lin-14 has a complex genomic DNA structure with 13 exons spanning over 20kb, and produces at least three different transcripts. The predicted proteins encoded by the three transcripts are all approximately 540aa but differ in their N-terminal regions. LIN-14 is a novel protein whose biochemical activity remains unknown. To investigate the structure and function of LIN-14, we tagged the LIN-14 carboxy terminus with GFP to make a LIN-14::GFP fusion protein, and the fusion protein was expressed under the control of
lin-14 native upstream genomic sequences (approximately 12kb). This LIN-14::GFP construct was found to fully rescue a
lin-14(null) mutant, indicating that the LIN-14::GFP fusion protein is fully functional. A series of deletions of the LIN-14::GFP were then constructed and expressed under the control of the hypodermal-specific
col-10 promoter. (The smaller and better-characterized
col-10 promoter was used instead of the
lin-14 promoter sequences for the convenience of making the constructs). We first asked what is the minimum segment of LIN-14 that is capable of rescuing
lin-14(
n179ts) at nonpermissive temperature. Judged by its rescuing activity in hypodermal cell lineages, an exon9-13 construct (containing
aa292-535 of LIN-14) appears to be the minimum rescuing contstruct that we have tested so far. An
aa284-466 construct, which contains an amino acid sequence well conserved between C. elegans, C. briggsae and C. vulgarus, can also rescue
n179 , though not as strongly as the exon9-13 construct. An exon10-13 construct does not rescue, arguing that exon9 is indispensable for executing LIN-14 function. In fact, the exon10-13 construct shows dominant negative activity, suggesting that this truncated LIN-14 polypeptide can interfere with endogenous
lin-14 activity, perhaps by interacting with targets, or with endogenous LIN-14 protein. The exon9-13 construct, expressed by the
col-10 promoter, can rescue a
lin-14(null) mutation in the hypodermis, suggesting that LIN-14 exons 1-8 are not strictly required for
lin-14 function. This result suggests that, at least in the hypodermis, the alternative LIN-14 products predicted to be generated by alternative utilization of 5' exons are also not essential. We also measured the efficiency of nuclear localization of the various truncated LIN-14::GFP proteins in transgenic worms. The results show that LIN-14 has an unconventional, extended NLS domain, rather than a more typical discrete NLS. The whole region from
aa284-466 is required for efficient nuclear localization and deletions from either end of this domain significantly reduce the efficiency of nuclear localization. Basic Arg/Lys clusters similar to a typical NLS consensus sequence can be found at both ends of this domain, but non of them is sufficient to bring LIN-14 to the nucleus. LIN-14 activity is not required for its nuclear localization since a mutant exon9-13 construct (containing the
n179 point mutation 303R->G*) is localized in nucleus equally well as the corresponding wild type exon9-13 construct, even though it fails to show any rescue activity. In the course of these experiments, we observed that deletion constructs that display
lin-14 activity (exon8-13, exon9-13,
aa284-466) show strong association with chromosomes during mitosis. Chromosomes show strong LIN-14::GFP fluorescence from metaphase to anaphase. As the chromosomes decondense after mitosis, LIN-14::GFP becomes strongly re-localized back to the reformed nucleus. In contrast, the mutant exon9-13 construct containing the
n179 mutation does not associate with metaphase or anaphase chromosomes (although it is nuclear localized). Other non-rescuing constructs (exon10-13 and exon11-13) also show no chromosomal association, suggesting the ability to associating with chromosomes is closely related to LIN-14 function. One model consistent with our data so far is that
lin-14 encodes a transcription factor with DNA binding activity in its more carboxy-terminal exons. *Thanks to Brenda Reinhart and Gary Ruvkun for sharing
n179 sequence data.