dpy-18 Encodes an -Subunit of Prolyl-4-Hydroxylase in Caenorhabditis elegans

Corresponding author: Steven W. L'Hernault, Department of Biology, Emory University, 1510 Clifton Rd., Atlanta, GA 30322., bioslh{at}biology.emory.edu (E-mail)

Communicating editor: R. K. HERMAN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Collagen is an extracellular matrix (ECM) component encoded by a large multigene family in multicellular animals. Procollagen is post-translationally modified by prolyl-4-hydroxylase (EC 1.14.11.2) before secretion and participation in ECM formation. Therefore, collagen processing and regulation can be studied by examining this required interaction of prolyl-4-hydroxylase with procollagen. High-resolution polymorphism mapping was used to place the Caenorhabditis elegans dpy-18 gene on the physical map, and we show that it encodes a prolyl-4-hydroxylase catalytic subunit. The Dpy phenotype of dpy-18(e364) amber mutants is more severe when this mutation is in trans to the noncomplementing deficiency tDf7, while the dpy-18(e499) deletion mutant exhibits the same phenotype as dpy-18(e499)/tDf7. Furthermore, dpy-18 RNA interference (RNAi) in wild-type worms results in Dpy progeny, while dpy-18 (RNAi) in dpy-18(e499) mutants does not alter the Dpy phenotype of their progeny. These observations suggest that the dpy-18 null phenotype is Dpy. A dpy-18::gfp promoter fusion construct is expressed throughout the hypodermis within the cells that abundantly produce the cuticle collagens, as well as in certain head and posterior neurons. While prolyl-4-hydroxylase has been studied extensively by biochemical techniques, this is the first report of a mutationally defined prolyl-4-hydroxylase in any animal.

COLLAGENS are an important structural component of the extracellular matrix (ECM) in all multicellular animals. About 30 genes encode collagen in mammals, and ~25% of total body protein is some form of collagen (reviewed by VAN DER REST and GARRONE 1991 ). The nematode Caenorhabditis elegans has >100 genes that encode collagen (COX et al. 1984 ), and at least 60 of these genes are known to be expressed (POLITZ and EDGAR 1984 ). Two of the C. elegans collagen genes, let-2 and emb-9, encode type IV basement membrane collagens (GUO and KRAMER 1989 ), while the remaining genes encode cuticle collagens. C. elegans expresses collagen throughout its life cycle, but the amount and type of collagen expressed vary greatly according to developmental stage. In general, expression of cuticle collagens increases during the four larval molts, when a new cuticle is synthesized (COX et al. 1981A ; COX and HIRSH 1985 ). In addition, several collagen-encoding genes are expressed only by dauer larvae, which are a stress-resistant developmental stage (COX et al. 1981B ; KRAMER et al. 1985 ).

A structural feature common to all collagens is the presence of at least one Gly-X-Y tripeptide repeat that interacts with the Gly-X-Y domains of two other collagen polypeptides to form a triple helix. The X and Y positions within the tripeptide repeat are often occupied by proline. Collagen proproteins undergo co- and post-translational modification within the endoplasmic reticulum (ER) before they are secreted from the cell to be incorporated into the ECM (for a review of collagen post-translational processing, see KIVIRIKKO et al. 1992 ). In particular, proline residues in the Y position are often modified to 4-hydroxyproline by the enzyme prolyl-4-hydroxylase. This modification allows collagen polypeptides to hydrogen bond to each other during triple-helix formation and is essential for triple-helix stability at physiological temperatures (reviewed by BRODSKY and RAMSHAW 1997 ). Moreover, an incompletely hydroxylated procollagen polypeptide remains bound to prolyl-4-hydroxylase, is retained within the ER lumen, and is unable to contribute to the ECM (WALMSLEY et al. 1999 ).

Some aspects of collagen biochemistry are similar in C. elegans and in vertebrates, but the distribution of collagen subtypes found in C. elegans is significantly different from that found in vertebrates. For instance, most vertebrate collagens are fibrillar (types I, II, and III), but C. elegans collagens have the highest sequence and predicted structural homology to two classes of vertebrate nonfribrillar collagen. The let-2 and emb-9 genes encode network-forming type IV basement membrane collagens, while the cuticle is composed principally of collagens that are most similar to the nonfibrillar vertebrate fibril-associated collagen with interrupted triple helix (FACIT; GORDON and OLSEN 1990 ; SHAW and OLSEN 1991 ; KRAMER 1994A , KRAMER 1994B ). Vertebrate FACIT collagens are hypothesized to control collagen fibril geometry, fibrillar collagen interactions with other ECM proteins, or the flexibility/compressibility of the ECM. Although their precise role(s) is uncertain, vertebrate FACITs appear to function in association with fibrillar collagens.

In vertebrates, collagen triple-helix stability requires post-translational modifications that include formation of hydroxyproline. C. elegans collagen contains ~12% hydroxyproline (KRAMER 1994A , KRAMER 1994B ), suggesting that this collagen post-translational modification is important for collagen function in this invertebrate. C. elegans lives at cooler temperatures than the warm-blooded vertebrates used for most collagen biochemical studies, and it is known that the requirement for hydroxyproline during collagen assembly can be affected by temperature. For instance, vertebrate fibrillar collagen that lacks hydroxyproline can be assembled into triple-helical filaments in vitro at low, nonphysiological temperatures, but these filaments fall apart at the normal body temperature. However, one vertebrate collagen type that is similar to that found in the C. elegans cuticle, FACIT type XII collagen, requires proline hydroxylation for assembly of the triple helix at all assayed temperatures (MAZZORANA et al. 1993 ). This suggests that proline hydroxylation should be required to assemble the C. elegans cuticle at the temperatures (16°–25°) normally employed to grow this organism in the laboratory.

Prolyl-4-hydroxylase consists of catalytically active -subunits and protein disulfide isomerase (PDI) ß-subunits (EC 5.3.4.1; KOIVU et al. 1987 ; PIHLAJANIEMI et al. 1987 ) that form an 2ß2 tetramer in most animals (reviewed by KIVIRIKKO and PIHLAJANIEMI 1998 ). The biochemistry of prolyl-4-hydroxylase, and in particular the -subunit, has been studied extensively because of its essential role in collagen biosynthesis. A cDNA for a C. elegans -subunit was shown to encode a protein that could form a catalytically active dimer with a human ß-subunit (VEIJOLA et al. 1994 ). This in vitro result was initially surprising because this enzyme is usually active only as a tetramer. However, it was later shown that the C. elegans active enzyme is also an ß-dimer in vitro (VEIJOLA et al. 1996 ).

