Highly efficient transgenesis with miniMos in Caenorhabditis briggsae

Abstract Caenorhabditis briggsae as a companion species for Caenorhabditis elegans has played an increasingly important role in study of evolution of development and genome and gene regulation. Aided by the isolation of its sister spices, it has recently been established as a model for speciation study. To take full advantage of the species for comparative study, an effective transgenesis method especially those with single-copy insertion is important for functional comparison. Here, we improved a transposon-based transgenesis methodology that had been originally developed in C. elegans but worked marginally in C. briggsae. By incorporation of a heat shock step, the transgenesis efficiency in C. briggsae with a single-copy insertion is comparable to that in C. elegans. We used the method to generate 54 independent insertions mostly consisting of a mCherry tag over the C. briggsae genome. We demonstrated the use of the tags in identifying interacting loci responsible for hybrid male sterility between C. briggsae and Caenorhabditis nigoni when combined with the GFP tags we generated previously. Finally, we demonstrated that C. briggsae tolerates the C. elegans toxin, PEEL-1, but not SUP-35, making the latter a potential negative selection marker against extrachromosomal array.


Introduction
As a comparative model of Caenorhabditis elegans, Caenorhabditis briggsae shares a similar morphology, carries a genome of comparable size (Stein et al. 2003;Ross et al. 2011;Ren et al. 2018;Yin et al. 2018;Li et al. 2020) and adopts similar developmental pattern to that of C. elegans (Zhao et al. 2008). Caenorhabditis elegans is a model organism. Its genome has been subject to intensive manipulations with various tools, including random mutagenesis with ultraviolet light coupled with trimethylpsoralen (UV/TMP) (Thompson et al. 2013), low copy insertion of transgenes with biolistic bombardment (Praitis et al. 2001;Hochbaum et al. 2010;Radman et al. 2013), targeted single-copy insertion using transcription activator-like effector nucleases system (Wood et al. 2011;Lo et al. 2013), or using CRISPR/Cas9 system (Chiu et al. 2013;Cho et al. 2013;Waaijers et al. 2013;Dickinson et al. 2015;Dickinson and Goldstein 2016), Mos1-mediated single-copy insertion (MosSCI) (Frøkjaer-Jensen et al. 2008) as well as random single-copy insertion with miniMos (Frøkjaer-Jensen et al. 2014). Many of these tools have been adopted in C. briggsae for genome editing with a comparable or a much-reduced successful rate. For example, biolistic bombardment was successfully adopted in C. briggsae transgenesis with comparable efficiency, whereas single-copy insertion with miniMos demonstrated a much lower efficiency in C. briggsae than in C. elegans (Praitis et al. 2001;Semple et al. 2010;Zhao et al. 2010;Frøkjaer-Jensen et al. 2014;Bi et al. 2015). An efficient transgenesis method with single-copy insertion is essential for comparative functional study of biology between C. elegans and C. briggsae. Recent work in Caenorhabditis species has demonstrated that heat shock treatment or compromised function in heat shock pathway significantly increased the frequency of transposition (Ryan et al. 2016), raising the possibility of improving miniMos-based transgenesis in C. briggsae by incorporation of a heat shock step or inhibition of activities of heat shock proteins.
Isolation of C. briggsae sister species, Caenorhabditis nigoni, with which it can mate and produce viable progeny, paves the way of speciation study using nematode as a model for the first time (Woodruff et al. 2010;Kozlowska et al. 2012). To isolate postzygotic hybrid incompatible (HI) loci between C. briggsae and C. nigoni, a dominant and visible marker is required for targeted introgression, in which the genome of one species is labeled with the marker and repeatedly backcrossed into the other to isolate the HI loci. The marker greatly facilitates tracing of its linked genomic fragment during backcrossing and its associated HI phenotype. Given numerous such markers are required over a genome, generation of such markers using targeted single-copy insertion becomes an enormous burden. This is because that the efficiency for targeted single-copy insertion is relatively low. As an alternative, over 100 GFP markers were inserted into the C. briggsae genome with biolistic bombardment, which allowed genome-wide mapping of HI loci between the 2 species (Yan et al. 2012;Bi et al. 2015;Ren et al. 2018).
