Depletion of a Novel SET-Domain Protein Enhances the Sterility of mes-3 and mes-4 Mutants of Caenorhabditis elegans

Corresponding author: Susan Strome, Department of Biology, Indiana University, Bloomington, IN 47401., sstrome{at}bio.indiana.edu (E-mail)

Communicating editor: T. C. KAUFMAN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Four maternal-effect sterile genes, mes-2, mes-3, mes-4, and mes-6, are essential for germline development in Caenorhabditis elegans. Homozygous mes progeny from heterozygous mothers are themselves fertile but produce sterile progeny with underproliferated and degenerated germlines. All four mes genes encode chromatin-associated proteins, two of which resemble known regulators of gene expression. To identify additional components in the MES pathway, we used RNA-mediated interference (RNAi) to test candidate genes for enhancement of the Mes mutant phenotype. Enhancement in this assay was induction of sterility a generation earlier, in the otherwise fertile homozygous progeny of heterozygous mothers, which previous results had suggested represent a sensitized genetic background. We tested seven genes predicted to encode regulators of chromatin organization for RNAi-induced enhancement of mes-3 sterility and identified one enhancer, called set-2 after the SET domain encoded by the gene. Depletion of SET-2 also enhances the sterile phenotype of mes-4 but not of mes-2 or mes-6. set-2 encodes two alternatively spliced transcripts, set-2_l and set-2_s, both of which are enriched in the germline of adults. In the adult germline, SET-2_L protein is localized in mitotic and mid-late-stage meiotic nuclei but is undetectable in early pachytene nuclei. SET-2_L protein is localized in all nuclei of embryos. The localization of SET-2_L does not depend on any of the four MES proteins, and none of the MES proteins depend on SET-2 for their normal localization. Our results suggest that SET-2 participates along with the MES proteins in promoting normal germline development.

GERM cells display numerous unique traits, among them the ability to undergo meiosis and the ability to generate offspring. These unique traits require unique control mechanisms. Among the specialized regulatory molecules required in the germline of Caenorhabditis elegans are the four maternal-effect sterile (MES) proteins, named after their maternal-effect sterile mutant phenotype (CAPOWSKI et al. 1991 ). In the progeny of homozygous mes mothers, the germline undergoes only limited proliferation and then degenerates, resulting in sterile but otherwise healthy adults (PAULSEN et al. 1995 ; GARVIN et al. 1998 ). The predicted MES protein sequences reveal that MES-2 and MES-6 are the C. elegans orthologs of the Drosophila Polycomb group (PcG) proteins, Enhancer of Zeste [E(Z)], and Extra Sex Combs (ESC), respectively (HOLDEMAN et al. 1998 ; KORF et al. 1998 ). MES-3 is a novel protein with no recognizable motifs (PAULSEN et al. 1995 ). MES-4 is a SET-domain protein with multiple PHD fingers (Y. FONG, L. BENDER and S. STROME, unpublished results).

The similarity of MES-2 and MES-6 to PcG proteins suggests that the MES proteins are likely to function similarly to PcG proteins. PcG members associate into multiprotein complexes to repress transcription of target genes (GOULD 1997 ; PIRROTTA 1997 ; PREUSS 1999 ). Their best known targets in Drosophila are homeotic genes, which function to determine the anterior-posterior body pattern (PIRROTTA 1995 ; SIMON 1995 ; AKAM 1998 ). After the initial expression patterns of homeotic genes are established in the early embryo (CARROLL et al. 1986 ; CELNIKER et al. 1989 ), maintenance of the expression patterns is controlled by the antagonistic functions of the PcG and the trxG (trithorax group). PcG proteins maintain repression of genes outside of their expression domains by promoting assembly of chromatin into a repressive state, while trxG proteins maintain active gene expression within the proper domains by remodeling chromatin into a transcriptionally active state (OWEN-HUGHES et al. 1996 ; SHAO et al. 1999 ).

Although PcG and trxG proteins are generally thought to function antagonistically in transcriptional regulation, the distinction between them is not always clear. E(Z), for example, is a classic PcG member. Mutations in E(Z) cause anterior to posterior segment transformations that are characteristic of loss of PcG gene function (JONES and GELBART 1990 ). However, E(Z) has also been implicated as a trxG protein (LAJEUNESSE and SHEARN 1996 ). Animals heterozygous for both a mutation in E(z) and a mutation in ash-1 (a trxG member) present homeotic transformation phenotypes similar to those displayed by trithorax and ash1 double heterozygous mutants. Furthermore, in hemizygous mutants for amorphic mutant alleles of E(z), accumulation of many homeotic gene products is lost. These results suggest that E(Z), a PcG protein, can in certain contexts display trxG characteristics and activate gene expression.

On the basis of analysis of transgene expression, the MES proteins participate in repression of gene expression in the germline of C. elegans, as PcG proteins do in the Drosophila soma. Transgenes present in multiple copies in extrachromosomal arrays are efficiently expressed in somatic tissues but silenced in the germline of wild-type C. elegans (KELLY et al. 1997 ). Such high-copy transgenic arrays are desilenced in the germlines of sterile mes mutants (KELLY and FIRE 1998 ). Transgenes in arrays also can be desilenced in wild-type germlines by reducing their copy number and placing them in the context of complex DNA (KELLY et al. 1997 ). Taken together, these findings suggest that the MES proteins participate in repressing transgene expression in wild-type germlines and that this is via an effect on chromatin state.

To identify new components in the MES regulatory pathway, a powerful approach is to search for enhancers of the Mes phenotype in a sensitized genetic background. We reasoned that M+Z- mes mutants (M, maternal; Z, zygotic or nonmaternal) might represent such a sensitized background for this kind of search. First, M+Z- mes hermarphrodites (i.e., mes/mes hermaphrodites from mes/+ mothers) are fertile but have reduced brood sizes, indicating that their germlines are compromised (CAPOWSKI et al. 1991 ). In the case of mes-3, M+Z- hermaphrodites are occasionally sterile. Second, transgene desilencing occurs in the germlines of M+Z- mes-3 and mes-4 hermaphrodites (KELLY and FIRE 1998 ), suggesting that the germlines of M+Z- mes mutants display some characteristics of the sterile M-Z- mes germlines. We hoped that additional mutations that enhance the Mes phenotype might induce sterility in these M+Z- mes worms. We used RNA-mediated interference (RNAi) to test candidate genes for enhancement of mes-3. We found that RNA-mediated interference with one gene, called set-2, caused sterility in M+Z- mes-3 and mes-4 mutants, but not in mes-2 or mes-6 mutants or in wild-type worms. set-2 encodes two alternative transcripts, set-2_l and set-2_s, both of which are germline enriched. The larger protein product, SET-2_L, is distributed similarly to the MES proteins in the nuclei of germ cells and embryos.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Alleles and strain maintenance:
N2 var. Bristol was used as the wild-type C. elegans strain in this report. The following mutant strains were used for RNAi analyses: mes-3(bn35) dpy-5 (e61)/mes-3(bn35)dpy-5(e61)I/hDp20(I;V, f), mes-4(bn67) dpy-11(e224)V/nT1[let(m435)](IV,V), mes-2(bn11) unc-4(e120)/mnC1[dpy-10(e128) unc-52(e444)]II, and mes-6(bn38)IV/DnT1[unc(n754)let](IV,V). Strains were maintained following standard procedures (BRENNER 1974 ).

