The EGL-13 SOX Domain Transcription Factor Affects the Uterine Cell Lineages in Caenorhabditis elegans

Corresponding author: Anna P. Newman, Department of Biochemistry and Molecular Biology, MS 319B, One Baylor Plaza, Houston, TX 77030., anewman{at}bcm.tmc.edu (E-mail)

Communicating editor: B. J. MEYER

	ABSTRACT

TOP ABSTRACT The n483, ty3, ty7,... The uterine {pi} cell... Expression of an egl-13::GFP... Presumptive {pi} cells undergo... Genetic control of uterine... LITERATURE CITED

We isolated egl-13 mutants in which the cells of the Caenorhabditis elegans uterus initially appeared to develop normally but then underwent an extra round of cell division. The data suggest that egl-13 is required for maintenance of the cell fate.

DURING metazoan development, cells respond to inductive signals by acquiring particular fates and differentiating. One broadly used signaling pathway, defined by the LIN-12/Notch receptors (KIMBLE and SIMPSON 1997 ; GREENWALD 1998 ; KADESCH 2000 ), involves receptor cleavage. The resulting cytoplasmic moiety translocates to the nucleus and forms a complex with proteins including CSL [an acronym for mammalian CBF-1, Drosophila Su(H), and Caenorhabditis elegans LAG-1], leading to transcription of downstream genes.

The C. elegans uterine-vulval connection is organized by the uterine anchor cell (AC; KIMBLE 1981 ; NEWMAN and STERNBERG 1996 ). The fates of two cells (Z1.ppp and Z4.aaa) are determined when they signal each other during the AC vs. ventral uterine (VU) precursor decision (KIMBLE 1981 ; SEYDOUX and GREENWALD 1989 ). The sisters of these two cells (Z1.ppa and Z4.aap) always become VU cells (KIMBLE and HIRSH 1979 ). The VU cell fate, specified by lin-12 activity (GREENWALD et al. 1983 ), is to divide twice to produce four VU intermediate precursor cells. Subsequently, the AC induces the three adjacent VU intermediate precursor cells on each side—via LIN-12—to adopt the (rather than ) cell fate (Fig 1; NEWMAN et al. 1995 ). cells generate the uterine cell types [utse (uterine-seam cell) and uv1 cell] that connect to the vulva (NEWMAN et al. 1996 ).

View larger version (29K): In this window In a new window Download PPT slide   Figure 1. Schematic of the developing and mature uterine-vulval connection. (A) VU intermediate precursor cell stage in a late L3 hermaphrodite. There are six VU intermediate precursor cells per side, surrounding a centrally placed AC (light gray). The intermediate precursor cells closest to the AC have the cell fate, whereas those that are more distal have the cell fate. Vulval cells of the ventral epithelium are represented by dark gray. (B) One round of cell division later. Each cell has divided along a d/v axis to produce a larger dorsal daughter and a smaller ventral one. The outer ventral daughters become uv1 cells, while all the remaining progeny fuse with each other and with the AC to form the utse syncytium. The cell daughters (white) are each the result of an a/p or l/r division. The cell daughters will undergo an additional round of cell division, but the cell daughters will not. (C) Mature uterine-vulval connection. Schematic of a transverse section through the central region of the hermaphrodite containing the vulva. The central portion of the utse is visible and forms a thin laminar process dorsal to the vulva. The uv1 cells connect to the vulva and to the utse. Adapted from NEWMAN et al. 2000 , with permission from Elsevier Science.

The AC also induces underlying vulval precursor cells (VPCs) to adopt vulval (1° and 2°) cell fates (STERNBERG and HORVITZ 1986 ). In mature hermaphrodites, the 1° cell descendants connect to the utse and uv1 cells (Fig 1; NEWMAN et al. 1996 ; SHARMA-KISHORE et al. 1999 ). The utse is H-shaped with extensions that connect the uterus to the lateral epidermis (seam) and a thin central process that separates the uterine and vulval lumens. This multinucleate syncytium is formed by fusion of 8 of the 12 daughters and the AC.

