Research Article

CeRh1 (rhr-1) is a dominant Rhesus gene essential for embryonic development and hypodermal function in Caenorhabditis elegans

Qiongmei Ji, Sarwar Hashmi, Zhi Liu, Jun Zhang, Ying Chen, and Cheng-Han Huang
PNAS April 11, 2006 103 (15) 5881-5886; https://doi.org/10.1073/pnas.0600901103

  1. Communicated by Sydney G. Kustu, University of California, Berkeley, CA, February 2, 2006

  2. ↵ †Q.J., S.H., and Z.L. contributed equally to this work. (received for review December 8, 2005)

Abstract

Rhesus (Rh) proteins share a conserved 12-transmembrane topology and specify a family of putative CO2 channels found in diverse species from microbes to human, but their functional essentiality and physiological importance in metazoans is unknown. To address this key issue and analyze Rh-engaged physiologic processes, we sought to explore model organisms with fewer Rh genes yet are tractable to genetic manipulations. In this article, we describe the identification in nematodes of two Rh homologues that are highly conserved and similar to human Rh glycoproteins, and we focus on their characterization in Caenorhabditis elegans. RNA analysis revealed that CeRh1 is abundantly expressed in all developmental stages, with highest levels in adults, whereas CeRh2 shows a differential and much lower expression pattern. In transient expression in human cells, both CeRh1 and CeRh2-GFP fusion proteins were routed to the plasma membrane. Transgenic analysis with GFP or LacZ-fusion reporters showed that CeRh1 is mainly expressed in hypodermal tissue, although it is also in other cell types. Mutagenesis analysis using deletion constructs mapped a minimal promoter region driving CeRh1 gene expression. Although CeRh2 was dispensable, RNA interference with CeRh1 caused a lethal phenotype mainly affecting late stages of C. elegans embryonic development, which could be rescued by the CbRh1 homologue from the worm Caenorhabditis briggsae. Taken together, our data provide direct evidence for the essentiality of the CeRh1 gene in C. elegans, establishing a useful animal model for investigating CO2 channel function by cross-species complementation.

Rhesus (Rh) defines a family of Rh blood group-related membrane proteins found in red cells and other tissues (1–3). It is now clear that this family is of ancient origin and has members not only spread in unicellular organisms from photosynthetic green algae (4) to social amoeba slime molds (5) but also is ubiquitous in metazoans. Although variable from one to six genes depending on species, Rh proteins are unified by their high primary sequence identities and their characteristic 12-transmembrane (TM)-spanning organization (1, 5). Such cross-phyla structural features pinpoint Rh family homologues as sharing a conserved transport function and constituting a unique division within the extended major facilitator superfamily (6).

The function of Rh proteins has long been puzzling, but increasing evidence indicates that they specify a family of gas channels for CO2, a discovery made by studying Rh1 in Chlamydomonas reinhardtii, a green alga (4, 7). In this organism, Rh1 is specifically induced by 3% CO2 (4), and its knockdown by RNA interference (RNAi) deters the cells from rapid growth at high CO2 levels (7). Consistent with this substrate specificity, Rh differs from related ammonium transporters (Amt) in evolutionary history and functional diversification (5), and the phenotypes of an AMT mutant strain and an RH RNAi strain differ in C. reinhardtii (8). Parallel to these studies, evidence that Amt is a gas channel for NH3 has also accumulated (9–13) and been substantiated by the high-resolution structures of two prokaryotic Amt proteins (14–16). Such atomic structures reveal that Amt forms a compact trimer with each monomer enclosing a hydrophobic pore for conducting NH3. This channel mechanism is contrary to the early thoughts that Amt is an active transporter with capacity to concentrate NH4 + (17, 18) and Rh-like Amt is an NH4 + cation transporter (19, 20). Furthermore, Rh gene disruption does not alter ammonium homeostasis in slime mold (21) or renal ammonium excretion in mice (22), challenging the view that Rh proteins transport ammonium (23–33). Thus, Rh and Amt specify two sequence-related yet substrate-distinct members of the biological gas channel family mediating diffusion of CO2 and NH3, respectively.

