenotype at all. In addition, the suppressor mutations do not affect in any way the rates of growth or behavior in either a clk-1(+) (wild-type) or clk-1(qm30) background (data not shown), indicating that the phenotypes we are observing are due to a specific suppression of clk-1(e2519). This is also the case for the life span phenotype. For example, the qm213 mutation, which fully suppresses the increased life span of clk-1(e2519) has no effect on life span in a wild-type background: it lives an average of 17.8 ± 5 days compared to 17.2 ± 2.5 days for the wild type (P = 0.7), suggesting that even the shortened life span phenotype is due to a specific suppression rather than a deleterious effect of the mutation. This complex pattern of suppression could be due to differences in the level, timing, and tissue specificity of expression of each suppressor (see Discussion).


Figure 2
View larger version (38K):
[in this window]
[in a new window]
  		FIG. 2. Developmental and behavioral phenotypes of N2, clk-1, and clk-1 suppressor strains grown on OP50 bacteria. Values that are significantly different from the data for N2 (#) and clk-1(e2519) (*) are indicated (see Materials and Methods for detailed explanations). (A) Rate of embryonic development. Each point represents the percentage of embryos that had hatched by that time point. For each genotype, there were 66 to 111 embryos. The means ± standard deviations [SD] are 13.28 ± 0.53 (*), 16.09 ± 1.87 (#), 14.03 ± 1.06 (#, *), 14.00 ± 0.60 (#, *), 15.02 ± 1.38 (#, *), and 14.27 ± 1.02 (#, *). (B) Rate of postembryonic development. Each point represents the percentage of larvae that had molted to adulthood by that time point. For each genotype, there were 83 to 116 larvae. The means ± SD are 49.10 ± 5.27 (*), 69.03 ± 2.30 (#), 58.26 ± 4.21 (#, *), 57.66 ± 2.68 (#, *), 69.92 ± 5.16 (#), and 59.31 ± 3.64 (#, *). (C) Life span. Each point represents the percentage of worms that were alive at that time point. For each genotype, there were 100 worms. The means ± SD are 18.42 ± 3.58 (*), 20.81 ± 5.49 (#), 18.66 ± 3.14 (*), 17.13 ± 3.12 (*), 21.04 ± 5.99 (#), and 16.62 ± 3.53 (*). (D) Rate of pharyngeal pumping. The bars represent the mean pumping rate of 10 animals, with each animal's rate based on the mean of five measures; the error bars represent the standard errors of the means. (E) Rate of defecation. The bars represent the mean defecation cycle length of 12 animals, with each animal's cycle length based on the mean of five cycles; the error bars represent the standard errors of the means.

 
Suppressors have both dominant and semidominant effects. During the characterization of the suppressors, we made several observations that indicated that the suppressors are dominant. In particular, we found that all the progeny produced from crosses between a suppressed strain and a nonsuppressed clk-1(e2519) strain produced progeny which were all capable of growing on UQ– bacteria. In order to quantify this more precisely, we compared the strength of suppression of different phenotypes conferred by one or two copies of the suppressor mutations (Fig. 3). When we scored the brood size of the suppressors on UQ– bacteria, we found that all the suppressors were semidominant, that is, that one copy of the suppressor resulted in a significantly lower brood size than two copies (Fig. 3A). However, for the behavioral phenotypes (the rates of pharyngeal pumping and defecation [Fig. 3B and C]), we found that for both the qm211 and qm213 suppressor mutations, one copy of the mutations suppressed as strongly as two copies did. Taken together, these observations suggest that the mutations are semidominant but that for some phenotypes, the threshold of suppressing activity required for maximal suppression can be met by one copy of the suppressor (see Discussion).


Figure 3
View larger version (14K):
[in this window]
[in a new window]
  		FIG. 3. Dosage effect of suppressor mutations. The number of copies of the suppressor mutation are indicated by the large white numbers on the black bars. (A) Brood sizes of N2, clk-1, and clk-1 suppressor strains grown on the UQ-deficient GD1 bacteria. The bars represent the mean brood sizes ± standard deviations produced by 10 to 12 animals. In all cases, one copy of the suppressor mutation resulted in a significantly lower brood size than two copies, indicating that the suppressors are semidominant. Values that are significantly different from the values for strains with two copies of the suppressor mutation are indicated with asterisks. (B) Rate of pharyngeal pumping. The bars represent the mean pumping rates of 12 animals, with each animal's rate based on the mean of three measures; the error bars represent the standard errors of the means. For both suppressors, there was no significant difference between the rate produced by one or two copies of the suppressor. (C) Rate of defecation. The bars represent the mean defecation cycle length of 12 animals, with each animal's cycle length based on the mean of three cycles; the error bars represent the standard errors of the means. For both suppressors, there was no significant difference between the rate produced by one or two copies of the suppressor.