The LIN-11 LIM domain and COG-2 SOX domain transcription factors are expressed in cells and their daughters (FREYD et al. 1990 ; HANNA-ROSE and HAN 1999 ; NEWMAN et al. 1999 ). The lin-11 gene is also expressed in the vulva, but cog-2 is not. The cell lineages are similar to wild type in lin-11 mutants. By contrast, the differentiation of the utse is abnormal in lin-11 mutants, the AC fails to fuse with the utse, and a functional uterine-vulval connection is not made (NEWMAN et al. 1999 ). In cog-2 mutants, the AC also fails to fuse with the utse. cells were found to divide along the correct axis in cog-2 mutants, and it was hypothesized that the cog-2 gene might function specifically in fusion of the AC with the utse (HANNA-ROSE and HAN 1999 ).

To identify mutants with defects in C. elegans uterine development, we performed an EMS mutagenesis of the N2 strain (BRENNER 1974 ). F₂ progeny with an egg-laying defective (Egl) phenotype were picked and their uterine anatomies were observed under Nomarski optics. cells differ from those of the alternative fate cells in that (1) cells divide along a dorsoventral (rather than longitudinal or transverse) axis; (2) cells produce two daughters rather than four; and (3) cell daughters connect to the vulva (NEWMAN et al. 1995 , NEWMAN et al. 1996 ). While continuous analysis of cell lineages is necessary to unambiguously determine whether a cell fate transformation has occurred, we used anatomical observation to identify seven mutants that appeared to affect cell development on the basis of one or more of the above characteristics. From 20,000 haploid genomes screened, we obtained one allele each of the presenilin gene sel-12 and the heterochronic gene lin-29 (NEWMAN et al. 2000 ; CINAR et al. 2001 ). Four mutants defined a distinct complementation group, comprising the ty3, ty7, ty8, and ty14 alleles. Three-factor crosses demonstrated that ty3 was located between 2.15 and 6.67 map units to the left of center on the X chromosome. This map position was roughly similar to that of the egl-13 gene identified in a previous screen for Egl mutants (TRENT et al. 1983 ). We therefore constructed the egl-13(n483)/ty3 trans-heterozygote. We found that hermaphrodites of this genotype were Egl, indicating that the alleles failed to complement and that ty3, ty7, ty8, and ty14 were alleles of the egl-13 gene.

	The n483, ty3, ty7, and ty8 alleles alter conserved residues of the EGL-13 SOX domain transcription factor

TOP ABSTRACT The n483, ty3, ty7,... The uterine {pi} cell... Expression of an egl-13::GFP... Presumptive {pi} cells undergo... Genetic control of uterine... LITERATURE CITED

Another genetic screen had identified the cog-2 gene, which encodes a SOX domain transcription factor (HANNA-ROSE and HAN 1999 ). egl-13 and cog-2 mapped to roughly similar positions. Furthermore, cog-2 mutants were Egl and had defects in the uterine-vulval connection similar to those observed in the ty3, ty7, ty8, and ty14 mutants. We therefore performed complementation analysis between cog-2 (ku194) and egl-13 (n483) and between cog-2 (ku194) and egl-13 (ty3). In both cases, the mutations failed to complement. The cog-2 gene is thus allelic to egl-13, which encodes a SOX domain transcription factor.

SOX domain transcription factors, which function broadly in development, have a conserved 79-amino-acid DNA-binding domain known as the SOX box (PEVNY and LOVELL-BADGE 1997 ; BOWLES et al. 2000 ). DNA sequence analysis of the n483 allele revealed a predicted protein with leucine substituted for a conserved proline at position 68 of the SOX box (Fig 2). The ty3 allele is a T to A transversion that converts a tyrosine within the SOX box to a stop codon. The ty7 allele leads to substitution of Lys for Arg at position 5 of the SOX box, which is conserved among SOX proteins and has been shown in human SRY to make a salt bridge to a phosphate of the DNA backbone (WERNER et al. 1995 ). While an Arg to Lys substitution should not affect the electrostatic interaction, the altered size of the side chain may affect protein conformation and function. Finally, the ty8 allele deletes approximately the N-terminal half of the SOX box (which is encoded by exons 8 and 9) and introduces a frameshift thereafter (egl-13 contains 10 exons). Specifically, nucleotides 4327–4623 of the wild-type sequence (starting with the first ATG) are deleted (corresponding to the region from the end of exon 7 through all but the last three nucleotides of exon 8) except for the tetranucleotide sequence TATT. Since the wild-type sequence from nucleotides 4327–4623 contains four copies of TATT, the ty8 allele could result from two small deletions separated by four nucleotides or from a single deletion with a TATT insertion at the junction. We did not find a mutation in the ty14 allele.