Expression of Rh genes and localization of Rh proteins has been studied in mice and humans, each harboring four paralogues (1). Rh30 and RhAG mainly reside in the red cell (34–36), whereas RhBG and RhCG are broadly expressed and show polarized sorting in various epithelial tissues (37–43). However, key questions remain whether these and other homologues are essential for any metazoan or whether they are involved in vital organismal processes. To address the Rh functional essentiality and explore its entailed physiological processes, we sought model organisms with fewer genes yet are tractable to genetic manipulation. In this article, we report the identification of two Rh genes in nematodes and focus on their characterization in Caenorhabditis elegans. We demonstrate that the CeRh1 gene is expressed in multiple tissues, particularly abundant in epithelial hypodermis, and plays essential roles in maturation of the embryo during C. elegans development. In addition, our studies establish an animal model for analyzing Rh functions via cross-species complementation and provide a useful model system in which to investigate human Rh glycoproteins.

Results

Worm Rh Proteins Are Conserved and Orthologous to Human Rh Glycoproteins.

To study Rh genes in nematode, we took advantage of the cloned C. elegans genes CeRh1 (rhr-1) and CeRh2 (rhr-2) (44) to screen worm genome databases and guide the cloning of their homologues in other species. By using the retrieved coding sequences, gene-specific primers were designed to derive full-length Rh proteins. As presented below, two Rh homologues were consistently identified as transcribed genes in both Caenorhabditis briggsae and Caenorhabditis remanei; we therefore designate the genes as rhr-1 and rhr-2 and the corresponding proteins as Rh1 and Rh2. Using cloned cDNA sequences as queries to perform genome database analysis revealed that rhr-1 has 6 or 7 exons and rhr-2 has 11 exons in each of the three worms (see Fig. 6, which is published as supporting information on the PNAS web site).

Fig. 1 A shows the alignment of six Rh proteins deduced from rhr-1 and rhr-2 mRNAs of C. elegans, C. briggsae, and C. remanei. Clearly, all worm proteins have a similar predicted 12-TM-spanning topologic organization, with the highest identity in TM spanning 2, 3, 6, 7, 9, and 11. All worm proteins, like their mammalian counterparts, have conserved negative-charged D/E residues in TM spanning 2, 4, 5, and 7 but have no predicted N-glycosylation site on extracellular loop 1. Phylogenetic analysis showed that worm Rh proteins form two groups (Fig. 1 B). Due to its multiexon structure (see Fig. 6) and sequence divergence, Rh2 may have arisen as a divergent copy of Rh1, suggesting that Rh1 retains the original function and Rh2 a new adaptive function. Members of Rh1 are 94–96% identical, and those of Rh2 are 93–94% identical. Compared with humans, the Rh1 group shares 40% identity (62% similarity) with RhAG and 36–38% identity (60–63% similarity) with RhBG/RhCG; the Rh2 group shares 35% identity with RhAG and 35–37% identity with RhBG/RhCG (Fig. 1 C). Hence, worm Rh1 and Rh2 are both conserved and orthologous to human Rh proteins.

Fig. 1.
Fig. 1.

Primary sequences of nematode Rh proteins and their comparison with human homologues. (A) Alignment of worm Rh proteins deduced from their cDNA sequences. Yellow boxes highlight TM domains, and asterisks distinguish identical residues. Conserved negative charge D/E residues in TM domains are bolded red. Ce, C. elegans; Cb, C. briggsae; Cr, C. remanei. (B) A joint tree of worm and human Rh proteins. The values at nodes are bootstrap proportions. The outgroup is Nitrosomonas europaea Rh (NeRhp). (Scale bar, 0.1 substitutions per nucleotide.) (C) Pairwise percentage of amino acid sequence identities (red) and similarities (blue) among Rh proteins.

