Caenorhabditis elegans lin-13, a Member of the LIN-35 Rb Class of Genes Involved in Vulval Development, Encodes a Protein With Zinc Fingers and an LXCXE Motif

Corresponding author: Iva Greenwald, HHMI/Biochemistry, 701 W. 168th St., Rm. 720, Columbia University College of Physicians and Surgeons, New York, NY 10032., greenwald{at}cuccfa.ccc.columbia.edu (E-mail)

Communicating editor: R. K. HERMAN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The SynMuv genes appear to be involved in providing a signal that inhibits vulval precursor cells from adopting vulval fates in Caenorhabditis elegans. One group of SynMuv genes, termed class B, includes genes encoding proteins related to the tumor suppressor Rb and RbAp48, a protein that binds Rb. Here, we provide genetic evidence that lin-13 behaves as a class B SynMuv gene. We show that null alleles of lin-13 are temperature sensitive and maternally rescued, resulting in phenotypes ranging in severity from L2 arrest (when both maternal and zygotic activities are removed at 25°), to sterile Multivulva (when only zygotic activity is removed at 25°), to sterile non-Multivulva (when both maternal and zygotic activities are removed at 15°), to wild-type/class B SynMuv (when only zygotic activity is removed at 15°). We also show that LIN-13 is a nuclear protein that contains multiple zinc fingers and a motif, LXCXE, that has been implicated in Rb binding. These results together suggest a role for LIN-13 in Rb-mediated repression of vulval fates.

THE cell-cell interactions that specify cell fates during Caenorhabditis elegans vulval development have been intensively studied (reviewed in GREENWALD 1997 ). The vulva is normally formed from the 22 descendants of three cells, P5.p, P6.p, and P7.p. However, each of six cells, numbered P3.p–P8.p, has the potential to generate vulval cells, and hence these six cells are named "vulval precursor cells" (VPCs). Each VPC adopts one of three fates, which can be recognized by the lineages they undergo. The "1°" and "2°" fates are termed "vulval" fates, because they lead to the production of vulval cells. The "3°" fate is termed the "nonvulval" or "hypodermal" fate, because it leads to the production of cells that join the hyp7 hypodermal syncytium. In wild-type hermaphrodites, P3.p–P8.p always adopt the same pattern of fates: 3°-3°-2°-1°-2°-3°.

Genetic screens have identified genes that are necessary for VPC patterning and vulval fates. These screens have relied on two vulval abnormalities. A "Vulvaless" phenotype results when P5.p, P6.p, and P7.p adopt nonvulval fates. The Vulvaless phenotype is readily visualized in the dissecting microscope when progeny hatch internally and form a "bag of worms." A "Multivulva" (Muv) phenotype results when P3.p, P4.p, and/or P8.p inappropriately adopt vulval (1° and 2°) fates. The Multivulva phenotype is readily visualized in the dissecting microscope by multiple pseudovulval protrusions, with each protrusion generally corresponding to progeny produced by a single VPC.

The analysis of Vulvaless and Multivulva mutants, in combination with cell ablation experiments, has revealed that three different signaling events specify the pattern of the VPC fates (reviewed in GREENWALD 1997 ). One event is referred to as the "inductive" signal, which promotes the adoption of 1° fates. The inductive signal is produced by the anchor cell of the gonad and is transduced by the LET-23 receptor tyrosine kinase in the VPCs. The inductive signal may countermand another signal, referred to as the "inhibitory" signal, which prevents VPCs from adopting vulval fates. The inhibitory signal is believed to emanate from hyp7, the syncytial hypodermal cell that encompasses much of the animal. A third signal, referred to as the "lateral" signal, is believed to be produced by the VPC that is adopting the 1° fate to cause an adjacent cell to adopt the 2° fate.

The absence of inhibitory signaling results in a Multivulva phenotype because extra VPCs adopt vulval fates. Studies of a group of genes termed the "SynMuv" genes revealed the existence of inhibitory signaling (HERMAN and HEDGECOCK 1990 ). These genes are called "SynMuv" genes because a Multivulva phenotype is visible only if mutations in two different genes are combined (a Synthetic Multivulva phenotype; HORVITZ and SULSTON 1980 ; FERGUSON and HORVITZ 1989 ). FERGUSON and HORVITZ 1989 identified two classes of SynMuv mutations, class A and class B; a Multivulva phenotype results only when a class A mutation is combined with a class B mutation. These genetic properties led them to suggest that there are two functionally redundant pathways, one composed of class A genes and the other composed of class B genes (FERGUSON and HORVITZ 1989 ). Other genetic studies have led to the suggestion that SynMuv genes act upstream of or in parallel to the LET-23 receptor tyrosine kinase (FERGUSON et al. 1987 ; HUANG et al. 1994 ).

Several SynMuv genes have been characterized in molecular detail. The first three SynMuv genes that were cloned were found to encode novel proteins. The lin-15 locus contains two genes, one class A and one class B, each encoding a novel protein product (CLARK et al. 1994 ; HUANG et al. 1994 ). The class B SynMuv gene lin-36 encodes a novel protein that is expressed in many cell types (CLARK et al. 1994 ; HUANG et al. 1994 ; THOMAS and HORVITZ 1999 ).

Recently, LU and HORVITZ 1998 found that other SynMuv genes encode homologs of proteins known to regulate gene expression. They found that lin-35, a class B SynMuv gene, encodes a protein related to the tumor suppressor Rb, and lin-53, another class B SynMuv gene, encodes a protein related to RbAp48, a protein that binds Rb. Furthermore, they also obtained evidence that the hda-1 histone deacetylase gene also acts as a class B SynMuv (LU and HORVITZ 1998 ), consistent with biochemical studies of the interactions of histone deacetylase and Rb in mammalian cells (BREHM et al. 1998 ; LUO et al. 1998 ; MAGNAGHI-JAULIN et al. 1998 ). These results suggest that other class B SynMuv genes may also be involved in Rb-mediated transcriptional regulation.

Here, extending an initial observation of FERGUSON and HORVITZ 1989 , we show that the lin-13 gene has genetic properties consistent with function as a SynMuv gene. We show that the null phenotype of lin-13 is temperature sensitive and that a Multivulva phenotype results from the absence of zygotic lin-13 activity at 25°, whereas the absence of zygotic activity at 15° causes a Multivulva phenotype only in the presence of a class A SynMuv mutation. In addition, we show that the predicted LIN-13 protein contains multiple zinc fingers and a motif (LXCXE) that has been implicated in Rb binding. Finally, we show that LIN-13 is a nuclear protein and is consistently expressed in many cell types, including hyp7, at the time of VPC specification. Our results are consistent with a role for LIN-13 in an Rb-mediated transcriptional control process that leads to repression of vulval fates.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

General methods and strains:
General methods for the handling and maintenance of C. elegans are described by BRENNER 1974 . Experiments were usually performed at both 15° and 25°, as indicated in the tables and text. Mutations used were as follows:

• LG I: lin-35(n745) (FERGUSON and HORVITZ 1989 ), smg-1(r861) and unc-54(r293) (HODGKIN et al. 1989 ).

