Worm Breeder's Gazette 13(2): 42 (February 1, 1994)
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.
We have been trying for a long time to achieve proper expression of a lin-12 :1acZreporter gene. We appear to have transgenic lines that express in the cells predicted by phenotypic analysis and are currently analyzing the pattern. Here, we share some of our observations, which we hope will be helpful to others working on expression projects.
Design of lin-12 :lacZfusions and use of different coiniection markers
We initially made 4 lacZ fusion constructs (Fire, et al., Gene 93:189-198(1990)) with 1, 1.7, 2.1, and 3.4 kb of lin-12 5' sequence. These were coinjected into wild-type animals with the rol-6 ( su1006 )gene (Mello et al., EMBO J 10: 3959-3970(1991)) or into dpy-20 ( el282ts )animals with the dpy-20 (+)gene(Han and Sternberg, Genes and Dev. 5: 2188-2198(1991)). We found that for these constructs some dpy-20 ( e1282ts );Ex[ dpy-20 (+) lin-20 :1acZ] lines stained better than their rol-6 (d)counterparts. We also integrated a number of the extrachromosomal arrays that we generated. We found it much easier to identify integration events with the rol 6(d) arrays than with the dpy-20 (+)marked arrays. We think that this is because the dpy-20 rescue is incompletely penetrant and so all lines that carried integration events segregated a few Dpy animals. We did observe staining in the Dpy animals segregating from the putative integrated lines which suggested that the arrays were present even though the animals were not rescued.
When we examined animals carrying integrated and extrachromosomal arrays with these lin-12 :1acZfusions, we did not observe expression in any stages other than the embryo and early L1 .Since lin-12 is known to be required in different subsets of cells in the L2 , L3 and L4 larval stages, we attempted to provide more non coding sequences that might be required for proper expression. We coinjected genomic fragments carrying the first four introns and also made new lacZ constructs in which lin-12 intron sequences were added 3' of the poly-A addition site in the 5' lin-12 :1acZreporter genes. We also used lin-12 (+)as a coinjection marker and injected lin-12 loss of function mutants. None of these approaches resulted in later larval staining and, in fact, addition of the introns or coinjection with the full-length lin-12 (+)gene seemed to inhibit the embryonic staining pattern as well.
Use of smg-1 to stabilize expression of a lin- t2 .IacZ fusion gene
Based on these results and other data we had concerning the requirement of 3' introns for proper rescuing activity, we designed a lin-12 :1acZfusion that includes all of the sequences that are present in the full-length rescuing genomic clone in their "proper" order. This construct is a transcriptional fusion that transcribes the entire lin-12 message, but is not predicted to translate it (see figure). When this construct was coinjected with rol-6 ( su1 O06)we observed no staining in a wild-type background. However, when the same construct was injected into a smg-1 ( r861 )strain (courtesy of Andy Fire) we were able to obtain lines that express lacZ in the predicted subsets of cells at all stages of development This expression is dependent on the presence of the smg-1 mutation because when integrated arrays are crossed out of that background the staining goes away. We believe that this effect on expression is due to the ability of smg mutations to stabilize large unstable RNAs (Pulak and Anderson, Genes and Dev. 7:1885-1897(1993)).
Behavior of mutations on Chromosome III in a smg-1 background
In working with the lacZ constructs in different backgrounds we have discovered some interactions between the smg mutation and other mutations that we do not think have been previously reported. For example, smg-1 ( r861 ); unc-36 ( e251 )/+grows slowly (the animals are wild-type in movement but lay few eggs) and when we generated the strain smg-1 ( r861 ); unc-36 ( e251 )/ arls11 [ rol-6 (d) lin-12 :1acZ]theseanimals did not roll even though in combination with other mutations arIs11 behaves as a dominant roller. We also discovered an interaction of smg-1 with qCl(a crossover suppressor for LGIII which carries dpy-19 ( e1259 )and glp-1 ( q339 ),J. Austin and J. Kimble, personal communication): smg-1 ( r861 );qCI/+ hermaphrodites are nearly sterile (a few animals produce a few eggs, but most have a Glp dissecting scope phenotype), and the Glp phenotype of qCI homozygotes is suppressed since smg-1 ( r861 );qCl is fertile.