RNAi analyses:
The cDNAs for RNAi tests were obtained from the expressed sequence tag (EST) project (Y. Kohara, National Institute of Genetics, Mishima, Japan) as ZAPII phagemid clones (yk273c7, yk11d12, yk25g10, yk40b5, yk112b11, yk52e6, yk250a4, and yk427g1). The cDNA inserts are flanked by T7 and T3 promoters in the context of linearized pBlueScript vectors. After the pBlueScript plasmids were excised from the phagemids by helper phages (SAMBROOK et al. 1989 ), they were transformed into Escherichia coli and amplified. Sense and antisense RNAs of each gene were transcribed separately in vitro from the cDNA by T7 or T3 polymerases, using the MEGAscript kit (Ambion, Austin, TX) followed by DNase treatment. The RNAs were heated to 85° for 5 min and cooled to room temperature to anneal.

To generate the double-stranded RNA that targets set-2_l only, an 1-kb set-2_l-specific cDNA fragment was amplifed by RT-PCR using poly(A)⁺ RNA prepared from wild-type worms as the template as described previously (SAMBROOK et al. 1989 ; CONRAD et al. 1991 ). The downstream primer for reverse transcription and for PCR was 5'-AAAAGCGGCCGCTCGGTTTTTCAGCTTC-3'. The upstream primer for PCR was 5'-AAAGGATCCGCATCGGGAAGCTCTTC-3'. The PCR product was cloned into pBlueScript. The sequence of the PCR product was verified by sequencing. Double-stranded RNA from this PCR product was produced as described above.

Double-stranded RNA (500 ng/µl) of each above cDNA was injected into heterozygous mes mutants that were raised at 20°. Embryos laid 12 hr and later after injection were grown to adult stage at 20° and scored for their sterility under a dissecting microscope. Worms with an empty uterus, indicating the absence of embryos, were scored as sterile.

Antibody production and immunostaining:
Anti-SET-2_L antibodies were generated against the N-terminal peptide of SET-2_L, MSTHDMNHHPPRKSHSKRDK. The peptide was synthesized and conjugated to keyhole limpet hemocyanin carrier protein by Research Genetics (Huntsville, AL). The conjugated peptide was injected into rats by Cocalico. The free peptide was crosslinked to immunopure epoxy-activated agarose (Pierce, Rockford, IL) for affinity purification of anti-SET-2_L antibodies. Anti-SET-2_L antibodies were eluted from the affinity column by 5 M KI and dialyzed against PBS overnight at 4°. Immunostaining was performed as previously described (STROME and WOOD 1983 ).

Northern blot hybridization analyses:
poly(A)⁺ RNA was prepared as previously described from N2, glp-4(bn2), and fem-2 (b245ts) mutant hermaphrodites raised at 25° (SAMBROOK et al. 1989 ; CONRAD et al. 1991 ). It was electrophoresed on a 0.8% agarose 6% formaldehyde gel and blotted to Hybond-Nylon membrane. The blots were hybridized with a set-2 probe, which was amplified by PCR using yk25g10 as the template and radiolabeled with [-³²P]dCTP using a Boehringer Mannheim (Indianapolis) random-primed labeling kit. rpp-1, a ribosomal protein gene (EVANS et al. 1997 ), served as a loading control. RNA bands were quantified using NIH Image software (National Institutes of Health). The relative intensity of each band was expressed as the hybridization intensity of the set-2 band relative to the rpp-1 band in each lane, and the ratio in the mutants was compared to that in N2, which was arbitrarily set to 1.

RT-PCR and Southern blot analyses:
RT-PCR reactions were performed as previously reported (SPIETH et al. 1993 ).Poly(A)⁺ RNA prepared as described above was used as the RNA template. The downstream primers used for reverse transcription and for PCR were 5'-GAAGATGGGCTCGACATAAG-3' for set-2_l and 5'-CGAAGCACGTCTTCGTCGC-3' for set-2_s. The upstream primers for PCR were SL1 (5'-GGTTTAATTACCCAAGTTTGAG-3') and SL2 (5'-GGTTTTAACCCAGTTACTCAAG-3'). The PCR products were electrophoresed on a 0.8% agarose gel and probed for set-2_l or set-2_s by Southern blot, following standard procedures (SAMBROOK et al. 1989 ). The probes used were 5'-CGTACTTCACAACGTCCC-3' to detect set-2_l and 5'-GACGACGTGGAAACTGTACG-3' to detect set-2_s. The PCR products were cloned into pGEM-T vector and sequenced using T7 and SP6 promoter sequence as the primers.

Sequence analyses of set-2_l and set-2_s cDNAs:
set-2_l-specific cDNA was amplified as three overlapping fragments by RT-PCR as described above. Fragment 1 was as described in RT-PCR and Southern blot analyses. For fragment 2, the downstream primer for reverse transcription and for PCR was 5'-GCTTCCCGATGCAGATTCGG-3' and the upstream primer for PCR was 5'-CTTATGTCGAGCCCATCTTC-3'. For fragment 3, the downstream primer for reverse transcription and for PCR was 5'-AAAAGCGGCCGCTCGGTTTTTCAGCTTC-3' and the upstream primer for PCR was 5'-AAAGGATCCGCATCGGGAAGCTCTTC-3'.

The 231 bp of set-2_s-specific cDNA was sequenced using yk25g10 as the template and the T7 promoter sequence as the primer. yk25g10 is a cDNA clone from the EST project, which contains the set-2_s-specific sequence as well as the common region between set-2_l and set-2_s.

The common region between set-2_l and set-2_s was sequenced using yk25g10 and yk11d12 as the templates. yk11d12 is a cDNA clone from the EST project, which contains 249 bp of the set-2_l-specific region and overlaps with fragment 3 of set-2_l (see above) and the common region between set-2_l and set-2_s. Sequencing was performed using an ABI PRISM DNA sequencing kit and ABI PRISM 310 genetic analyzer (PE Applied Biosystems).