View larger version (12K): In this window In a new window Download PPT slide   Figure 2. Schematic of the EGL-13 protein. The 79 amino acids of the protein's SOX box (HANNA-ROSE and HAN 1999 ) are indicated below, as are the amino acid and nucleotide changes of the alleles sequenced in this report. Seven primer pairs were used to PCR amplify the entire egl-13 genomic region from 130 bp upstream of the predicted start site through the end of the coding sequence (with the exception of the intron between exons 1 and 2) and the PCR products were sequenced. Each sequence was determined from at least two sequencing reactions. Not shown is the ty8 allele, which deletes approximately the first half of the SOX box and introduces a frameshift thereafter (see text). The SOX box is defined as in BOWLES et al. 2000 .

It was previously argued that ku194, which truncates the protein prior to the SOX box, was likely to be a molecular null (HANNA-ROSE and HAN 1999 ). On the basis of the above analysis, the ty3 and ty8 alleles are also likely to be null.

	The uterine cell lineages are abnormal in egl-13 mutants

TOP ABSTRACT The n483, ty3, ty7,... The uterine {pi} cell... Expression of an egl-13::GFP... Presumptive {pi} cells undergo... Genetic control of uterine... LITERATURE CITED

As discussed above, cells differ from cells in dividing along a dorsoventral axis, undergoing one less round of cell division, and expressing egl-13 and lin-11. An earlier study found that, in egl-13 mutants, presumptive cells were essentially normal in division axis and expression of egl-13 and lin-11 (HANNA-ROSE and HAN 1999 ). Similarly, we did not observe significant alterations in the axis of cell division in the n483, ty3, or ty8 mutants (data not shown). By contrast, there was a striking change in number of divisions. Specifically, we found that many presumptive daughters divided in each of these three alleles (Table 1). In the previous analysis of ku194, the presumptive cells had not been observed for an additional round of cell division to determine whether their daughters divided. When we observed the uterine cell lineages of ku194 and ku241 mutant animals, we found that most cell daughters divided (Table 1). Thus, in all five egl-13 alleles whose cell lineages were observed, presumptive cells initially divided along a dorsoventral axis as in the wild type, but then often underwent an additional round of cell division. We conclude that, in egl-13 mutants, the cell lineage is initiated but not completed correctly.

View this table: In this window In a new window   Table 1. Divisions of presumptive cell daughters

	Expression of an egl-13::GFP reporter construct is abnormal in egl-13 mutants

TOP ABSTRACT The n483, ty3, ty7,... The uterine {pi} cell... Expression of an egl-13::GFP... Presumptive {pi} cells undergo... Genetic control of uterine... LITERATURE CITED

As discussed above, an egl-13::GFP transcriptional fusion is expressed in the uterine cells and their daughters. We created additional strains with the egl-13::GFP construct (pWH17) integrated into the C. elegans genome and utilized tyIs4 (see Table 2 footnote d) as well as the previously characterized kuIs29 to conduct genetic analyses of mutants involved in cell development. It was previously observed that the kuIs29 expression pattern was the same in egl-13 (ku194) and egl-13 (+) backgrounds during the early L4 stage (HANNA-ROSE and HAN 1999 ). Similarly, we found that, during the late L3 through early L4 stages, the number of uterine cells that fluoresced as a result of either the kuIs29 or the tyIs4 transgenic array was essentially the same in an egl-13 mutant or wild-type background (data not shown). This spans the time from when cell fates are being specified among the VU intermediate precursor population to just prior to the final round of uterine division (when cell daughters normally divide and maintenance-defective cell daughters would abnormally divide). However, it was also noted that egl-13 mutant animals sometimes contained extra cells with weak fluorescence during the late L4 stage (HANNA-ROSE and HAN 1999 ). To investigate these issues further, we performed close observations of uterine cell green fluorescent protein (GFP) expression during the early-mid- and mid-L4 stages in wild-type and egl-13 mutant animals.

