20 Years of unc-119 as a transgene marker

Abstract

This fall marks 20 years since the cloning of unc-119 was reported. Despite having a strong phenotype that makes animals somewhat difficult to grow and handle, unc-119 mutant rescue has become one of the most frequently-used markers for C. elegans transformation. In this Commentary, I describe the history of how unc-119 rescue traveled through the worm community, contributing to the development of transgene methods in C. elegans.

Abbreviations

Unc

uncoordinated

CGC

Caenorhabditis Genetics Center

MRC

Medical Research Council

YAC

Yeast Artificial Chromosome

GFP

Green Fluorescent Protein

HRG4

Human Retinal Gene 4

NIH

National Institutes of Health

MosTIC

Mos1 excision-induced transgene-instructed gene conversion

MosSCI

Mos1-mediated single-copy insertion

MosDEL

Mos1-mediated deletion

CRISPR

clustered regularly interspaced short palindromic repeats

Cas9

CRISPR-associated protein 9

sgRNA

single-guide ribonucleic acid

DSLR

Digital Single Lens Reflex

Introduction

Transgenes introduced into C. elegans can be followed by many methods.¹ A phenotypic way of distinguishing transformants from non-transformants makes the job much easier under the dissecting microscope. While there are many markers that work well, unc-119 rescue has clearly become widespread in the field: A search of the Caenorhabditis Genetics Center (CGC) database (http://www.cgc.cbs.umn.edu) finds some 2000 strains that contain at least one transgene rescued by unc-119; a search of the Materials section of published C. elegans papers using Textpresso (http://www.textpresso.org/celegans/) finds approximately 1000 that contain “unc-119”; and the plasmid deposition site Addgene (http://www.addgene.org) contains some 50 plasmids that refer to unc-119. As described below, the popularity of unc-119 rescue resulted from the connectivity of the C. elegans community, coupled with the suitability of this marker for a wide variety of applications.

Identification of unc-119

In the fall of 1991, my graduate research project was to work with a visible uncoordinated (Unc) mutant isolated by my thesis advisor, David Pilgrim (University of Alberta, Canada) when he was a postdoc with Jonathan Hodgkin (MRC Laboratory of Molecular Biology, Cambridge, UK). After performing crosses in which the transposable element Tc1 was free to mobilize, Pilgrim recovered a mutant, e2498, that defined a new locus, subsequently named unc-119. This was not the first time the gene had been mutated: Acetaldehyde mutagenesis had previously disrupted unc-119 via a chromosomal break that created 2 independently segregating pieces of LG III, the free duplication eDp6, and the deficiency chromosome, eDf2.² The eDf2; eDp6 strain has the same phenotype as unc-119(e2498), and the 2 fail to complement. Working with the small deficiency tDf2, I obtained 3 additional EMS-induced alleles of the locus in a classic F₁ noncomplementation screen: ed3, ed4 and ed9. All failed to complement e2498 and had identical phenotypes³ (Fig. 1).

Figure 1.

Appearance of adult unc-119 mutant and wild-type (rescued) hermaphrodites on agar plates and summary of phenotypes. (A) unc-119(ed4) adults displaying short morphology and curled appearance. Mutants cannot form dauer larvae. (B) unc-119 mutant rescued with an integrated Cbr-unc-119(+) single-copy transgene, displaying normal morphology and sinusoidal locomotion. Rescued animals can also form dauer larvae. Images were taken at the same scale. The rescued animal is approximately 1mm long. Phase-contrast images were taken with a Canon DSLR through a 10x objective on an upright Olympus BX-51 compound light microscope.

At first, molecular identification of unc-119 turned out to be more challenging than it ought to have been. Although e2498 was likely to be a Tc1 insertion, extra bands that might have contained the gene were not apparent in Southern blots probed with Tc1. Using strains containing eDf2 or eDp6 and cosmid clones on either side, I was able to localize unc-119 to the overlap between 2 YACs, Y39A1 and Y60D9, by the summer of 1993. I subcloned the YACs by purifying them from pulsed-field gels and making SstI fragment libraries in pBluescript. Fortuitously, one of the first clones picked, pDP#MM008, happened to contain most of the unc-119 gene, though not enough to rescue. As I was about to clone a larger fragment from a genomic plasmid library, I read in the February, 1994 Worm Breeder's Gazette that Julie Ahringer (also at the MRC at the time, now at the University of Cambridge, UK) had cloned the nearby gene vab-7. She sent me a vab-7-rescuing cosmid, M142, with which I achieved the first transformation rescue of unc-119. The gene was narrowed down to a 5.7-kbp HindIII-XbaI fragment cloned in plasmid pDP#MM016 (Fig. 2). As expected, the e2498 allele contained a Tc1 insertion, most of the gene was missing from eDf2 and eDp6, and the point mutants were all GC to AT transitions: ed9 mutated a splice acceptor resulting in a −1 frameshift, while ed3 and ed4 were nonsense mutations within 30 base pairs of each other. By 1994, we knew that unc-119 encoded an apparently novel protein of 219 amino acids. Making use of a newly-available GFP plasmid from Martin Chalfie (Columbia University, NY), I found that unc-119::GFP was expressed broadly throughout the nervous system with some additional expression in the head. The identification of unc-119, and its expression using an unc-119::lacZ reporter made from an expression plasmid from Andrew Fire (now at Stanford University, CA), were reported in Genetics in November of 1995.³ I also cloned the C. briggsae ortholog from a λ genomic library made by Terry Snutch (University of British Columbia, Canada) and David Baillie (Simon Fraser University, Canada), and showed that Cbr-unc-119 could rescue the C. elegans mutant phenotype as an extrachromosomal array.⁴ The C. briggsae gene was much more compact, owing to a reduction in size of the introns (Fig. 2).