Bioinformatics:
The predicted amino acid sequences of TRX, BRM, ASH-1, FSH, E(PC), and NURF p55 in Drosophila were obtained by searching GenBank (http://www3.ncbi.nlm.nih.gov/Entrez/). They were used to search for potential homologs in C. elegans using the Wu-BLAST program (http://www.sanger.ac.uk/Projects/C_elegans/blast_server.shtml). Information on the cDNA clones available for each homolog was provided by GenBank.

The RNA recognition motif (RRM) in SET-2_L was predicted by the Pfam program (http://www.sanger.ac.uk/Software/Pfam/search.shtml). The secondary structure of this putative RRM in SET-2_L was predicted by the following programs: SSP (http://searchlauncher.bcm.tmc.edu/seq-search/struc-predict.html), SSpro (http://promoter.ics.uci.edu/BRNN-PRED/), Predator (http://www.embl-heidelberg.de/cgi/predator_serv.pl), and PSIpred (http://insulin.brunel.ac.uk/psiform.html).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Test for enhancers of mes-3:
To identify new components in the MES regulatory pathway, we tested candidate genes for whether depletion of gene product by RNAi enhances mes-3 sterility. RNAi, which involves injection of gene-specific double-stranded RNA (dsRNA) into hermaphrodites, is an extremely potent and specific method of reducing gene expression and has been demonstrated to phenocopy strong or null mutations in many genes (ROCHELEAU et al. 1997 ; FIRE et al. 1998 ). Consequently, RNAi has been used extensively to examine the functions of genes with known nucleotide sequences but unknown mutant phenotypes. The parent strain we used to test for RNAi-induced enhancement of the Mes phenotype contains the mes-3(bn35) mutation on both chromosomes balanced by a duplication, hDp20, which contains a wild-type copy of mes-3 (i.e., bn35/bn35/hDp20; Fig 1). This strain is similar to a heterozygous mes-3 strain and produces predominantly fertile F₁ progeny. The fertile bn35/bn35 F₁ hermaphrodites produce all sterile F₂ progeny. We reasoned that RNAi depletion of a protein that functions with MES-3 might induce sterility in the bn35/bn35 F₁'s (Fig 1). The bn35/+ siblings, however, might remain fertile as a result of zygotic expression of MES-3.

View larger version (14K): In this window In a new window Download PPT slide   Figure 1. Strategy used to search for enhancers of mes-3. The strain mes-3 dpy-5/mes-3 dpy-5/hDp20 was used to search for enhancers of mes-3. (Left) The genotypes and phenotypes of the progeny from mothers without RNA injection. (Right) The genotypes and phenotypes of the progeny from mothers injected with double-stranded RNA made from a gene whose depletion enhances the Mes-3 phenotype. RNAi depletion of such an enhancer will cause the homozygous mes F1 progeny to develop into sterile adults, while their heterozygous siblings remain fertile.

The genes that we selected to test for a genetic interaction with mes-3 encode potential chromatin regulators. They included a gene encoding a protein that shares a domain with the Drosophila PcG protein Enhancer of Polycomb [E(PC)] (STANKUNAS et al. 1998 ); a gene encoding a homolog of the chromatin remodeling factor, nucleosomal remodeling factor p55 (NURF p55; MARTINEZ-BALBAS et al. 1998 ); and five C. elegans homologs of trxG genes. As described above, although PcG and trxG proteins are generally thought to function antagonistically in Drosophila, there is evidence that certain proteins belong to both groups. In particular, the PcG member, E(Z), sometimes presents trxG characteristics. Therefore, the C. elegans trxG homologs were considered candidates to genetically interact with the MES proteins, two of which (MES-2 and MES-6) are PcG homologs. The C. elegans genome contains two genes with significant sequence similarity to trithorax (STASSEN et al. 1995 ). One of them, T12D8.1, encodes a predicted protein that is similar to TRX throughout its length; it contains three PHD fingers and a SET domain, as does TRX. The other gene, C26E6.9, encodes two predicted proteins that have sequence similarity with TRX only in the SET domain. We also found homologs of the trxG genes ash-1 (TRIPOULAS et al. 1996 ), brm (TAMKUN et al. 1992 ), and fsh (HAYNES et al. 1989 ).

Double-stranded RNA to each of the above genes was synthesized and injected into heterozygous mes-3 mutants, and the F₁ progeny were scored for sterility. As summarized in Table 1, three types of RNAi effects were observed:

1. No enhancement of sterility in the mes-3 F₁'s: We did not pursue these genes further.

   
   

   View this table: In this window In a new window   Table 1. Tests for enhancement of mes-3 sterility by RNAi depletion of selected genes

2. Embryonic lethality among embryos of the injected worms and also enhancement of sterility in the mes-3 F₁'s: The embryonic lethality phenotype indicated that RNAi depletion of these genes generated early somatic defects in the F₁ progeny, which led to the concern that the sterile phenotype may be a secondary consequence of somatic defects or of general sickliness. Therefore, we did not pursue these genes further.

3. Enhancement of the sterility of mes-3 F₁'s, no significant enhancement of the sterility of mes-3/+ F₁'s, and no embryonic lethality: Among the seven genes tested, one gene displayed this genetic enhancement of the mes-3 phenotype. This gene, C26E6.9, was named set-2 and is the focus of the remainder of this article.

set-2 is an enhancer of mes-3 and mes-4:
set-2 was named for the SET domain in the predicted protein products. According to the GeneFinder prediction program, the locus encodes two overlapping transcripts; this was confirmed by Northern hybridization (see below). We call the longer one [open reading frame (ORF) no. C26E6.9a] set-2_l and the shorter one (C26E6.9b) set-2_s. When both transcripts were depleted by RNAi, the sterility of F₁ mes-3 worms was increased from 15 to 92% (Fig 2). RNAi depletion of set-2 in wild-type worms, however, did not cause sterility or any other significant defects. The sterile germlines observed in mes-3; set-2(RNAi) worms were quite variable in appearance. Approximately 15% of sterile adults contained a severely underproliferated germline, resembling the phenotype of sterile mes-3 worms from homozygous mes-3 mothers. However, the remaining sterile worms contained fairly well-proliferated germlines and some contained gametes; none of them produced fertilized embryos. We do not know whether this variability in germline phenotype reflects variability in RNAi depletion of SET-2 or variable effects of full depletion of SET-2. This issue, as well as whether set-2 RNAi enhancement of sterility is due to depletion of maternal or zygotic SET-2 or both, will be addressed when a null mutant of set-2 is isolated.