Biolistic bombardment using cbr-unc-119 (Zhao et al. 2010;Bi et al. 2015) generates low copy number of transgene into genome but requires tedious animal preparations. It also needs to wait for several weeks before screening for transformed animal. In addition, presumably due to dramatic mechanical shearing, the transgenic animals often suffer from chromosomal rearrangements (Bi et al. 2015;Ren et al. 2018;Tyson et al. 2018;Ding et al. 2022). Importantly, the insertion site cannot be precisely determined due to unknown copy number of the transgene and its arrangement within host genome. As such, the transgene insertion sites were estimated by genotyping introgression boundary with PCR using species-specific primers (Yan et al. 2012;Bi et al. 2015). Unfortunately, the divergence between the C. briggsae and C. nigoni genomes is too far to allow efficient recombination during backcrossing, resulting in a poor mapping resolution of the insertion site (Yan et al. 2012;Bi et al. 2015Bi et al. , 2019. We have recently demonstrated that inter-chromosomal interactions are involved in hybrid male sterility between C. briggsae and C. nigoni (Li et al. 2016;Bi et al. 2019). For example, a C. briggsae X chromosome fragment in an otherwise C. nigoni background as an introgression leads to male sterility. Hybrid F1 male sterility can be rescued by the presence of the introgression and another C. briggsae genomic fragment on its Chromosome II. We speculate that such interacting loci might be common in producing HI. Availability of the dual-color labeling system makes it possible for systematic mapping such interacting loci. This is because it facilitates tracking of cosegregation of the 2 loci. It is worth noting that most existing markers in the C. briggsae genome were derived from GFP, preventing effective screening for such loci. Therefore, C. briggsae animals bearing a visible marker other than the GFP are desired.
Transposon-mediated single-copy insertion at random genomic position with high efficiency has been successfully developed in C. elegans (Frøkjaer-Jensen et al. 2014). A fly transposon Mos1 was truncated with minimal transposon sequences, termed as miniMos cargo vector, in order to boost its capacity of carrying foreign DNA (Frøkjaer-Jensen et al. 2014). Single-copy of transgene can be inserted into host genome at high frequency via coinjecting a transposaseexpressing vector and miniMos cargo vector into C. elegans gonad. The method was also adopted in C. briggsae but with a much lower efficiency with unknown reasons (Frøkjaer-Jensen et al. 2014). Attempts have been made to improve insertion frequency in C. briggsae. For example, the promoter (cel-Peft-3) driving transposase expression was substituted with C. briggsae promoter cbr-Peft-3 or cbr-Ppie-1 in order to boost transposase expression. However, the insertion frequency did not improve as expected. A highly efficient method for inserting single-copy transgene has yet to be established in C. briggsae. To facilitate screen for single-copy insertion, a negative selection marker cel-Phsp-16.41::peel-1 was co-injected into animals to kill extrachromosomal array-bearing worms during screening (Frøkjaer-Jensen et al. 2012). However, the killing efficiency of the negative selection marker peel-1 has not been investigated in C. briggsae.
This study established a highly efficient methodology for inserting single-copy transgene into the C. briggsae genome based on miniMos. Negative selection markers against transgene existing only in extrachromosomal array were also explored with limited success.
Plasmids for microinjection were extracted using the PureLink HQ Mini Plasmid Purification Kit or HiPure Plasmid Midiprep kit (Invitrogen). Genomic DNAs were extracted from mix-staged animals with PureLink Genomic DNA Mini Kit (Invitrogen). Injection mixture for transformation consisted of miniMos cargo vector at 20 ng/ml, Mos1 transposase-expressing plasmid at 50 ng/ml, red or green fluorescent coinjection markers each at 10 ng/ml. pZZ203 was added to 80 ng/ml with a total DNA concentration of 170ng/ml. Plasmids pMA122, pZZ185, and pZZ196 were built to test their use as a negative selection marker. Neomycin stock solution of 12.5 mg/ ml was made with G418 powder (ThermoFisher) in water.

