Skip to main content
NCBI home page
Genetics logoLink to Genetics
. 2015 Jul 10;201(1):39–46. doi: 10.1534/genetics.115.179648

Computer-Assisted Transgenesis of Caenorhabditis elegans for Deep Phenotyping

Cody L Gilleland *, Adam T Falls *, James Noraky *, Maxwell G Heiman †,1, Mehmet F Yanik *,‡,§,1
PMCID: PMC4566274  PMID: 26163188

Abstract

A major goal in the study of human diseases is to assign functions to genes or genetic variants. The model organism Caenorhabditis elegans provides a powerful tool because homologs of many human genes are identifiable, and large collections of genetic vectors and mutant strains are available. However, the delivery of such vector libraries into mutant strains remains a long-standing experimental bottleneck for phenotypic analysis. Here, we present a computer-assisted microinjection platform to streamline the production of transgenic C. elegans with multiple vectors for deep phenotyping. Briefly, animals are immobilized in a temperature-sensitive hydrogel using a standard multiwell platform. Microinjections are then performed under control of an automated microscope using precision robotics driven by customized computer vision algorithms. We demonstrate utility by phenotyping the morphology of 12 neuronal classes in six mutant backgrounds using combinations of neuron-type-specific fluorescent reporters. This technology can industrialize the assignment of in vivo gene function by enabling large-scale transgenic engineering.

Keywords: C. elegans, high throughput, transgenesis, deep phenotyping, automation

RECENT advances in sequencing technology have made it feasible to genotype hundreds of thousands of patients to identify disease-relevant human gene variants. To take advantage of this new information and rapidly probe gene function there is a need for large-scale high-throughput genetic modification and phenotyping of model organisms. Its small size, ease of culture, rapid development, facile genetics, and large mutant libraries make Caenorhabditis elegans a powerful model organism for assigning functions to genes relevant to human disease (Shaye and Greenwald 2011). This organism is also a paramount model for discovering basic biological mechanisms. For example, the neurodevelopmental guidance molecule Netrin (Ishii et al. 1992), the programmed cell death caspases and their regulators (Yuan and Horvitz 1990; Yuan et al. 1993), the role of small RNAs (Lee et al. 1993; Wightman et al. 1993), regulators of organismal aging (Kenyon et al. 1993; Kimura et al. 1997), and many other key pieces of biology have emerged first from studies in this simple model organism.

A major strength of C. elegans is the wealth of vector and strain resources available. Vector resources include the Promoterome (Dupuy et al. 2004), a collection of 5526 predicted promoter sequences; the ORFeome (Reboul et al. 2003; Lamesch et al. 2004), a collection of 12,625 cDNAs; and the TransgeneOme (Sarov et al. 2012), a collection of 16,102 genomic DNA clones in which protein coding sequences have been tagged with GFP and an affinity epitope. Strain resources include 6841 “knockout” strains generated by the C. elegans Deletion Mutant Consortium (C. elegans Deletion Mutant Consortium 2012), each of which carries a small genetic deletion disrupting 1 of 6013 genes; 2007 fully sequenced strains generated by the Million Mutation Project (Thompson et al. 2013), which together carry 183,327 nonsynonymous mutations in 19,666 protein-coding genes, including 12,594 predicted knockout mutations in 8150 genes; and ∼10,000 other strains that have been generated by independent laboratories and made publicly available through the C. elegans Genetics Center (https://www.cbs.umn.edu/research/resources/cgc). These resources, in addition to tools such as RNAi (Fire et al. 1998; Kamath and Ahringer 2003), CRISPR/Cas9 (Friedland et al. 2013; Waaijers et al. 2013), and cell-specific fluorescent markers—including optogenetic (Nagel et al. 2005) and calcium-imaging reagents (Kerr et al. 2000; Tian et al. 2009)—have made C. elegans one of the richest genetic models available.

To fully exploit these resources, it is necessary to systematically introduce collections of genetic vectors into panels of mutant strains (Figure 1A). For example, to assign functions to genes, one could introduce a collection of cell-type-specific fluorescent reporters into a panel of mutants for “deep phenotyping” of multiple cellular features simultaneously as demonstrated in Figure 1B. Or, a collection of epitope-tagged DNA-binding factors could be introduced into a panel of mutants and assayed using chromatin immunoprecipitation and DNA sequencing (ChIP-Seq) to detect altered patterns of DNA binding. Or, a collection of neuron-specific optogenetic reagents could be introduced into a panel of synaptic mutants to identify behavior-relevant deficits. The possibilities for this kind of experiment are limitless.

Figure 1.

Figure 1

Open in a new tab

C. elegans transgenesis process and deep phenotyping. (A) Large-scale resources of genetic vectors and mutant strains are available and must be brought together via transgenesis (microinjection) to enable high-throughput phenotypic analysis. (B) Deep phenotyping of sensory neuron morphology in the head of wild-type animals generated by our computer-assisted microinjection (CAMI) platform using CFP, YFP, and mCherry cell-type-specific reporters. Bar, 50 µm.