Figure 2.

The C. elegans and C. briggsae unc-119 loci.^{3, 4} The Ce-unc-119 genomic region is the 5.7-kbp HindIII-XbaI rescuing fragment found in the pBluescript KS- clone pDP#MM016, while the Cbr-unc-119 genomic region corresponds to the 2.1-kbp rescuing fragment found plasmid pCFJ151.²⁸ The derived clone pDP#MM016B contains 2 point mutations that introduce a BglII restriction site at the start of the coding region: ATG AAG GCA GAG CAA CAA ATG AAG GCA GAt CtA CAA (BglII site underlined, changed bases in lowercase). The ed9 allele changes the 3'G of an intron, such that the subsequent G becomes used as a splice acceptor and results in a (-1) frameshift.³ The coding region in the Ce-unc-119 transcript shown corresponds to M142.1b (Wormbase WS246). A longer gene model found in WormBase (M142.1a) contains an exon upstream of the region shown here. The coding region for the C. briggsae transcript corresponds to CBG18291 (Wormbase WS246). The C. briggsae nm67 allele was made in the laboratory of Eric Haag (University of Maryland College Park, MD)³⁷; 3 additional deletion alleles, not shown, were made in the laboratory of Robert Waterston (University of Washington, Seattle, WA).³⁸

Shortly after the first report on Ce-unc-119, a sequence ortholog of UNC-119, Human Retinal Gene 4 (HRG4), appeared in Genbank from a paper on genes enriched in the human retina.⁵ HRG4 can fully rescue the unc-119 mutant defects when expressed from the C. elegans unc-119 promoter.⁶ Orthologues are now known in many species including Drosophila,⁶ zebrafish,⁷ mouse,⁸ and the protozoan Naegleria,⁹ suggesting it is found in all animals. An ortholog, POC7, was also identified in proteomic analysis of centrioles from the green alga Chlamydomonas.¹⁰ UNC-119 proteins are lipid-binding chaperones that control trafficking of myristoylated G protein α-subunits and Src tyrosine kinases.¹¹ In C. elegans, UNC-119 is known to be important for axon structure^12,13, localization of the G protein α subunits ODR-3 and GPA-13,¹⁴ and other aspects of ciliogenesis.¹⁵

Use of unc-119 for Extrachromosomal Transformation

Transgenesis in C. elegans got its start in the mid-1980s when Andrew Fire (now at Stanford University, CA) reported success with gonadal injection of a plasmid carrying an amber suppressor tRNA, sup-7, to suppress a tra-3 nonsense mutation.¹⁶ Soon after, Craig Mello (now at University of Massachusetts Medical School in Worcester, MA) reported the use of the dominant rol-6(su1006D) allele, carried on plasmid pRF4, for injection of the gonad syncytium.¹⁷ Cloning of genes, and generation of reporter fusions (at first to lacZ,¹⁸ then to GFP¹⁹), became widespread as the physical map and subclones of the C. elegans genome became available.²⁰ Use of rol-6D to make arrays by gonadal microinjection became the “dominant” way to make reporter transgenes for many years, because it could be introduced into otherwise wild-type worms to cause a strong Roller phenotype. However, this phenotype affects late larval and adult body shape, strongly reduces male mating efficiency, and reduces fecundity relative to non-transgenic siblings with wild-type locomotion.²¹

Widespread use of unc-119 rescue as a marker started in the late 1990s as a direct result of the community's sharing of information over the emerging Internet. The USENET newsgroup CELEGANS/bionet.celegans, started in 1994, became a popular form of communicating and posting questions directly to researchers. (It still exists today in Google Groups, http://groups.google.com.) In March of 1996, in response to an inquiry about uncoordinated mutants that might be useful alternatives for transformation, I posted a description of the suitability of unc-119 rescue for transgenesis. Within a few months, I had sent a modified rescuing clone (pDP#MM016B) and an unc-119 mutant to over 50 C. elegans labs around the world. Although the original e2498 allele was sent out initially, different labs may have received ed3, ed4 or ed9. Aside from the first papers on unc-119 and its C. briggsae ortholog,^{3, 4} the first to report use of unc-119-rescued transgenes began to appear in 1998.^22–25 The observation that many transgenes seemed to work well with unc-119 as a coinjection marker suggested that in general, unc-119 rescue does not interfere with the vast majority of applications. This was no doubt an important consideration for labs initiating transgenesis experiments, and those developing newer technologies.

