A new miniMOS tool kit capable of visualizing single copy insertion in C. elegans

The miniMOS technique has been widely used in the C. elegans community to generate single copy insertions. A worm is considered as a potential insertion candidate if it is resistant to G418 antibiotics and does not express a co-injected fluorescence marker. If the expression of the extrachromosomal array is very low, it is possible for a worm to be mistakenly identified as a miniMOS candidate, as this low expression level can still confer resistance to G418 without producing a detectable fluorescence signal from the co-injection marker. This may increase the workload for identifying the insertion locus in the subsequent steps. In the present study, we modified the plasmid platform for miniMOS insertion by incorporating a myo-2 promoter-driven TagRFP or a ubiquitous H2B::GFP expression cassette into the targeting vector and introducing two loxP sites flanking the selection cassettes. Based on this new miniMOS tool kit, the removable fluorescence reporters can be used to visualize the single copy insertions, greatly reducing insertion locus identification efforts. In our experience, this new platform greatly facilitates the isolation of the miniMOS mutants.


INTRODUCTION
A transposon is a DNA fragment that can relocate within the genome of a single cell, which is discovered by the Nobel laureate Barbara McClintock (McClintock, 1950(McClintock, , 1942. Mos1 transposase belongs to the Tc1/mariner family, which is relatively conserved among most eukaryotic species and was initially isolated in Drosophila mauritiana. It can catalyze the relocation of the transposon (Hartl, 2001;Richardson et al., 2006). This naturally existing phenomenon, namely transposase-mediated relocation of a transposon, has led to the development of several widely used techniques that help the researchers to modify the genome of different species in many studies (Coates et al., 2000;Frokjaer-Jensen et al., 2010, 2014Kawakami, 2005). However, techniques based on Mos1 transposase could not be applied to C. elegans research as C. elegans does not naturally harbor Mos1 elements in their genome until the Jorgensen Lab introduced Mos1 transposase and elements into C. elegans (Bessereau et al., 2001). They further applied the Mos1-related techniques to generate single copy insertions and deletions, namely mosSCI and mosDEL techniques, respectively (Frokjaer-Jensen et al., 2010. They also improved the Mos1 transposon by testing Mos1 elements with different sizes, and finally identified a minimal Mos1 element of 550 bp in length with comparable or even better efficiency for Mos1mediated insertion (Frokjaer-Jensen et al., 2014). This minimal essential element was named miniMOS thereafter. Nowadays, Mos1-related techniques are popular genome editing methods within the C. elegans research community.
The pCFJ910 plasmid contains a miniMOS element (Frokjaer-Jensen et al., 2014), as shown in Fig 1. We used the plasmid to generate single-copy insertion mutants in our previous projects (Hao et al., 2018;Wu et al., 2019). The insertion line was initially isolated based on the following criteria: worms were resistant to G418 antibiotics and did not express the co-injection marker. However, we sometimes encountered difficulties in isolating miniMOS lines. It is possible that worms fitting the above criteria may only carry an extrachromosomal array with an extremely low level of plasmids, rather than a single copy insertion. This extrachromosomal array may be sufficient to confer G418 resistance, while its expression of the co-injection marker may be too low to be detected by the microscope. Therefore, we wondered if we can incorporate a visible reporter for miniMOS insertion to increase our chances of isolating mutants with single copy insertion. In this study, we present a modified plasmid tool kit for miniMOS insertion. This kit includes specific fluorescence markers that can be used to visualize single-copy insertions and flanking loxP sites that make the selection cassettes removable. Based on our experience, this new platform greatly facilitates the isolation of strains carrying miniMOS insertions.

Animals
Wildtype strain (N2) of C. elegans was obtained from Caenorhabditis Genetics Center (CGC), which was maintained at 20 C in a 6-cm NGM-agar (nematode growth medium) plate, following previously published protocol (Brenner, 1974). Hermaphrodites were used in the study. Except for N2, all other strains in this study were generated by our lab (Table S1). All strains are available upon request.

Plasmids construction
All newly generated plasmids were constructed by Gibson assembly and sequenced, including pYW347, pYW249, pYW241, pYW242, pYW146, and pYW186. The detailed strategy for molecular cloning and the primers used for PCR are listed in Tables S2 and S3. For constructing pYW241 (Fig. 1), a 2,832 bp fragment upstream of the ATG codon of the tph-1 gene was used as the tph-1 promoter. For constructing pYW242 (Fig. 1), ser-1 promoter and cDNA were amplified from a gift plasmid (Guo et al., 2018) from Dr. Taihong Wu (Harvard University, Cambridge, MA, USA). For constructing pYW249 (Figs. 1 and 2), a 5,820 bp fragment covering the whole genomic region of the zipt-17 gene, starting from −3,914 bp relative to ATG to +503 bp relative to TAA of the zipt-17 gene, was amplified. For constructing pYW146, a 1,300 bp fragment upstream to the ATG codon of the lin-44 gene was used as the lin-44 promoter (Ge et al., 2020). The pYW347 and pYW249 plasmids are available in Addgene, with plasmid IDs 196064 and 196063 respectively. All sequences of the above plasmids are listed in Table S2.
All plasmids are available upon request. The authors affirm that all data necessary for confirming the conclusions of the article are present within the article, figures, and tables.