Here we report that this previously identified C. elegans -subunit of prolyl-4-hydroxylase (VEIJOLA et al. 1996 ) is encoded by dpy-18, and this is the first mutationally defined prolyl-4-hydroxylase in any animal. We have sequenced two dpy-18 mutations and show that the null phenotype for dpy-18 is Dpy, demonstrating that this prolyl-4-hydroxylase -subunit plays an important role during morphogenesis. The dpy-18 promoter is active in the hypodermis, consistent with the production of cuticle collagens in this tissue throughout the C. elegans life cycle. C. elegans produces many different collagens that are probably all post-translationally modified by prolyl-4-hydroxylase. Therefore, DPY-18 might provide both new insights into the regulation of collagen biosynthesis and a target for antinematode pharmaceuticals.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains, culture conditions, nomenclature, and genetic markers:
Bristol N2 is the reference wild-type strain used in all C. elegans experiments (BRENNER 1974 ). The strains used in this study, with the exception of CB4856, were all derived from N2. The C. elegans strain CB4856 was isolated in 1972 from a pineapple field in Hawaii (HODGKIN and DONIACH 1997 ) and was provided by T. Schedl. All genetic manipulations and culturing of C. elegans were performed as previously described (BRENNER 1974 ), and standard nomenclature was used (HORVITZ et al. 1979 ). The following genes and mutations were used in this study: LGIII: vab-7(e1562) (HODGKIN 1983 ), fer-2(hc2ts) (WARD et al. 1981 ), spe-16(hc54ts, eb35ts) (SHAKES 1988 ; K. L. HILL and S. W. L'HERNAULT, unpublished data), dpy-18(e499, e364) (BRENNER 1974 ); LGIV: him-3(e1147) (HODGKIN et al. 1979 ); LGV: him-5(e1490) (HODGKIN et al. 1979 ). The balanced deficiency strain unc-32(e189) tDf7/qC1 dpy-19(e1259) glp-1(q339); him-3(e1147) was provided by H. and R. Schnabel.

Body length measurement of dpy-18 homo- and hemizygotes:
dpy-18(e499) hemizygotes were obtained by crossing homozygous dpy-18(e499) L4 hermaphrodites to heterozygous males carrying the noncomplementing deficiency tDf7. The P₀ adults were allowed to lay eggs on plates for 1 day and they were transferred individually to fresh plates on each successive day. After ascertaining the presence of outcrossed progeny on the first-day mate plates, L4 Dpy dpy-18(e499)/tDf7 hermaphrodites were picked from the second-day mate plates to fresh plates, incubated at 16° overnight, and measured the following day. Age-matched homozygous dpy-18(e499) control hermaphrodites were also picked and incubated at 16° at the same time. dpy-18(e364) hemizygotes and homozygous controls were obtained in a similar manner, using spe-16(hc54ts) dpy-18(e364).

Hemi- and homozygous animals were immobilized in 2.5 mM levamisole and mounted on 2% agar pads under coverslips. Worms were observed by differential interference contrast (DIC) at x10 on a Zeiss Axiophot compound microscope with a DAGE CCD300T-RC camera (Dage-MTI Inc., Michigan City, IN). Resulting images were captured with a Scion LG3 Frame capture board (Scion Corp., Frederick, MD) and manipulated with NIH Image software (Wayne Rasband, NIH). Images were subsequently displayed on a monitor with Adobe Photoshop PC v.5.0 (Adobe Systems, San Jose, CA). A length of dental floss was stretched from the tip of the pharynx to the tip of the tail of each worm, and the length in centimeters was recorded from the dental floss on a sheet of acetate overhead projector film. An image of a stage micrometer slide captured at the same magnification and manipulated in the same fashion allowed measurements of worms in centimeters to be converted to microns. Data analysis was performed using the Microsoft Excel 98 program (Microsoft, Redmond, WA); all t-tests are two-tailed, assuming unequal variance.

DNA nomenclature and sequencing:
As described below, dpy-18 encodes an -subunit of prolyl-4-hydroxylase, and a cDNA sequence for this gene has been published without knowledge of its association with dpy-18 (VEIJOLA et al. 1994 ). Additionally, the genomic sequence of the chromosome III -subunit of prolyl-4-hydroxylase and several corresponding expressed sequence tags (ESTs) have been described; the predicted gene is Y47D3B.10 (Fig 1A). Y47D3B.10 is encoded by the yeast artificial chromosome (YAC) Y47D3 and is partially present on the cosmid T28D6. In this article, the first A of the start AUG (as proposed by VEIJOLA et al. 1994 and for Y47D3B.10) is numbered 1 and the last A of the UAA stop codon is numbered 4934 (based on the genomic sequence that encodes Y47D3B.10). This numbering scheme includes all introns, and positive numbering continues past the stop codon into the 3' untranslated region (UTR). Nucleotides that are 5' to the start AUG are indicated by negative numbers.

View larger version (21K): In this window In a new window Download PPT slide   Figure 1. (A) dpy-18 region of chromosome IIIR. The deficiency tDf7 fails to complement dpy-18, spe-16, and fer-2, but complements vab-7 and nob-1. The ~7 map units between the cloned genes nearest to dpy-18, tra-1 (left), and nob-1 (right) correspond to ~645 kb. The prolyl-4-hydroxylase -subunit gene encoded by dpy-18 is wholly within the YAC Y47D3 and is partially present on the cosmid T28D6, and the 2-map-unit interval between tra-1 and dpy-18 is ~270 kb. ebP1 and ebP2 denote PCR product-size polymorphisms that lie within and left of dpy-18, respectively. (B) Agarose gel electrophoresis of DNA size standards (lane 1) and PCR products (lanes 2–5). The primer pair 5 and 6 (see Table 1) was used to detect ebP1, which is ~1.9 kb in Bristol N2 control genomic DNA (lane 2) and 1.7 kb in CB4856 (lane 3). The primer pair 9 and 10 (see Table 1) was used to detect ebP2, which is ~1.8 kb in Bristol N2 control genomic DNA (lane 4) and 1.65 kb in CB4856 (lane 5). The dpy-18 gene is transcribed from right to left.

All DNA sequencing in this study was performed by the DNA sequencing facility at Iowa State University, using standard ABI (Perkin-Elmer Corp., Foster City, CA) automated fluorescent sequencing methods for polymerase chain reaction (PCR) products. The chromosome III prolyl-4-hydroxylase -subunit gene was PCR-amplified from genomic DNA derived from dpy-18(e364) and dpy-18(e499). The PCR products from four independent reactions for each genotype were pooled, purified using phenol-chloroform extraction or the Qiaquick PCR purification kit (QIAGEN, Valencia, CA), and sequenced to identify the mutant lesions. Sequence traces from the dpy-18 mutants were compared with the corresponding control sequence traces from N2 and also to the sequence available from the Sanger Centre (Hinxton, England).