Use of unc-119 for Low-Copy Chromosomal Insertions

In the lab of Judith Austin (University of Chicago, IL), Vida Praitis (now at Grinnell College, IA) and colleagues developed microparticle bombardment methods for C. elegans²⁶ around the year 2000. Initially they had planned to suppress the phenotype of rol-6 mutants with a transgene containing the dominant sqt-1(sc1) mutation.²¹ After contacting the lab of Jim Kramer at the same institution, they were told by Brian Ackley (in the Kramer lab at the time, now at the University of Kansas, KS) about unc-119 rescue. The Kramer lab had originally obtained unc-119 mutants and rescuing DNA directly from David Pilgrim in the late 1990s. For bombardment, the inability of mutants to form dauer larvae meant that rare transformants, rescued for unc-119, would be able to outgrow the thousands of non-transformed mutants on a plate and be recovered after the bacteria were depleted.

Bombardment was initially performed using a single plasmid containing both unc-119 rescue and the transgene of interest (Fig. 3A).²⁶ Praitis et al. inserted transgenes directly into pDP#MM016, a plasmid in which the size of the C. elegans gene made routine cloning somewhat of an inconvenience: The unc-119 rescuing fragment is 5.7 kbp (Fig. 2), and the plasmid backbone 2.9 kbp, making the starting vector already 8.6 kbp in size. Recalling that the C. briggsae unc-119 gene has smaller introns, Andrew Singson and Barth Grant (both at Rutgers, NJ) asked me to send a subclone of Cbr-unc-119 to try for bombardment experiments around the year 2003. I sent them pMM571, a plasmid containing a 2.1-kbp Cbr-unc-119(+) fragment (Fig. 2). Both groups demonstrated single-copy rescue of C. elegans unc-119 mutants with pMM571. Likely hearing about their work from Andrew Singson, Andy Golden (NIH, Bethesda, MD) requested pMM571 from me and it was used by Christopher Richie in his lab to generate the plasmid pCR39 for their bombardment experiments.

Figure 3.

Multiple applications that use unc-119 rescue. All approaches involve transgene delivery into unc-119 mutant animals. While injected DNA can be linear or circular, features of injected DNA are shown as linear for simplicity. In all of the applications except co-CRISPR, “unc-119(+)” can be either Ce-unc-119(+) or Cbr-unc-119(+). (A) For conventional transgenesis, unc-119 mutants are coinjected with unc-119(+) and a transgene (e.g. gene::GFP) to generate extrachromosomal arrays. Both unc-119(+) and the transgene of interest can be on the same plasmid. In biolistic transformation, microparticle bombardment is used with a single plasmid, although it is possible to use 2 different plasmids.³⁹ Bombardments produce extrachromosomal lines, single- and low-copy insertions, a small fraction of which can occur at an endogenous locus.⁴⁰ (B-D) Methods that use excision of the Mos1 transposon. Note that if a single plasmid containing unc-119(+) and a transgene of interest is constructed for MosSCI or miniMos transposition, it can be also used for conventional transgenesis or bombardment. (B) In MosSCI, Mos1 excision from a known chromosomal site creates a double-stranded break that gets repaired by homology-directed repair (HR). This is guided by the left and right flanking regions around the break, typically ˜1.5 kbp in the original MosSCI protocol.²⁸ The original injected DNA can form an array that must be selected against, usually through inclusion of negative selection (neg. selection) markers such as mCherry reporters and/or an inducible toxic hs-peel-1 plasmid.²⁹ (C) In MosDEL, unc-119(+) is inserted inside a deleted region.³⁰ (D) In miniMos transposition, a modified Mos transposon inserts de novo at a random genomic site.³¹ (E) In one version of a CRISPR/Cas9-mediated approach,³³ a GFP reporter and unc-119(+) are inserted by HR at a double-stranded break created by Cas9. An optional step is the removal of the unc-119(+) marker through germline expression of Cre recombinase, which recombines loxP sites flanking unc-119(+)³³. Recent work suggests that the homology can be as short as 30 base pairs.⁴¹ (F) In a co-CRISPR strategy, an unc-119 point mutant can be reverted, while a gene of interest is edited in its endogenous site, using 2 guide RNAs (sgRNA1 and sgRNA2).^{34, 36, 42}

