We have generated a novel null allele,
ot900, of the C. elegans LIM homeodomain gene
ceh-14. All existing deletion alleles of
ceh-14 were generated using a Tc1 transposon insertion, one of which,
ch3, is a putative null based on the resulting frame-shift and a reduction in detectable mRNA and protein levels (Cassata et al. 2000). However, in our hands we have noticed a propensity of the
ch3 allele to revert to wild-type after several generations, perhaps due to a more complex rearrangement of the locus caused by the Tc1 transposon than previously appreciated.
To avoid potential complexities of working with the
ch3 allele, we used CRISPR/Cas9 to generate a new null allele of
ceh-14. The
ot900 allele was isolated with gRNAs targeted to the first and last exons of
ceh-14 (gRNA #1: CTTGGCGAGTGCGATGAGC, gRNA #2: GTACTGTGGAGTCATGTGT; see Fig. 1), and then PCR screening for resulting large deletion alleles. Our primers for PCR genotyping of the deletion are 5-CTCAACTAAATCGTCAGAATCGTC-3 and 5-GAGATGTATGAAACGAGCGAGCG-3 which generate a 620bp product in the context of the
ot900 deletion. The
ceh-14 locus in the
ot900 allele is only capable of generating an 11aa peptide due to the remaining bases in the first exon, before reaching a premature stop codon (Fig 1B).
Previous studies have implicated
ceh-14 in differentiation and function of the phasmid neurons, but have shown incomplete penetrance and variability in phenotypes resulting from the
ceh-14 (
ch3) allele (Kagoshima et al. 2013, Serrano-Saiz et al. 2015). We used the
srg-13 marker of PHA identity (which was expressed in 20/20 wild-type animals) to assess whether phasmid neuron differentiation is more severely affected in
ot900 than
ch3, and found that this marker was indeed lost in 100% of animals (expressed in 0/15 animals), despite not finding an appreciable effect in
ceh-14 (
ch3) (expressed in 15/15 animals). This suggests that some of the variability in phenotype may be the result of the
ch3 allele itself, rather than a property of CEH-14 function.