• LG II: lin-8(n111) (FERGUSON and HORVITZ 1985 ), lin-38(n751) (FERGUSON and HORVITZ 1989 ).

• LG III: unc-93(e1500) (GREENWALD and HORVITZ 1980 ), unc-36(e251) (BRENNER 1974 ), lin-13(n387, n388) (FERGUSON and HORVITZ 1985 ), qC1 (AUSTIN and KIMBLE 1987 ).

• LG X: lin-15(n767), lin-15(n374), and lin-15(n745) (FERGUSON and HORVITZ 1989 ); lin-15(sy197) (cited in HUANG et al. 1994 ).

The transgenes arIs30, used for lin-13::lacZ studies, and arIs48 and arIs49, used for LIN-13::GFP (green fluorescent protein) studies, are described below.

Transformation rescue:
Transgenic lines were generated by microinjecting unc-36 lin-13(n387)/qC1 hermaphrodites with cosmid (10 µg/ml) or plasmid DNA (2 µg/ml) along with the dominant rol-6 marker pRF4 at a concentration of 100 µg/ml (MELLO et al. 1991 ). Stable Rol lines were reared at 25°, and individual Unc Rol hermaphrodites segregating from each line were analyzed for the Muv and sterility defects.

Plasmids:
The following plasmids were generated during the course of this study and used for experiments described in the text.

• pM16 (rescues lin-13): A 16-kb MluI fragment from cosmid C03B8 was cloned into the MluI site of pGem (Promega, Madison, WI).

• pMB13: A 13-kb BstEII to MluI fragment from cosmid C03B8 was cloned into the BstEII and MluI sites of Litmus 38 (New England Biolabs, Beverly, MA).

• pMS12 (rescues lin-13): A 12-kb MluI to SalI site of cosmid C03B8 was cloned into the MluI and SalI sites of Litmus 38 (New England Biolabs). This plasmid served as the basis for pMSN12, pMS12, pNlin-13::GFP, and pClin-13::GFP.

• pMSN12 (rescues lin-13): A NotI site was engineered by PCR at the predicted ATG of C03B8.4 in pMS12.

• pMSN12: The NotI site of pMSN12 was cut, treated with Klenow to fill in the overhangs, and religated to destroy the ATG site.

• pNlin-13::lacZ (transcription reporter): pNlin-13::lacZ contains a modified lacZ gene (encoding ß-galactosidase carrying a nuclear localization signal) derived from pPD95.11 (FIRE et al. 1990 ). A 3.8-kb MscI to StuI fragment containing the lacZ gene (including a stop codon) was inserted in frame into the NotI site of plasmid pMSN12.

• pNlin-13::GFP (N terminus tagged with GFP): A PCR product encoding GFP (variant S65T, I167T) without its own stop codon was cloned in frame at the predicted ATG where the NotI was inserted in pMS12. The resulting hybrid protein has GFP at the N terminus fused to the entire LIN-13 protein except for its initial methionine.

• pClin-13::GFP (C terminus tagged with GFP): A PCR product encoding GFP (variant S65T, I167T) without its own stop codon was cloned in frame into the unique KpnI site (codon 2063) of pMS12. The resulting hybrid protein has GFP inserted into LIN-13 near the C terminus, so that the LIN-13 C terminus is preserved.

Generation of lines expressing lin-13::lacZ:
pNlin-13::lacZ was coinjected at a concentration of 50 µg/ml with pRF4 [rol-6(su1006)] (MELLO et al. 1991 ) at a concentration of 100 µg/ml into smg-1 unc-54 hermaphrodites. Four independent lines containing extrachromosomal arrays were established and each was integrated. All of the analysis described in this article was performed using the integrated array arIs30, which displayed a consistent staining pattern.

Generation of lines expressing LIN-13::GFP:
pNlin-13::GFP was coinjected at a concentration of 100 µg/ml with pRF4 [rol-6(su1006)] (MELLO et al. 1991 ) at a concentration of 100 µg/ml into N2 hermaphrodites. Two lines carrying extrachromosomal arrays were examined using a Zeiss Axiophot microscope equipped with fluorescence optics and both showed GFP expression in embryos still in utero. One extrachromosomal array was integrated to generate arIs48.

pClin-13::GFP was coinjected at a concentration of 100 µg/ml with pNC4-21 [unc-4(+)] (MILLER and NIEMEYER 1995 ) at a concentration of 50 µg/ml into unc-4(e120) hermaphrodites. Four lines carrying extrachromosomal arrays were examined using a Zeiss Axiophot microscope equipped with fluorescence optics and all displayed the same expression pattern as the lines carrying extrachromosomal arrays of pNlin-13::GFP. One extrachromosomal array was integrated to generate arIs49.

Antibody staining:
For indirect immunofluorescence of fixed larval populations, hermaphrodites were synchronized by treating a mixed population with hypochlorite solution to isolate embryos. Staged preparations were fixed as described in BETTINGER et al. 1996 . Fixed worms were incubated with a polyclonal anti-ß-galactosidase antibody (CAPELL) or a polyclonal anti-GFP antibody (Clontech, Palo Alto, CA), diluted 1:200. In some cases, monoclonal antibody MH27 (PRIESS and HIRSH 1986 ) diluted 1:1000 in PTB (1x PBS, 1% BSA, 0.5% Triton X-100) was also added. The primary incubation proceeded overnight at 4°. Worms were washed with 1x PBS, 0.1% BSA, 0.5% Triton-X 100 with several changes of buffer for 4 hr at room temperature. Cy3-conjugated goat anti-rabbit secondary antibody and FITC-conjugated goat anti-mouse secondary antibody (both from Jackson Immunoresearch, West Grove, PA) were diluted 1:300 in PTB and incubated with the fixed worms overnight at 4°. Worms were washed as before, for 4 hr at room temperature. 4',6-diamidino-2-phenylindole (DAPI) was added to the last wash at a final concentration of 1 µg/ml. Worms were mounted on a 2% agarose pad with 3 µl of 10% N-propyl gallate and viewed with a Zeiss LSM 410 laser scanning confocal attachment on a Zeiss Axiovert 100 microscope.