Transgenesis
Animal injected with pZZ160 (GFP) or pZZ184 (mCherry) and pCFJ601 was individually placed on a 55-mm crossing plate seeded with OP50 ($6.7 ml NGM per plate), and allowed to recover at 25 C overnight (roughly 12-16 h) (Fig. 1). For heat shock treatment, plates with injected animals were sealed with parafilm and incubated on the water bath preheated to 35 C for 3 h. Remove parafilm and any water droplet on the inner lid. Incubate plates at room temperature ($22 C) for half an hour. Add 0.5 ml 12.5 mg/ml neomycin (G418) solution prewarmed up at room temperature onto each plate. Gently rotate plates to allow neomycin solution to fully cover plate surface. Air-dry the plates and incubate the dried plates in a 25 C incubator for 6-7 days. Due to the presence of selection marker NeoR in the injected miniMos plasmids (pZZ160 and pZZ184), single 10-12 survived L4 or young adult animals from the plates crowded with animals that do not express coinjection marker onto individual plates to be incubated at 25 C. After 2-3 days, harvest animals with homozygous transgene based on the reporter expression and lack of expression of visible coinjection marker for genotyping.

Quantification of P 0 insertion frequency
Three days after microinjection, plates without F 1 transgenic progeny were discarded and excluded from the number of injected P 0 animals. The F 2 animals were screened for miniMos insertion under stereo-microscope based on reporter expression or phenotypic rescue, and the absence of coinjection markers. Animals confirmed with transgene insertion from the same P 0 plate were treated as an independent single-copy insertion. The P 0 insertion frequency was measured through dividing the number of independent insertions by the total number of injected P 0 animals. Mapping of insertion site was performed as described (Frøkjaer-Jensen et al. 2014). The C. briggsae cb4 genome (Ross et al. 2011) was used to report the chromosomal coordinates.

Selection marker for transformation
For selection with Neomycin (NeoR), injected animals were allowed to recover overnight (12-16 h) followed by heat shock treatment on a 35 C water bath for 3 h. The heat shock treated plates were incubated at room temperature for half hour to recover. Then 500 ll of 12.5 mg/ml G418 solution was added to each plate. Plates were incubated at 25 C for 6-7 days. Twelve F 2 animals were singled onto a new NGM plate from the P 0 plates crowded with animals. Homozygosity was checked based on reporter expression in F 3 .
For selection with cbr-dpy-5, all rescued dpy-5(þ) F 1 animals were singled onto individual plates. Twelve F 2 animals were singled onto new plate from each F 1 plate in which over 50% animals were dpy-5(þ) ones. F 2 plates with 100% dpy-5(þ) animals that did not express visible coinjection markers were kept for genotyping.

Development of negative selection marker for extrachromosomal array
A fusion PCR product was made between a cassette consisting of peel-1::tbb-2 UTR or gfp::tbb-2 UTR and the C. briggsae syntenic region of hsp-16.41 promoter with the primers listed in Supplementary Table 2. Additional details of C. briggsae syntenic region of hsp-16.41 promoter sequence screening are available in Supplementary Fig. 2. Injection mixture was made with pMA122, pZZ185, pZZ196, or the fusion PCR product at 70 ng/ml, fluorescent coinjection marker vector pZZ161 at 20 ng/ml, pGH8, pCFJ90, pCFJ104 at 10 ng/ml each, and pZZ203 at 50 ng/ml.
To examine the lethality in C. briggsae after heat shock, an array-bearing line was generated and embryos harvested for synchronization. The synchronized L1 animals were divided into 2 groups, one was subjected to heat shock treatment at 35 C for 3 h before transferring into 25 C incubator, and the other was incubated at 25 C without heat shock treatment as a control. Caenorhabditis elegans heat shock treatment was performed at 34 C for 3 h. The ratio of surviving adults carrying an array out of all progeny were calculated in both control and heat-shock treated animals.

