Comparative analysis of new mScarlet-based red fluorescent tags in Caenorhabditis elegans

Abstract

One problem that has hampered the use of red fluorescent proteins in the fast-developing nematode Caenorhabditis elegans has been the substantial time delay in maturation of several generations of red fluorophores. The recently described mScarlet-I3 protein has properties that may overcome this limitation. We compare here the brightness and onset of expression of CRISPR/Cas9 genome-engineered mScarlet, mScarlet3, mScarlet-I3, and GFP reporter knock-ins. Comparing the onset and brightness of expression of reporter alleles of C. elegans golg-4, encoding a broadly expressed Golgi resident protein, we found that the onset of detection of mScarlet-I3 in the embryo is several hours earlier than older versions of mScarlet and comparable to GFP. These findings were further supported by comparing mScarlet-I3 and GFP reporter alleles for pks-1, a gene expressed in the CAN neuron and cells of the alimentary system, as well as reporter alleles for the pan-neuronal, nuclear marker unc-75. Hence, the relative properties of mScarlet-I3 and GFP do not depend on cellular or subcellular context. In all cases, mScarlet-I3 reporters also show improved signal-to-noise ratio compared to GFP.

Keywords: Caenorhabditis elegans; RFP; fluorophore; marker.

Fig. 1.

Analogous CRISPR-mediated insertion of fluorophores at the N-terminus of the GOLG-4 locus. a) WormBase genome browser snapshot of the golg-4 locus and the site of CRIPSR fluorophore insert. The operon on which golg-4 is located is schematically indicated and includes a third gene located further downstream that is not shown here. Codon-optimized fluorophores are schematized below: derivatives of mScarlet—mScarlet3 and mScarlet-I3—show amino acid mutations relative to mScarlet in black. GFP cloned from pPD95.75 expression plasmid is also included for comparison. Introns are also annotated and were added as recommended by the MPI C. elegans codon adapter (Redemann et al. 2011). b) Representative expression of red fluorophore-tagged GOLG-4 in the head is shown. Average fluorescence intensity of the pharyngeal terminal bulb (outlined in dotted white circle) of individual animals is quantified and plotted on the right, and the median values are indicated. ***P < 0.001, ****P < 0.0001, two-tailed t test. c and d) Photobleaching of the red fluorophores were measured through a time course of repeated imaging of RFP-tagged GOLG-4 expression in the same region and using the same imaging parameters as in panel b. Worms were imaged once per second over 300 s, and the fluorescence intensity was plotted over time, relative to starting values (c), or as average fluorescence intensity of the region of interest (d). Means and standard deviations of at least 10 individuals per strain are shown.

Fig. 2.

Expression of fluorophore-tagged GOLG-4 through embryonic development. Representative images of single focal planew through the middle of each embryo are shown, comparing the expression dynamics between the mScarlet variants and GFP. Strains are those shown in Fig. 1. For comparison to GFP, imaging setting was chosen to match the fluorescence intensity of GFP::GOLG-4 to that of mScarlet-I3::GOLG-4 in the adult, quantified in Fig. 1. Eight embryo at developmental stages from the 4-cell stage to the 4-fold stage, spanning most of embryogenesis, are represented in columns and compared across fluorophores, in rows. Embryo stages were determined by DIC (not shown), and eggshells are traced in dashed lines. The detected expression of GFP::GOLG-4 and mScarlet-I3::GOLG-4 in 4 cell stage embryos suggests maternal contribution of either protein or transcript.

Fig. 3.

Expression of pks-1::SL2::GFP::H2B and pks-1::SL2::mScarlet-I3::H2B. a) WormBase genome browser snapshot of the psk-1 locus and the site of CRISPR-mediated insertion at the C-terminus, after the stop codon. Schematic of fluorophore constructs is shown below. b) Expression of fluorophores in the L3 larval head and midbody. Both GFP and mScarlet-I3 reporters are expressed in the nuclei of CANL/R and intestinal (int) cells. Lower expression is also observed in several nuclei within the pharynx. An additional nucleus in the head, presumed to be the head mesodermal cell (HMC), is also labeled. Images are of a single plane containing 1 of the 2 CAN nuclei in focus. c) Expression of fluorophores in various stages of the embryo. Focal plane containing 1 of the 2 nuclei is shown for each embryo. Fluorescent signal can be visualized from the comma stage onwards, for both GFP and mScarlet-I3 constructs (arrowheads). Note the pks-1 reporter constructs specifically label CAN nuclei at these embryonic stages shown.

Fig. 4.

Expression of fluorophore-tagged UNC-75, comparing GFP to mScarlet-I3. a) WormBase genome browser snapshot of the unc-75 locus, showing the site of CRISPR-mediated insertion at the N-terminus. Fluorophore constructs are the same as in previous figures. The fluorophores were inserted at the N-terminus to avoid disruption of a C-terminal nuclear localization sequence (Loria et al. 2003). b) Expression of fluorophore-tagged UNC-75 in the adult worm. Images are maximum intensity projections through the head, midbody, and tail sections, capturing dense regions of neuron nuclei. c) Expression of fluorophore-tagged UNC-75 at 5 stages in embryo development, determined by DIC. Single Z planes through the approximate midpoint of each embryo are shown, and eggshells are traced with dashed lines.