When we observed kuIs29 and tyIs4 control lines during the early-mid-L4 stage, we saw an average of six to seven fluorescent uterine nuclei per side (Table 2; Fig 3). This is about the range expected for the daughters of three cells on each side plus or minus the AC nucleus (which may become fluorescent at this early stage and has an equal chance of moving to the left or right in the process of fusing with the utse). An average close to 6.5 GFP+ cells per side persisted through the mid-L4 (Table 2) and late L4 stages, with the uv1 nuclei often showing brighter fluorescence than the utse nuclei as previously reported (HANNA-ROSE and HAN 1999 ). By contrast, our observations of the egl-13 (ku194) and egl-13 (ty3) alleles revealed the presence of roughly two additional GFP+ uterine nuclei at the early-mid-L4 stage and four additional GFP+ nuclei at the mid-L4 stage (Table 2; Fig 3). Also, significantly fainter yet detectable nuclei were noted in the uterine-vulval and surrounding uterine regions in egl-13 mutants. We scored these animals at isolated time points, not by continuous observation, and thus cannot definitively conclude that each fluorescent uterine nucleus represents the descendant of a presumptive cell daughter that divided. However, the presence of additional GFP-positive nuclei is consistent with the cell lineage data presented above, which demonstrates that presumptive cells often undergo an extra round of cell division in egl-13 mutants.

View larger version (74K): In this window In a new window Download PPT slide   Figure 3. egl-13::GFP expression in early-mid-L4 stage animals. (A) Left lateral view of egl-13 (+) animal containing the tyIs4 transgenic array. Six cell daughter nuclei are visible. (B) Right lateral view of egl-13 (ty3) mutant containing tyIs4. Nine fluorescent presumptive lineage nuclei are evident. Brackets demarcate the region containing fluorescent uterine nuclei. Bright fluorescent cells outside brackets are body wall muscles (which also express the gene fusion; HANNA-ROSE and HAN 1999 ). Animal in B is slightly older than animal in A. Bars, 10 µm.

It was previously observed that in egl-13 (+) animals containing the kuIs29 array, the AC becomes GFP+ upon fusion with the utse (HANNA-ROSE and HAN 1999 ). Here, we report that the unfused and often very prominent AC nucleus in egl-13 mutants consistently fails to express egl-13::GFP. This is presumably an additional consequence of the cell defect revealed by the altered cell lineage.

	Presumptive cells undergo an extra round of division in egl-13; lin-12(d) double mutants

TOP ABSTRACT The n483, ty3, ty7,... The uterine {pi} cell... Expression of an egl-13::GFP... Presumptive {pi} cells undergo... Genetic control of uterine... LITERATURE CITED

In lin-12 gain-of-function mutants [lin-12(d)], excess cell fates are specified, and most VU intermediate precursor cells divide only once (NEWMAN et al. 1995 ). (Only the most distal cells sometimes undergo a second round of division.) Since egl-13 and lin-12(d) mutants have opposite phenotypes with respect to cell division, we constructed a mutant of genotype dpy-19 lin-12 (n137); lon-2 egl-13 (ty3). We found that of 39 cell daughters observed, 38 divided (Table 1). Thus the cell lineages of the double mutant are similar to those of egl-13 alone. This is consistent with the egl-13 gene's functioning downstream of lin-12, which is also suggested by the fact that lin-12 mutants have an earlier cell developmental defect than do egl-13 mutants.