View larger version (34K): In this window In a new window Download PPT slide   Figure 2. RNAi analysis of set-2. Double-stranded RNA was made from a region specific to set-2l or common to set-2s and set-2l and was injected into N2, mes-2/+, mes-3/+, mes-4/+, or mes-6/+ young adult hermaphrodites. The percentage of sterile F1 progeny was calculated for the various genotypes (A) and graphed (B).

To test whether set-2_l and set-2_s both function to enhance mes-3, we injected mes-3 heterozygous worms with dsRNA that targets both isoforms or with dsRNA that targets only set-2_l. RNAi depletion of set-2_l alone caused 53% sterility in mes-3 F₁'s, whereas RNAi depletion of both isoforms of set-2 caused 92% sterility in mes-3 F₁'s (Fig 2). Therefore, the enhancement was more severe when both transcripts were depleted by RNAi than when only the larger one was depleted. This indicates that depletion of either set-2_l or set-2_s partially enhances the mes-3 mutant phenotype.

We also tested whether RNAi depletion of set-2 enhances the sterility caused by mutations in the other three mes genes. As described above, we injected set-2 dsRNA into heterozygous mes mothers and analyzed their mes and mes/+ F₁ progeny. RNAi depletion of set-2 enhanced the sterility of mes-4 F₁'s to a similar degree as seen in mes-3 but did not enhance the sterility of mes-2 or mes-6 F₁'s (Fig 2).

set-2_l is trans-spliced to SL1:
Northern hybridization analysis demonstrated that the set-2 locus encodes two transcripts, estimated to be 6.5 and 3.4 kb (Fig 3A). According to GeneFinder, the first four exons of set-2_l encode a GTP-binding domain and the last three exons encode a SET domain. To verify this prediction, the 5' end of set-2_l was determined by RT-PCR followed by sequence analysis (see MATERIALS AND METHODS). We found that at least some set-2_l transcripts are trans-spliced to SL1 but not to SL2 (Fig 4A). set-2_s is not trans-spliced to either SL1 or SL2. Sequencing of the set-2_l RT-PCR product showed that, different from the GeneFinder prediction, the 5' end of set-2_l starts after the predicted GTP-binding domain-containing exon. Consistent with our determined 5' end, we found that RNAi using RNA directed against the GTP-binding domain did not enhance mes-3 sterility (data not shown). Consequently, we conclude that the SET-2_L protein does not include the GTP-binding domain; this domain might be encoded by the gene upstream of set-2_l. The set-2_l cDNA was subsequently sequenced (see MATERIALS AND METHODS), and its gene structure is shown in Fig 4B. Since set-2_s appears not to be trans-spliced to either SL1 or SL2 (data not shown), we did not verify its 5' end. However, our sequencing of a cDNA clone of set-2_s (yk25g10) suggests that it encodes the full-length SET-2_S protein; it includes a region of predicted 5' UTR, the entire ORF predicted by GeneFinder, and a 3' UTR.

View larger version (32K): In this window In a new window Download PPT slide   Figure 3. Northern blot hybridization analysis of set-2. (A) poly(A)+ RNA was prepared from N2, glp-4(bn2ts), and fem-2(b245ts) adult hermaphrodites grown at 25° and probed with a region common to set-2s and set-2l on Northern blots. After growth at 25°, glp-4 adults have a severely underproliferated germline, and fem-2 adults produce oocytes but no sperm. The probe recognizes two bands estimated to be 6.5 and 3.4 kb. The predicted sizes of set-2l and set-2s mRNAs without poly(A) tails are 5.2 and 2.8 kb, respectively. The same blot was probed with the C. elegans ribosomal protein gene, rpp-1, as a loading control. The numbers on the right of the blot are in kilobases. (B) The levels of set-2l and set-2s in each strain were normalized to the loading control and are shown in the histogram with the level in N2 arbitrarily set to 1.

View larger version (23K): In this window In a new window Download PPT slide   Figure 4. The 5' end of set-2l and the gene structures of the set-2 isoforms. (A) RT-PCR products of set-2l, generated using SL1 or SL2 as the upstream primer, were probed with a set-2l-specific probe. The results show that set-2l is trans-spliced to SL1 and not to SL2. (B) Gene structures of the two predicted isoforms of set-2, set-2l and set-2s. set-2s starts within intron 9 of set-2l and shares the 3' 3 kb with set-2l. The regions used for RNAi analysis are indicated.

Protein motifs in SET-2:
The predicted 1507-amino- acid SET-2_L and 739-amino-acid SET-2_s proteins both contain a C-terminal SET domain that displays 58% amino acid identity with the SET domain in TRX (Fig 5). The SET domain was first identified as a shared motif among Suppressor of variegation 3-9 [SU(VAR) 3-9] (TSCHIERSCH et al. 1994 ), E(Z) (JONES and GELBART 1993 ), and TRX (STASSEN et al. 1995 ) in Drosophila. All SET domain-containing proteins that have been characterized are associated with chromatin (KUZIN et al. 1994 ; TSCHIERSCH et al. 1994 ; CHINWALLA et al. 1995 ; CARRINGTON and JONES 1996 ; TRIPOULAS et al. 1996 ), suggesting that SET domain proteins are generally involved in chromatin regulation. However, the SET domains in different proteins appear to function through different mechanisms. For example, the SET domains in vertebrate homologs of SU(VAR)3-9 have histone H3 methyltransferase (HMTase) activity, but the SET domains in TRX and E(Z) do not display HMTase activity (REA et al. 2000 ).

View larger version (40K): In this window In a new window Download PPT slide   Figure 5. Protein motifs in SET-2. (A) Schematic alignment of SET-2 isoforms and TRX, showing the relative protein lengths and motifs. (B) Alignment of the SET domains in TRX and SET-2. Identical residues are in black boxes. The alignment was done using NCSA BIOLOGY WORKBENCH software (http://workbench.sdsc.edu).

SET-2_L also contains two consecutive proline-rich regions (Fig 5A). Since proline residues enhance the flexibility of protein structures, they often appear in proteins that are involved in dynamic processes. Many proline-rich proteins function in signal transduction pathways, and the proline-rich regions in them are important for protein-protein interactions (KAY et al. 2000 ).