CeRh1 and CeRh2 Are Differentially Expressed in C. elegans at Various Stages of Development.

Given the high degree of conservation with respect to each of the two groups in worms (Fig. 1), we selected to study C. elegans CeRh1 and CeRh2 as our model systems. The phylogenetic separation of CeRh1 from CeRh2 and the sequence divergence between them suggest that they might play different physiological roles; we therefore characterized their expression in detail.

We first performed semiquantitative RT-PCR to resolve the temporal expression of CeRh1 vs. CeRh2 using total RNA isolated from worms at different stages of development. The results showed that CeRh1 and CeRh2 both were expressed in all stages from eggs (time point 0) to 40 h during worm development (Fig. 2 A). Nevertheless, they were differentially expressed, with CeRh1 being much more abundant than CeRh2 in most stages, except for nonhatching eggs in which the CeRh2 level was higher (Fig. 2 A). Furthermore, CeRh1 expression was rapidly up-regulated 2 h after fertilization, and this high-level expression was maintained throughout development until the 40-h time point. In contrast, the level of CeRh2 expression was constantly lower and variable as well, showing a slight increase at later stages. These data identify CeRh1 and CeRh2 as the high- and low-abundance gene, respectively, supporting the view that they play distinct roles in the worm C. elegans.

Fig. 2.
Fig. 2.

Analysis of CeRh1 vs. CeRh2 expression during development, dauer state, and mixed stages. (A) Time course of CeRh1 vs. CeRh2 expression from eggs to 40 h of development. Lanes are as follows: M, HaeIII X174 size marker; 1, CeRh1; 2, CeRh2; 3, ama-1. (B) CeRh1 vs. CeRh2 expression in dauer larvae shown by a representative gel. (C) Northern blots of CeRh1 vs. CeRh2 expression in mixed stage worms. Lane 1, CeRh1 (probe = 803 bp, nucleotides 643-1445); lane 2, CeRh2 (probe = 749 bp; nucleotides 685-1433). Total RNA loading, 20 μg per lane; exposure time, 16 h at −80°C.

We next assessed the expression patterns of CeRh1 vs. CeRh2 after 40 h and in other states of C. elegans, given its 3-day life cycle and branch route to dauer formation (45). Interestingly, CeRh1 was reduced and CeRh2 increased in dauer worms (Fig. 2 B), suggesting that their regulated expression plays a part in facultative diapause and that CeRh2 might be a dauer-phase preferential gene. To delineate their temporal expression further, we used Northern blots to assay RNA from mixed stage adult worms. The results showed that CeRh1 was highly expressed, but CeRh2 was barely detectable (Fig. 2 C). Extended exposure of the blot to 3 days still revealed little presence of CeRh2 (data not shown). These data together suggest that CeRh1 is a growth-phase preferential gene, and CeRh2 is a dauer-phase preferential gene.

CeRh1 and CeRh2 Are Routed to Plasma Membranes When Expressed in Human Epithelial Cells.

Because of a lack of nematode cell lines, we used human cells to address whether CeRh1 and CeRh2 could be destined to the plasma membrane, like their mammalian counterparts (2, 3). Transient expression of GFP::CeRh1 fusion protein in human embryonic kidney epithelial cells (HEK-293) followed by confocal microscopy showed that the GFP signal was mainly confined to cell periphery (Fig. 3 A Right). In control cells transfected with pEGFP vector, the signal was evenly dispersed in cytoplasm (Fig. 3 A Left). These results indicated the localization of GFP::CeRh1 fusion protein to the plasma membrane. GFP::CeRh2 fusion protein was also routed to the plasma membrane (Fig. 3 B). Thus, both CeRh1 and CeRh2 retain intrinsic signals to direct their plasma membrane location, pointing to a conserved transport mechanism from worms to human.

Fig. 3.
Fig. 3.