Introgression
Introgression was performed as described (Bi et al. 2015). Specifically, C. briggsae transgenic strains, ZZY0777 and ZZY0782 each expressing a single-copy of Pmyo-2::mCherry located on the X and chromosome II, respectively, were backcrossed to C. nigoni (JU1421) for 15 generations to give rise to ZZY10377 and ZZY10382. Introgression boundaries were mapped as described (Yan et al. 2012).

Heat shock treatment significantly increased the efficiency of transgene insertion
To improve the transgenesis efficiency using miniMos in C. briggsae, we started the transgenesis by repeating the steps basically as described previously (Frøkjaer-Jensen et al. 2014), i.e. injection of a vector carrying cbr-Pmyo-2::gfp along with a transposaseexpressing vector into cbr-unc-119 deletion mutant strain RW20000 (Zhao et al. 2010). One of major complications associated with the cbr-unc-119 as an injection selection marker was the low transformation efficiency even for the formation of extrachromosomal array. This was the case using the rescuing fragment from either C. elegans or C. briggsae (data not shown). We speculated that the low insertion rate associated with the selection marker could be related to the low transformation efficiency of single-copy insertion.
To improve transformation efficiency, we generated another injection selection marker cbr-dpy-5 using CRISPR/Cas9. This was based on the observation that dpy-5 had been used as a very pZZ160 (Pmyo-2::GFP) or ZZ184 (Pmyo-2::mCherry). The expressed transposase is expected to cut the myo-2 promoter fusion DNA fragment and randomly insert into the host genome. A heat shock was included to increase transgenesis efficiency. A neomycin resistance gene, NeoR, was included in the cargo vector so that only transgenic animals carrying NeoR as either extrachromosomal array or single-copy transgene can grow on the NeoRþ plates. The categories of transgenic animals can be roughly differentiated by population growth rate on NeoRþ plates. As an option, a second visible marker, for example, a second fluorescence maker with different color, can be included in the injection mixture to distinguish the array-containing animals from those without array. efficient selection marker in C. elegans microinjection (Zhao et al. 2004;Hunt-Newbury et al. 2007). As expected, cbr-dpy-5 worked as an effective selection marker for microinjection. We successfully obtained 3 independent transgenic lines with single-copy insertion using the marker after injecting 35 animals. However, due to lack of negative selection maker against extrachromosomal array, it is tedious to screen for the rescued animals out of all progeny. We then performed miniMos-based single-copy insertion using neomycin as a selection marker as described (Giordano-Santini et al. 2010;Frøkjaer-Jensen et al. 2014). The method provided a key advantage in simplifying selection of the transformed animals because all unrescued animals were killed, but the overall efficiency of obtaining single-copy insertion was not improved compared with previous studies (Fig. 2).
Given that heat shock treatment significantly increased transposition frequency (Ryan et al. 2016), we reasoned that inclusion of a heat shock step might boost the successful rate of transgenesis in C. briggsae based on miniMos, which is a modified transposon. As expected, inclusion of a step of heat shock treatment, i.e. 35 C for 3 h, we were able to significantly increase the frequency of single-copy insertion of Pmyo-2::gfp or Pmyo-2::mCherry (Figs. 1 and 2). Insertion frequency increased from lower than 5% without heat shock to $25% with heat shock for both constructs, indicating that the heat shock treatment was essential for increasing transposition rate.
A large collection of single-copy insertions expressing Pmyo-2::mCherry were generated in C. briggsae To complement the genetic resources mainly consisting of GFP insertions over the C. briggsae we generated previously (Bi et al. 2015), we decided to generate a cohort of insertions randomly distributed over the C. briggsae genome with a focus on generation of insertion expressing a fluorescent protein other than GFP, i.e. Pmyo-2::mCherry. This would be particularly useful for isolation of interacting loci responsible for hybrid male sterility as detailed below. To this end, we generated a total of 54 insertions consisting of 9 Pmyo-2::gfp and 45 Pmyo-2::mCherry that were able to be uniquely mapped to the C. briggsae genome ( Fig. 3; Supplementary Table 1). The markers show roughly random distribution over the C. briggsae genome.