Sequence analysis:
Standard molecular biology protocols were performed as described in SAMBROOK et al. 1989 . The DNA sequence of C03B8 was obtained from the C. elegans genome sequencing project (WATERSTON et al. 1997 ). The exons of lin-13 were predicted by GENEFINDER (EDGLEY et al. 1997 ). To examine the accuracy of this prediction, we analyzed cDNAs isolated by screening a mixed-stage cDNA library (Stratagene, La Jolla, CA) with the ³²P-labeled 12-kb MluI-SalI fragment. Four cDNAs were identified among 120,000 PFUs screened. These cDNAs were sequenced on one strand in their entirety and spanned the 3' end and somewhat more than half of the predicted coding region. To determine the structure of the 5' end and the remaining coding region, we generated PCR products using DNA prepared by Q. Dong from a mixed-stage cDNA library (BARSTEAD and WATERSTON 1989 ). The 5' end was determined from a PCR product generated using the SL1 trans-spliced leader sequence (KRAUSE and HIRSH 1987 ) and an internal lin-13 primer (data not shown).

The lesions associated with lin-13 mutations were found by sequencing the lin-13 coding region of the mutants. We amplified the lin-13 genomic region by PCR reactions from individual lin-13 homozygous mutant hermaphrodites segregating from unc-36 lin-13/qC1 parents. The PCR products were directly sequenced using established methods.

RNA-mediated interference:
Double-stranded RNA was prepared using the RNA transcription kit (Stratagene) and injected without dilution according to FIRE et al. 1998 . Injected hermaphrodites were transferred every day for 3 days. A total of 82 injected hermaphrodites were incubated at 25° and their progeny were examined daily for up to 9 days. A total of 24 injected hermaphrodites were incubated at 15° and their progeny were examined daily until they reached adulthood.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

lin-13 genetic interactions with SynMuv genes:
lin-13(n387) and lin-13(n388) were identified in screens for vulval mutants and genetically characterized (FERGUSON and HORVITZ 1985 ). Both mutations, which cause similar phenotypes, are heat sensitive and display maternal effects: at 25°, lin-13 homozygous hermaphrodites segregating from a lin-13/+ mother are sterile and display a Muv phenotype, but at 15° they are fertile, non-Muv, and grandchildless (i.e., their progeny are sterile; FERGUSON and HORVITZ 1985 ). The focus of the work described in this article is the role of lin-13 in VPC specification.

FERGUSON and HORVITZ 1989 identified two classes of mutations, class A and class B, that behave as SynMuv mutations: mutations in class A are non-Muv, and mutations in class B are non-Muv, but double mutants carrying a mutation in class A along with a mutation in class B are Muv. FERGUSON and HORVITZ 1989 also reported that lin-13(n387); lin-15A(n767) hermaphrodites are Muv at 15°, suggesting that lin-13 may be a class B SynMuv gene. However, they did not report any combinations of lin-13 alleles with any other class A alleles or any class B alleles, raising the possibility that the interaction with lin-15A(n767) is specific for the lin-15A gene or for the n767 allele.

We therefore examined double mutants between lin-13(n387) and several different class A and class B SynMuv mutations at 15° (Table 1). Double mutants containing lin-13(n387) and the class A mutations lin-8(n111), lin-15(sy197), lin-15(n767), or lin-38(n751) are Muv at 15°. Double mutants containing lin-13(n387) and the class B mutations lin-15(n374) or lin-35(n745) are non-Muv at 15°. The double mutant analysis therefore indicates that lin-13(n387) consistently behaves as if it has reduced class B SynMuv gene activity at 15°.

In addition to these genetic interactions at 15°, we note also that for the class A mutation lin-38(n751), a strong semidominant interaction was seen at 25°: 22% (111/507) of hermaphrodites of genotype lin-38(n751); lin-13(n387) unc-36(e251)/dpy-17(e164) were Muv at 25°. The interaction with lin-13(n387)/+ was weak or not detectable for other class A mutations and not detectable for class B mutations (data not shown).

Molecular cloning and sequence analysis of lin-13:
lin-13 was known to map very near and to the left of the cloned gene mab-5 (COSTA et al. 1988 and personal communication). Cosmids for the implicated genomic region were tested in germline transformation experiments (G. KAO, A. MELÉNDEZ and I. GREENWALD, unpublished observations), and cosmid C03B8 (C. ELEGANS SEQUENCING CONSORTIUM 1998) was found to rescue both the sterility and Muv phenotypes of lin-13(n387) (Fig 1). We analyzed subclones of C03B8 in rescue assays and identified a 16-kb MluI fragment (pM16) and a smaller 12-kb MluI-SalI fragment (pMS12) that have rescuing activity (Fig 1). These fragments include a predicted gene, C03B8.4, which encodes a predicted zinc-finger protein; the only predicted gene present on the 12-kb MluI-SalI fragment, and the only complete gene present on the 16-kb MluI fragment, is C03B8.4. The introduction of a frameshift mutation at the predicted start ATG of C03B8.4 on the 12-kb MluI fragment (pMS–DN12) abolished rescuing activity (Fig 1). These results suggested that C03B8.4 corresponds to lin-13, and this inference was confirmed by finding sequence changes in both lin-13 alleles (see below).

View larger version (9K): In this window In a new window Download PPT slide   Figure 1. Molecular cloning of lin-13. Rescue of lin-13 mutant phenotypes was assessed as described in MATERIALS AND METHODS. Relevant restriction sites mentioned in the text are indicated: M, MluI; N, NotI; B, BstEII; S, SalI. (A) Simplified genetic map. (B) Schematic physical map.

We isolated and characterized cDNAs corresponding to the zinc-finger protein and verified the exon/intron junctions as predicted by GENEFINDER (EDGLEY et al. 1997 ). Our results were in agreement with the predicted sequence except for the following: the GENEFINDER prediction contained two putative small exons that were not identified by the cDNA analysis, one exon that was instead found to be longer by the cDNA analysis, and a predicted intron that was instead found to be in cDNA. The sequence we deduced from our cDNA analysis has been submitted to GenBank (accession no. AF245435). Northern and 5' rapid amplification of cDNA ends analyses indicate that the lin-13 transcript is ~7 kb in length and appears to be SL1 trans-spliced at the 5' end (data not shown).

The predicted LIN-13 protein contains 2248 amino acids and includes recognizable amino acid sequence motifs (Fig 2). Using the sequence analysis program SMART (SCHULTZ et al. 1998 ) and inspection, we identified 24 apparent zinc fingers of the C2-H2 class [C-X(2-4)-C-X(3)-F-X(5)-L-X(2)-H-X(3)-H] and 1 apparent zinc finger of the C4 class (KLUG and SCHWABE 1995 ). All of these apparent zinc fingers appear to be conserved in C. briggsae lin-13 (based on a sequence available via the Washington University genome sequencing consortium; data not shown). Zinc-finger motifs have been implicated in DNA binding and in protein-protein interactions (BERG 1990 ; KLUG and SCHWABE 1995 ).