Polymorphism mapping:
Previously, dpy-18 was mapped ~2 map units right of tra-1 on chromosome III (HODGKIN 1997 ). We narrowed the DNA region that contains dpy-18 by mapping relative to sequence polymorphisms (JAKUBOWSKI and KORNFELD 1999 ) because this region is characterized by unstable recombinant DNA clones (HODGKIN 1993 ; K. L. HILL and S. W. L'HERNAULT, unpublished data) and there are no cloned genes close to dpy-18. This approach requires locating suitable sequence polymorphisms, and our new data, together with previously determined sequence data (see above), guided design of single-stranded DNA primer pairs. Each primer was designed to a predicted exon of a predicted open reading frame, and a primer pair was designed for every ~10 kb of genomic sequence. Each primer pair allows amplification of an ~1–2-kb DNA fragment that includes at least one predicted intron, because noncoding regions usually have more frequent sequence alterations than exons. PCR fragments were amplified from genomic DNA prepared from Bristol N2 and the Hawaiian C. elegans isolate CB4856 by standard methods (BARSTEAD and WATERSTON 1991 ). Some primer pairs generated PCR products that showed a size difference between the two different C. elegans strains, and one such polymorphism, ebP1, was used to identify a right limit on the position of dpy-18. Primer pair 5 and 6 reveals fragment size polymorphism ebP1 while primer pair 9 and 10 reveals an additional fragment size polymorphism, ebP2, when PCR products are fractionated by agarose gel electrophoresis (Table 1; Fig 1B).

CB4856 males were crossed to vab-7(e1562) fer-2(hc2ts) spe-16(hc54ts) dpy-18(e364) self-sterile hermaphrodites (raised restrictively at 25°); F₁ outcross heterozygous hermaphrodites were raised at 20°, and their F₂ progeny were screened for Vab non-Dpy recombinants. We obtained 21 Vab non-Dpy recombinants, and homozygous recombinant lines were used to prepare genomic template DNA (BARSTEAD and WATERSTON 1991 ) for polymorphism analysis. PCR products from each recombinant template were typed by agarose gel electrophoresis to map ebP1 with respect to dpy-18.

RNA interference:
Three EST clones (yk339d8, yk276b3, and yk247d7) corresponding to the C. elegans prolyl-4-hydroxylase -subunit gene on chromosome III were obtained from Yuji Kohara (Genome Biology Lab, Center for Genetic Resource Information, National Institute of Genetics, Mishima, Japan). The cDNA yk247d7, which contains a dpy-18 insert that lacks only the first 19 nucleotides of protein-coding sequence, was used for these studies. RNA interference (RNAi; FIRE et al. 1998 ), performed using dsRNA generated from this cDNA, produced an embryonic lethal phenotype substantially different from that of any known dpy-18 alleles. The yk247d7 cDNA contains several regions of high sequence homology to a second prolyl-4-hydroxylase -subunit present in the C. elegans genome (F35G2.4, CAA93466–CAA93470, F35G2 cosmid accession no. Z69637); therefore the dsRNA generated from yk247d7 may have targeted both prolyl-4-hydroxylase -subunits. To prevent RNAi crosstalk from affecting expression of both genes, a specific template derived from yk247d7 was prepared. yk247d7 was truncated with the restriction enzyme KpnI (Fig 2A), and the resulting 337-bp fragment of 5' dpy-18 sequence does not show significant nucleotide homology to F35G2.4. The resulting cloned construct, pKH19, was the template for in vitro synthesis of a double-stranded RNA that was injected into either N2 or dpy-18(e499) hermaphrodites by standard techniques (FIRE et al. 1998 ). The injected animals were incubated at 25°, transferred to fresh plates at 24-hr intervals for 2 days following injection, and the resulting progeny were examined after an additional 24–48 hr of incubation. Embryos were scored as arrested if they had not hatched 24 hr after removal of the P₀. The age-matched sibs of injected dpy-18(e499) hermaphrodites were picked and transferred in parallel to provide control progeny for comparison. Offspring of both injected and uninjected dpy-18(e499) animals fell into four categories: Dumpy (Dpy), Lumpy Dumpy (LpyDpy; medium to strong Dpy with mild swellings and/or constrictions), severe Lumpy Dumpy (strong Dpy with gross abnormalities in body shape), and dead eggs. Because there was no qualitative difference in phenotype between dpy-18(e499) and dpy-18(e499); (dpy-18(RNAi)), the percentage of total progeny falling into each class was compared between injected and uninjected animals, using Student's t-test to determine if there was a quantitative difference in phenotype (see above). Selected progeny in M-9 buffer were mounted on 2% agar pads under coverslips, examined by DIC with a Zeiss Axiophot compound microscope, and digital images were prepared as described above to document each phenotypic class.

View larger version (25K): In this window In a new window Download PPT slide   Figure 2. dpy-18 rescuing and expression constructs. Thin horizontal lines represent 5' UTR, introns, and 3' UTR; numbered boxes represent exons. (A) The 7140-bp chromosome III region cloned in pKH9 that rescues dpy-18 mutants. The AUG indicated in exon 2 is the proposed start codon under normal conditions, and the A of this codon is numbered 1. The numbered positions of the 5' and 3' ends of exons 1, 2, and 3 are indicated below the line. There are at least two known variants for exon 1 and both are spliced to position -1 of exon 2. The first is cis spliced and has been described previously (VEIJOLA et al. 1994 ), while the second is a trans-spliced SL1 exon. The positions of the dpy-18(e364) nonsense amber mutation in exon 3 and the UAA stop codon in exon 8 are indicated above the line. (B) The 6348-bp dpy-18(e499) region, using the same numbered coordinates as for wild type in A. The coordinates of the two regions deleted () by the dpy-18(e499) compound deletion are indicated below the line. The first deletion removes 5' UTR sequence, the cis exon 1, and part of exon 2, including the proposed start AUG codon while the second deletion removes a second region of exon 2. (C) The structure of plasmid pBHdpy-18::gfp2. The wild-type region of dpy-18 from -2008 through +3 was ligated to the PstI site of plasmid pPD95.69, placing dpy-18 sequence 5' to the SV40 nuclear localization signal (NLS), followed by GFP coding sequence and an unc-54 3' UTR. For additional information on plasmid construction, see MATERIALS AND METHODS.

Reverse transcription PCR:
Total RNA was prepared from him-5(e1490) using the QIAGEN (Chatsworth, CA) RNEasy kit, treated with RQ1 RNase-free DNase (Promega Corp., Madison, WI), and repurified using the RNEasy kit. Reverse transcription (RT)-PCR was performed using Ready-to-Go You-Prime first-strand synthesis beads (Pharmacia Biotech, Piscataway, NJ) and an oligo(dT) primer (Promega). PCR on the resulting cDNA was carried out using Ready-to-Go PCR beads (Pharmacia). Primer pair 5 and 6 (Table 1) was used as a positive control to confirm the presence of dpy-18 cDNA in the first-strand RT pools. Primer 7 was used with either primer 11 or 12 (Table 1) to test for the presence of a trans-spliced exon at the 5' end of dpy-18 mRNA.