 
Suppressors restore UQ biosynthesis. HPLC analysis revealed that all the suppressors restore UQ biosynthesis (Fig. 4E to H and data not shown), although not to wild-type levels. As can be seen from a quantification of the levels of the quinones present in the various strains (Fig. 4G and H), the predominant Q species in the wild type is UQ9, whereas in clk-1(e2519), as well as in the suppressors, the predominant Q species is DMQ9, with only a very small amount of UQ9 detectable in the suppressors. In fact, the suppressor that makes the most UQ9 (qm213) makes 20 times less than the wild type, and the suppressor that makes the least UQ9 (qm210) makes 63 times less than the wild type, yet this appears to be sufficient for growth and fertility on UQ-deficient bacteria, and even for almost wild-type rates of growth and for some behaviors.


Figure 4
View larger version (47K):
[in this window]
[in a new window]
  		FIG. 4. HPLC analysis of quinone content. Representative chromatograms of quinones extracted from (B) N2; (C) clk-1(e2519); (E) qm210;clk-1(e2519), the suppressor that produces the least UQ; and (F) qm213;clk-1(e2519), the suppressor that produces the most UQ. All worms were grown on OP50 bacteria. The traces are lined up with the standards (A and D) run on that same day. In panels E and F, the arrows indicate the small amount of UQ9 detected in the suppressors. (G and H) Quantification of quinone content. (G) Quinone content of N2, clk-1, and clk-1 suppressor strains. Worms were grown on OP50 with the exception of the rightmost bar, for which the worms were grown for more than five generations on the UQ9-producing strain KO229(pSN18). Each bar indicates the mean ± standard deviation (error bar) from three independent extractions, with the exception of the last bar, for which the data come from two independent extractions. (H) Expanded view of the results shown in panel G for UQ9, which allows visualization of the small amount of UQ9 present in the suppressors and clk-1(e2159) when grown on KO229(pSN18). The black arrow indicates that, even at this scale, there is no detectable UQ9 in clk-1 mutants.

 
In order to test whether this small amount of UQ9 could really be responsible for the observed suppression, we chose to examine the defecation rate of the wild type, clk-1(e2519), and qm213;clk-1(e2519) on a transgenic E. coli strain that has been engineered to produce UQ9 (Fig. 5). We chose the defecation rate because it is one of the phenotypes that can be fully rescued by some of the suppressors. The KO229 strain lacks the octaprenyl diphosphate synthase encoded by the ispB gene, which is required for the synthesis of the UQ side chain. The pKA3 plasmid contains a wild-type ispB gene from E. coli and restores UQ8 biosynthesis, whereas the pSN18 plasmid contains a Synechocystis sp. strain PCC6803 ispB homolog, which results in the synthesis of UQ9 as well as some UQ8 (32).


Figure 5
View larger version (31K):
[in this window]
[in a new window]
  		FIG. 5. Defecation rates of N2, clk-1, and a clk-1 suppressor strain grown on UQ8- and UQ9-producing bacteria. UQ8 is the predominant Q species in E. coli OP50 and KO229(pKA3), and Q9 is the predominant A species in E. coli KO229(pSN18) (32; also data not shown). The bars represent the mean defecation cycle length of 10 animals, with each animal's cycle length based on the mean of three cycles; the error bars represent the standard errors of the means of the individual animals. The clk-1(e2519) mutant is significantly slower on KO229(pKA3) than on OP50 (P < 0.001) and significantly lower on KO229(pSN18) than on KO229(pKA3) (P < 0.01). The qm213;clk-1(e2519) is also significantly slower on KO229(pKA3) and on KO229(pSN18) than on OP50 (P < 0.0001), but there is no difference between KO229(pKA3) and pSN18. qm213;clk-1(e2519) is significantly faster than clk-1(e2519) on all bacteria: OP50 (P < 0.0001), KO229(pKA3) (P < 0.001), and KO229(pSN18) (P < 0.0001). qm213;clk-1(e2519) grown on KO229(pKA3) is significantly faster than clk-1(e2519) grown on KO229(pSN18) (P < 0.0001) although they contain the same amount of UQ9 (Fig. 4), suggesting that UQ9 obtained exogenously is not equivalent to UQ9 synthesized endogenously.