View larger version (146K): In this window In a new window Download PPT slide   Figure 2. Features of the LIN-13 predicted protein sequence. (A) LIN-13 predicted protein. All cysteines are indicated in bold. The 24 predicted zinc fingers of the C2-H2 class are boxed. Predicted zinc fingers boxed in solid lines are closer to the consensus sequence (KLUG and RHODES 1987 ; BERG 1990 ); the remainder are boxed in shaded lines (see Fig 2B). The single predicted zinc finger of the C4 class is underlined. The LXCXE motif is outlined by an oval. See text for further information. (B) C. elegans LIN-13 zinc fingers. The 24 predicted zinc fingers of the C2-H2 class are aligned, along with a consensus sequence. Amino acids that are diagnostic of the zinc-finger motif (KLUG and RHODES 1987 ; BERG 1990 ) are in boldface. An asterisk indicates the zinc fingers boxed in solid lines in Fig 2A. (C) LXCXE motif alignments. The LXCXE motifs from C. elegans LIN-13 (this study), C. briggsae LIN-13 (CbLIN-13; Washington University genome sequence consortium) and several mammalian and viral proteins that are known to interact with Rb via the LXCXE motif are shown (see JONES et al. 1990 ; FATTAEY et al. 1993 ; BUYSE et al. 1995 ; TAUNTON et al. 1996 ; TEVOSIAN et al. 1997 ). (D) Molecular alterations identified in lin-13 alleles. The LIN-13 protein is depicted schematically. Solid bars represent the C2-H2 zinc fingers and a shaded bar represents the C4 zinc finger. The position of the LXCXE motif is indicated by a triangle. lin-13(n387) is a C to G alteration (S524stop), and lin-13(n388) is a C to T alteration (R857stop).

By inspection of the LIN-13 sequence, we also identified an LXCXE motif. The LXCXE motif was identified in several Rb-binding proteins and has been shown to mediate the binding of such proteins to Rb (FATTAEY et al. 1993 ; LEE et al. 1998 ; MAGNAGHI-JAULIN et al. 1998 ). The sequence data available for the apparent C. briggsae lin-13 indicate that this LXCXE motif is conserved (Fig 2).

lin-13 mutations:
We expected that the two existing lin-13 alleles would be missense mutations, as both alleles are temperature sensitive, a property that is not usually associated with null alleles. Furthermore, FERGUSON and HORVITZ 1989 reported that lin-13(n387)/nDf16 larvae produced when lin-13(n387) heterozygous males were mated with nDf16 heterozygous hermaphrodites at 25° arrested at what appeared by size to be the L2 stage. This result suggested that the null phenotype of lin-13 is likely to be zygotic lethal.

However, when we identified the molecular lesions associated with lin-13(n387) and lin-13(n388), we found that both are early nonsense mutations. lin-13(n387) is a T-to-A (S524stop) change that disturbs the third zinc finger and would encode a predicted protein of 523 amino acids (see Fig 3). lin-13(n388) is a C-to-T (R857stop) change immediately after the fifth zinc finger and would encode a predicted protein of 856 amino acids (see Fig 3). Thus, both nonsense mutations are predicted to result in greatly truncated LIN-13 protein products that are likely to eliminate or strongly reduce lin-13 activity. Furthermore, lin-13 mRNAs containing such early premature stop codons are likely to be unstable in C. elegans (PULAK and ANDERSON 1993 ). In the next section, we describe results that are consistent with these lin-13 mutations being null alleles.

View larger version (85K): In this window In a new window Download PPT slide   Figure 3. Analysis of lin-13::lacZ and LIN-13::GFP. (A–C) lin-13::lacZ expression in hyp7 in L3 larvae. Larvae were fixed and stained with an anti-LacZ antibody (red) as well as DAPI to visualize nuclei. (A) Anti-LacZ; (B) DAPI; (C) merged anti-LacZ antibody (red) and DAPI. (D and E) Nuclear localization of LIN-13::GFP. (D) Living embryo. Fluorescence is nuclear. Note also the discrete fluorescent foci within the nuclei. (E) Lateral hypodermal seam cells in an L3 larva. Larvae were fixed and stained with an anti-GFP antibody to visualize LIN-13::GFP (red). Larvae were also stained with MH27, a monoclonal antibody that detects adherens junctions (green; PRIESS and HIRSH 1986 ).

The lin-13 null phenotype:
As noted above, one salient feature of lin-13 alleles is that they are temperature sensitive. Although temperature sensitivity is not usually associated with null alleles, temperature-sensitive null alleles have been identified for other genes (e.g., daf-4; ESTEVEZ et al. 1993 ) and have been interpreted as an indication that a process rather than a mutant gene product is temperature sensitive. To investigate whether the process that requires lin-13 is temperature sensitive, as opposed to whether truncated gene products encoded by lin-13(n387) and lin-13(n388) are temperature sensitive, we used RNA-mediated interference (RNAi; FIRE et al. 1998 ) to inactivate endogenous lin-13 activity.

N2 hermaphrodites were injected with lin-13 double-stranded RNA (dsRNA) and individual hermaphrodites were incubated on separate plates at 25° or at 15°. We consistently observed stronger phenotypes at 25° than at 15°. After injection of dsRNA into 82 N2 hermaphrodites at 25°, we observed four different kinds of broods: 13 broods were composed entirely of progeny that appeared to be arrested at the L2 stage (see below), 18 broods were composed entirely of sterile Muv hermaphrodites, 31 broods were composed of a mixture of sterile Muv, arrested larvae, and wild-type hermaphrodites, and 9 broods were composed of entirely wild-type hermaphrodites. After injection of dsRNA into 24 N2 hermaphrodites at 15°, we observed two different kinds of broods: 18 broods were composed entirely of progeny that were sterile and generally had a protruding vulva, and 6 broods were composed of a mixture of wild-type hermaphrodites and hermaphrodites with a protruding vulva.

With respect to VPC specification, the observation that when the progeny were raised at 25°, some were Muv, whereas when progeny were raised at 15°, none were Muv suggests that the process that requires lin-13 activity is temperature sensitive and is consistent with the interpretation that existing lin-13 alleles strongly reduce or eliminate lin-13 activity. Thus, with respect to VPC specification, lin-13 activity appears to be necessary only at higher temperatures. We do not know the precise nature of the cellular defects underlying the protruding vulva phenotype observed at 15°. As the hermaphrodites that have a protruding vulva are generally sterile, their phenotype is reminiscent of the phenotype caused by somatic gonad defects that cause abnormal vulval eversion (SEYDOUX et al. 1993 ). Preliminary observations suggest that sterile lin-13 hermaphrodites do have somatic gonadal abnormalities (A. MELÉNDEZ, unpublished observations; J. HUBBARD, personal communication).