Transgenic rescue of dpy-18:
Two overlapping PCR fragments were amplified from Bristol N2 genomic DNA using primer pairs 1 and 2, and 3 and 4 (Table 1). Resulting 5' and 3' dpy-18-containing fragments were cloned, cut with compatible enzymes, and the inserts were ligated into Bluescript (Stratagene, La Jolla, CA) to create pKH9. pKH9 is a 7140-bp construct that contains 2008 bp of sequence 5' to the predicted start AUG, the entire dpy-18 gene sequence, and 198 bp 3' to the stop codon. pKH9 was coinjected with the dominantly expressed fluorescent marker plasmid pPD118.33 [a myo-2 pharyngeal promoter driving green fluorescent protein (GFP; CHALFIE et al. 1994 ; A. FIRE, S. XU, J. AHNN and G. SEYDOUX, personal communication)] into either dpy-18(e499) or unc-49(e382) spe-16(eb35ts) dpy-18(e364)/qC1 [dpy-19(e1259) glp-1(q339)] young adult hermaphrodites following standard techniques (FIRE 1986 ; MELLO et al. 1991 ). The mean body length of GFP(+) F₁ progeny from injected animals was compared to the mean length of their GFP(-) sibs to assess rescue of dpy-18(e499) (see above). dpy-18(e364) rescue was inferred from the presence of GFP(+) Unc non-Dpy progeny in broods of injected unc-49(e382) spe-16(eb35ts) dpy-18(e364)/qC1 [dpy-19(e1259) glp-1(q339)] hermaphrodites. This strain does not normally segregate Unc non-Dpy animals because the balancer qC1 effectively suppresses recombination between unc-49 and dpy-18 (EDGLEY et al. 1995 ).

dpy-18::gfp expression construct:
PCR was used to amplify a 2011-bp genomic fragment from N2 DNA with primer pair 1 and 8 (Table 1). This PCR fragment, which contains 2008 bp of sequence 5' to the predicted start AUG and has a restriction enzyme linker at each end, was restriction digested and ligated in frame to the promoterless gfp expression vector pPD95.69 (CHALFIE et al. 1994 ; A. FIRE, S. XU, J. AHNN and G. SEYDOUX, personal communication) to create pBHdpy-18::gfp2 (Fig 2C). This construct was restriction mapped and the dpy-18-pPD95.69 junction was sequenced to verify the clone construction. The expression construct was coinjected with the dominant transformation marker pRF4 [rol-6(su1006); MELLO et al. 1991] by standard methods (FIRE 1986 ; MELLO et al. 1991 ), and stable transgenic lines were established before assessing the pattern of GFP expression in transgenic animals. Adult Rol animals were anesthetized and images were acquired as described above.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Polymorphism mapping of dpy-18:
The dpy-18 gene is located on the right arm of C. elegans chromosome III, between the cloned genes tra-1 (to the left) and nob-1 (to the right; Y75B8A.2; see Fig 1). This ~7-map-unit interval contains ~645 kb of DNA, based on the sequence information available from the C. elegans Genomic Center (Sanger Centre). Conventional genetic analyses could not be used to further refine the dpy-18 map position because this region has few mutationally defined or cloned genes. Instead, polymorphism mapping was used to narrow the interval in which dpy-18 must reside. This technique allows direct correlation of the physical and genetic maps by ordering sequence polymorphisms with respect to visible genetic markers in mapping crosses (COX et al. 1985 ; RUVKUN et al. 1989 ; JAKUBOWSKI and KORNFELD 1999 ). The CB4856 C. elegans isolate from Hawaii (HODGKIN and DONIACH 1997 ) was used for this purpose because it has been shown to contain a sequence polymorphism (vs. Bristol N2 sequence) every ~1.5–2 kb (see http://genome.wustl.edu/gsc/CEpolymorph/snp.shtml).

The nearly complete genome sequence of C. elegans Bristol N2 facilitated our search for CB4856 polymorphisms within the dpy-18 region. Consistent with prior data (see Web site in the previous paragraph), some CB4856 polymorphisms discovered in the dpy-18 region were single-base changes (data not shown). However, several CB4856 polymorphisms, including ebP1 and ebP2, changed the sizes of PCR products from those obtained with a Bristol N2 template, which allowed recombinant typing by agarose gel electrophoresis (Fig 1B). A total of 21 Vab non-Dpy recombinants, including 5 within the 0.3-map-unit spe-16-dpy-18 interval, had the CB4856 allele of ebP1 and did not have the dpy-18(e364) mutation, indicating that this polymorphism lies to the right of, or within, the dpy-18 gene. This right-hand limit for the position of dpy-18 is ~270 kb to the right of tra-1. Many previously studied dpy genes affect collagen (KRAMER 1997 ), so this genomic sequence interval was examined for genes involved in collagen biosynthesis. This interval contains a prolyl-4-hydroxylase -subunit gene, and functional prolyl-4-hydroxylase is required for collagen assembly and stability (MAZZORANA et al. 1993 ; KIVIRIKKO and PIHLAJANIEMI 1998 ). Since the ebP1 polymorphism was located within intron 6 of this prolyl-4-hydroxylase -subunit gene, this gene was a plausible dpy-18 candidate (Fig 2).

Identification of dpy-18 as an -subunit of prolyl-4-hydroxylase:
Two overlapping genomic DNA-derived PCR products, which together contained the chromosome III prolyl-4-hydroxylase -subunit gene and flanking sequence, were cloned to create the ~7.1-kb insert-containing pKH9 plasmid (Fig 2A). Homozygous dpy-18(e499) worms that were transgenic for pKH9 had significantly longer bodies than their nontransformed Dpy sibs and were not significantly shorter than wild-type controls (Table 2); similar results were obtained with dpy-18(e364) (not shown; see MATERIALS AND METHODS). The rescue of the Dpy phenotype of both dpy-18(e499) and dpy-18(e364) suggested that the chromosome III -subunit of prolyl-4-hydroxylase is the dpy-18 gene.

The sequences of two dpy-18 mutant alleles were determined to confirm that this locus encodes the chromosome III -subunit of prolyl-4-hydroxylase. Prior work showed that dpy-18(e364) was suppressible by amber-suppressing tRNA mutations such as sup-5 and sup-7 (WATERSTON and BRENNER 1978 ; WATERSTON 1981 ). Since amber mutations usually alter coding sequence, PCR fragments containing the exon sequence of dpy-18(e364) and dpy-18(e499) were sequenced and compared to N2. This analysis showed that the dpy-18(e364) mutation is a GA transition at position 1851 (the fourth codon of exon 3) that changes a UGG tryptophan codon to a UAG amber stop codon (Fig 2A). One PCR product that included the 5' end of the prolyl-4-hydroxylase -subunit gene from dpy-18(e499) was significantly smaller than the comparable wild-type product (data not shown). The sequence of this region revealed that dpy-18(e499) contains a compound mutation that is two deletions that both affect the coding sequence (Fig 2C). The first deletion removes 776 bp starting at -684, extending through the start AUG and ending at +92. The second deletion removes 18 bp of exon 1, starting at 192 and ending at 207. These transformation rescue data and the presence of sequence alterations in two independent dpy-18 alleles confirm that dpy-18 encodes this -subunit of prolyl-4-hydroxylase.

A cDNA corresponding to the C. elegans dpy-18-encoded -subunit of prolyl-4-hydroxylase has been described previously (accession no. U12762; VEIJOLA et al. 1994 ). Our genomic sequence results, and those of the C. elegans Genome Project (T28D6.1, accession no. Z81134, or Y47D3B.10, accession no. AL031635) differ from this previously published sequence. The first difference is the 5' six nucleotides of the published cDNA (VEIJOLA et al. 1994 ) that do not correspond to genomic sequence and do not appear to be due to splicing. The second region of difference occurs within exon 6, starting at position 3260 and extending to position 3307: the encoded amino acid sequence is changed from VSRRHLRLYCYYLAGPSFL (VEIJOLA et al. 1994 ) to VSQKDISRLYCYYKRDRPFL. A cartoon illustrating the exon and intron structure of dpy-18 is shown in Fig 2A.