 
We found that both clk-1(e2519) and qm213;clk-1(e2519) mutants have significantly slower defecation on both KO229-based strains than on strain OP50. Possibly, the mutants are sensitive to the lack of menaquinone that is a feature of the KO229-based strains (15) or to some other difference in the strain that is unrelated to UQ. Alternatively, the clk-1(e2519) mutants could be sensitive to the level of UQ in their food source. Indeed, under our growth conditions, both the pKA3- and pSN18-containing strains produce less total UQ than OP50, approximately three and nine times less than OP50, respectively (data not shown).

We also observed that the low defecation rate of clk-1(e2519) is not suppressed when fed the KO229(pSN18) strain, although this strain makes UQ9. On the other hand, the suppressed qm213;clk-1(e2519) strain is much less sensitive to the level of UQ, as it has a higher defecation rate than clk-1(e2519) on all food sources. Thus, the defecation rate of qm213;clk-1(e2519) on KO229(pKA3) (which makes only UQ8) is significantly higher than that of clk-1(e2519) on KO229(pSN18) (which makes UQ9). This suggests that exogenously obtained UQ9 is not equivalent to endogenously synthesized UQ9. Although clk-1(e2519) mutants grown on KO229(pSN18) assimilate more UQ9 than the suppressors produce (Fig. 4), it is possible that this UQ9 does not reach the required cells or subcellular location. However, it has in fact been shown that, at least for UQ8, most of the UQ obtained exogenously can reach the mitochondria (17) (see Discussion).

Suppressors encode tRNAGlu genes. Using linkage analyses, we determined that seven of the suppressors map to LGX (qm194 to qm198, qm199, qm211), one to LGI (qm210), and one to LGII (qm213) (data not shown). Using two- and three-point mapping strategies, we further refined the genetic positions of six of the suppressors (the key data are summarized in Table 1). We determined that in each of the intervals to which a suppressor had been mapped, there was at least one gene coding for a tRNAGlu. Given that the e2519 mutation results in the substitution of a Glu codon for a Lys codon (GAG to AAG) one possible mechanism of suppression could be the mutation of a tRNAGlu. We sequenced the candidate tRNA genes in each of these six mutants and in each case identified one tRNA gene that contained a C-to-T transition at position 36 of the gene. qm195 and qm199 contain the same mutation in the same tRNAGlu gene, although they were isolated independently in two different screens, whereas the other four contain the same mutation but in distinct tRNAGlu genes. These genes have been named rte-1 to rte-5 for RNA, transfer, glutamic acid (E).

The C36T mutation found in all the suppressors alters the anticodon from CTC to CTT and would allow for the decoding of an AAG Lys codon as a Glu codon. This suggests that the suppressors act by restoring the production of some wild-type CLK-1(+) protein, which in turn is capable of restoring a small amount of UQ biosynthesis. The other three suppressors have been linked to the X chromosome, which contains multiple tRNAGlu genes. Although the molecular identities of these three mutations have not been formally determined, we believe that they also correspond to tRNAGlu genes, given that they are similar to the other suppressors in heritability and phenotype (data not shown).


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Pleiotropic phenotype and dietary requirements of clk-1 mutants. We have identified nine suppressors of the missense mutation clk-1(e2519). The six suppressors cloned all encode tRNAGlu genes, whose anticodons have been altered to read the altered Glu codon of the e2519 mutant. The suppressors still predominantly accumulate the UQ intermediate DMQ, although they restore a very small amount of UQ biosynthesis. This small amount of UQ appears to be sufficient for continuous growth and fertility on UQ-deficient bacteria. The suppressors also suppress some phenotypes on UQ-replete bacteria, although different subsets of Clk-1 phenotypes are suppressed by different mutations. These observations allow us to resolve some questions about the clk-1 mutants.