Microinjection, miniMOS mutant isolation, and insertion locus identification
Microinjection was performed as previously reported (Rieckher & Tavernarakis, 2017), using a ZGENEBIO microinjector (ZGene Biotech Inc., Taipei, China), under an inverted microscope. The information on the injection mix is listed in Table S4. All the plasmids Figure 1 Plasmids generated to facilitate miniMOS mutant isolation. All plasmids were constructed using pCFJ910 as an initial template and are ampicillin resistant. The figure shows the relative locations of each element in each plasmid. The length of each fragment in the figure is not proportional to its actual length. The plasmid backbone, containing elements for antibiotic resistance and plasmid replication, is not shown in the figure.
Full-size  DOI: 10.7717/peerj.15433/ fig-1 were kept at −20 C before mixing. We used a freshly prepared injection mix for each injection and did not reuse it. All steps from microinjection to mutant isolation can be found in Fig. 3

Fluorescence imaging
For Figs. 4A and 4H, the images were taken using an Olympus IX71 fluorescence inverted microscope with a 20× objective. As the original images showed a dim signal, the exposure of these images was adjusted by +4, to facilitate the presentation. For Fig. 4H, five images covering the whole body of the worm were taken separately, and then manually aligned. For taking images in Figs. 4B-4G and 4I-4N, a 6-cm NGM plate was coated with 100 mL of 25 mM potassium azide, then the worm for imaging was picked to the plate and left for free moving. When the worm was immobilized, an image was taken using a MshOt stereo fluorescence microscope MS23, which is composed of an Olympus SZX7 stereo fluorescence microscope equipped with an LED fluorescence light source. All images were pseudo-colored according to the fluorescence.

Quantification of insertion ratio
The quantification of the insertion ratio was performed based on a previously published article with some modifications (Frokjaer-Jensen et al., 2014). Briefly, we used two different injection mixes (Table S4). Both mixes contained a miniMos plasmid, a Mos1 transposase plasmid, two co-injection markers, and pUC19. One of the mixes contained the negative PEEL-1 selection plasmid. We picked F1 transgenic animals to individual plates and allowed them to starve at 25 C. These F1 animals were resistant to G418 and carried either P myo-2 ::mCherry or P myo-3 ::mCherry fluorescence, or both. Insertion candidates were initially identified by their G418 resistance and the absence of the injection marker, as previously described (Frokjaer-Jensen et al., 2014). Among these candidates, those with regularly expressed fluorescence were the candidates met our criteria. The insertion ratio was calculated by dividing the number of the candidates by the number of F1 animals and multiplying the result by 100%. For peel-1 selection, we heat-shocked the starved plates for 2 h at 34 C.
Based on the newly constructed plasmid platform, we further made three targeting vectors for our ongoing projects: one for showing the tph-1 gene expression pattern by tph-1 promoter-driven GFP (pYW241 in Fig. 1), one for labeling ser-1 gene by GFP (pYW242 in Fig. 1), and one for introducing wildtype zipt-17 genomic region into the zipt-17 (ok745) mutant (pYW249 in Fig. 1). Unfortunately, neither a single copy inserted P tph-1 ::GFP reporter nor P ser-1 ::SER-1::GFP reporter can be seen under the inverted fluorescence microscope.

Molecular cloning strategy using our new plasmid platform
We recommend using Gibson assembly to do the plasmid construction. Locations of the primers and their sequences were shown in Figs. 2A and 2B. The main backbones of the plasmids, the 7,447 base pair (bp) fragment from pYW347 or the 6,871 bp fragment from pYW249, can be amplified using primer sets P1/P2 or P1/P3, respectively (Fig 2A, Tables  S2 and S3). The fragment of interest in your project can be amplified using specific primers carrying 15 to 25 bp homologous arms to the above fragments, as shown in Fig 2C. Alternatively, a traditional molecular cloning strategy based on T4 ligase-mediated DNA ligation can be conducted, using restriction sites SnaBI in pYW347, or BamHI, AbsI, and StuI in pYW249 ( Fig. 2A).