As mentioned above, substantial larval arrest was observed among the progeny of dsRNA-injected hermaphrodites raised at 25°. This phenotype was unexpected, as larval arrest is not observed among the progeny of lin-13/+ parents. Larval arrest appears to occur in the L2 stage on the basis of body size and gonadal morphology. The apparent L2 arrest occurred either transiently (for several days) or "permanently" (for at least 9 days at 25°). The arrested L2 larvae show normal motility and are not rod-like and so do not resemble null mutations in Ras pathway components (e.g., BEITEL et al. 1990 ).

As RNAi reduces both maternal and zygotic gene activity (FIRE et al. 1998 ), we hypothesized that this L2 arrest phenotype reflected the concomitant depletion of both maternal and zygotic lin-13 activity. To test this hypothesis, we allowed lin-13(n387) homozygous hermaphrodites to segregate from lin-13/+ mothers at 15°. These hermaphrodites are fertile and non-Muv. If lin-13(n387) is a null allele, then these hermaphrodites should be unable to provide maternal lin-13(+) activity to their progeny. Indeed, we have found that when their progeny are raised at 25°, they arrest as L2 larvae. This observation, in conjunction with the sequence analysis and RNAi data, suggests that the null phenotype of lin-13 is temperature sensitive and that the phenotype resulting from the absence of both maternal and zygotic lin-13 activity is L2 arrest.

As described above, FERGUSON and HORVITZ 1989 observed that lin-13(n387)/nDf16 arrest as larvae at 25°. It is not clear how their observation relates to the null phenotype we have now described. The lin-13(n387)/nDf16 larvae were obtained from nDf16/+ mothers, so perhaps maternal lin-13 activity was somewhat depleted. In this context, it is perhaps worth noting that at 15°, lin-13(n387)/nDf16 hermaphrodites have a single vulval protrusion (FERGUSON and HORVITZ 1989 ), as observed in our RNAi experiments. As lin-13(n387) hermaphrodites segregating from heterozygous parents at 15° do not have a protrusion, the difference may be attributable to reduced maternal lin-13 activity in nDf16/+ parents. Alternatively, the lin-13(n387)/nDf16 larvae may arrest for a different reason: as nDf16 is relatively large, it is possible that the deficiency removes other genes whose dosage becomes critical in the absence of lin-13 activity.

The behavior of lin-13 null alleles is summarized in Table 2. The phenotypes observed range in severity from L2 arrest (when both maternal and zygotic activities are removed at 25°), to sterile Multivulva (when only zygotic activity is removed at 25°), to sterile non-Multivulva (when both maternal and zygotic activities are removed at 15°), to wild type (when only zygotic activity is removed at 15°).

lin-13::lacZ expression:
The inductive and lateral signals specify VPCs to adopt vulval fates during the L3 stage. To examine the cell types in which lin-13 is expressed at that time, we constructed a lin-13::lacZ transcriptional reporter gene in which lacZ was inserted in frame at a site engineered at the ATG codon. This reporter gene contains all lin-13 genomic rescuing sequences and was designed to be analogous to the smg-1-dependent lin-12::lacZ reporter described in WILKINSON and GREENWALD 1995 (see MATERIALS AND METHODS for details). L3 hermaphrodites showed consistent expression of LacZ in hyp7 nuclei (Fig 3). Consistent expression was also seen in body wall muscles, intestinal nuclei, distal tip cells, and many neurons (data not shown). In contrast, only occasional expression was observed in the VPCs. We do not know whether this occasional staining reflects artifactual expression of LacZ from the multicopy transgene or a very low level of normal expression in the VPCs.

LIN-13 subcellular localization:
We constructed LIN-13::GFP hybrid proteins in which GFP (CHALFIE et al. 1994 ) is fused in frame at either the predicted amino terminus or carboxy terminus (MATERIALS AND METHODS). Transgenes expressing either of these LIN-13::GFP hybrid proteins fully rescue the Muv and sterility defects of lin-13(n387) and appear to have identical expression patterns (data not shown). The LIN-13::GFP hybrid proteins localize to the nucleus (Fig 3D and Fig E), consistent with a role in transcriptional regulation. In the nucleus, there are also discrete bodies or speckles of fluorescence; other nuclear proteins have been found to localize in such patterns, but in most cases the functional significance of these structures is not clear (LAMOND and EARNSHAW 1998 ). Recently, FUKUSHIGE et al. 1999 have visualized binding of the GFP-tagged transcription factor ELT-2::GFP to an array carrying known ELT-2 binding sequences in living embryos as discrete foci of fluorescence that resemble the speckles we have observed, raising the intriguing possibility that they reflect binding of LIN-13::GFP to specific binding sites for LIN-13.

In living animals, we see significant GFP fluorescence in embryos (Fig 3D); we have not determined the cell types expressing GFP in embryos. In living larvae, however, expression is very low, and in particular we could not detect expression during the L3 stage. To improve the sensitivity of LIN-13::GFP detection, we fixed and stained hermaphrodites with antibodies to GFP. In such preparations, we see consistent LIN-13::GFP accumulation in the nuclei of hyp7 and in seam cells throughout development, including the L3 stage. We also see nuclear accumulation in other cell types; we did not characterize these in detail, but the expression pattern of LIN-13::GFP seems roughly equivalent to that of lin-13::lacZ. As with lin-13::lacZ, we do not see a reliable expression of LIN-13::GFP in the VPCs, although we occasionally detect GFP in one or two VPCs in fixed and antibody-stained hermaphrodites (data not shown).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Genetic properties of lin-13:
We have obtained evidence that the two existing temperature-sensitive lin-13 mutations are null alleles. Both lin-13 alleles contain premature stop codons that are predicted to result in severely truncated LIN-13 proteins. RNA-mediated interference causes a temperature-sensitive phenotype similar to that caused by mutations in lin-13. The phenotypes observed are variable and are influenced by maternal rescue. These phenotypes range in severity from L2 arrest (when both maternal and zygotic activities are removed at 25°), to sterile Multivulva (when only zygotic activity is removed at 25°), to sterile non-Multivulva (when both maternal and zygotic activities are removed at 15°), to wild-type/class B SynMuv (when only zygotic activity is removed at 15°). The apparent temperature-sensitive null phenotype suggests that the process or complex involving LIN-13 is temperature sensitive.