First, the fact that different subsets of phenotypes are suppressed by the different mutants indicates that the phenotypes can be decoupled from each other (Fig. 2). From an analysis of the patterns of suppression, we can deduce that the increased life span of clk-1 is not due to and does not require slow embryonic development or slow behaviors. In particular, it is interesting to note that a low rate of pharyngeal pumping is not necessary for an increased life span, as the clk-1(e2519);qm211 mutants have a wild-type rate of pumping but a long life span that is indistinguishable from that of clk-1(e2519) mutants. This observation suggests that the mechanism of life span extension of clk-1 mutants is not simply caloric restriction, although slow pumping can lead to caloric restriction in worms and increase life span in worms (24). On the other hand, the suppressor mutants cannot tell us whether slow postembryonic development is required for the increased life span. However, previous results have suggested that this is not the case (5).

Second, the quinone contents of the mutants allow us to address some questions about DMQ9. Previously we had speculated that the large amount of DMQ9 present in clk-1 mutants but not in the wild type, could be contributing to the phenotype. Indeed, there is in vitro and in vivo evidence that DMQ can act in the electron transport chain (25, 30, 39). There has also been some in vitro evidence to suggest that DMQ could be a lesser prooxidant than UQ (31), which could explain the lowered levels of reactive oxygen species observed in the clk-1 mutants (19, 35). However, the biological role of DMQ is somewhat controversial, given that DMQ does not seem to support respiration in S. cerevisiae, although it appears to act as a prooxidant in this system (33). Our previous observations suggest that DMQ contributes to the viability of the clk-1 mutants on UQ-replete bacteria, as mutants completely devoid of endogenous quinones are inviable, even on UQ-replete bacteria (12). Our present observations, however, suggest that the Clk-1 phenotypes are not due to the presence of DMQ9, as the suppressors still accumulate a very large amount of DMQ9, yet most of the Clk-1 phenotypes are suppressed. Thus, the presence of DMQ does not contribute to those phenotypes that are fully rescued by the suppressors (life span, defecation rate, and pharyngeal pumping). However, the developmental rates are only partially suppressed, so it remains possible that the presence of DMQ affects these phenotypes. Alternatively, it could be that these phenotypes have a higher threshold, requiring more CLK-1(+) activity.

Third, the suppressors allow us to address whether there is a specific requirement for endogenously synthesized UQ versus exogenous, dietary, UQ. clk-1 mutants grown on the standard UQ8-replete bacterial strain (OP50) assimilate UQ8 and transport it all the way to their mitochondria (17). However, the mutants contain significantly less UQ8 than the amount of UQ9 in the wild type. Therefore, the Clk-1 phenotypes on UQ-replete bacteria could be due to (i) too little UQ, (ii) a nonequivalence of UQ8 and UQ9, or (iii) a subtle nonequivalence of endogenously synthesized versus exogenously acquired UQ. The amount of UQ9 synthesized by the suppressors is dramatically less than the amount of UQ8 obtained from the bacteria, yet the phenotypes are suppressed, suggesting the Clk phenotypes are not due to too little UQ. clk-1(e2519) is not rescued by the UQ9-producing bacterial strain, indicating that even UQ9 cannot rescue the mutants when it is obtained exogenously, so it is not the difference in chain length (UQ8 versus UQ9) that results in the Clk-1 phenotypes on OP50. In fact, this observation suggests that exogenously obtained UQ cannot entirely substitute for endogenously synthesized UQ. Therefore, it appears as though viability can be supported by exogenous UQ but that wild-type rates of development and behavior require endogenous UQ biosynthesis. At this time we cannot distinguish whether there is any specific requirement for UQ9 in C. elegans or whether UQ8 would also be sufficient if it were synthesized endogenously.

What the functionally important difference between exogenous and endogenous UQ might be is not immediately obvious. UQ is found in all cellular membranes, with the highest levels in the mitochondria, where it is involved in electron transport (reviewed in references 7 and 38). In general, the majority of ubiquinone derived from the diet remains outside of the mitochondria (reviewed in reference 8). However, it has been shown in worms and in cultured cells that exogenously acquired UQ can reach the inner mitochondrial membrane, and at least in cells, UQ can then participate in electron transport (11, 17, 27). Possibly, in worms, not enough of the UQ acquired from bacteria reaches the site of electron transport in the mitochondria to influence mitochondrial function, or the acquired UQ does not assimilate into all tissues and cell types, some of which may be crucial for the regulation of physiological rates. Alternatively, it could be that electron transport or some other important mitochondrial function requires that UQ be synthesized de novo in the inner mitochondrial membrane.