Characterization of residual dpy-18 function in mutant alleles and determination of the dpy-18 null phenotype:
Double-stranded RNA (FIRE et al. 1998 ) was created from the EST clone yk247d7-derived plasmid pKH19, which contains 337 bp of the 5' coding sequence corresponding to the chromosome III prolyl-4-hydroxylase -subunit gene (see MATERIALS AND METHODS). This RNA was microinjected into N2 hermaphrodites, and resulting progeny were examined for phenotypic defects relative to uninjected N2 wild type. Progeny of the injected animals showed a primarily Dpy phenotype, similar to that of dpy-18(e364) or dpy-18(e499) homozygotes (data not shown), suggesting that the dpy-18(null) phenotype is Dpy.

The mutations found in dpy-18(e364) and dpy-18(e499) suggest that each should be null for the encoded -subunit of prolyl-4-hydroxylase. To determine whether the alleles dpy-18(e364) and dpy-18(e499) are null for DPY-18 function, each was placed in trans to the noncomplementing deficiency tDf7 (Fig 1). The lengths of the resulting hemizygous animals were measured to assess whether the mutant dpy-18 gene products possess any residual catalytic activity. Worms with the genotype spe-16(hc54ts) dpy-18(e364)/tDf7 were significantly shorter than those homozygous for spe-16(hc54ts) dpy-18(e364), suggesting that the dpy-18(e364) mutation does not entirely abolish DPY-18 function (Table 2). While dpy-18(e499) homozygotes were shorter than wild type or dpy-18(e364), dpy-18(e499)/tDf7 hemizygous animals were not significantly more Dpy than dpy-18(e499) homozygotes (Table 2), suggesting that the compound deletion dpy-18(e499) is null for DPY-18 activity.

To confirm that dpy-18(e499) is null, homozygous dpy-18(e499) hermaphrodites were injected with the pool of pKH19-derived dsRNA that produced a Dpy phenotype when injected into N2 (see above). Progeny of dpy-18(e499) homozygotes are mostly Dpy, but some show more extreme phenotypes: Lumpy Dumpy, severe Lumpy Dpy, or embryonic lethal. When the broods of dpy-18(e499) animals injected with dpy-18 dsRNA were compared to the broods of uninjected sibs, the same four phenotypic classes were observed, demonstrating no qualitative difference in phenotype. To determine whether a quantitative difference in phenotype existed, the percentage of offspring falling into each phenotypic category from injected and uninjected animals was compared using Student's t-test. This analysis found no statistical difference in the distribution of phenotypes between the two populations (P = 0.06–0.6), except that there were extra dead eggs. However, the physical trauma associated with injection of foreign material into the C. elegans germline can cause some embryonic lethality, so the presence of dead eggs cannot be attributed solely to the dpy-18(RNAi). Therefore, dpy-18 RNAi in a dpy-18(e499) background did not noticeably change the phenotype of resulting progeny, and we conclude that dpy-18(e499) is null.

dpy-18 expression pattern:
The dpy-18 expression construct pBHdpy-18::gfp2 was made by ligating 2008 bp of wild-type upstream sequence and the proposed start AUG in frame to GFP (Fig 2C). N2 animals transformed to GFP(+) after injection of this construct together with the dominant rol-6 plasmid pRF4 (FIRE 1986 ; MELLO et al. 1991 ) were examined by DIC and fluorescence microscopy to determine where the dpy-18 promoter is active. GFP expression was first observed at the threefold embryonic stage (not shown) and was continuously expressed thereafter throughout embryonic and postembryonic development and into adulthood. The GFP signal was observed in the hypodermis (Fig 3), two pairs of bilaterally symmetrical head neurons that extend processes to the tip of the pharynx (small arrows in Fig 3A and Fig B), and one posteriorly located neuron that sends a process along the ventral nerve cord to the head (arrowheads, Fig 3B).

View larger version (52K): In this window In a new window Download PPT slide   Figure 3. (A and B) Composite images of pBHdpy-18::gfp2 expression in the hypodermis of adult hermaphrodites. Helical twisting of the hypodermis is due to the presence of the dominant transgenic marker rol-6 (su1006). Large arrows indicate the labeled hypodermal cells of the vulva. Small arrows indicate head neurons. Arrowheads in B indicate a posterior neuron and its process extending anteriorly along the ventral nerve cord; large arrowhead, cell body; small arrowheads, process.

dpy-18 is SL1 trans spliced:
Comparison of the 5' sequence of the dpy-18 cDNA published by VEIJOLA et al. 1994 with the dpy-18 genomic sequence revealed a splicing event immediately upstream of the putative start AUG, indicating that the dpy-18 mRNA contains at least one exon 5' to the one containing the proposed start AUG. To investigate the splicing of dpy-18 mRNA, we performed RT-PCR using SL1 (Table 1, primer 11) and SL2 (Table 1, primer 12) spliced leader forward primers, and a reverse primer within exon 4 (Table 1, primer 7). RT-PCR using the SL1 forward primer generated abundant product, but no product was obtained using the SL2 forward primer, demonstrating that dpy-18 is SL1 trans spliced (data not shown). Sequencing of the SL1 RT-PCR product revealed that the SL1 spliced leader sequence is spliced to the dpy-18 transcript 1 bp 5' to the proposed start AUG codon (data not shown). This transcript does not contain the upstream sequence present in the published cDNA (VEIJOLA et al. 1994 ), suggesting that alternative splicing may occur.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Two lines of evidence indicate that dpy-18 encodes the chromosome III-encoded -subunit of prolyl-4-hydroxylase in C. elegans. First, a 7140-bp genomic construct containing the -subunit coding region and 5' and 3' flanking sequences restores a wild-type body length to two dpy-18 mutants (Fig 2, Table 2, and MATERIALS AND METHODS). Consequently, this sequence encodes dpy-18 and contains sufficient promoter sequence to allow dpy-18 expression. Second, we sequenced the chromosome III prolyl-4-hydroxylase -subunit coding region of dpy-18(e499) and dpy-18(e364) mutants and found that each has a mutation in this gene (Fig 2).

The dpy-18 gene has eight exons, and several of its introns are large by C. elegans standards (Fig 2). In particular, the second intron is >1 kb in length, suggesting that regulatory sequences might be present within this intron (e.g., OKKEMA et al. 1993 ). While pBHdpy-18::gfp2, which contains only 2008 bp of upstream sequence and the putative start AUG codon, showed robust expression (Fig 3), other dpy-18::gfp fusions that included intron 2 (and other dpy-18 sequences) were expressed inconsistently and at low levels (data not shown). This suggests that dpy-18 intron 2 might contain elements that negatively regulate expression from the dpy-18 promoter. Despite inconsistent expression, several intron 2-containing dpy-18::gfp fusions caused transgenic animals to occasionally produce dead embryos and Dpy progeny. This suggests that these dpy-18::gfp fusion constructs may have caused an RNAi effect, as has been seen for other genes (e.g., FIRE et al. 1990 ).