Extending an earlier observation of FERGUSON and HORVITZ 1989 , we have found that lin-13 has genetic properties consistent with function as a class B SynMuv. At 15°, lin-13 alleles consistently behave as class B SynMuv alleles; i.e., when segregating from a lin-13/+ mother, they display a Muv phenotype in combination with all class A mutations tested, but they did not display a Muv phenotype in combination with any class B mutation tested. The consistent behavior of lin-13 as a class B SynMuv gene implicates lin-13 in inhibitory signaling.

lin-13 appears to be absolutely required for inhibitory signaling. At 25° lin-13 homozygous hermaphrodites segregating from a lin-13/+ mother are Muv even in the presence of class A gene activity (FERGUSON and HORVITZ 1985 ). One plausible explanation for this behavior is that lin-13 might also possess class A activity. If so, then lin-13 might provide a physical or functional link between the class A and class B genes.

While we were revising this article, SOLARI and AHRINGER 2000 published another linkage between class A and class B SynMuv genes. Using RNA-mediated interference, they obtained evidence that five components of the chromatin remodeling complex NuRD (TONG et al. 1998 ; XUE et al. 1998 ) have both class A and class B SynMuv gene activity. Interestingly, these components include HDA-1 (histone deacetylase) and LIN-53 (RbAp48), which have been physically as well as functionally associated with Rb (BREHM et al. 1998 ; LUO et al. 1998 ; MAGNAGHI-JAULIN et al. 1998 ; see also below).

The nature of the lin-13 mutations and the lin-13 RNAi phenotype suggests that the putative class A activity, or other activity, of lin-13 is dispensable at 15°. In this DISCUSSION, we focus on the class B SynMuv activity of lin-13, as it is this activity that is apparent at all temperatures.

LIN-13 and Rb:
We have shown that LIN-13 is a nuclear protein that is predicted to contain 24 zinc fingers. These features are consistent with a role for LIN-13 in transcriptional regulation of target genes. The zinc fingers of LIN-13 might bind DNA directly or function in protein-protein interactions (BERG 1990 ; KLUG and SCHWABE 1995 ).

LIN-13 also contains an LXCXE motif. The LXCXE motif has been identified in several different Rb-binding proteins and furthermore has in numerous cases been shown to be necessary and sufficient for binding to the "pocket" of Rb and Rb-related proteins (FATTAEY et al. 1993; LEE et al. 1998 ; MAGNAGHI-JAULIN et al. 1998 ). For example, Rb has been found to interact via the pocket domain with the LXCXE motif of HDAC1 (MAGNAGHI-JAULIN et al. 1998 ), leading to repression of target genes. Rb has also been found to interact via the pocket domain with the LXCXE motif of various viral oncoproteins (see Fig 2C), leading to activation of target genes (FATTAEY et al. 1993 ; LEE et al. 1998 ). The presence of this motif in LIN-13, in the context of the class B SynMuv genetic activity of lin-13, suggests that LIN-13 may directly interact with LIN-35 Rb, the product of another class B SynMuv gene (LU and HORVITZ 1998 ).

Loss of the activity of any one class B SynMuv gene can result in a Multivulva phenotype if class A activity is also removed. These results suggest that class B SynMuv genes function within a single pathway, process, or complex. Available biochemical data for the class B SynMuv gene products LIN-35 Rb, LIN-53 Rbp48, and HDA-1 histone deacetylase support this interpretation, as they all appear to be constituents of a single complex (LU and HORVITZ 1998 ).

The presence of zinc fingers and its nuclear localization suggest that LIN-13 plays a role in gene regulation. In C. elegans both LIN-53 RbAp48 and HDA-1 HDAC lack LXCXE motifs. Although these proteins are able to interact with LIN-35 Rb in vitro (LU and HORVITZ 1998 ), it is possible that in vivo LIN-13 serves as a bridge among these proteins. Another possible function for LIN-13 is to serve as a transcription factor that, like E2F, is regulated by Rb binding. Biochemical characterization of LIN-13, including an examination of potential physical interactions with various components of the LIN-35 Rb complex, will be necessary to distinguish among these possibilities.

Recently, HSIEH et al. 1999 showed that several class B SynMuv genes, including lin-35 Rb, are required for the silencing of genes present in tandemly repeated transgenic arrays. HSIEH et al. 1999 showed that the expression of several repetitive transgenes is reduced in class B SynMuv mutant backgrounds. Although we have not yet tested the effect of eliminating lin-13 activity on the expression of the same arrays examined by HSIEH et al. 1999 , we note that we have never observed reduced expression of any of the various repetitive transgenes we have placed in a lin-13 background in the course of other studies (A. MELÉNDEZ and I. GREENWALD, unpublished observations). If further analysis demonstrates that lin-13 indeed behaves differently from other class B SynMuvs with respect to context-dependent gene silencing, we might speculate that LIN-13 serves to target the LIN-35 Rb/HDAC complex to specific target genes involved in inhibition of vulval fates.

The role of SynMuv genes in VPC specification:
A Multivulva phenotype can in principle result from activation of a pathway or process that promotes vulval fates or from the inhibition of a pathway or process that represses vulval fates. Genes that influence this process can in principle act in an external signaling cell or in the VPCs. There have been two views to date. One view is that SynMuv genes act in an external signaling cell, hyp7, to control expression of a factor that antagonizes the effects of the inductive signal or otherwise to promote nonvulval fates (HERMAN and HEDGECOCK 1990 ). The other view is that SynMuv genes act in the VPCs to repress transcription of genes that are involved in vulval differentation (LU and HORVITZ 1998 ).

The proposal that SynMuv genes act in hyp7 was based on genetic mosaic studies of a lin-15 mutation that concomitantly removes both class A and class B activities (HERMAN and HEDGECOCK 1990 ) and was supported by mosaic analysis of the class B SynMuv gene lin-37 (HEDGECOCK and HERMAN 1995 ). Our findings that lin-13::lacZ and LIN-13::GFP are consistently expressed in hyp7 and are generally undetectable in VPCs at the time of VPC specification are consistent with this view. If SynMuv genes act in hyp7 and not in the VPCs, they may promote the expression in hyp7 of an inhibitory factor that represses vulval fates. This hypothetical factor may activate an as-yet-unknown but independent signal transduction pathway in the VPCs, so that whether a VPC adopts a vulval fate or not depends on the relative strengths of the inductive and inhibitory signaling inputs. Alternatively, the inhibitory factor may be a direct antagonist of the LET-23 epidermal growth factor (EGF) receptor. Such antagonists have been identified in Drosophila: the secreted protein Argos and the transmembrane protein Kekkon 1 appear to inhibit EGF receptor signaling via direct binding mechanisms (FREEMAN et al. 1992 ; GHIGLIONE et al. 1999 ).