We are also considering another, altogether distinct, possibility to explain the high degree of rescue provided by the suppressors. Given that the suppressors encode suppressor tRNAs, they act by restoring the production of some CLK-1(+) protein, which in turn restores a small amount of UQ biosynthesis. Thus, it remains possible that it is the presence of wild-type CLK-1(+) protein, rather than the small amount of accumulated UQ9, that confers suppression. It is possible that CLK-1 carries out some other function in addition to ubiquinone biosynthesis, and that it is this other function that is important for the Clk-1 phenotypes. This would be consistent with the lack of rescue conferred by the Q9-producing bacteria, and the observation that clk-1(e2519) mutants have a weaker phenotype than clk-1(qm30) mutants do, despite the fact that neither mutant produces any UQ. The e2519 mutant produces a full-length protein, which could retain some activity for a function that is not related to UQ biosynthesis, thereby resulting in a milder phenotype. Unfortunately, at the present time we have no tool to disrupt UQ biosynthesis in the presence of CLK-1(+) function or vice versa.

C. elegans tRNA missense suppressors. All six suppressors that we have cloned, and likely all nine that we have isolated, encode tRNAGlu genes whose anticodons are altered to read the substituted Lys codon of clk-1(e2519). To our knowledge, these suppressors represent the first missense suppressors identified in C. elegans and in any metazoan. In fact, C. elegans is the only metazoan in which any tRNA suppressors have been identified in forward genetic screens, and to date, all are nonsense amber suppressors. Amber mutations introduce TAG nonsense codons. All eight of the amber suppressors cloned so far contain mutations in the anticodon of tRNATrp genes (CCA to CTA) that can decode this nonsense codon as Trp.

Amber suppressors have several properties, including that the strength of suppression is dose and often temperature dependent, that the suppressors are tissue specific, that different suppressors exhibit different strengths of suppression, and that strong suppression is deleterious (reviewed in reference 20). Our analysis suggests that the tRNAGlu suppressors have some similarities to the amber suppressors but are also different in some interesting ways. For example, similar to the amber suppressors, we observe a strong dosage effect for the suppressors on at least one phenotype, brood size, suggesting that the suppressors are semidominant for this phenotype. However, they can also be fully dominant for some other phenotypes. We interpret this to mean that the level of UQ restored by a single copy of the suppressor may reach the threshold requirement for some phenotypes but not for others. It is remarkable that, although there are 24 genes that decode the Glu codon (GAG) (http://lowelab.ucsc.edu/GtRNAdb/Celeg/), a single copy of one mutant gene is sufficient for a dramatic phenotypic suppression.

In contrast to the amber suppressors, we have not observed any deleterious effects with our suppressors. They can be maintained indefinitely in the homozygous state and at any temperature without exhibiting obvious growth defects or sterility. There might be a number of reasons for this. For example, producing elongated proteins, which could act as dominant negatives, may be more deleterious than replacing some lysines by glutamic acids. Moreover, the suppressors described here cannot simply be classified as weak or strong. Indeed, different suppressors suppress different subsets of phenotypes, which cannot be clearly related to the amount of UQ biosynthesis they restore. Instead, the observation that they suppress different sets of phenotypes may suggest that some phenotypes have a stage- or tissue-specific focus of action and that the suppressors are stage or tissue specific as well, as are the amber suppressors (21, 22, 26). This would be consistent with our finding that even one copy of a particular suppressor can almost fully rescue the defecation phenotype, while even two copies of other suppressors can have no effect on this particular phenotype. Indeed, both of the alleles of rte-1, which were isolated independently, do not suppress the low defecation rate phenotype, indicating that this is a particular characteristic of rte-1 and not a genetic background effect. Finally, differences in the levels of expression might also play a role (26). For example, given that the qm211 mutation has a weaker effect on the developmental phenotypes as well as on the life span phenotype, the affected gene could be expressed at a lower level. Given that it has been shown that there are important internal promoter sequences in C. elegans tRNA genes (6), it is interesting to note that all of these suppressors have identical coding sequences. This suggests the importance of additional elements outside of the coding region for regulating the expression of the various tRNAGlu genes, as has been shown for the tRNATrp genes (21, 22, 26).

Finally, we can speculate on why missense suppressors have not been isolated before in C. elegans. The almost full suppression we observe, compared to the very small amount of UQ biosynthesis restored, suggests that in the case of clk-1, uniquely small amounts of wild-type protein are sufficient for a nearly wild-type phenotype. It is likely that other genes may require a higher proportion of wild-type protein for a wild-type phenotype.