Previous work showed that dpy-18(e364) is an amber nonsense mutation, because it is suppressible by amber-suppressing tRNAs (WATERSTON and BRENNER 1978 ; WATERSTON 1981 ). While amber mutations are frequently null, dpy-18(e364)/tDf7 are more Dpy than dpy-18(e364) homozygotes, showing that this allele is hypomorphic (Table 2). The hypomorphic nature of dpy-18(e364) indicates that some catalytically active protein is synthesized; this is likely due to occasional read-through of the stop codon, which has also been seen, for instance, in the nonnull spe-9(eb23) ochre nonsense mutant (SINGSON et al. 1998 ).

The dpy-18(e499) mutant contains two deletions that remove upstream, exon 1 and exon 2 sequence, including the putative start AUG codon and the putative ER signal sequence (Fig 2B; see VEIJOLA et al. 1994 for discussion of ER signal sequence). The absence of the start AUG should abolish translation or cause an N-terminal deletion in the resulting protein. Unlike dpy-18(e364), the mean body length of dpy-18(e499)/dpy-18(e499) is the same as that of dpy-18(e499)/tDf7. This result implies that dpy-18(e499) is null, and is confirmed by dpy-18(RNAi) results.

Double-stranded RNAi is well-established in C. elegans and other organisms as a way to block the function of the gene that encodes that RNA and mimic its null phenotype (FIRE et al. 1998 ; reviewed by MONTGOMERY and FIRE 1998 ; FIRE 1999 ; SHARP 1999 ). We found that dpy-18(RNAi) produced a Dpy phenotype in N2 animals, and had no obvious effect on dpy-18(e499) animals, demonstrating that the null phenotype for dpy-18 is Dpy, and dpy-18(e499) is null. However, dpy-18(e499) homozygotes sometimes arrest as embryos or have a severely disrupted body morphology. RNAi experiments performed using dsRNA with homology to both dpy-18 and the second prolyl-4-hydroxylase -subunit gene found on chromosome IV (F35G2.4) produced extensive embryonic lethality (not shown; see MATERIALS AND METHODS), demonstrating the essential nature of prolyl-4-hydroxylase function. These RNAi-arrested embryos show phenotypes that are very similar to the occasional dpy-18(e499) dead embryo. This suggests that prolyl-4-hydroxylase activity is essential, and that these genes are partially redundant with respect to function during embryogenesis. Activity of F35G2.4 is usually sufficient to allow survival of dpy-18 embryos, but both F35G2.4 and dpy-18 must function if the embryonic lethality associated with insufficient prolyl hydroxylase -subunit is to be reliably avoided.

Expression of a dpy-18::gfp fusion construct in the hypodermis is consistent with cuticle collagen expression in that tissue (KRAMER 1994A , KRAMER 1994B ). However, type IV collagens are produced primarily in muscle (GRAHAM et al. 1997 ), and no dpy-18::gfp expression was seen in muscle at any time during the life cycle. One possible explanation is that the pBHdpy-18::gfp2 fusion construct lacks regulatory elements required for expression in muscle cells, or expression in muscle might be below the level required for detection of GFP fluorescence. Alternatively, the F35G2.4 prolyl-4-hydroxylase -subunit (discussed above) might be the major form employed within muscle cells. Although the dpy-18 promoter fusion is expressed in head and posterior neurons, no neural defects are obvious in dpy-18 mutants, so the significance of this expression pattern is unclear. Perhaps dpy-18 mutants have subtle behavioral defects that would be revealed by specific assays (e.g., BARGMANN and MORI 1997 ).

There are at least two different transcripts expressed by the dpy-18 gene and both contain sequence 5' to the proposed start AUG. The previously published cDNA sequence contains a 79-bp noncoding exon that is spliced one nucleotide 5' to the proposed initiating AUG codon (VEIJOLA et al. 1994 ). Our experiments indicate that SL1 trans splicing to this same splice acceptor can also occur. Therefore, while both cDNA sequences suggest that the same AUG is utilized to initiate translation, they do not agree with regard to the 5' end of the dpy-18 transcript. The potential roles of these sequences 5' to the proposed start AUG and whether they are a consequence of tissue- or stage-specific regulation remain to be established.

Collagen mutations have been studied extensively in C. elegans, and they are associated frequently with either altered body morphology or lethality (reviewed by KRAMER 1994A , KRAMER 1994B ). Mutation of cuticle collagen genes leads to a Dumpy body in the case of col-2 or dpy-13, a Long body in the case of sqt-1(sc101), helical twisting of the long body axis in sqt-1 or rol-6, and occasional larval lethality in the case of restrictively raised sqt-3(e2117ts) (KRAMER et al. 1982 , KRAMER et al. 1988 , KRAMER et al. 1990 ; VON MENDE et al. 1988 ; VAN DER KEYL et al. 1994 ). Mutations in the emb-9- or let-2-encoded type IV collagens, which are the principal components of basement membranes, cause failure of elongation, hypodermal rupture, and embryonic lethality (GUO et al. 1991 ). Prior to the work described in this article, none of the enzymes involved in the post-translational modification of collagen had been mutationally defined in C. elegans.

Human syndromes caused by disruption of collagen biosynthesis have also been extensively studied, and they cause phenotypes ranging from mild skin abnormalities and joint hyperflexibility to pre- or perinatal lethality (reviewed by PROCKOP 1992 ; SAUBERLICH 1994 ). These disorders are most frequently caused by mutations in genes that encode collagen polypeptides, leading to abnormal, reduced, or absent collagen helix formation. An exception is Ehlers-Danlos syndrome type VI, which is usually due to impaired lysyl hydroxylase activity (reviewed by BURROWS 1999 ). Prolyl-4-hydroxylase has not been implicated in any inherited human disorder, but it is known that proper function of this enzyme is required in humans. For instance, insufficient dietary intake of the essential cofactor ascorbate (vitamin C) reduces prolyl-4-hydroxylase activity and causes the disease scurvy. Ascorbate is required for formation of hydroxyproline and, in its absence, procollagen remains bound to prolyl-4-hydroxylase and is retained within the ER (WALMSLEY et al. 1999 ), leading to deterioration of the ECM; chronic cases of scurvy can be lethal (reviewed by SAUBERLICH 1994 ).

The C. elegans dpy-18 gene is the first mutationally defined prolyl-4-hydroxylase -subunit described in any animal. Because of its unique ability to be mutated without causing lethality, dpy-18 may provide a powerful, natural in vivo system for analysis of prolyl-4-hydroxylase function and regulation, which may lead to novel treatments for collagen disorders in humans. In addition, the lethality in C. elegans associated with absence of prolyl-4-hydroxylase function may provide a target for pharmaceutical control of nematodes through disruption of cuticle formation.