Dual-colored marking system facilitates screen for interacting loci responsible for HI loci
We demonstrated that genetic interaction between X Chromosome and autosome was essential for hybrid male sterility between C. briggsae and C. nigoni (Bi et al. 2019). Identification of such interacting loci is challenging without proper markers on interacting chromosomes. We showed that the interaction between independent GFP-labeled C. briggsae introgression fragments on 2 different chromosomes was essential for male fertility in C. nigoni, but it was difficult to pinpoint these interacting loci systematically. Availability of such dual-color labelled strains would greatly facilitate the identification of such interactions. To help illustrate this point, we tried to recapitulate the interaction we identified earlier by using 2 C. nigoni strains ZZY10377 and ZZY10382 that carry a mCherry-labeled introgression fragment derived from the right arm of the C. briggsae X and chromosome II (Fig. 4), respectively, between which an interaction was found (Bi et al. 2019). We crossed 2 introgression strains, one labeled with GFP and the other with mCherry in reciprocal way. We then examined the fertility of the males that simultaneously carried both introgressions. We found that the males carrying both loci were mostly fertile, whereas the males carrying a single introgression of GFP or mCherry inserted on the X chromosome were sterile, indicating that the dual-color labeled introgressions did recapitulate the interaction. Presence of the dual-colored markers paves the way for genome-wide identification of any other interacting loci responsible for the hybrid incompatibilities.
Caenorhabditis briggsae appeared to develop native immunity against PEEL-1 For transgene insertion using MosSCI or miniMos in C. elegans, a sperm-derived toxin gene, peel-1, is used as a negative selection marker to effectively kill extrachromosomal array-bearing worms (Seidel et al. 2011;Frøkjaer-Jensen et al. 2012, greatly reducing the burden of screening for transgenic strains carrying an insertion out of all transgenic animals, including those carrying an extrachromosomal array. To adopt the negative selection marker in C. briggsae, we first compared the killing effect of PEEL-1 in C. elegans and C. briggsae through its forced expression driven by a C. elegans heat shock promoter, Phsp-16.41. We generated transgenic lines carrying extrachromosomal array consisting of the PEEL-1-expressing vector in both C. elegans and C. briggsae. The synchronized L1 animals were subjected to heat shock at 34 C and 35 C for C. elegans and C. briggsae, respectively. The killing effect in C. elegans was comparable to that reported previously (Seidel et al. 2011), but no apparent killing was observed in C. briggsae after heat shock treatment (Fig. 5a).
To further investigate what caused the failure of PEEL-1 to kill C. briggsae, we replaced C. elegans heat shock promoter, , with its C. briggsae equivalent (Supplementary Fig. 2) and generated the transgenic lines. Again, we did not observe significant increase in killing (Fig. 5a). We reasoned that the C. briggsae heat shock promoter might not be a functional equivalent of . To examine whether the C. briggsae syntenic heat shock promoter responds to heat shock treatment, we generated transgenic lines carrying extrachromosomal array consisting of a fusion between the C. briggsae heat shock promoter and GFP. We did see induced GFP expression in the transgenic animals after . Y-axis denotes insertion frequency in P 0 animals. X-axis indicates various promoters used in the previous and this study that drive Mos1 transposase expression. Each fluorescent marker in this study was tested for at least 3 replicates. Animals with or without heat shock are indicated with hs(þ) and hs(À), respectively. Student's t-test was performed for comparison between animals or without heat shock. Error bars represent SE.
heat shock treatment in both C. elegans and C. briggsae (data not shown), suggesting that C. briggsae somehow develops immunity against PEEL-1. To further investigate C. briggsae's immunity against PEEL-1, we generated transgenic C. briggsae animals expressing peel-1 driven by an effective promoter, Peft-3, which was known to able to drive expression in C. briggsae (Frøkjaer-Jensen et al. 2012. Again, we did not observe a significant killing effect (Fig. 5b). Taken together, it seems that C. briggsae develops native immunity against PEEL-1, preventing it from being used as a negative selection marker against extrachromosomal array.
Limited success of using sup-35 as a negative selection marker for extrachromosomal array In addition to peel-1, a maternal-effect toxin, SUP-35, that kills developing embryos, has recently been identified in C. elegans (Ben-David et al. 2017). To test the killing efficiency of SUP-35 in C. briggsae, we made a sup-35-expressing vector driven by the C. elegans eft-3 promoter, Peft-3 ( Supplementary Fig. 1). We injected the construct along with fluorescence coinjection markers into both C. elegans and C. briggsae. As expected, nearly all F 1 embryos expressing the fluorescence markers arrested at late embryogenesis or as early larvae for both C. elegans and C. briggsae (Fig. 5c), indicating that SUP-35 is an effective toxin in C. briggsae and has a potential to be developed as a negative selection marker against extrachromosomal array in C. briggsae. To this end, we generated transgenic strains in C. elegans and C. briggsae that carry an extrachromosomal array, which consists of the sup-35-expressing vector driven again by C. briggsae equivalent of hsp-16.41 promoter ( Supplementary Fig. 2). However, we did not observe any significant killing effect in C. elegans after heat shock treatment (Fig. 5d). Caenorhabditis elegans expresses both PEEL-1 and its antidote ZEEL-1 only transiently in embryo, but expresses SUP-35 and its antidote PHA-1 throughout its life cycle (Gerstein et al. 2010). It is possible that the postembryonic PHA-1 is sufficient to neutralize the toxicity of SUP-35 induced by heat shock treatment in C. elegans. It is also possible that the C. briggsae heat shock promoter may not respond to heat shock as effectively as its C. elegans equivalent in C. elegans. Consistent with this, the heat shock treatment showed a significant killing effect in C. briggsae carrying the same construct, i.e. from 66.4% to 39.6% though it is not efficient enough to serve as a negative selection marker.