Transient expression of GFP-CeRh1 and GFP-CeRh2 in HEK-293 cells. Full-length CeRh1 or CeRh2 cDNA was cloned in pEGFP vector in 5′gfp::rhr-1-3′ or 5′gfp::rhr-2-3′ fusion. The plasmid constructs were transfected into HEK-293 cells, and the green fluorescence signal was observed by using confocal microscopy. (A) GFP-CeRh1 fusion protein expression. Images are shown for the cells transfected with GFP-CeRh1 construct (Right) and pEGFP vector (Left). (B) GFP-CeRh2 fusion protein expression. Images are shown for the cells transfected with pEGFP vector (Left) or the GFP-CeRh2 construct (Right).

CeRh1 Is Expressed in Multiple Sites: 5′ Promoter Dissection and Transgenic Localization Studies.

We focused on CeRh1 in subsequent studies because of its abundant expression. Notably, CeRh1 has a relatively long 5′ region (≈4.8 kb) containing no expressed genes (data not shown); we therefore made rhr-1::gfp and a series of rhr-1::lacZ reporters (Fig. 4 A) to test their transgenic expression at various stages. This analysis mapped the 5′ promoter to the 1.9-kb region upstream of the ATG initiation codon that directs efficient and faithful expression of rhr-1 in C. elegans, as indicated by the same and robust β-gal-staining patterns by using no. 444 and 459 reporters.

Fig. 4.
Fig. 4.

Dissection of the promoter region of the rhr-1 gene and localization of CeRh1 protein expression. (A) Diagram of rhr-1 organization (six exons and five introns) and structure of the four transgenic constructs. A minus number shows the nucleotide from the +1 position of initiation ATG. All constructs contain the same coding sequence from the N terminus to TM 6, fused with lacZ or gfp via multiple cloning sites (MCS) of the pPD90.23 or pPD95.77 vector. (B) Transgenic analysis and localization of CeRh1 expression by using the four constructs. Representative images are shown for each reporter in β-gal staining. For CeRh1::GFP (no. 416), hypodermis (a and b), intestine (c), whole body (d).

The extensive transgenic analysis with the four reporters again demonstrated that CeRh1 protein was abundantly and broadly expressed in many locations (Fig. 4 B). Of these sites, the single most prominent is hypodermis, an epithelial tissue in C. elegans, which is densely decorated by β-gal staining or GFP signals. The expression of rhr-1 was detected in all developmental stages. In postembryonic stages, β-gal staining was seen in major hypodermal cells of all larval stages and young adults until they developed to fully mature gravid hermaphrodites [7–10 days postlarval stage 4 (L4) adult hermaphrodite]. The rhr-1:lacZ expression in all stages was specifically detected in the large hypodermal syncytium covering most of the worm (hyp7) and in the hypodermal cells of the head (hyp4 and hyp6) and hypodermal cells in the tail (Fig. 4 B). In addition, CeRh1 was found in gonads and intestine as well as the nervous system (tail neurons and occasionally head neurons) and enriched in developing embryos when reporter no. 446 was used (Fig. 4 B). The latter finding suggests that the 1.0-kb 5′ region may contain cis-regulatory elements to direct embryonic stage-specific expression. These data support the view that CeRh1 is the dominant, growth-phase preferential gene and moreover that these expression sites mimic some of those of Rh expression in mice and human (2, 3).

CeRh1 Function Is Required for Proper Embryonic Development and CeRh2 Gene Dispensable.

We next asked whether the two genes participate in organismal processes, e.g., animal development. RNAi was used to examine the requirement of rhr-1 or rhr-2 function for worm development. RNA soaking showed that rhr-2 is not essential for embryonic development (data not shown), consistent with its dauer-phase preferential expression. Although the adaptive function of CeRh2 has yet to be explored, a thorough analysis of the deletion mutant of rhr-2 (ok403 strain RB651) revealed a normal embryonic and adult phenotype, directly demonstrating its dispensability for development.