However, a major bottleneck to these approaches is the rate-limiting step of transgenesis by microinjection (Mello et al. 1991), an immensely useful technology that has not changed since it was developed >20 years ago. Although C. elegans offers the fastest transgenesis of any animal model, manual injections remain laborious and low throughput, and thus are not amenable to the kind of large-scale screening applications available for bacteria, yeast, or tissue culture. Conventional C. elegans transgenesis involves (1) using a dissecting microscope to mount 1–10 animals in a drop of oil on an agar pad on a coverslip, (2) transferring the coverslip to an inverted microscope equipped with a 40× objective and micromanipulator, (3) using a fine glass needle loaded with DNA solution to penetrate and fill the gonad of each animal, and (4) returning to the dissecting microscope and recovering the injected animals to standard growth medium. This procedure must be performed quickly, as animals begin to desiccate within ∼10 min under oil. New trainees typically require 2–3 weeks to learn this technique, and even experienced investigators rarely inject more than four to six strains per day due to fatigue from the labor-intensive process. A high-throughput platform for transgenesis could have a major amplifying effect by allowing researchers to fully exploit the massive vector and strain resources available in this powerful model system.

Materials and Methods

Maintenance and imaging of C. elegans

Strains were constructed in the N2 background and cultured under standard conditions (Brenner 1974; Stiernagle 2006). Plasmids, transgenes, and strains are listed in Supporting Information, Table S2, Table S3, and Table S4. Some strains were provided by the Caenorhabditis Genetics Center (CGC), which is funded by National Institutes of Health Office of Research Infrastructure Programs (P40 OD010440). Images were collected using a Deltavision Core Imaging System (Applied Precision) using a 40× 1.35 NA objective (Olympus). They were deconvolved and maximum-intensity projections were generated using softWoRx Suite 1.2. The resulting images were pseudocolored and assembled using Photoshop CS5 (Adobe).

Mounting gel preparation

The mounting gel preparation is composed of a mixture of Pluronic F-127 hydrogel (25%) and sodium azide anesthetic (10 mM) in deionized water. The hydrogel (pluronic F-127, poloxamer 407) and anesthetic (sodium azide) reagents are acquired in powder form. A stock solution of 100 mM sodium azide is prepared for later addition to the pluronic solution by adding 650 mg of sodium azide powder (caution: toxic, use gloves and avoid contact with skin) to 100 ml of deionized water. The pluronic F-127 solution is prepared by adding 80 g of powder into 288 ml of chilled deionized water (4°) in a 500-ml flask. The pluronic F-127 powder is dissolved by storing the flask at 4° for 2 days and shaking vigorously by hand twice per day or by storing in a chilled shaking incubator (4°) overnight. Once the pluronic powder is fully dissolved, add 32 ml of the 100 mM sodium azide solution. The flask is gently swirled to mix the contents yet avoid the creation of bubbles. Store the flask at 4° between each use. This prepared batch can be used for roughly 100 sets of injections.

Large-scale immobilization of C. elegans flat against glass surface

The procedure for mounting animals involves dispensing 3 ml of chilled pluronic gel (4°) onto a small agar dish containing synchronized animals, pouring the resulting mixture into a multiwell plate with a glass bottom, incubating the well plate at 15° (6 min) to allow the animals to settle against the glass, and then applying heat from below via a peltier warmer inside the incubator to quickly harden the pluronic gel and immobilize the animals flat against the glass. The peltier unit is positioned between two copper blocks (1/4 inch thickness) to ensure uniform heat distribution and a plastic plate cover is placed on top of the multiwell plate to retain heat and moisture during warming with a water level on top to ensure uniform distribution of the hydrogel. Alternatively, a programmable thermal cycler commonly used in PCR can be used to harden the gel during the immobilization process (See Supplementary). The multiwell plate is then transferred to the microscope for microinjection without disrupting animal positions. A temperature-controlled incubator maintains the hydrogel at a constant 25° throughout the injection process by feeding warm air to the bottom of the stage surrounded by a plastic sheet ensuring uniform heat distribution. We developed custom multiwell plate configurations with expanded width (three wells: 2 × 12 area of 96-well format, see Figure 3A). This configuration provides large needle-accessible areas and enables simultaneous animal mounting for three independent experiments with large populations. However, commercially available multiwell plates providing sufficient clearance could be substituted.

Figure 3.

Figure 3

Open in a new tab

Algorithms for precision mapping and targeting of C. elegans gonads at large scale. (A) The entire multiwell plate of immobilized worms is scanned at high speed (blue and red arrows) and stitched to create a high-resolution montage to map worm positions with single-micrometer accuracy under 2× magnification. (B) The montage is generated by rapidly stitching individual rows (red and blue areas) in open-loop via precision timing of XY stage movement and camera frame rate. Each compiled row is overlapped (purple areas) with the lower compiled row by 20% and stitched using computer vision registration algorithms. (C) Automated identification of gonad targets from stitched images. ***, stitch line. The gonads are targeted by first drawing a spline (red) down the center of the animals then creating a region of interest (green box) around a specified distance along the spline (yellow circle). The regions of interest are further processed to detect the gonad targets for microinjection. Top right box indicates preferred orientation for injection. (D) The gonad target locations are then mapped and calibrated to XY stage coordinates and a greedy nearest neighbor algorithm determines the optimal path.