   ACKNOWLEDGMENTS
 
We thank David Shanks for his work in the early stages of the project, particularly for helping to isolate mutants. We gratefully acknowledge C. F. Clarke, P. N. Rather, and M. Kawamukai for kindly providing bacterial strains and the Caenorhabditis Genetics Centre, which is funded by the National Institute of Health National Center for Research Resources (NIH NCRR), for providing nematode strains.

This work was funded in part by a research contract from Chronogen Inc. R.B. was supported by a Mary Louise Taylor McGill Major Fellowship, P.A.T.N. by a McGill Faculty of Graduate Studies Fellowship, and S.H. is a Strathcona professor of zoology.


   FOOTNOTES
 
* Corresponding author. Mailing address: Department of Biology, McGill University, 1205 Ave. Docteur Penfield, Montreal, Quebec, Canada H3A 1B1. Phone: (514) 398-6440. Fax: (514) 398-1674. E-mail: siegfried.hekimi{at}mcgill.ca. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Bolten, S. L., P. Powell-Abel, D. A. Fischhoff, and R. H. Waterston. 1984. The sup-7(st5) X gene of Caenorhabditis elegans encodes a Formula amber suppressor. Proc. Natl. Acad. Sci. USA 81:6784-6788.[Abstract/Free Full Text]

2. Braeckman, B. P., K. Houthoofd, K. Brys, I. Lenaerts, A. De Vreese, S. Van Eygen, H. Raes, and J. R. Vanfleteren. 2002. No reduction of energy metabolism in Clk mutants. Mech. Ageing Dev. 123:1447-1456.[CrossRef][Medline]

3. Braeckman, B. P., K. Houthoofd, A. De Vreese, and J. R. Vanfleteren. 1999. Apparent uncoupling of energy production and consumption in long-lived Clk mutants of Caenorhabditis elegans. Curr. Biol. 9:493-496.[CrossRef][Medline]

4. Branicky, R., Y. Shibata, J. Feng, and S. Hekimi. 2001. Phenotypic and suppressor analysis of defecation in clk-1 mutants reveals that reaction to changes in temperature is an active process in Caenorhabditis elegans. Genetics 159:997-1006.[Abstract/Free Full Text]

5. Burgess, J., A. K. Hihi, C. Y. Benard, R. Branicky, and S. Hekimi. 2003. Molecular mechanism of maternal rescue in the clk-1 mutants of Caenorhabditis elegans. J. Biol. Chem. 278:49555-49562.[Abstract/Free Full Text]

6. Ciliberto, G., L. Castagnoli, D. A. Melton, and R. Cortese. 1982. Promoter of a eukaryotic tRNAPro gene is composed of three noncontiguous regions. Proc. Natl. Acad. Sci. USA 79:1195-1199.[Abstract/Free Full Text]

7. Crane, F. L. 2001. Biochemical functions of coenzyme Q10. J. Am. Coll. Nutr. 20:591-598.[Abstract/Free Full Text]

8. Dallner, G., and P. J. Sindelar. 2000. Regulation of ubiquinone metabolism. Free Radic. Biol. Med. 29:285-294.[CrossRef][Medline]

9. Ewbank, J. J., T. M. Barnes, B. Lakowski, M. Lussier, H. Bussey, and S. Hekimi. 1997. Structural and functional conservation of the Caenorhabditis elegans timing gene clk-1. Science 275:980-983.[Abstract/Free Full Text]

10. Felkai, S., J. J. Ewbank, J. Lemieux, J. C. Labbe, G. G. Brown, and S. Hekimi. 1999. CLK-1 controls respiration, behavior and aging in the nematode Caenorhabditis elegans. EMBO J. 18:1783-1792.[CrossRef][Medline]

11. Fernandez-Ayala, D. J., G. Lopez-Lluch, M. Garcia-Valdes, A. Arroyo, and P. Navas. 2005. Specificity of coenzyme Q10 for a balanced function of respiratory chain and endogenous ubiquinone biosynthesis in human cells. Biochim. Biophys. Acta 1706:174-183.[Medline]

12. Hihi, A. K., Y. Gao, and S. Hekimi. 2002. Ubiquinone is necessary for Caenorhabditis elegans development at mitochondrial and non-mitochondrial sites. J. Biol. Chem. 277:2202-2206.[Abstract/Free Full Text]