Discussion
Given the similar morphology, physiology and developmental program between C. elegans and C. briggsae, methods for C. elegans transgenesis are expected to be transferrable to C. briggsae with minimal modification. However, there is an exception to this, which is the case for miniMos mediated single-copy insertion (Frøkjaer-Jensen et al. 2014).

Potential cause of low efficiency of transgenesis in C. briggsae mediated by miniMos
It has been well established that transposon mobility can be significantly increased across species after exposure to biotic or abiotic stress (Liu et al. 1995;Walbot 1999;Bouvet et al. 2008). Heat shock proteins serve as molecular chaperone to facilitate folding of other proteins into their appropriate conformations following heat exposure, thus buffering the subsequent phenotypic changes (Erlejman et al. 2014). Perturbation of heat shock protein also increases transposition frequency in a similar way to that of heat treatment itself. It is possible that under optimal growth condition, C. briggsae has a more robust system to maintain genome integrity in its germline than C. elegans, which prevents its genome from environment-induced damage or editing. Improvement of transgenesis efficiency by heat shock treatment or inhibition of heat shock pathway could be at the cost of compromised genome integrity. One important way to preserve genome integrity in the germline is through Piwi-interacting RNA (piRNA), which constitutes one of the major regulatory molecules that curb the transposon activities (Ishizu et al. 2012). Consistent with this, functional perturbation of Hsp90 attenuates the piRNA silencing mechanism, leading to transposon activation and the induction of morphological mutants in Drosophila (Specchia et al. 2010). It is unclear in C. elegans whether the heat shock treatment leading to increased transposon mobilization. This is because the PRG-1 piRNA mutant does not show increased transposon mobilization even after many generations (Wahba et al. 2021).
Differential responses to PEEL-1 and SUP-35 between C. elegans and C. briggsae It is intriguing that C. briggsae responds differentially to the overexpressed paternal toxin PEEL-1 and maternal toxin SUP-35 from C. elegans. In C. elegans, PEEL-1 was suspected to function as a calcium pump on the cell membrane by generating a membrane pore, which leads to the release of intracellular calcium (Seidel  Table 1. Insertion site was determined using 2 rounds inverse PCR as described previously. et al. 2011). The maternal-effect toxin, SUP-35, kills developing embryos, but the molecular mechanism of its toxicity remains unclear (Ben-David et al. 2017). Our data showed that C. briggsae somehow develops native immunity against PEEL-1 but not SUP-35 (Fig. 5). However, forced expression of SUP-35 driven by the C. briggsae heat shock promoter did not mediate complete killing of C. briggsae after heat shock treatment (Fig. 5d). Two possible reasons account for the ineffective killing. First, time windows for SUP-35 killing seem to be narrow, i.e. at certain stage of embryogenesis. This is why a constitutive promoter, i.e. eft-3 promoter is able to drive SUP-35 expression and mediate complete killing. For the heat shock promoter, its response to heat shock may be more  Fig. 4. Use of dual-color fluorescent markers in mapping interacting loci responsible for rescue of HI phenotype. a) Schematics