However, the majority (80%) of the progeny of the WT young hermaphrodite worms fed with rhr-1-harboring bacteria died during embryonic progression (Fig. 5), whereas the F1 progeny of N2 animals fed in PBS without rhr-1 plasmid were unaffected. We believe the 80% death rate is an underestimate, given the abundant expression of rhr-1 and feeding variables. Indeed, the few animals (≈10%) that progressed to adulthood were normal. For detailed analysis of the embryonic lethal phenotype, the arrested embryos of the rhr-1 RNAi worms were examined under Nomarski optics. The F1 embryos (n = 100) progressed at least to the gastrulation stage (≈100 min after first cleavage) before arrest at or just before the comma stage embryo (Fig. 5) during embryogenesis. Approximately 10% of the arrested embryos also arrested at a later “pretzel” stage embryo. Arrested embryos exhibited very severe tissue extrusion that would have blocked morphogenesis. Thus, our data demonstrate that CeRh1 plays an essential role during late-stage embryogenesis.

Fig. 5.
Fig. 5.

Mutant phenotypes caused by rhr-1 RNA interference: requirement of rhr-1 gene for normal embryonic development in C. elegans. Shown for comparison are WT embryos at comma stage (A) and pretzel stage (B). Mutant embryo showed developmental arrest at comma stage embryo with tissue extrusion (C); few percentage of RNAi embryos, which escaped arrest at comma stage arrested at pretzel-stage embryo (D). All photographs were taken by using Nomarski optics. (Magnification, ×400.)

Embryonic Lethality of CeRh1 RNAi Mutants Can Be Rescued by C. briggsae CbRh1.

Having shown the essentiality of the rhr-1 gene for C. elegans, we tested whether its RNAi-induced embryonic lethality could be rescued by rhr-1 of C. briggsae. Transgenic C. elegans were made with the rescue construct having the CbRh1 gene fused to CeRh1 promoter and UTR along pRF4 injection followed by double-stranded RNA (dsRNA) soaking. The results showed that F1 progeny of CbRh1-transgenic worms soaked with the 749-bp dsRNA for CeRh1 only showed 10% embryonic lethality (Table 1). In contrast, L4 stage N2 young adults soaked with CeRh1 dsRNA resulted in 70% embryonic lethality in F1 progeny embryos, and a similar level in F1 progeny was seen after rhr-1 RNAi of L4 worms carrying the pRF4 marker only (Table 1). Thus, C. briggsae rhr-1 expression compensated for C. elegans rhr-1 knockdown and rescued the embryonic phenotypes, proving that CbRh1 is functionally equivalent to CeRh1. This rescue also showed that the 1,880-bp 5′ promoter and 509-bp 3′ UTR of CeRh1 retain all of the cis-regulatory elements necessary to drive the faithful expression of CbRh1 gene in C. elegans.

View this table:
Table 1.

Rescue of CeRhr-1 RNAi-induced embryonic lethality with CbRhr-1

Discussion

In this article, we addressed two key questions about Rh proteins, a family of putative CO2 channels (4, 7) of ancient origin, broad distribution, and strong purifying selection (5). Are Rh proteins functionally essential for any metazoan? And if so, what are the organismal or physiologic processes they participate in? Using the C. elegans model, we cloned two homologues in nematodes, characterized their spatial temporal expression, and obtained evidence showing that CeRh1 plays an essential role in the maturation of embryos. The remarkable structural conservation predicts that worm Rh proteins are functional orthologues of human proteins and the broadly, highly expressed rhr-1 serves an important function in C. elegans. The routing of worm Rh proteins to the plasma membrane is similar to that in humans. The prominent location of Rh1 in C. elegans hypodermal epithelia mimics that of a variety of epithelial linings including skin (3) and vital organs of humans or rodents (37–40). The other sites of Rh expression between worm and mammals, the nervous system vs. brain (2), gonad vs. testis (2), or intestine vs. gastrointestinal tracts (41, 42) are also comparable. Altogether, these salient features define C. elegans as a useful model in which rescue of the lethal phenotype by cross-species complementation opens a new avenue to studying Rh functions in vivo.