13. Hihi, A. K., H. Kebir, and S. Hekimi. 2003. Sensitivity of Caenorhabditis elegans clk-1 mutants to ubiquinone side-chain length reveals multiple ubiquinone-dependent processes. J. Biol. Chem. 278:41013-41018.[Abstract/Free Full Text]

14. Hsu, A. Y., W. W. Poon, J. A. Shepherd, D. C. Myles, and C. F. Clarke. 1996. Complementation of coq3 mutant yeast by mitochondrial targeting of the Escherichia coli UbiG polypeptide: evidence that UbiG catalyzes both O-methylation steps in ubiquinone biosynthesis. Biochemistry 35:9797-9806.[CrossRef][Medline]

15. Jonassen, T., D. E. Davis, P. L. Larsen, and C. F. Clarke. 2003. Reproductive fitness and quinone content of Caenorhabditis elegans clk-1 mutants fed coenzyme Q isoforms of varying length. J. Biol. Chem. 278:51735-51742.[Abstract/Free Full Text]

16. Jonassen, T., P. L. Larsen, and C. F. Clarke. 2001. A dietary source of coenzyme Q is essential for growth of long-lived Caenorhabditis elegans clk-1 mutants. Proc. Natl. Acad. Sci. USA 98:421-426.[Abstract/Free Full Text]

17. Jonassen, T., B. N. Marbois, K. F. Faull, C. F. Clarke, and P. L. Larsen. 2002. Development and fertility in Caenorhabditis elegans clk-1 mutants depend upon transport of dietary coenzyme Q8 to mitochondria. J. Biol. Chem. 277:45020-45027.[Abstract/Free Full Text]

18. Jonassen, T., M. Proft, F. Randez-Gil, J. R. Schultz, B. N. Marbois, K. D. Entian, and C. F. Clarke. 1998. Yeast Clk-1 homologue (Coq7/Cat5) is a mitochondrial protein in coenzyme Q synthesis. J. Biol. Chem. 273:3351-3357.[Abstract/Free Full Text]

19. Kayser, E. B., M. M. Sedensky, and P. G. Morgan. 2004. The effects of complex I function and oxidative damage on lifespan and anesthetic sensitivity in Caenorhabditis elegans. Mech. Ageing Dev. 125:455-464.[CrossRef][Medline]

20. Kimble, P. A. A. J. 1997. mRNA and translation, p. 185-208. In D. L. Riddle, T. Blumenthal, B. J. Meyer, and J. R. Priess (ed.), C. elegans II. Cold Spring Harbor monograph series 33. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

21. Kondo, K., J. Hodgkin, and R. H. Waterston. 1988. Differential expression of five Formulaamber suppressors in Caenorhabditis elegans. Mol. Cell. Biol. 8:3627-3635.[Abstract/Free Full Text]

22. Kondo, K., B. Makovec, R. H. Waterston, and J. Hodgkin. 1990. Genetic and molecular analysis of eight tRNATrp amber suppressors in Caenorhabditis elegans. J. Mol. Biol. 215:7-19.[CrossRef][Medline]

23. Lakowski, B., and S. Hekimi. 1996. Determination of life-span in Caenorhabditis elegans by four clock genes. Science 272:1010-1013.[Abstract]

24. Lakowski, B., and S. Hekimi. 1998. The genetics of caloric restriction in Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 95:13091-13096.[Abstract/Free Full Text]

25. Levavasseur, F., H. Miyadera, J. Sirois, M. L. Tremblay, K. Kita, E. Shoubridge, and S. Hekimi. 2001. Ubiquinone is necessary for mouse embryonic development but is not essential for mitochondrial respiration. J. Biol. Chem. 276:46160-46164.[Abstract/Free Full Text]

26. Li, L., R. M. Linning, K. Kondo, and B. M. Honda. 1998. Differential expression of individual suppressor tRNATrp gene family members in vitro and in vivo in the nematode Caenorhabditis elegans. Mol. Cell. Biol. 18:703-709.[Abstract/Free Full Text]

27. Liu, X., N. Jiang, B. Hughes, E. Bigras, E. Shoubridge, and S. Hekimi. 2005. Evolutionary conservation of the clk-1-dependent mechanism of longevity: loss of mclk1 increases cellular fitness and lifespan in mice. Genes Dev. 19:2424-2434.[Abstract/Free Full Text]

