Description

In C. elegans, the gpa-4 gene encodes a G-protein α-subunit, part of an evolutionary conserved Gα family expressed in muscle and neuronal cells (Jansen et al. 1999). Previous studies using fluorescent reporters have suggested that the gpa-4 promoter drives expression exclusively in the head chemosensory neurons ASIL and ASIR (Jansen et al. 1999). This led subsequent studies to consider the gpa-4 promoter as ASI-specific (see Table 1 for all studies using the gpa-4 promoter to drive gene expression).

Recently, red-to-green fluorescent reporter switches have been developed to validate promoter-driven transcriptional patterns and tissue specificity. These systems use either Cre/lox or FLP/FRT recombination. Their mode-of-operation, in principle, is identical and relies on two different transgenes. First, the recombinase is placed under transcriptional control of the promoter to be tested. Second, a ubiquitous promoter drives the expression of a red fluorophore, which is excised by the recombinase leading to the expression of a green fluorescent protein (Ruijtenberg and van den Heuvel 2015; Muñoz-Jiménez et al. 2017). As the red-to-green switch is irreversible, any cell or lineage in which the promoter has been active during development will be labeled green, while the rest of the animal will be red.

To examine the expression pattern of the gpa-4 promoter, we generated a gpa-4p::Cre transgene integrated on Chr X using MosSCI (Frøkjær-Jensen et al. 2012; Figure 1A). Recombination efficiency was first tested by PCR with primers encompassing the floxed mCherry cassette (Figure 1A,B). In the absence of recombination (mCherry::gfp cassette), a single product of 1.7 kb was amplified. When recombination occurs, two products will be visible, as most cells in the animal have the intact mCherry::gfp cassette while in the target tissue of the Cre driver construct, the mCherry cassette is deleted. This is indeed the case for both the positive control intestinal Cre driver (elt-2p::Cre) and the gpa-4p::Cre one, demonstrating that Cre is expressed and active when expressed under transcriptional control of the gpa-4 promoter. The intensity of the lower, recombined band was, however, surprisingly intense for an event which should only happen in the two ASI neurons per animal. Indeed, microscopic examination of animals showed clear and reproducible expression of GFP in many tissues, such as the somatic gonad, vulval precursor cells, seam cells, body wall muscle cells, tail and head neurons, as well as, the P lineage (Figure 1C). Both negative and positive controls showed the expected fluorescence pattern (no GFP and GFP expression only in the intestine, respectively; Figure 1D,E). Recombination by Cre driven by the gpa-4 promoter occurs therefore in many more cells than the published gpa-4 transcriptional fusion (Jansen et al. 1999) in which ASI-specific expression was reported.

One limitation of our approach could be that the insertion site of the gpa-4p::Cre influences its expression. However, there are two reasons arguing against this possibility. First, the same single copy insertion site was used for a variety of Cre drivers (muscle, intestine, mesoblast, vulva and XXX neuroendocrine cells; Ruijtenberg and van den Heuvel 2015; Gómez-Saldivar et al. 2020). Expression patterns for the different promoters did in each case correspond to the expected expression pattern. Second, single cell RNA-seq in L1 larvae (Cao et al. 2017) detected the endogenous gpa-4 transcript in cells identified as somatic gonad precursors, vulval precursors, sex myoblasts and ciliated sensory neurons (Extended data). These tissues overlap with the cell types in which we identified gpa-4 promoter activity and Cre expression. Taken together, we conclude that the previously thought ASI-specific transgenic gpa-4 promoter is widely active during embryonic and larval development.