SET-2_L has a putative RRM on the basis of analysis by the Pfam program (see MATERIALS AND METHODS). RRMs have been found in a variety of RNA-binding proteins (SWANSON et al. 1987 ; BANDZIULIS et al. 1989 ; KENAN et al. 1991 ). The primary sequences of RRMs are not well conserved. However, their structures typically contain four ß sheets and two helices, which form two consecutive ßß sandwiches (BURD and DREYFUSS 1994 ). The second ßß sandwich, especially the last ß sheet, is less well conserved. Several structure prediction programs (see MATERIALS AND METHODS) predicted that the putative RRM of SET-2_L contains the first ßß sandwich and the ß portion of the second ßß sandwich. Another ß sheet was predicted to be adjacent to the C terminus of the predicted RRM domain. This may serve as the last ß sheet to form the second ßß sandwich and a functional RRM in SET-2_L.

set-2 mRNAs are germline enriched:
To gain insight into the pattern of set-2 expression, a set-2 cDNA fragment common to both isoforms was used to probe Northern blots of poly(A)⁺ RNA prepared from wild-type, glp-4(bn2ts), and fem-2(b245ts) adult hermaphrodites (Fig 3). glp-4 adult hermaphrodites, which have a severely underproliferated germline (BEANAN and STROME 1992 ), contain <10% of the level of set-2_l transcripts present in wild-type adult hermaphrodites. set-2_l is present at a higher level in fem-2 hermaphrodites (145% of wild type), which produce only oocytes (KIMBLE et al. 1984 ). However, the level of set-2_s is lower in fem-2 hermaphrodites than in wild type. These results suggest that set-2 mRNA is expressed predominantly in the germline. Consistent with this, in situ hybridization analysis showed germline-enriched accumulation of set-2 transcript in larvae and adults (Y. KOHARA, personal communication).

SET-2_L protein is localized in the nuclei of embryos and germ cells:
To determine the distribution of SET-2_L protein, antibodies were raised against the N-terminal peptide of SET-2_L. The distribution of SET-2_L in wild-type worms and embryos was determined by immunostaining. SET-2_L protein is localized in the nuclei of all cells in embryos at all stages of embryogenesis (Fig 6). In L1 larvae, SET-2_L remains visible in the nuclei of most cell types but is most prominent in Z2 and Z3, the primordial germ cells. In adults, SET-2_L staining is nuclear and is strongest in the germline (Fig 7), although it is also detectable in other cell types, such as intestinal, pharyngeal, and neuronal cells (data not shown). Interestingly, the level of SET-2_L varies in different portions of the gonad. It is present in the distal mitotic cells, diminishes to very low levels in the transition zone and early pachytene regions, and increases dramatically in mid-pachytene nuclei. In oocytes, the level of SET-2_L appears relatively low. The specificity of the antibodies was demonstrated by the observation that SET-2_L immunostaining is reduced to below detection in set-2(RNAi) embryos and worms (Fig 6).

View larger version (46K): In this window In a new window Download PPT slide   Figure 6. Immunolocalization of SET-2L in embryos and L1 larvae. Wild-type worms were stained with affinity-purified anti-SET-2L antibody and 4',6-diamidino-2-phenylindole (DAPI) to visualize DNA. During all stages of embryogenesis, SET-2L is localized in the nuclei of all cells. In L1 larvae, it appears to be present at higher levels in the primordial germ cells, Z2 and Z3 (arrow). (Bottom) SET-2L staining is not detectable in set-2l(RNAi) embryos. Bar, 10 µm.

View larger version (50K): In this window In a new window Download PPT slide   Figure 7. Immunolocalization of SET-2L in the germline. Extruded gonads of wild-type worms were stained with affinity-purified anti-SET-2L antibody and DAPI to visualize DNA. The distal region of the gonad, where germ nuclei divide mitotically and enter meiosis, is to the left, and oocytes are in the upper right. SET-2L is visible in the distal mitotic region of the gonad, undetectable in early pachytene nuclei, and present at high levels in mid- and late-pachytene nuclei. Its level decreases again in oocytes. Bar, 20 µm.

SET-2_L and the MES proteins do not depend upon each other for correct nuclear localization:
One hypothesis to explain set-2 RNAi enhancement of the sterility of mes-3 and mes-4 is that depletion of SET-2 causes defects in the accumulation or distribution of MES-3 and/or MES-4. To test this possibility, we investigated whether the localization of MES proteins is altered in the absence of SET-2. Wild-type worms injected with set-2_l-specific dsRNA were stained with anti-MES-2, anti-MES-3, anti-MES-4, and anti-MES-6. The staining patterns of all four MES proteins resembled the staining patterns in uninjected worms (data not shown). Thus, the enhancement of sterility in mes-3 and mes-4 mutants by RNAi depletion of SET-2_L does not result from significant destabilization or mislocalization of MES proteins.

Conversely, to test whether the localization of SET-2 depends on any of the MES proteins, we compared the staining pattern of SET-2_L in mes-2, mes-3, mes-4, and mes-6 worms and embryos with the pattern in wild type. We found that SET-2_L staining in embryos and germlines is not altered in any of the mes mutants (data not shown).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Screen for enhancers in a sensitized genetic background:
Screening for enhancers or suppressors of mutant phenotypes in a sensitized genetic background has proven to be a powerful approach to identify new components in a regulatory pathway. For example, KARIM et al. 1996 took this approach to identify downstream genes of Ras1 signal transduction in Drosophila, using eye-specific expression of a mutant form of Ras1 to render eye development particularly sensitive to reduction in levels of other pathway components (KARIM et al. 1996 ). As reported here, by searching for inducers of mes-3 sterility in the M+Z- generation, which we consider to be a sensitized genetic background, we identified SET-2 as an additional participant in MES regulation of germline development. Furthermore, using RNAi as the method of gene depletion greatly accelerated the testing process. Our search for enhancers could be expanded to many more C. elegans genes, either by undertaking genetic screens for enhancers of mes sterility or by testing many genes by RNAi. The RNAi approach could be extended to the whole-genome level by systematically feeding mes/+ worms bacteria that express dsRNA directed against nearly every C. elegans gene (FRASER et al. 2000 ) and screening for sterility among the mes progeny.

SET-2 may function redundantly with zygotic MES-3 and MES-4:
MES proteins have been considered as strict maternal-effect factors. However, transgenic studies have demonstrated that zygotically synthesized MES-3 and MES-4 participate in transgene repression in the C. elegans germline (KELLY and FIRE 1998 ). Furthermore, although M+Z- mes worms are fertile, they have reduced brood sizes (CAPOWSKI et al. 1991 ); in the case of mes-3, 15% of M+Z- worms are sterile. Our findings in this article strengthen the view that zygotic MES-3 and MES-4 play important roles in the germline. RNAi depletion of set-2 causes the majority of mes-3 or mes-4 homozygous progeny from heterozygous mothers to be sterile, whereas their heterozygous siblings remain fertile. Thus, depletion of SET-2 leads to sterility only when MES-3 or MES-4 is not zygotically synthesized.