An updated protocol for generating miniMOS line
With our new plasmid platform, we further updated the protocol for isolating miniMOS mutants, based on a previously published protocol (Frokjaer-Jensen et al., 2014). We summarized our protocol in Fig. 3, starting from plasmid construction. Our protocol is composed of seven steps, which usually take about 1 month to generate mutant animals without the indispensable elements for mutant screening, e.g., TagRFP, neoR, H2B::GFP (Fig. 3). Some unique details are reported as follows. We usually use young adult worms carrying about 10 eggs for injection. According to our experience, worms at this age are not as fragile as the younger adult worms and are capable of generating more progeny than the older ones. With the help of our fluorescence expression cassettes, P myo-2 ::TagRFP or ubiquitous H2B::GFP, the insertion candidates can be primarily identified by the expression of dim and regularly-distributed fluorescence in the pharynx or all over the body of the worms (Fig. 3 step 4, and Fig. 4). Because of the existence of a single copy insertion of fluorescence reporter, the homozygote candidate worm may be isolated by picking up worms with a relatively brighter fluorescence signal (Fig. 3, step 6). These visible reporters are also useful for further verifying the homozygous animals, whose progeny should all carry fluorescent reporters (Fig. 3, step 6). An optional step for excising these indispensable elements for mutant screening can be conducted by injection of P eft-3 ::Cre plasmid (Fig. 3, step 7).

Facilitated isolation of miniMOS candidate by the fluorescence markers
Here we present our results and criteria for the isolation of insertion candidates based on fluorescence reporters, i.e., P myo-2 ::TagRFP or ubiquitous H2B::GFP. The signals from the P myo-2 ::TagRFP cassette were restricted in the pharynx of the C. elegans (Figs. 4A-4G), which is consistent with previous publications (Frokjaer-Jensen et al., 2014;Semple, Garcia-Verdugo & Lehner, 2010;Taylor & Dillin, 2013;Toker et al., 2022). The signal from a single copy inserted P myo-2 ::TagRFP cassette was smoothly distributed in the pharynx, they were not so bright, and nearly symmetrical along the anterior-posterior axis of the pharynx, especially in the pharyngeal bulbs of the worm (Fig. 4A). After the injection, the non-transgenic animals, which carried no extrachromosomal array and single copy insertion, were killed by G418 antibiotics. More than 95% of the surviving animals carried an extrachromosomal array. The expression pattern of extrachromosomal P myo-2 ::TagRFP cassette varied among worms (Figs. 4B-4E), which we refer to as irregular patterns hereafter in this article. The irregular patterns of the P myo-2 ::TagRFP cassette expression are diverse, they can be saturated fluorescent signals in one or two pharyngeal bulbs (Figs. 4B and 4C), or hardly visible in the isthmus and terminal bulb of the pharynx (Fig. 4D), or absent from half of the procorpus of the pharynx (Fig. 4E). The signals of the P myo-2 :: TagRFP cassette in these worms were not evenly distributed in the pharynx. In addition to these irregular patterns, we also observed two regular patterns, with the red fluorescence signals distributed smoothly in the pharynx at different strengths (Figs. 4F-4G). We speculated that the intensities of the signals in these regular patterns possibly indicate the genotype of the insertion locus, or may be caused by the position effect variegation as a result of miniMOS insertions at different loci (Frokjaer-Jensen et al., 2016). In our plasmid platform, we included two loxP sites flanking the fluorescence reporter (Fig. 1). To test if the fluorescence reporter is removable, we injected the worms with a Cre recombinaseexpressing plasmid and picked their progeny without P myo-2 ::TagRFP fluorescence as candidates for genotype analysis. A specific PCR band confirmed the genotypes of the loxP-fragment-deleted locus (Fig. S1).
In our plasmid platform, the H2B::GFP fusion gene was driven by the P rps-27 promoter, which led to the ubiquitous labeling of nuclei of the worms (Figs. 4H-4N). The signals from a single copy-inserted P rps-27 ::H2B::GFP cassette gave the worms with a distinct regular pattern (Fig. 4H), consistent with previous publications (Adikes et al., 2020;Frokjaer-Jensen et al., 2012). Similar to our observation in single copy-inserted P myo-2 :: TagRFP signals, these signals were also not so bright (Figs. 4H, 4M and 4N). The extrachromosomal P rps-27 ::H2B::GFP showed saturated signals, its expression pattern was quite different from that of the single copy-inserted one and seemed to follow no regularity (Figs. 4I-4L). We also observed two regular patterns carrying green fluorescence signals with different strengths (Figs. 4M and 4N), which may result from different genotypes or position effect variegation. The loxP-flanked H2B::GFP fragment can be deleted by injecting the worms with a Cre recombinase-expressing plasmid, and its genotype was confirmed by a specific PCR band (Fig. S1).