	ACKNOWLEDGMENTS

We thank Alan Coulson, Andy Fire, Yuji Kohara, Tim Schedl, Sam Ward, and Heinke and Ralf Schnabel for providing C. elegans strains, DNA clones, or PCR primers. We thank Tim Schedl for suggesting the polymorphism mapping procedure utilized in this article and Shawn Ahmed for useful discussions. We also thank two anonymous reviewers for numerous suggestions that resulted in a much-improved manuscript. The Caenorhabditis Genetic Center provided some nematode strains, and it is funded by the National Institutes of Health National Center for Research Resources. This work was supported by U.S. Public Health Service grant GM RO1 GM40697 and National Science Foundation grant IBN-9631102 to S.W.L.

Manuscript received September 21, 1999; Accepted for publication March 27, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BARGMANN, C. I., and I. MORI, 1997 Chemotaxis and thermotaxis, pp. 717–737 in C. elegans II, edited by D. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BARSTEAD, R. and R. H. WATERSTON, 1991  Vinculin is essential for muscle function in the nematode. J. Cell Biol. 114:715-724[Abstract/Free Full Text].

BRENNER, S., 1974  The genetics of Caenorhabditis elegans.. Genetics 77:71-94[Abstract/Free Full Text].

BRODSKY, B. and J. A. RAMSHAW, 1997  The collagen triple-helix structure. Matrix Biol. 15:545-554[Medline].

BURROWS, N. P., 1999  The molecular genetics of the Ehlers-Danlos syndrome. Exp. Dermatol. 24:99-106.

CHALFIE, M., Y. TU, G. EUSKIRCHEN, W. W. WARD, and D. C. PRASHER, 1994  Green fluorescent protein as a marker for gene expression. Science 263:802-805[Abstract/Free Full Text].

COX, G. N. and D. HIRSH, 1985  Stage-specific patterns of collagen gene expression during development of C. elegans.. Mol. Cell. Biol. 5:363-372[Abstract/Free Full Text].

COX, G. N., M. KUSCH, K. DENEVI, and R. S. EDGAR, 1981a  Temporal regulation of cuticle synthesis during development of Caenorhabditis elegans.. Dev. Biol. 84:277-285.

COX, G. N., S. STRAPANS, and R. S. EDGAR, 1981b  The cuticle of Caenorhabditis elegans II. Stage-specific changes in ultrastructure and protein composition during postembryonic development. Dev. Biol. 86:456-470[Medline].

COX, G. N., J. M. KRAMER, and D. HIRSH, 1984  Number and organization of collagen genes in C. elegans.. Mol. Cell. Biol. 4:2389-2395[Abstract/Free Full Text].

COX, G. N., S. CARR, J. M. KRAMER, and D. HIRSH, 1985  Genetic mapping of C. elegans collagen genes using DNA polymorphisms as phenotypic markers. Genetics 109:513-528[Abstract/Free Full Text].

EDGLEY, M., D. L. BAILLIE, D. L. RIDDLE and A. M. ROSE, 1995 Genetic balancers, pp. 148–184 in Caenorhabditis elegans: Modern Biological Analysis of an Organism, edited by H. F. EPSTEIN and D. C. SHAKES. Academic Press, San Diego.

FIRE, A., 1986  Integrative transformation of Caenorhabditis elegans.. EMBO J. 5:2673-2680[Medline].

FIRE, A., 1999  RNA-triggered gene silencing. Trends Genet. 15:358-363[Medline].

FIRE, A., S. W. HARRISON, and D. DIXON, 1990  A modular set of lacZ fusion vectors for studying gene expression in Caenorhabditis elegans.. Gene 93:189-198[Medline].

FIRE, A., S. XU, M. K. MONTGOMERY, S. A. KOSTAS, and S. E. DRIVER et al., 1998  Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans.. Nature 391:806-811[Medline].

GORDON, M. K. and B. R. OLSEN, 1990  The contribution of collagenous proteins to tissue-specific matrix assemblies. Curr. Opin. Cell Biol. 2:833-838[Medline].

GRAHAM, P. L., J. J. JOHNSON, S. WANG, M. H. SIBLEY, and M. C. GUPTA et al., 1997  Type IV collagen is detectable in most, but not all, basement membranes of Caenorhabditis elegans and assembles on tissues that do not express it. J. Cell Biol. 137:1171-1183[Abstract/Free Full Text].

GUO, X. and J. M. KRAMER, 1989  The two Caenorhabditis elegans basement membrane (type IV) collagen genes are located on separate chromosomes. J. Biol. Chem. 264:17574-17582[Abstract/Free Full Text].

GUO, X., J. J. JOHNSON, and J. M. KRAMER, 1991  Embryonic lethality caused by mutations in basement membrane collagen of C. elegans.. Nature 349:707-709[Medline].

HODGKIN, J., 1983  Male phenotypes and mating efficiency in C. elegans.. Genetics 103:43-64[Abstract/Free Full Text].

HODGKIN, J., 1993  Molecular cloning and duplication of the nematode sex-determining gene tra-1.. Genetics 133:543-560[Abstract].

HODGKIN, J., 1997 Appendix 1, Genetics, pp. 881–1047 in C. elegans II, edited by D. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

HODGKIN, J. and T. DONIACH, 1997  Natural variation and copulatory plug formation in Caenorhabditis elegans.. Genetics 146:149-164[Abstract].

HODGKIN, J., H. R. HORVITZ, and S. BRENNER, 1979  Nondisjunction mutants of the nematode C. elegans.. Genetics 91:67-94[Abstract/Free Full Text].

HORVITZ, H. R., S. BRENNER, J. HODGKIN, and R. K. HERMAN, 1979  A uniform genetic nomenclature for the nematode C. elegans.. Mol. Gen. Genet. 175:129-133[Medline].

JAKUBOWSKI, J. and K. KORNFELD, 1999  A local, high-density, single nucleotide polymorphism map used to clone Caenorhabditis cdf-1.. Genetics 153:743-752[Abstract/Free Full Text].

KIVIRIKKO, K. and T. PIHLAJANIEMI, 1998  Collagen hydroxylases and the protein disulfide isomerase subunit of prolyl 4-hydroxylases. Adv. Enzyme Mol. Biol. 72:325-398.

KIVIRIKKO, K., R. MYLLYLA and T. PIHLAJANIEMI, 1992 Hydroxylation of proline and lysine residues in collagens and other animal and plant proteins, pp. 1–51 in Post-Translational Modification of Proteins, edited by J. J. HARDING and M. J. C. CRABBE. CRC Press, Boca Raton, FL.

KOIVU, J., R. MYLLYLA, T. HELAAKOSKI, T. PIHLAJANIEMI, and K. TASANEN et al., 1987  A single polypeptide acts both as the ß subunit of prolyl 4-hydroxylase and as a protein disulfide-isomerase. J. Biol. Chem. 262:6447-6449[Abstract/Free Full Text].

KRAMER, J. M., 1994a  Structures and functions of collagens in Caenorhabditis elegans.. FASEB J. 8:329-336[Abstract].