	Genetic control of uterine cell development

TOP ABSTRACT The n483, ty3, ty7,... The uterine {pi} cell... Expression of an egl-13::GFP... Presumptive {pi} cells undergo... Genetic control of uterine... LITERATURE CITED

We have previously shown that lin-12 is required for specification of the uterine cell fate, while the lin-11 LIM domain transcription factor is required for differentiation of cell daughters (NEWMAN et al. 1995 , NEWMAN et al. 1999 ). In lin-11 mutants, uterine cell lineages are essentially wild type, although the daughters occasionally divide. By contrast, differentiation of the utse is defective; this includes a failure to fuse with the AC. lin-11 appears to be a direct target of LIN-12 signaling in the cells (GUPTA and STERNBERG 2002 ).

We have shown here that egl-13 mutants disrupt cell development at an earlier stage than do lin-11 mutants. Thus, while lin-11 mutants have only a slight effect on the cell lineages, egl-13 mutants have a striking cell lineage defect in which most of the presumptive cell daughters divide. In addition, the utse does not fuse with the AC in egl-13 mutants. Thus, egl-13 appears required for the maintenance of the cell fate and subsequent differentiation of its daughters. egl-13 and lin-11 appear not to be required for each other's expression (HANNA-ROSE and HAN 1999 ; NEWMAN et al. 1999 ). Further analysis will help to elucidate precisely how these two transcription factors collaborate to mediate proper development of cells in the C. elegans uterus.

	FOOTNOTES

¹ Present address: Sinsheimer Laboratories, Department of MCD Biology, University of California, Santa Cruz, CA 95064.

	ACKNOWLEDGMENTS

We thank the Caenorhabditis Genetics Center, Wendy Hanna-Rose, Bob Horvitz, and Paul Sternberg for strains. Thanks go to Zheng Zhou and Xiaomeng Yu for their comments on the manuscript. This work was supported by grants from the William Stamps Farish Fund and the National Institutes of Health (NIH; GM-57462) to A.P.N. This publication was also made possible by grant no. T32 ES07332 from the NIH National Institute of Environmental Health Sciences to K.S.O.

Manuscript received December 30, 2002; Accepted for publication August 4, 2003.

	LITERATURE CITED

TOP ABSTRACT The n483, ty3, ty7,... The uterine {pi} cell... Expression of an egl-13::GFP... Presumptive {pi} cells undergo... Genetic control of uterine... LITERATURE CITED

BOWLES, J., G. SCHEPERS, and P. KOOPMAN, 2000  Phylogeny of the SOX family of developmental transcription factors based on sequence and structural indicators. Dev. Biol. 227:239-255.[Medline]

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

CINAR, H. N., K. L. SWEET, K. E. HOSEMANN, K. EARLEY, and A. P. NEWMAN, 2001  The SEL-12 presenilin mediates induction of the Caenorhabditis elegans uterine cell fate. Dev. Biol. 237:173-182.[Medline]

FREYD, G., S. K. KIM, and H. R. HORVITZ, 1990  Novel cysteine-rich motif and homeodomain in the product of the Caenorhabditis elegans cell lineage gene lin-11.. Nature 344:876-879.[Medline]

GREENWALD, I., 1998  LIN-12/Notch signaling: lessons from worms and flies. Genes Dev. 12:1751-1762.[Free Full Text]

GREENWALD, I. S., P. W. STERNBERG, and H. R. HORVITZ, 1983  The lin-12 locus specifies cell fates in Caenorhabditis elegans.. Cell 34:435-444.[Medline]

GUPTA, B. P. and P. W. STERNBERG, 2002  Tissue-specific regulation of the LIM homeobox gene lin-11 during development of the Caenorhabditis elegans egg-laying system. Dev. Biol. 247:102-115.[Medline]

HANNA-ROSE, W. and M. HAN, 1999  COG-2, a Sox domain protein necessary for establishing a functional vulval-uterine connection in Caenorhabditis elegans.. Development 126:169-179.[Abstract]

KADESCH, T., 2000  Notch signaling: a dance of proteins changing partners. Exp. Cell Res. 260:1-8.[Medline]

KIMBLE, J., 1981  Lineage alterations after ablation of cells in the somatic gonad of Caenorhabditis elegans.. Dev. Biol. 87:286-300.[Medline]

KIMBLE, J. and D. HIRSH, 1979  Post-embryonic cell lineages of the hermaphrodite and male gonads in Caenorhabditis elegans.. Dev. Biol. 70:396-417.[Medline]