of introgression strains expressing GFP marker (green) generated previously with biolistic bombardment or mCherry marker (red) generated with miniMos in this study. Strains ZZY10330 and ZZY10377 carry an X-linked introgression derived from C. briggsae that produces GFP or RFP expression, respectively, and HI phenotype in an otherwise C. nigoni background, i.e. hybrid male sterility. Strains ZZY10353 and ZZY10382 carry a Chromosome II-linked introgression derived from C. briggsae that produces GFP or RFP expression, respectively, and HI phenotype in an otherwise C. nigoni background, i.e. homozygous inviable (data not shown). b) Schematics of crossing strategy in mapping interacting loci between X chromosome and Chromosome II. We previously demonstrated that presence of the introgression fragment from ZZY10353 rescued the male sterility of ZZY10330. However, this demands tedious genotyping of ZZY10353 to ensure its presence because it is impossible to distinguish the 2 introgressions both expressing GFP. Substitution of ZZY10353 with ZZY10382 expressing RFP greatly facilitated the process for screening for simultaneous presence of the 2 introgressions. c) Reciprocal crossing between strains with autosome-and X-linked introgressions expressing GFP and RFP, respectively, serves as the same purpose as in (b).
effective during larval development than embryogenesis. Second, heat-shock induced expression of SUP-35 may not provide enough dosage to kill C. briggsae larvae. Further optimization of heat shock condition or choice of a different heat-shock promoter is necessary to develop an effective negative selection marker against extrachromosomal array in C. briggsae.

Data availability
Strains and plasmids are available upon request. Supplementary  Figure 1 contains configurations of plasmids used in the study. Supplementary Figure 2 shows the schematics of steps in identifying C. briggsae heat shock promoter and the sequence of C. briggsae heat shock promoter tested in this study. Supplementary

Conflicts of interest
None declared.

Literature cited
Ben n.s. ** Fig. 5. Comparison of killing efficiencies of PEEL-1 (a and b) and SUP-35 (c and d) in C. elegans (N2) and C. briggsae (AF16). a) Fraction of surviving arraybearing animals with (grey) and without (black) heat shock in C. briggsae (AF16), which carries a peel-1 driven by heat shock promoter as an extrachromosomal array as indicated. Note, no significant change in survival using hsp-16.41 promoter of from C. elegans and CBG19187(cbr-hsp-16.60) promoter from C. briggsae. b) Fraction of arrested F 1 animals presumably carrying an injection marker derived from P 0 , which was injected with peel-1 driven by a constitutively expressing promoter, Peft-3. Note the incomplete killing of PEEL-1 expressing animals in C. briggsae. c) Fraction of arrested F 1 animals presumably carrying a co-injection marker derived from P 0 , which was injected with sup-35 driven by Peft-3. Note the complete killing of sup-35 expressing animals in both C. elegans and C. briggsae. d) Fraction of surviving array-bearing animals with (gray) and without (black) heat shock in C. elegans (N2) and C. briggsae (AF16), which carries a sup-35 driven by a heat shock promoter from CBG19187(cbr-hsp-16.60) as an extrachromosomal array as indicated. Note, no significant change in survival in C. elegans and a significant drop in C. briggsae survival after heat shock.