Despite the uncertainty of their essential nature before this study, Rh proteins are known to play critical roles in unicellular microbes and mammalian cells. Down-regulation of Rh1 results in C. reinhardtii growth defects at high CO2 levels (7), thus relating Rh function to CO2 as a nutrient substrate in this organism (4). In humans, rare natural mutations of Rh30 or RhAG cause Rhnull disease, a genetic disorder that shows spherocytosis and shortened red cell life (31, 32), pointing to the structural and functional significance of the two proteins. Although there is no natural mutation described in human RhBG or RhCG, a recent study showed that RhCG down-regulation or loss-of-function is correlated with oesophageal tumor progression (37). Additionally, mouse Rhbg and Rhcg genes both have been identified as “hotspots” susceptible to retroviral insertions causing brain tumors (46). The results of our study reinforce the fundamental importance of Rh proteins in organismal physiology.

The expansion of one gene to six genes during the evolution from microbes to fish, followed by contraction to four genes in mice and higher species, underlies the building and remodeling of Rh functional differentiation or modification (5). This redundancy may well explain the seemingly normal phenotype seen in single knockouts made in slime mold (21) and mice (22), which harbor two and four Rh genes, respectively (1, 5). Our studies here demonstrate in worms that the Rh1 gene is essential and the Rh2 gene is dispensable. It will be of interest to find out whether disrupting the other Rh gene in slime mold produces a phenotype. Clearly, Rh genes have undergone functional modification since the first duplication because CeRh1 loss-of-function can be rescued by exogenous CbRh1 and not by endogenous CeRh2. Given the different spatial temporal patterns, we postulate that CeRh1 is the major functional protein and CeRh2 plays an adaptive role in dauer C. elegans, coincident with the linkage of rhr-2 to daf-11 (guanylyl cyclase) (47) and their mapping to the same pathway (48).

Rh and Amt coexist in unicellular eukaryotes through invertebrates before their exclusive emergence in the vertebrate lineage and vascular plants, respectively (4, 5). Nematodes harbor multiple Amt genes besides the Rh genes characterized here. Similar to Rh, these genes form two groups, one close to microbial and the other to invertebrate Amt (5). Thus, during the evolutionary period of their co-occurrence, Rh and Amt had been subject to independent divergent changes, probably driven by their own functional adaptation or selection. Indeed, the knockdown or knockout of Rh vs. Amt found in same organisms produced different phenotypes, as seen in green alga (7, 8) or slime mold (21, 49). It should be noted that the embryonic phenotypes caused by rhr-1 RNAi cannot be compensated by endogenously expressed amt genes. Altogether, these data provide evidence that Rh is a CO2 channel (4, 7) and functionally is not equivalent to Amt in the animal kingdom (19, 20).

Our studies clarify the controversy over the role of CeRh1 by genome-wide RNAi analyses (50–52) and reveal its essential nature in normal embryogenesis and hypodermal function in C. elegans. In C. elegans, the transformation of embryo from a ball of cells into a vermiform shape begins by the movement of hypodermal cells. The earliest detected requirement for rhr-1 function is during embryonic development. RNAi affected embryos failed to progress beyond the cell proliferation stage and arrested before morphogenesis, suggesting the role of CeRh1 in cell-fate specification and organization of tissues or organs during development. The hypodermis plays an essential role in changing the shape of the embryo during morphogenesis (53), and its cells originate on the dorsal surface of the embryo and migrate ventrally to enclose the embryo at the beginning of morphogenesis (54). The earliest phenotype in rhr-1 RNAi worms was observed at the comma stage of embryogenesis, which showed an extrusion of internal tissue. These findings suggest that the differentiation of the hypodermis had occurred and that rhr-1 expression is required for the function and/or integrity of the hypodermal tissue.