KIMBLE, J. and P. SIMPSON, 1997  The LIN-12/Notch signaling pathway and its regulation. Annu. Rev. Cell Dev. Biol. 13:333-361.[Medline]

NEWMAN, A. P. and P. W. STERNBERG, 1996  Coordinated morphogenesis of epithelia during development of the Caenorhabditis elegans uterine-vulval connection. Proc. Natl. Acad. Sci. USA 93:9329-9333.[Abstract/Free Full Text]

NEWMAN, A. P., J. G. WHITE, and P. W. STERNBERG, 1995  The C. elegans lin-12 gene mediates induction of ventral uterine specialization by the anchor cell. Development 121:263-271.[Abstract]

NEWMAN, A. P., J. G. WHITE, and P. W. STERNBERG, 1996  Morphogenesis of the C. elegans hermaphrodite uterus. Development 122:3617-3626.[Abstract]

NEWMAN, A. P., G. Z. ACTON, E. HARTWIEG, H. R. HORVITZ, and P. W. STERNBERG, 1999  The lin-11 LIM domain transcription factor is necessary for morphogenesis of C. elegans uterine cells. Development 126:5319-5326.[Abstract]

NEWMAN, A. P., T. INOUE, M. WANG, and P. W. STERNBERG, 2000  The Caenorhabditis elegans heterochronic gene lin-29 coordinates the vulval-uterine-epidermal connections. Curr. Biol. 10:1479-1488.[Medline]

PEVNY, L. A. and R. LOVELL-BADGE, 1997  Sox genes find their feet. Curr. Opin. Genet. Dev. 7:338-344.[Medline]

SEYDOUX, G. and I. GREENWALD, 1989  Cell autonomy of lin-12 function in a cell fate decision in C. elegans.. Cell 57:1237-1245.[Medline]

SHARMA-KISHORE, R., J. G. WHITE, E. SOUTHGATE, and B. PODBILEWICZ, 1999  Formation of the vulva in Caenorhabditis elegans: a paradigm for organogenesis. Development 126:691-699.[Abstract]

STERNBERG, P. W. and H. R. HORVITZ, 1986  Pattern formation during vulval development in Caenorhabditis elegans.. Cell 44:761-772.[Medline]

TRENT, C., N. TSUNG, and H. R. HORVITZ, 1983  Egg-laying defective mutants of the nematode Caenorhabditis elegans.. Genetics 104:619-647.[Abstract/Free Full Text]

WERNER, M. H., J. R. HUTH, A. M. GRONENBORN, and G. M. CLORE, 1995  Molecular basis of human 46X, Y sex reversal revealed from the three-dimensional solution structure of the human SRY-DNA complex. Cell 81:705-714.[Medline]

This article has been cited by other articles:

				K. S. Oommen and A. P. Newman Co-regulation by Notch and Fos is required for cell fate specification of intermediate precursors during C. elegans uterine development Development, November 15, 2007; 134(22): 3999 - 4009. [Abstract] [Full Text] [PDF]

				H. Kagoshima, R. Nimmo, N. Saad, J. Tanaka, Y. Miwa, S. Mitani, Y. Kohara, and A. Woollard The C. elegans CBF{beta} homologue BRO-1 interacts with the Runx factor, RNT-1, to promote stem cell proliferation and self-renewal Development, November 1, 2007; 134(21): 3905 - 3915. [Abstract] [Full Text] [PDF]

				H. Sun, B. L. Nelms, S. F. Sleiman, H. M. Chamberlin, and W. Hanna-Rose Modulation of Caenorhabditis elegans Transcription Factor Activity by HIM-8 and the Related Zinc-Finger ZIM Proteins Genetics, October 1, 2007; 177(2): 1221 - 1226. [Abstract] [Full Text] [PDF]

				V. J.P. Robert, T. Sijen, J. van Wolfswinkel, and R. H.A. Plasterk Chromatin and RNAi factors protect the C. elegans germline against repetitive sequences Genes & Dev., April 1, 2005; 19(7): 782 - 787. [Abstract] [Full Text] [PDF]