Worm Breeder's Gazette 7(2): 35
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
DNA sequences involved in chromosomal replication, segregation and structure can be identified by transformation of the yeast, Saccharomyces cerevisiae (1-4). Specific DNA fragments from a wide variety of eukaryotes including C. elegans, permit autonomous replication of hybrid plasmids in yeast cells (5). In addition, the termini of the linear extrachromosomal rDNA molecule from Tetrahymena are capable of functioning as chromosome ends when introduced into yeast by transformation (4). This conservation of signals responsible for chromosome behavior encouraged us to search for C. elegans DNA sequences that would bestow proper segregation behavior upon yeast hybrid plasmids. So far, we have identified and characterized two such segregator (SEG) functions from the worm genome. Segregator functions rescue aberrant segregation of autonomously replicating hybrid molecules in yeast cells. In spite of their high copy number and their ability to replicate during each cell division, ARS (autonomously replicating sequence) hybrid molecules are not propagated efficiently during mitosis. After 10 generations of growth in non-selective media, only 0.1 - 1.0% of the yeast cells retain the transformed phenotype. This mitotic instability can be alleviated by the addition of yeast centromeric sequences (2,3). To search for similar segregation functions in the C. elegans genome, we inserted random worm DNA fragments into the yeast vector, YRp17. YRp17 contains a putative yeast chromosomal origin of replication (ARS1) and a yeast selectable marker: the URA3 gene. We transformed a ura3 yeast strain with the collection of YRp17/worm hybrid plasmids and then enriched for mitotically stable Ura+ transformants. In this fashion, we isolated two hybrid plasmids that are stably propagated in yeast. Both of the original hybrid molecules contained one fragment bearing SEG function: SEG1, 7.1 kilobase pairs (kb) and SEG2, 3.2 kb. We then used the Ura+ phenotype of yeast cells carrying the YRp17-SEG hybrids to assess mitotic and meiotic behavior of the C. elegans SEGs. In comparison to yeast centromeres, worm SEGs behave aberrantly. After 10 generations of growth in non-selective media, 15 - 40% of the cells contain the SEG-bearing molecules while hybrid molecules with yeast centromeres are present in 40 - 90% of the cells. Diploids (ura3 /ura3 YRp17 hybrid/O) were induced to undergo meiosis and the four products of meiosis were analyzed after tetrad dissection. If the URA3+ hybrid molecule is present in one copy per cell, if it is then replicated during premeiotic S phase and subsequently segregated properly, it should be found in two of the four meiotic products. Yeast centromere plasmids segregate 2 Ura+:2 Ura- in 57% of the tetrads. SEG1 and SEG2 hybrids segregate 2+:2 in 19% and 43% of the tetrads, respectively. In contrast, YRp17 alone produces no 2+:2 tetrads. In 98% of the yeast centromere 2+:2- , the sister chromatids do not separate until the equational, second meiotic division. Only 65% and 35% of the SEG1 and SEG2 2+:2- tetrads show normal second division disjunction; the remainder separate precociously. SEG1 and SEG2 are both linked to sequences that are repeated in the C. elegans genome. SEG1 hybridizes weakly to 6 or 7 EcoRI fragments in addition to the homologous 7.1 kb fragment. SEG2 hybridizes to some 30 other bands. The two patterns of hybridization are different; likewise, SEG1 and SEG2 do not cross-hybridize. Neither SEG1 nor SEG2 hybridizes to yeast centromere sequences. If the repeated sequence is associated with SEG function (and we have yet to demonstrate a strict correlation) then SEG1 and SEG2 may be members of two different repeated classes of worm segregators. SEG1 and SEG2 represent putative C. elegans centromeres. They were found at reasonably high frequency: one every 500 - 2000 kb (a very rough estimate, replete with error-prone assumptions). Such a high frequency is consistent with the observation that C. elegans chromosomes are polycentric. However, other functions for segregators are imaginable (e.g. nuclear membrane or matrix binding). Furthermore, SEG1 and SEG2 may not behave as segregators in C. elegans. Microinjection will provide the means for assaying their segregation in worm cells. If the SEG sequences allow hybrid molecules to segregate properly, they may be ideal vectors for the DNA transformation of C. elegans.