To understand the channel function of Rh proteins, we analyzed the site of action of rhr-1 in C. elegans. Three lines of evidence indicate that CeRh1 acts primarily in the hypodermis to regulate epithelial function: (i) it occurs widely in this epithelial tissue, as shown by its expression pattern in transgenic worms; (ii) its level of expression is constantly high throughout the worm life cycle; and (iii) rhr-1 RNAi results in cellular defects that arrest hypodermal morphogenesis. The hypodermis is a complex polarized layer with well defined apical and basolateral domains, functioning in a plethora of physiologic processes, e.g., nutrient storage, cuticle secretion, waste disposal, and pH buffering. Consistently, our findings indicate that CeRh1 is a master player in maintaining this multicellular architecture, probably by functioning for CO2 transport and acid–base balance across the plasma membrane. This central theme provides an important foundation from which to explore molecular mechanisms by which rhr-1 functions in C. elegans and other members in metazoan organisms.

Materials and Methods

Nematode Strains and Culturing.

WT C. elegans N2 and C. briggsae AF16 and C. elegans rhr-2 mutant (ok403) (Caenorhabditis Genetics Center, University of Minnesota, Minneapolis) were cultured by using standard methods (55). Genomic DNA or total RNA was isolated from mixed stage worms by using commercial kits (Qiagen). Cosmids F8F3.3 (rhr-1) and B0240.1 (rhr-2) (47) were from the Wellcome Trust Sanger Institute (Cambridge, U.K.).

Cloning of Worm Rh Genes.

C. elegans Rh1 and Rh2 were derived as described in ref. 44. C. briggsae Rh1 and Rh2 genes were derived from RT-PCR by using gene-specific primers. By using CeRh1 or CeRh2 as a query, C. remanei Rh1 or Rh2 was assembled in silico (www.ncbi.nlm.nih.gov/blast/Blast.cgi) (see Fig. 6).

Sequence Alignment, Database Search, and Phylogeny Analysis.

The primary sequences of worm Rh proteins were aligned by using muscle (56) and used with human sequences to compute pairwise identity and to build the tree by using the neighbor-joining method (57).

Plasmids.

Plasmids used for transient expression, transgenic animal analysis, RNAi knockdown, and genetic rescue were sequenced and are collectively shown in Supporting Text, which is published as supporting information on the PNAS web site. Their features and construction procedures are also detailed there.

Developmental Stage-Specific Expression and Northern Blot Analysis.

To analyze stage-specific expression of CeRh1 vs. CeRh2, total RNA was prepared from worms at defined time points of development and converted to single-stranded cDNA. Semiquantitative RT-PCR was carried out by using gene-specific primers and ama-1 as an internal control. The ama-1 gene encodes the large subunit of RNA polymerase II and its expression is fairly constant during development (58). For blot analysis, total RNA of mixed stage worms was separated on 1.0% agarose gel, blotted to a nylon filter, and hybridized with specific probes (44). Primers and probes are listed in Table 2, which is published as supporting information on the PNAS web site.

Transient Expression of Worm Genes in Human Cells.

HEK-293 cells (American Type Culture Collection) were cultured as described (2), at 37°C in DMEM containing 10% FBS and antibiotics in a 5% CO2 incubator. After a 24-h culture period, ≈1 × 105 cells in a four-well chamber were transfected with 0.85 μg of gfp::rhr-1 or gfp::rhr-2 plasmid plus Lipofectamine2000 (Invitrogen) and incubated for 24 h. Control cells were transfected with pEGFP vector. GFP::CeRh1 or GFP::CeRh2 fluorescent cells were observed under a Zeiss LSM510 confocal microscope. Images were processed by using the browser software.

Microinjection and Transgenic Analyses.

Plasmid for rhr-1::gfp or rhr-1::lacZ (final, 50 μg/ml) was injected into the gonads of young adult N2 worms. Plasmid pRF4 (80 μg/ml), which has a dominant mutation in rol-6, was coinjected for selection. Transgenic animals were screened for a GFP signal under a fluorescent microscope or for LacZ activity by β-gal staining.