Alternatively, LU and HORVITZ 1998 have proposed that lin-35 Rb and lin-53 Rbp48 function in the VPCs in response to inhibitory signaling from hyp7 to antagonize transcription factors that are activated by Ras. They favored a model in which these genes function in the VPCs because an anti-LIN-35 antibody appeared to stain VPC nuclei but not hyp7 nuclei and because genetic mosaic analysis of the class B SynMuv gene lin-36 had been interpreted as indicating a focus in ABp, the ancestor of the VPCs (THOMAS and HORVITZ 1999 ). However, LIN-53::GFP is detected in hyp7 as well as in the VPCs (LU and HORVITZ 1998 ), and about half of the mosaic individuals lacking lin-36 activity in ABp were non-Muv (THOMAS and HORVITZ 1999 ), consistent with the possibility that lin-36 also has a complex focus for VPC specification. As it remains possible that some SynMuv genes act in hyp7 while others act in the VPCs, and as expression data can be misleading as to cellular focus, systematic genetic mosaic analyses of all the SynMuv genes, including lin-35 and lin-13, will be necessary before this issue can be resolved.

	ACKNOWLEDGMENTS

We thank Gautam Kao for performing initial rescue experiments with candidate lin-13 cosmids. We thank the Washington University Sequencing Center for communication of sequence data prior to publication. We also thank Paul Sternberg for discussion, Beirong Lu for help integrating lin-13::lacZ, Jessica Greenwood for help integrating lin-13::gfp, and Richard Ruiz and Ilya Temkin for expert technical assistance. We are grateful to all members of our lab for advice and discussion, and to Johnny Fares, Barth Grant, Oliver Hobert, Sophie Jarriault, Simon Tuck, and Chenhui Wen for insightful comments on the manuscript. A.M. is a Postdoctoral Associate and I.G. is an Investigator of the Howard Hughes Medical Institute.

Manuscript received December 8, 1999; Accepted for publication March 24, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AUSTIN, J. and J. KIMBLE, 1987  glp-1 is required in the germ line for the regulation of the decision between mitosis and meiosis in C. elegans.. Cell 51:589-599[Medline].

BARSTEAD, R. J. and R. H. WATERSTON, 1989  The basal component of the nematode dense-body is vinculin. J. Biol. Chem. 264:10177-10185[Abstract/Free Full Text].

BEITEL, G. J., S. G. CLARK, and H. R. HORVITZ, 1990  Caenorhabditis elegans ras gene let-60 acts as a switch in the pathway of vulval induction. Nature 348:503-509[Medline].

BERG, J. M., 1990  Zinc fingers and other metal-binding domains. J. Biol. Chem. 265:6513-6516[Free Full Text].

BETTINGER, J. C., K. LEE, and A. E. ROUGVIE, 1996  Stage-specific accumulation of the terminal differentiation factor LIN-29 during Caenorhabditis elegans development. Development 122:2517-2527[Abstract].

BREHM, A., E. A. MISKA, D. J. MCCANCE, J. L. REID, and A. J. BANNISTER et al., 1998  Retinoblastoma protein recruits histone deacetylase to repress transcription. Nature 391:597-601[Medline].

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

BUYSE, I. M., G. SHAO, and S. HUANG, 1995  The retinoblastoma protein binds to RIZ, a zinc-finger protein that shares an epitope with the adenovirus E1A protein. Proc. Natl. Acad. Sci. USA 92:4467-4471[Abstract/Free Full Text].

Genome sequence of the nematode C. elegans: a platform for investigating biology. (1998) Science 282:2012-2018[Abstract/Free Full Text].

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].

CLARK, S. G., X. LU, and H. R. HORVITZ, 1994  The Caenorhabditis elegans locus lin-15, a negative regulator of a tyrosine kinase signaling pathway, encodes two different proteins. Genetics 137:987-997[Abstract].

COSTA, M., M. WEIR, A. COULSON, J. SULSTON, and C. KENYON, 1988  Posterior pattern formation in C. elegans involves position-specific expression of a gene containing a homeobox. Cell 55:747-756[Medline].

EDGLEY, M. L., C. A. TURNER and D. L. RIDDLE, 1997 On-line C. elegans resources, pp. 1059–1062 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Plainview, NY.

ESTEVEZ, M., L. ATTISANO, J. L. WRANA, P. S. ALBERT, and J. MASSAGUE et al., 1993  The daf-4 gene encodes a bone morphogenetic protein receptor controlling C. elegans dauer larva development. Nature 365:644-649[Medline].

FATTAEY, A. R., E. HARLOW, and K. HELIN, 1993  Independent regions of adenovirus E1A are required for binding to and dissociation of E2F-protein complexes. Mol. Cell. Biol. 13:7267-7277[Abstract/Free Full Text].

FERGUSON, E. L. and H. R. HORVITZ, 1985  Identification and characterization of 22 genes that affect the vulval cell lineages of the nematode Caenorhabditis elegans.. Genetics 110:17-72[Abstract/Free Full Text].

FERGUSON, E. L. and H. R. HORVITZ, 1989  The Multivulva phenotype of certain Caenorhabditis elegans mutants results from defects in two functionally redundant pathways. Genetics 123:109-121[Abstract/Free Full Text].

FERGUSON, E. L., P. W. STERNBERG, and H. R. HORVITZ, 1987  A genetic pathway for the specification of the vulval cell lineages of Caenorhabditis elegans.. Nature 326:259-267[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].

FREEMAN, M., C. KLAMBT, C. S. GOODMAN, and G. M. RUBIN, 1992  The argos gene encodes a diffusible factor that regulates cell fate decisions in the Drosophila eye. Cell 69:963-975[Medline].

FUKUSHIGE, T., M. J. HENDZEL, D. P. BAZETT-JONES, and J. D. MCGHEE, 1999  Direct visualization of the elt-2 gut-specific GATA factor binding to a target promoter inside the living Caenorhabditis elegans embryo. Proc. Natl. Acad. Sci. USA 96:11883-11888[Abstract/Free Full Text].

GHIGLIONE, C., K. L. CARRAWAY, L. T. AMUNDADOTTIR, R. E. BOSWELL, and N. PERRIMON et al., 1999  The transmembrane molecule Kekkon 1 acts in a feedback loop to negatively regulate the activity of the Drosophila EGF receptor during oogenesis. Cell 96:847-856[Medline].

GREENWALD, I., 1997 Development of the vulva, pp. 519–542 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Plainview, NY.

GREENWALD, I. S. and H. R. HORVITZ, 1980  unc-93(e1500): a behavioral mutant of Caenorhabditis elegans that defines a gene with a wild-type null phenotype. Genetics 96:147-164[Abstract/Free Full Text].

HEDGECOCK, E. M. and R. K. HERMAN, 1995  The ncl-1 gene and genetic mosaics of Caenorhabditis elegans.. Genetics 141:989-1006[Abstract].

HERMAN, R. K. and E. M. HEDGECOCK, 1990  Limitation of the size of the vulval primordium of Caenorhabditis elegans by lin-15 expression in the surrounding hypodermis. Nature 348:169-171[Medline].