The C. briggsae ortholog would become the fragment of choice for rescuing unc-119 as newer methods were developed. Christian Frøkjær-Jensen and colleagues in the lab of Erik Jorgensen (University of Utah, UT) were modifying the MosTIC (Mos1 excision-induced transgene-instructed gene conversion) procedure developed by Valerie Robert and Jean-Louis Bessereau (Biologie cellulaire de la synapse, Paris, France),²⁷ to use rescue of unc-119 as a marker in a technique they named MosSCI (Mos1-mediated single-copy insertion).²⁸ This approach uses a targeting plasmid carrying both unc-119(+) and a transgene of interest between flanking genomic sequences to direct homologous targeting and insertion (Fig. 3B). Wayne Davis in the Jorgensen lab used pCR39, obtained from Christopher Richie, as a source to amplify Cbr-unc-119(+); this fragment subsequently became incorporated into many vectors used for MosSCI.²⁸ As shown in Figure 3B, unc-119 mutants are injected with a mixture of DNAs that provide a Mos transposase source, the targeting plasmid and various negative selection markers. While the reproductive advantage of chromosomally-rescued unc-119 animals alone helps in recovery of bona fide insertion strains, the negative selection markers enable identification (and/or elimination) of undesired animals rescued by extrachromosomal arrays derived from the injection mixture.^{28, 29} Rescue of unc-119 has been subsequently used in newer Mos-based methods, including MosDEL, in which Cbr-unc-119(+) is inserted while generating a targeted deletion of a gene³⁰ (Fig. 3C), and miniMos, in which DNA segments up to 45 kbp are inserted into random locations in the genome³¹ (Fig. 3D).

Future Applications of unc-119 Rescue

The development of the CRISPR/Cas9 system in C. elegans may have removed the need for transformation markers for most experiments, as genome editing events can be recovered at high efficiency even in the absence of array transmission.³² Rescue of unc-119 can still be used in specialized CRISPR/Cas9-based approaches. Daniel Dickinson and colleagues in the lab of Bob Goldstein (University of North Carolina, NC) developed a version of CRISPR/Cas9 mutagenesis that uses unc-119 rescue to identify gene-targeted events similar to MosSCI³³ (Fig. 3E). In their approach, LoxP sites flanking the Cbr-unc-119(+) insertion enable subsequent removal of the rescue marker through germline expression of Cre recombinase, in what could be called an 'un-rescue' of unc-119. Using the method of Dickinson et al., insertion of unc-119(+) could also be used to insertionally disrupt genes similar to MosDEL.

Rescue of unc-119 can be used to aid recovery of any CRISPR/Cas9-mediated targeting event. It was recently shown that pairs of chromosome editing events driven by CRISPR/Cas9 are more likely to occur concomitantly, such that successful editing for one gene predicts a higher recovery of editing of a second locus in the same progeny animals.³⁴ Using this co-CRISPR strategy, recovery of chromosomally-targeted gene edits could be facilitated by simultaneous reversion of unc-119 (Fig. 3F). An unc-119(ed3) mutant has been successfully reverted using CRISPR/Cas9 and DNA oligonucleotides.^{35, 36} As reversion can be achieved with a short region of unc-119(+) DNA as a repair template and not the intact gene, non-Unc progeny would represent bona fide chromosomal reversions, eliminating the need for negative array selection. Furthermore, as the 3 point mutations ed3, ed4 and ed9 are found within a 167-bp region, a single sgRNA and repair template could potentially revert any of the 3 alleles. Hence, through co-CRISPR, unc-119 rescue may continue to find applications in modern C. elegans genome editing methods.

Reagent Availability

The unc-119 alleles shown in Figure 2 are available from the Caenorhabditis Genetics Center (CGC, http://www.cgc.cbs.umn.edu). Plasmids that rescue unc-119 can be obtained from Addgene (https://www.addgene.org/). Advice on growing unc-119 mutants and optimizing injection can be found on the Wormbuilder site (http://www.wormbuilder.org).

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

I thank Brian Ackley, Wayne Davis, Christian Frøkjær-Jensen, Andy Golden, Barth Grant, Erik Jorgensen, Vida Praitis, Christopher Richie and Andy Singson for helping me reconstruct the history of the use of unc-119 rescue; Christian Frøkjær-Jensen for additional suggestions; Eric Haag for information about the nm67 allele of Cbr-unc-119; and David Pilgrim, for having given me the opportunity to clone unc-119 those many years ago.

Funding

Work in the Maduro Lab is currently funded by NSF IOS#1258054.