KRAMER, J. M., 1994b  Genetic analysis of extracellular matrix in C. elegans.. Annu. Rev. Genet. 28:95-116[Medline].

KRAMER, J. M., 1997 Extracellular matrix, pp. 471–500 in C. elegans II, edited by D. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

KRAMER, J. M., G. N. COX, and D. HIRSH, 1982  Comparisons of the complete sequences of two collagen genes from Caenorhabditis elegans.. Cell 30:599-606[Medline].

KRAMER, J. M., G. N. COX, and D. HIRSH, 1985  Expression of the C. elegans collagen genes col-1 and col-2 is developmentally regulated. J. Biol. Chem. 260:1945-1951[Abstract/Free Full Text].

KRAMER, J. M., J. J. JOHNSON, R. S. EDGAR, C. BASCH, and S. ROBERTS, 1988  The sqt-1 gene of C. elegans encodes a collagen critical for organismal morphogenesis. Cell 55:555-565[Medline].

KRAMER, J. M., R. P. FRENCH, E.-C. PARK, and J. J. JOHNSON, 1990  The Caenorhabditis elegans rol-6 gene, which interacts with the sqt-1 collagen gene to determine organismal morphology, encodes a collagen. Mol. Cell. Biol. 10:2081-2089[Abstract/Free Full Text].

MAZZORANA, M., H. GRUFFAT, A. SERGEANT, and M. VAN DER REST, 1993  Mechanisms of collagen trimer formation. Construction and expression of a recombinant minigene in HeLa cells reveals a direct effect of prolyl hydroxylation on chain assembly of type XII collagen. J. Biol. Chem. 268:3029-3032[Abstract/Free Full Text].

MELLO, C. C., J. M. KRAMER, D. STINCHCOMB, and V. AMBROS, 1991  Efficient gene transfer in C. elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J. 10:3959-3970[Medline].

MONTGOMERY, M. K. and A. FIRE, 1998  Double-stranded RNA as a mediator in sequence-specific genetic silencing and co-suppression. Trends Genet. 14:255-258[Medline].

OKKEMA, P. G., S. W. HARRISON, V. PLUNGER, A. ARYANA, and A. FIRE, 1993  Sequence requirements for myosin gene expression and regulation in Caenorhabditis elegans.. Genetics 135:385-404[Abstract].

PIHLAJANIEMI, T., T. HELAAKOSKI, K. TASANEN, R. MYLLYLA, and M. L. HUHTALA et al., 1987  Molecular cloning of the ß subunit of human prolyl 4-hydroxylase. This subunit and protein disulfide isomerase are products of the same gene. EMBO J. 6:643-649[Medline].

POLITZ, S. and R. S. EDGAR, 1984  Overlapping stage-specific sets of numerous small collagenous polypeptides are translated in vitro from Caenorhabditis elegans RNA. Cell 37:853-860[Medline].

PROCKOP, D. J., 1992  Mutations in collagen genes as a cause of connective-tissue diseases. N. Engl. J. Med. 326:540-546[Medline].

RUVKUN, G., V. AMBROS, A. COULSON, R. WATERSTON, and J. SULSTON et al., 1989  Molecular genetics of the Caenorhabditis elegans heterochronic gene lin-14.. Genetics 121:501-516[Abstract/Free Full Text].

SAUBERLICH, H. E., 1994  Pharmacology of vitamin C. Annu. Rev. Nutr. 14:371-391[Medline].

SHAKES, D. C., 1988 A genetic and pharmacological analysis of spermatogenesis in the nematode Caenorhabditis elegans. Ph.D. Thesis, Johns Hopkins University, Baltimore.

SHARP, P. A., 1999  RNAi and double-strand RNA. Genes Dev. 13:139-141[Free Full Text].

SHAW, L. M. and B. R. OLSEN, 1991  FACIT collagens: diverse molecular bridges in extracellular matrices. Trends Biochem. Sci. 16:191-194[Medline].

SINGSON, A., K. B. MERCER, and S. W. L'HERNAULT, 1998  The C. elegans spe-9 gene encodes a sperm transmembrane protein that contains EGF-like repeats and is required for fertilization. Cell 93:71-79[Medline].

VAN DER KEYL, H., H. KIM, R. ESPEY, C. V. OKE, and M. K. EDWARDS, 1994  Caenorhabditis elegans sqt-3 mutants have mutations in the col-1 collagen gene. Dev. Dyn. 201:86-94[Medline].

VAN DER REST, M. and R. GARRONE, 1991  Collagen family of proteins. FASEB J. 5:2814-2823[Abstract].

VEIJOLA, J., P. KOIVUNEN, P. ANNUNEN, T. PIHLAJANIEMI, and K. I. KIVIRIKKO, 1994  Cloning, baculovirus expression and characterization of the subunit of prolyl 4-hydroxylase from the nematode Caenorhabditis elegans.. J. Biol. Chem. 269:26746-26753[Abstract/Free Full Text].

VEIJOLA, J., P. ANNUNEN, P. KOIVUNEN, A. P. PAGE, and T. PIHLAJANIEMI et al., 1996  Baculovirus expression of two protein disulfide isomerase isoforms from Caenorhabditis elegans and characterization of prolyl 4-hydroxylases containing one of these polypeptides as their ß subunit. Biochem. J. 317:721-729.

VON MENDE, N., D. M. BIRD, P. S. ALBERT, and D. L. RIDDLE, 1988  dpy-13: a nematode collagen gene that affects body shape. Cell 55:567-576[Medline].

WALMSLEY, A. R., M. R. BATTEN, U. LAD, and N. J. BULLEID, 1999  Intracellular retention of procollagen within the endoplasmic reticulum is mediated by prolyl 4-hydroxylase. J. Biol. Chem. 274:14884-14892[Abstract/Free Full Text].

WARD, S., Y. ARGON, and G. A. NELSON, 1981  Sperm morphogenesis in wild-type and fertilization-defective mutants of C. elegans.. J. Cell Biol. 91:26-44[Abstract/Free Full Text].

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

WATERSTON, R. H. and S. BRENNER, 1978  A suppressor mutation in the nematode acting on specific alleles of many genes. Nature 275:715-719[Medline].

This article has been cited by other articles:

				K. Keskiaho, L. Kukkola, A. P. Page, A. D. Winter, J. Vuoristo, R. Sormunen, R. Nissi, P. Riihimaa, and J. Myllyharju Characterization of a Novel Caenorhabditis elegans Prolyl 4-Hydroxylase with a Unique Substrate Specificity and Restricted Expression in the Pharynx and Excretory Duct J. Biol. Chem., April 18, 2008; 283(16): 10679 - 10689. [Abstract] [Full Text] [PDF]

				K. B. Mercer, R. K. Miller, T. L. Tinley, S. Sheth, H. Qadota, and G. M. Benian Caenorhabditis elegans UNC-96 Is a New Component of M-Lines That Interacts with UNC-98 and Paramyosin and Is Required in Adult Muscle for Assembly and/or Maintenance of Thick Filaments Mol. Biol. Cell, September 1, 2006; 17(9): 3832 - 3847. [Abstract] [Full Text] [PDF]