One explanation for the sterility observed in set-2 (RNAi); mes-3 and set-2(RNAi); mes-4 hermaphrodites is that SET-2 functions redundantly with zygotic MES-3 and MES-4. When zygotic MES-3 and MES-4 are present, SET-2 is not necessary for development of a fertile germline. Conversely, when SET-2 is present, zygotic MES-3 and MES-4 are not necessary for fertility. When a combination of SET-2 and either MES-3 or MES-4 is absent, then sterility ensues. As described in the accompanying article (XU et al. 2001B ), analysis of the temperature-sensitive period for mes-3 suggests that MES-3 function is not required after embryogenesis; the presence of functional MES-3 in the maternal germline and during embryogenesis is sufficient to ensure that those embryos develop into fertile adults. One scenario is that maternally supplied MES-3 induces a special state of chromatin in the germline blastomeres, which can be propagated during larval development in the absence of MES-3. The results in this article suggest that SET-2 may be involved in this epigenetic propagation. In the absence of SET-2, MES-3 may be able to participate in propagation of the germline state. In the absence of both SET-2 and MES-3, chromatin defects in the germline lead to sterility. Thus far, we have no evidence that SET-2 interacts physically with MES-3. The distributions of MES-3 and SET-2_L do not depend upon one another, and the estimated size of the MES-3/MES-2/MES-6 complex in embryo extracts (255 kD; XU et al. 2001A ) predicts that the complex does not contain an additional large protein, such as SET-2_S or SET-2_L.

It is interesting that SET-2 depletion enhances the sterility of both mes-3 and mes-4 but not of mes-2 or mes-6. Because the mutant phenotypes of the four mes genes are very similar in most respects (CAPOWSKI et al. 1991 ; PAULSEN et al. 1995 ; GARVIN et al. 1998 ) and because MES-3, MES-2, and MES-6 are assembled into and probably function in a complex in embryos, we expected that an enhancer of mes-3 would also enhance mes-2 and mes-6. This is not observed for SET-2. Instead, enhancement of sterility by SET-2 depletion appears to correlate with zygotic effects on transgene expression in the germline: mes-3 and mes-4 M+Z- worms show transgene desilencing (KELLY and FIRE 1998 ) and, when depleted for set-2, are sterile, whereas mes-2 and mes-6 M+Z- worms rarely show transgene desilencing (KELLY and FIRE 1998 ) and, when depleted for set-2, are not sterile. These results may be explained by different degrees of perdurance of maternally supplied mes gene product. Persistence of mes-2 and mes-6 maternal products in mes-2 and mes-6 M+Z- worms may enable their germlines to develop relatively normally and maintain transgene silencing, whereas more rapid disappearance of mes-3 and mes-4 maternal product in mes-3 and mes-4 M+Z- worms may compromise germline function, release transgenes from silencing, and render germline development more sensitive to depletion of other required factors, such as SET-2.

The two isoforms of SET-2, SET-2_L and SET-2_S, appear to function redundantly in enhancing the mes-3 and mes-4 mutant phentoype, since depletion of both SET-2_L and SET-2_S by RNAi created a more severe enhancement effect than did depletion of SET-2_L alone. Both SET-2 isoforms contain a SET domain, but only SET-2_L contains a putative RRM, which might be involved in RNA processing of target genes. It would be interesting to identify the downstream targets of SET-2_L and SET-2_S, because they may be downstream targets of zygotic MES-3 and MES-4 as well and may contribute to the sterility in mes mutants.

SET-2_L protein distribution:
Immunostaining showed that SET-2_L is most prominently localized in the germline and in embryos, where its distribution closely parallels the distribution of MES-3 (see accompanying article). Interestingly, the levels of both proteins dramatically drop in the early meiotic region of the adult hermaphrodite germline. In the case of MES-3, this drop appears to depend on GLD-1 regulation [see accompanying article (XU et al. 2001B )]. The decline of SET-2 level does not appear to depend on either GLD-1 or MES-3; the SET-2 mRNAs do not contain consensus GLD-1-binding sites, and the distribution looks normal in M+Z- mes-3 mutant germlines. We speculate that the drop in MES-3 and SET-2 levels as germ cells enter meiosis allows remodeling of chromatin and alteration of gene expression patterns for oogenesis.

SET-2_L is also present at easily detectable levels in various somatic cells in adults. The somatic localization of SET-2_L raises the possibility that SET-2 functions in a process common to the germline and soma but through different partners. In the germline, it may participate along with MES proteins in maintenance of proper gene expression patterns. In the soma, it may cooperate with other factors to ensure proper development of somatic tissues. SET-2_L contains proline-rich regions, which are thought to be involved in protein-protein interactions (KAY et al. 2000 ). It also contains an RRM, suggesting that its activity may involve binding RNA.

The SET domain in SET-2:
SET-2 was named after the SET domain at the C termini of its two isoforms, SET-2_L and SET-2_S. The SET domains in two vertebrate homologs of SU(VAR)3-9 were recently shown to have histone HMTase activity (REA et al. 2000 ). A catalytic motif in the SET domain and two cysteine-rich regions flanking the SET domain are important for this activity. The SET domain in EZH2 and HRX, the human homologs of E(Z) and TRX, lacks the C-terminal cysteine-rich region and the cysteine-rich region upstream of the SET domain, respectively. They do not display HMTase activity. Like TRX, SET-2 lacks the cysteine-rich region upstream of the SET domain. Therefore, SET-2 is unlikely to have HMTase activity. Instead, the SET domain in SET-2 might mediate protein-protein interactions, as the SET domain in TRX does (CUI et al. 1998 ; ROZENBLATT-ROSEN et al. 1998 ; ROZOVSKAIA et al. 1999 ).

	FOOTNOTES

¹ Present address: Center for Cancer Research, Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139.

	ACKNOWLEDGMENTS

We thank Yuji Kohara for cDNA clones and Lucy Cherbas, Roger Innes, Beth Raff, and Claire Walczak for critical reading of this manuscript. This work was supported by National Institutes of Health grant GM-34059 (to S.S.).

Manuscript received June 22, 2001; Accepted for publication August 24, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AKAM, M., 1998  Hox genes, homeosis and the evolution of segment identity: no need for hopeless monsters. Int. J. Dev. Biol. 42:445-451[Medline].