RNAi by Bacterial Feeding and dsRNA Soaking.

C. elegans rhr-1 was cloned into L4440 vector and transformed into Escherichia coli HT115 (59). Feeding worms with rhr-1 harboring bacteria was done as described in ref. 60. A soaking plasmid was also made by cloning a 749-bp sequence (nucleotides 643-1392) of rhr-1 in pCRII-TOPO for in vitro RNA transcription by using SP6 and T7 promoters. C. elegans rhr-2 gene was cloned into pCRScript for in vitro RNA transcription by using T3 and T7 promoters. dsRNA was formed by annealing two single-stranded RNAs, gel-purified, and used for worm soaking.

Rescue of CeRh1 Knockdown Mutants with CbRh1 Gene.

The rescue construct had CbRh1-coding sequence fused to the 1,880-bp promoter and 509-bp 3′ UTR of CeRh1. This plasmid (15–20 μg/ml) plus the pRF4 marker (80 μg/ml) was injected into the distal arm of N2 hermaphrodite gonads. pRF4 was injected alone as a control. CbRh1/pRF4 transformants were detected by the roller phenotype and single-worm PCR. These transgenic worms, pRF4 worms, and L4 hermaphrodites of N2 worm were washed in PBS/sucrose solution and soaked in the same buffer with 5 μl of dsRNA (4 mg/ml) in separate siliconized tubes. After 24 h, they were transferred to an agar plate with E. coli OP50 and cultured until midadulthood and F1 progeny were examined for embryonic development.

Acknowledgments

We thank Jianbin Peng for bioinformatics analysis, Tian Ye for expression of worm genes in human cells, Michal Tarnawski for confocal microscopy imaging, Alan Coulson for F08F3 and B0240 cosmids, Andy Fire for worm vectors, and Theresa Stiernagle and the C. elegans Knockout Consortium for the worm mutant ok403 (strain RB651). This work was supported by National Institutes of Health Grant HD62704.

Footnotes

  • To whom correspondence should be addressed. E-mail: chuang{at}nybloodcenter.org

  • ↵ §Present address: Department of Protein Science, Amgen Inc., 1201 Amgen Court West, Seattle, WA 98119.

  • Author contributions: C.-H.H. designed research; Q.J., S.H., Z.L., J.Z., Y.C., and C.-H.H. performed research; C.-H.H., S.H., and Q.J. analyzed data; and C.-H.H. wrote the paper.

  • Conflict of interest statement: No conflicts declared.

  • Data deposition: The sequences reported in this paper have been deposited in the GenBank database (accession nos. AF183390, AF183391, AY363116, AY363117, DQ013065, and DQ013066).

  • Abbreviations:

    Abbreviations:

    Rh,
    Rhesus;
    TM,
    transmembrane;
    RNAi,
    RNA interference;
    Amt,
    ammonium transporter;
    L4,
    larval stage 4;
    dsRNA,
    double-stranded RNA;
    HEK,
    human embryonic kidney.
View Abstract
Back to top
Article Alerts
Email Article
Citation Tools
Request Permissions
CeRh1 (rhr-1) is a dominant Rhesus gene essential for embryonic development and hypodermal function in Caenorhabditis elegans
Qiongmei Ji, Sarwar Hashmi, Zhi Liu, Jun Zhang, Ying Chen, Cheng-Han Huang
Proceedings of the National Academy of Sciences Apr 2006, 103 (15) 5881-5886; DOI: 10.1073/pnas.0600901103
Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo
Proceedings of the National Academy of Sciences: 103 (15)
Table of Contents

Submit

Jump to section

You May Also be Interested in

Large piece of gold
News Feature: Tracing gold's cosmic origins
Astronomers thought they’d finally figured out where gold and other heavy elements in the universe came from. In light of recent results, they’re not so sure.
Image credit: Science Source/Tom McHugh.

Similar Articles