HODGKIN, J., A. PAPP, R. PULAK, V. AMBROS, and P. ANDERSON, 1989  A new kind of informational suppression in the nematode Caenorhabditis elegans.. Genetics 123:301-313[Abstract/Free Full Text].

HORVITZ, H. R. and J. E. SULSTON, 1980  Isolation and characterization of cell-lineage mutants of the nematode Caenorhabditis elegans.. Genetics 96:435-454[Abstract/Free Full Text].

HSIEH, J., J. LIU, S. A. KOSTAS, C. CHANG, and P. W. STERNBERG et al., 1999  The RING finger/B-box factor TAM-1 and a retinoblastoma-like protein LIN-35 modulate context-dependent gene silencing in Caenorhabditis elegans.. Genes Dev. 13:2958-2970[Abstract/Free Full Text].

HUANG, L. S., P. TZOU, and P. W. STERNBERG, 1994  The lin-15 locus encodes two negative regulators of Caenorhabditis elegans vulval development. Mol. Biol. Cell 5:395-412[Abstract].

JONES, R. E., R. J. WEGRZYN, D. R. PATRICK, N. L. BALISHIN, and G. A. VUOCOLO et al., 1990  Identification of HPV-16 E7 peptides that are potent antagonists of E7 binding to the retinoblastoma suppressor protein. J. Biol. Chem. 265:12782-12785[Abstract/Free Full Text].

KLUG, A. and D. RHODES, 1987  "Zinc fingers": a novel protein motif for nucleic acid recognition. Trends Biochem. 12:464-469.

KLUG, A. and J. W. SCHWABE, 1995  Protein motifs 5: zinc fingers. FASEB J. 9:597-604[Abstract].

KRAUSE, M. and D. HIRSH, 1987  A trans-spliced leader sequence on actin mRNA in C. elegans.. Cell 49:753-761[Medline].

LAMOND, A. I. and W. C. EARNSHAW, 1998  Structure and function in the nucleus. Science 280:547-553[Abstract/Free Full Text].

LEE, J.-O., A. A. RUSSON, and N. P. PAVLETICH, 1998  Structure of the retinoblastoma tumour-suppressor pocket domain bound to a peptide from HPV E7. Nature 391:859-865[Medline].

LU, X. and H. R. HORVITZ, 1998  lin-35 and lin-53, two genes that antagonize a C. elegans Ras pathway, encode proteins similar to Rb and its binding protein RbAp48. Cell 95:981-991[Medline].

LUO, R. X., A. A. POSTIGO, and D. C. DEAN, 1998  Rb interacts with histone deacetylase to repress transcription. Cell 92:463-473[Medline].

MAGNAGHI-JAULIN, L., R. GROISMAN, I. NAGUIBNEVA, P. ROBIN, and S. LORAIN et al., 1998  Retinoblastoma protein represses transcription by recruiting a histone deacetylase. Nature 391:601-605[Medline].

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].

MILLER, D. M. and C. J. NIEMEYER, 1995  Expression of the UNC-4 homeoprotein in Caenorhabditis elegans motor neurons specifies presynaptic input. Development 121:2877-2886[Abstract].

PRIESS, J. and D. HIRSH, 1986  C. elegans morphogenesis: the role of the cytoskeleton in elongation of the embryo. Dev. Biol. 117:156-173[Medline].

PULAK, R. and P. ANDERSON, 1993  mRNA surveillance by the Caenorhabditis elegans smg genes. Genes Dev. 7:1885-1897[Abstract/Free Full Text].

SAMBROOK, J., E. F. FRITCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual, Ed. 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SCHULTZ, J., F. MILPETZ, P. BORK, and C. P. PONTING, 1998  SMART, a simple modular architecture research tool: identification of signalling domains. Proc. Natl. Acad. Sci. USA 95:5857-5864[Abstract/Free Full Text].

SEYDOUX, G., C. SAVAGE, and I. GREENWALD, 1993  Isolation and characterization of mutations causing abnormal eversion of the vulva in C. elegans.. Dev. Biol. 157:423-436[Medline].

SOLARI, F. and J. AHRINGER, 2000  NURD-complex genes antagonise Ras-induced vulval development in Caenorhabditis elegans.. Curr. Biol. 10:223-226[Medline].

TAUNTON, J., C. A. HASSIG, and S. L. SCHREIBER, 1996  A mammalian histone deacetylase related to the yeast transcriptional regulator Rpd3p. Science 272:371-372[Medline].

TEVOSIAN, S. G., H. H. SHIH, K. G. MENDELSON, K. A. SHEPPARD, and K. E. PAULSON et al., 1997  HBP1: a HMG box transcriptional repressor that is targeted by the retinoblastoma family. Genes Dev. 11:383-396[Abstract/Free Full Text].

THOMAS, J. H. and H. R. HORVITZ, 1999  The C. elegans gene lin-36 acts cell autonomously in the lin-35 Rb pathway. Development 126:3449-3456[Abstract].

TONG, J. K., C. A. HASSIG, G. R. SCHNITZLER, R. E. KINGSTON, and S. L. SCHREIBER, 1998  Chromatin deacetylation by an ATP-dependent nucleosome remodelling complex. Nature 395:917-921[Medline].

XUE, Y., J. WONG, G. T. MORENO, M. K. YOUNG, and J. COTE et al., 1998  NURD, a novel complex with both ATP-dependent chromatin-remodeling and histone deacetylase activities. Mol. Cell 2:851-861[Medline].

WATERSTON, R. H., J. E. SULSTON and A. R. COULSON, 1997 The genome, pp. 23–46 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Plainview, NY.

WILKINSON, H. A. and I. GREENWALD, 1995  Spatial and temporal patterns of lin-12 expression during C. elegans hermaphrodite development. Genetics 141:513-526[Abstract].

This article has been cited by other articles:

				M. M. Harrison, X. Lu, and H. R. Horvitz LIN-61, One of Two Caenorhabditis elegans Malignant-Brain-Tumor-Repeat-Containing Proteins, Acts With the DRM and NuRD-Like Protein Complexes in Vulval Development but Not in Certain Other Biological Processes Genetics, May 1, 2007; 176(1): 255 - 271. [Abstract] [Full Text] [PDF]

				M. M. Harrison, C. J. Ceol, X. Lu, and H. R. Horvitz Some C. elegans class B synthetic multivulva proteins encode a conserved LIN-35 Rb-containing complex distinct from a NuRD-like complex PNAS, November 7, 2006; 103(45): 16782 - 16787. [Abstract] [Full Text] [PDF]

E. C. Andersen, X. Lu, and H. R. Horvitz C. elegans ISWI and NURF301 antagonize an Rb-like pathway in the determination of multiple cell fates Development,