BANDZIULIS, R. J., M. S. SWANSON, and G. DREYFUSS, 1989  RNA-binding proteins as developmental regulators. Genes Dev. 3:431-437[Free Full Text].

BEANAN, M. J. and S. STROME, 1992  Characterization of a germ-line proliferation mutation in C. elegans. Development 116:755-766[Abstract].

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

BURD, C. G. and G. DREYFUSS, 1994  Conserved structures and diversity of functions of RNA-binding proteins. Science 265:615-621[Abstract/Free Full Text].

CAPOWSKI, E. E., P. MARTIN, C. GARVIN, and S. STROME, 1991  Identification of grandchildless loci whose products are required for normal germ-line development in the nematode Caenorhabditis elegans. Genetics 129:1061-1072[Abstract].

CARRINGTON, E. A. and R. S. JONES, 1996  The Drosophila Enhancer of zeste gene encodes a chromosomal protein: examination of wild-type and mutant protein distribution. Development 122:4073-4083[Abstract].

CARROLL, S. B., R. A. LAYMON, M. A. MCCUTCHEON, P. D. RILEY, and M. P. SCOTT, 1986  The localization and regulation of the Antennapedia protein expression in Drosophila embryos. Cell 47:113-122[Medline].

CELNIKER, S. E., D. J. KEELAN, and E. B. LEWIS, 1989  The molecular genetics of the bithorax complex of Drosophila: characterization of the product of the Abdominal-B domain. Genes Dev. 3:1424-1436[Abstract/Free Full Text].

CHINWALLA, V., E. P. JANE, and P. J. HARTE, 1995  The Drosophila trithorax protein binds to specific chromosomal sites and is co-localized with Polycomb at many sites. EMBO J. 14:2056-2065[Medline].

CONRAD, R., J. THOMAS, J. SPIETH, and T. BLUMENTHAL, 1991  Insertion of part of an intron into the 5' untranslated region of a Caenorhabditis elegans gene converts it into a trans-spliced gene. Mol. Cell. Biol. 11:1921-1926[Abstract/Free Full Text].

CUI, X., I. DE VIVO, R. SLANY, A. MIYAMOTO, and R. FIRESTEIN et al., 1998  Association of SET domain and myotubularin-related proteins modulates growth control. Nat. Genet. 18:331-337[Medline].

EVANS, D., D. ZORIO, M. MACMORRIS, C. E. WINTER, and K. LEA et al., 1997  Operons and SL2 trans-splicing exist in nematodes outside the genus Caenorhabditis. Proc. Natl. Acad. Sci. USA 94:9751-9756[Abstract/Free Full Text].

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].

FRASER, A. G., R. S. KAMATH, P. ZIPPERLEN, M. MARTINEZ-CAMPOS, and M. SOHRMANN et al., 2000  Functional genomic analysis of C. elegans chromosome I by systematic RNA interference. Nature 408:325-330[Medline].

GARVIN, C., R. HOLDEMAN, and S. STROME, 1998  The phenotype of mes-2, mes-3, mes-4, and mes-6, maternal-effect genes required for survival of the germline in Caenorhabditis elegans, is sensitive to chromosome dosage. Genetics 148:1-20[Free Full Text].

GOULD, A., 1997  Functions of mammalian Polycomb group and trithorax group related genes. Curr. Opin. Genet. Dev. 7:488-494[Medline].

HAYNES, S. R., B. A. MOZER, N. BHATIA-DEY, and I. B. DAWID, 1989  The Drosophila fsh locus, a maternal effect homeotic gene, encodes apparent membrane proteins. Dev. Biol. 134:246-257[Medline].

HOLDEMAN, R., S. NEHRT, and S. STROME, 1998  MES-2, a maternal protein essential for viability of the germline in Caenorhabditis elegans, is homologous to a Drosophila Polycomb group protein. Development 125:2457-2467[Abstract].

JONES, R. S. and W. M. GELBART, 1990  Genetic analysis of the Enhancer of zeste locus and its role in gene regulation in Drosophila melanogaster. Genetics 126:185-199[Abstract].

JONES, R. S. and W. M. GELBART, 1993  The Drosophila Polycomb-group gene Enhancer of zeste contains a region with sequence similarity to trithorax. Mol. Cell. Biol. 13:6357-6366[Abstract/Free Full Text].

KARIM, F. D., H. C. CHANG, M. THERRIEN, D. A. WASSARMAN, and T. LAVERTY et al., 1996  A screen for genes that function downstream of Ras1 during Drosophila eye development. Genetics 143:315-329[Abstract].

KAY, B. K., M. P. WILLIAMSON, and M. SUDOL, 2000  The importance of being proline: the interaction of proline-rich motifs in signaling proteins with their cognate domains. FASEB J. 14:231-241[Abstract/Free Full Text].

KELLY, W. G. and A. FIRE, 1998  Chromatin silencing and the maintenance of a functional germline in Caenorhabditis elegans. Development 125:2451-2456[Abstract].

KELLY, W. G., S. XU, M. K. MONTGOMERY, and A. FIRE, 1997  Distinct requirements for somatic and germline expression of a generally expressed Caernorhabditis elegans gene. Genetics 146:227-238[Abstract].

KENAN, D. J., C. C. QUERY, and J. D. KEENE, 1991  RNA recognition: towards identifying determinants of specificity. Trends Biochem. Sci. 16:214-220[Medline].

KIMBLE, J., L. EDGAR, and D. HIRSH, 1984  Specification of male development in Caenorhabditis elegans: the fem genes. Dev. Biol. 105:234-239[Medline].

KORF, I., Y. FAN, and S. STROME, 1998  The Polycomb group in Caenorhabditis elegans and maternal control of germline development. Development 125:2469-2478[Abstract].

KUZIN, B., S. TILLIB, Y. SEDKOV, L. MIZROKHI, and A. MAZO, 1994  The Drosophila trithorax gene encodes a chromosomal protein and directly regulates the region-specific homeotic gene fork head. Genes Dev. 8:2478-2490[Abstract/Free Full Text].

LAJEUNESSE, D. and A. SHEARN, 1996  E(z): a polycomb group gene or a trithorax group gene? Development 122:2189-2197[Abstract].

MARTINEZ-BALBAS, M. A., T. TSUKIYAMA, D. GDULA, and C. WU, 1998  Drosophila NURF-55, a WD repeat protein involved in histone metabolism. Proc. Natl. Acad. Sci. USA 95:132-137[Abstract/Free Full Text].