Facilitated characterization of insertion locus
To test whether our method is more efficient, we compared our method with a previously published one by the Jorgensen EM group, which is referred to as the JEM method or protocol hereinafter. From December 2020 to November 2022, we injected a total of 19 plasmid mixtures for different ongoing projects in our lab, 12 using the JEM method and seven using our protocol. Firstly, we compared the number of worms used for insertion locus identification by inverse PCR. On average, the JEM method used about seven worms per injection mix, and our method used about three worms per injection mix (Fig. 5A). Considering each worm would be used for an independent inverse PCR experiment to identify the insertion site, this result indicates that our method is less labor-intensive. Secondly, we compared the success rate in insertion locus identification using these two methods, which was 23.8% for the JEM method and 75.7% for our method on average (Fig. 5B). This result suggests that our success rate in identifying the insertion site is almost three times as efficient as the JEM method. Furthermore, we also compared the insertion efficiency using injection mixtures with or without PEEL-1 selection (Table S5). We observed that the PEEL-1 selection by heat shock killed the transgenic worms in most cases, but we didn't find a significant difference in insertion ratios between the injections with and without PEEL-1 selection (Table S5, 10.1% vs 9.7% by JEM method, or 6.8% vs 6.2% by our method). We also noticed a tendency of our protocol to yield a comparatively lower insertion rate, compared to the JEM method (Table S5, 10.1% vs 6.8% for injections without peel-1 selection, or 9.7% vs 6.2% for injections with PEEL-1 selection). However, even with this potentially lower insertion ratio, our method still had a very high efficiency, about 75.7%, in insertion locus identification. Therefore, the above data collectively suggest that our method for generating miniMOS insertion lines is more effective for insertion identification. The fluorescence reporter facilitates the isolation of homozygote animals If all the progeny of the candidate worm are fluorescent and have similar expression patterns as the candidate worm, they are likely homozygotes. Our newly generated strain PSC190 met this criterion. We performed a series of genotyping PCRs to further verify their genotypes. A correct PCR band from primer set 1 suggests the presence of the insertion, and that from primer set 2 suggests the presence of the wildtype allele (Figs. 6A-6C). Genotyping PCR of five PSC190 worms with primer set 1 all generated a clear band, indicating insertion (Fig. 6D). In contrast, genotyping PCR of these worms with primer set 2 generated a series of bands. The band with a similar length to the positive control was confirmed to be nonspecific by sequencing, indicating the absence of wildtype locus in PSC190 (Fig. 6D). Since the progeny of PSC190 are 100% fluorescent, the above results collectively suggest that these randomly picked worms were all homozygotes.

Single copy inserted fluorescence reporters facilitate the identification of miniMOS mutants
In the present study, we established a new platform for generating miniMOS insertions, by incorporating fluorescent expression cassettes into the targeting vector pCFJ910. This new platform is composed of two plasmids, pYW347 and pYW249, using P myo-2 ::TagRFP or ubiquitous H2B::GFP as a visible reporter, respectively. Our new miniMOS platform has several advantages. Firstly, these fluorescence reporters can greatly facilitate miniMOS mutant isolation. As we presented in Fig. 4, the miniMOS candidate lines generated by our plasmid platform have regular expression patterns, compared to the extrachromosomal array-containing lines. These expression patterns were easily recognizable. This visible reporter facilitated the isolation of the candidate worms. Secondly, the insertion locus identification using our protocol is less labor-consuming. The number of worms used for the inverse PCR step is reduced by more than 50%, and the success rate for identifying the insertion locus is about 75%, which is nearly tripled. These findings suggest that our method greatly reduced the labor for identifying the insertion site. Thirdly, with the aid of the fluorescent reporter, homozygote isolation can be easily confirmed by whether their progeny are 100% fluorescent, which is more straightforward and saves some efforts from genotyping PCR, thus facilitating homozygote isolation.
We also acknowledge some limitations of our platform. Firstly, the fluorescence cassettes are 1.4 to 1.7 kb in length, which increase the length of the final plasmids. This may potentially increase the difficulty of molecular cloning. Secondly, even if the selection cassettes are removed, one loxP site would permanently remain in the genome. Although the impact of one loxP site on animal physiology might be tiny, it can be troublesome when you cross this worm into another genomic background carrying both loxP (or its mutant) sites and Cre recombinase. Thirdly, P myo-2 promoter-driven fluorescence marker can be lethal when a higher concentration of this plasmid exists in the injection mixture (Rieckher & Tavernarakis, 2017). When you use pYW347-based plasmid as a template for your miniMOS work, we recommend a working concentration lower than 17 ng/mL. Finally, according to our experience, the ubiquitous H2B::GFP signals are more sensitive to our eyes than P myo-2 ::TagRFP signals under the stereo fluorescence microscope, when searching for the candidate worms from a mixed population containing miniMOS worms and transgenic worms; while the loss of the P myo-2 ::TagRFP signals is more sensitive to our eyes when searching for worms without fluorescence. Thus, which is better depends on whether you plan to remove the loxP-flanked fluorescence cassettes.