28. Macinga, D. R., G. M. Cook, R. K. Poole, and P. N. Rather. 1998. Identification and characterization of aarF, a locus required for production of ubiquinone in Providencia stuartii and Escherichia coli and for expression of 2'-N-acetyltransferase in P. stuartii. J. Bacteriol. 180:128-135.[Abstract/Free Full Text]

29. Marbois, B. N., and C. F. Clarke. 1996. The COQ7 gene encodes a protein in Saccharomyces cerevisiae necessary for ubiquinone biosynthesis. J. Biol. Chem. 271:2995-3004.[Abstract/Free Full Text]

30. Miyadera, H., H. Amino, A. Hiraishi, H. Taka, K. Murayama, H. Miyoshi, K. Sakamoto, N. Ishii, S. Hekimi, and K. Kita. 2001. Altered quinone biosynthesis in the long-lived clk-1 mutants of Caenorhabditis elegans. J. Biol. Chem. 276:7713-7716.[Abstract/Free Full Text]

31. Miyadera, H., K. Kano, H. Miyoshi, N. Ishii, S. Hekimi, and K. Kita. 2002. Quinones in long-lived clk-1 mutants of Caenorhabditis elegans. FEBS Lett. 512:33-37.[CrossRef][Medline]

32. Okada, K., M. Minehira, X. Zhu, K. Suzuki, T. Nakagawa, H. Matsuda, and M. Kawamukai. 1997. The ispB gene encoding octaprenyl diphosphate synthase is essential for growth of Escherichia coli. J. Bacteriol. 179:3058-3060.[Abstract/Free Full Text]

33. Padilla, S., T. Jonassen, M. A. Jimenez-Hidalgo, D. J. Fernandez-Ayala, G. Lopez-Lluch, B. Marbois, P. Navas, C. F. Clarke, and C. Santos-Ocana. 2004. Demethoxy-Q, an intermediate of coenzyme Q biosynthesis, fails to support respiration in Saccharomyces cerevisiae and lacks antioxidant activity. J. Biol. Chem. 279:25995-26004.[Abstract/Free Full Text]

34. Rousseau, G., and F. Varin. 1998. Determination of ubiquinone-9 and 10 levels in rat tissues and blood by high-performance liquid chromatography with ultraviolet detection. J. Chromatogr. Sci. 36:247-252.[Medline]

35. Shibata, Y., R. Branicky, I. O. Landaverde, and S. Hekimi. 2003. Redox regulation of germline and vulval development in Caenorhabditis elegans. Science 302:1779-1782.[Abstract/Free Full Text]

36. Stenmark, P., J. Grunler, J. Mattsson, P. J. Sindelar, P. Nordlund, and D. A. Berthold. 2001. A new member of the family of di-iron carboxylate proteins. Coq7 (clk-1), a membrane-bound hydroxylase involved in ubiquinone biosynthesis. J. Biol. Chem. 276:33297-33300.[Abstract/Free Full Text]

37. Sulston, J., and J. Hodgkin. 1988. Methods, p. 587-606. In W. B. Wood (ed.), The nematode Caenorhabditis elegans. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

38. Turunen, M., J. Olsson, and G. Dallner. 2004. Metabolism and function of coenzyme Q. Biochim. Biophys. Acta 1660:171-199.[Medline]

39. Wallace, B. J., and I. G. Young. 1977. Aerobic respiration in mutants of Escherichia coli accumulating quinone analogues of ubiquinone. Biochim. Biophys. Acta 461:75-83.[Medline]

40. Waterston, R. H. 1981. A second informational suppressor, SUP-7 X, in Caenorhabditis elegans. Genetics 97:307-325.[Abstract/Free Full Text]

41. Wong, A., P. Boutis, and S. Hekimi. 1995. Mutations in the clk-1 gene of Caenorhabditis elegans affect developmental and behavioral timing. Genetics 139:1247-1259.[Abstract]

Molecular and Cellular Biology, May 2006, p. 3976-3985, Vol. 26, No. 10
0270-7306/06/$08.00+0     doi:10.1128/MCB.26.10.3976-3985.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Branicky, R.
Right arrow Articles by Hekimi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Branicky, R.
Right arrow Articles by Hekimi, S.

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
J. Bacteriol. J. Virol. Eukaryot. Cell
Microbiol. Mol. Biol. Rev. Clin. Vaccine Immunol. All ASM Journals