OWEN-HUGHES, T., R. T. UTLEY, J. COTE, C. L. PETERSON, and J. L. WORKMAN, 1996  Persistent site-specific remodeling of a nucleosome array by transient action of the SWI/SNF complex. Science 273:513-516[Abstract].

PAULSEN, J. E., E. E. CAPOWSKI, and S. STROME, 1995  Phenotypic and molecular analysis of mes-3, a maternal-effect gene required for proliferation and viability of the germ line in C. elegans. Genetics 141:1383-1398[Abstract].

PIRROTTA, V., 1995  Chromatin complexes regulating gene expression in Drosophila. Curr. Opin. Genet. Dev. 5:466-472[Medline].

PIRROTTA, V., 1997  PcG complexes and chromatin silencing. Curr. Opin. Genet. Dev. 7:249-258[Medline].

PREUSS, D., 1999  Chromatin silencing and Arabidopsis development: a role for polycomb proteins. Plant Cell 11:765-768[Free Full Text].

REA, S., F. EISENHABER, D. O'CARROLL, B. D. STRAHL, and Z. W. SUN et al., 2000  Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature 406:593-599. [see comments][Medline].

ROCHELEAU, C. E., W. D. DOWNS, R. LIN, C. WITTMANN, and Y. BEI et al., 1997  Wnt signaling and an APC-related gene specify endoderm in early C. elegans embryos. Cell 90:707-716[Medline].

ROZENBLATT-ROSEN, O., T. ROZOVSKAIA, D. BURAKOV, Y. SEDKOV, and S. TILLIB et al., 1998  The C-terminal SET domains of ALL-1 and TRITHORAX interact with the INI1 and SNR1 proteins, components of the SWI/SNF complex. Proc. Natl. Acad. Sci. USA 95:4152-4157[Abstract/Free Full Text].

ROZOVSKAIA, T., S. TILLIB, S. SMITH, Y. SEDKOV, and O. ROZENBLATT-ROSEN et al., 1999  Trithorax and ASH1 interact directly and associate with the trithorax group-responsive bxd region of the Ultrabithorax promoter. Mol. Cell. Biol. 19:6441-6447[Abstract/Free Full Text].

SAMBROOK, J. E., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SHAO, Z., F. RAIBLE, R. MOLLAAGHABABA, J. R. GUYON, and C. T. WU et al., 1999  Stabilization of chromatin structure by PRC1, a Polycomb complex. Cell 98:37-46[Medline].

SIMON, J., 1995  Locking in stable states of gene expression: transcriptional control during Drosophila development. Curr. Opin. Cell Biol. 7:376-385[Medline].

SPIETH, J., S. BROOKE, S. KUERSTEN, K. LEA, and T. BLUMENTHAL, 1993  Operons in C. elegans: polycistronic mRNA precursors are processed by trans-splicing of SL2 to downstream coding regions. Cell 73:521-532[Medline].

STANKUNAS, K., J. BERGER, C. RUSE, D. A. SINCLAIR, and F. RANDAZZO et al., 1998  The enhancer of polycomb gene of Drosophila encodes a chromatin protein conserved in yeast and mammals. Development 125:4055-4066[Abstract].

STASSEN, M. J., D. BAILEY, S. NELSON, V. CHINWALLA, and P. HARTE, 1995  The Drosophila trithorax proteins contain a novel variant of the nuclear receptor type DNA binding domain and an ancient conserved motif found in other chromosomal proteins. Mech. Dev. 52:209-223[Medline].

STROME, S. and W. B. WOOD, 1983  Generation of asymmetry and segregation of germ-line granules in early C. elegans embryos. Cell 35:15-25[Medline].

SWANSON, M. S., T. Y. NAKAGAWA, K. LEVAN, and G. DREYFUSS, 1987  Primary structure of human nuclear ribonucleoprotein particle C proteins: conservation of sequence and domain structures in heterogeneous nuclear RNA, mRNA, and pre-rRNA-binding proteins. Mol. Cell. Biol. 7:1731-1739[Abstract/Free Full Text].

TAMKUN, J. W., R. DEURING, M. P. SCOTT, M. KISSINGER, and A. M. PATTATUCCI et al., 1992  brahma: a regulator of Drosophila homeotic genes structurally related to the yeast transcriptional activator SNF2/SWI2. Cell 68:561-572[Medline].

TRIPOULAS, N., D. LAJEUNESSE, J. GILDEA, and A. SHEARN, 1996  The Drosophila ash1 gene product, which is localized at specific sites on polytene chromosomes, contains a SET domain and a PHD finger. Genetics 143:913-928[Abstract].

TSCHIERSCH, B., A. HOFMANN, V. KRAUSS, R. DORN, and G. KORGE et al., 1994  The protein encoded by the Drosophila position-effect variegation suppressor gene Su(var)3–9 combines domains of antagonistic regulators of homeotic gene complexes. EMBO J. 13:3822-3831[Medline].

XU, L., Y. FONG, and S. STROME, 2001a  The Caenorhabditis elegans maternal-effect sterile proteins, MES-2, MES-3, and MES-6, are associated in a complex in embryos. Proc. Natl. Acad. Sci. USA 98:5061-5066[Abstract/Free Full Text].

XU, L., J. PAULSEN, Y. YOO, E. B. GOODWIN, and S. STROME, 2001b  Caenorhabditis elegans MES-3 is a target of GLD-1 and functions epigenetically in germline development. Genetics 159:1007-1017[Abstract/Free Full Text].

	NOTE ADDED IN PROOF

In a recent update to GenBank, the exons encoding a GTP-binding domain were removed from C26E6.9a, and C26E6.9c was added to the list of C26E6.9 transcripts. C26E6.9c is predicted to encode a protein with three extra amino acids compared to C26E6.9a. Consequently, SET-2_L is likely encoded by both C26E6.9a and C26E6.9c.

This article has been cited by other articles:

				L. B. Bender, J. Suh, C. R. Carroll, Y. Fong, I. M. Fingerman, S. D. Briggs, R. Cao, Y. Zhang, V. Reinke, and S. Strome MES-4: an autosome-associated histone methyltransferase that participates in silencing the X chromosomes in the C. elegans germ line. Development, October 1, 2006; 133(19): 3907 - 3917. [Abstract] [Full Text] [PDF]

				M. A. Jedrusik and E. Schulze Telomeric Position Effect Variegation in Saccharomyces cerevisiae by Caenorhabditis elegans Linker Histones Suggests a Mechanistic Connection between Germ Line and Telomeric Silencing Mol. Cell. Biol., May 15, 2003; 23(10): 3681 - 3691. [Abstract] [Full Text] [PDF]