Clearing plasmid of debris

To allow use of needles with submicrometer inner diameters and avoid interruptions due to clogging, all plasmids were preprocessed with a centrifugation procedure prior to loading needles. Specifically, each plasmid mixture was centrifuged for 20 min (25,000 relative centrifugal force, rcf, 4°), and the top 90% of the supernatant was removed and transferred to a clean tube. This centrifugation/transfer step was repeated a second time. Plasmids were stored at −20° between uses. Each day a plasmid was subsequently used for microinjections, it was thawed and centrifuged for 10 min (25,000 rcf, 4°). Using a thin-tipped pipette, 1 µl of plasmid is extracted from the top 25% of the solution and then loaded from the back of the needle directly into the tip.

Clearing a clogged needle

The constant back pressure (∼3 Psi) and submicron tip diameter prevent debris from entering the front of needle while the above plasmid cleaning protocol prevents clogging of the needle by the debris within the plasmid mixture. The high injection dispense pressure (∼60 Psi) and piezo vibration also help to remove any material from the tip during normal operation. An injection needle that is left outside the hydrogel after injections can be clogged with dry hydrogel around the tip. To remove dried hydrogel, soak the needle tip in a droplet of deionized water to rehydrate the dry gel and remove the clog. To prevent clogging from dried hydrogel the system automatically performs a dispense operation before and after the needle enters the gel. If the needle clogs during injections, visualize the clog by adjusting the focus knob, manually adjust the injection dispense pressure to maximum pressure (∼90 Psi) and the dispense duration to 10 sec. Dispense the needle repeatedly until the clog is removed, then return to normal operation after adjusting the injection pressure back to normal operating pressure (∼60 Psi). If this does not remove the clog, then replace the needle.

Hardware components

See File S1, Table S1 for detailed parts list with sources. A summary of parts include Prior ProScanII stage with controller, Nikon Ti Eclipse microscope with Perfect Focus laser system and DIC optics, Sutter MP-285 manipulator, pressure regulators and gauges (Parker Watts model no. R364-02C), Eppendorf universal capillary holder (cat. no. 920007392), high-speed camera (Allied Vision GX2300), temperature-regulated incubator (In Vivo Scientific), piezoelectric vibrator (RadioShack item no. 273-059), antivibration air table (Newport RS4000), digital output card (National Instruments, USB-9162), DC power supply (30V-5A), computer (PowerSpec G212, 16 GB RAM), cover glass-bottom well plates (In Vitro Scientific item no. P06-1.5H-N), needle puller (Sutter P-97), microneedles (World Precision Instruments, standard glass capillaries, item no. 1B100F-4).

Data availability

Plasmids and strains are available upon request. File S1 contains instructions for assembling and using the system. Table S1 contains a purchasing list for all hardware components. Software is available through GitHub (https://github.com/CodyLGilleland/CAMI_Gilleland_2015_GENETICS.git). Plasmids, transgenes, and strains used in this study are listed in Tables S2, S3, and S4 respectively.

Results

To address the need for rapid transgenics, first we developed the computer-assisted microinjection (CAMI) system. Next, we demonstrated the utility of the CAMI method by creating a collection of cell-type-specific fluorescent transgenes, labeling 12 classes of sensory neurons (Figure 1B) and using them to screen a small panel of mutants for defects in neuronal morphogenesis. CAMI is scalable and thus allows transgenesis to be industrialized for large-scale applications. The overall workflow of CAMI resembles the conventional method—animals are mounted on a slide and transferred to an inverted microscope, a glass needle filled with DNA solution is used to microinject the gonad, and then animals are recovered—but CAMI introduces a scalable method of mounting and recovering animals, and a computer-assisted gonad targeting and microinjection system.

In contrast to the conventional oil-mounting protocol, CAMI makes use of a temperature-sensitive hydrogel (Ko and Van Gundy 1988) (poloxamer 407, pluronic F-127) to pipette animals from standard culture on agar plates to a multiwell plate with a thin cover glass bottom. Because standard liquid-handling approaches and multiwell plates are used, this gel-mounting procedure can be automated by commercially available robotics. Recently, capturing C. elegans in hydrogels by optical heating has been used for subsequent manual microinjection but the transgenesis method remained low throughput and labor intensive (Hwang et al. 2014). In the CAMI platform, first all animals are simultaneously and easily confined to a uniform image plane near the cover glass: To achieve this, the multiwell plate is placed on a peltier heating unit at 15°, then animals are allowed to settle flat by gravity, and finally the hydrogel is quickly heated to 25° causing it to harden, immobilizing the animals, and trapping them against the surface of the slide (Figure 2A). The hardened gel, combined with 10 mM sodium azide as anesthetic, causes the animals to adopt a straightened conformation, simplifying gonad targeting and needle penetration. The gel concentration is optimized to be stiff enough to restrict animal movement while still permitting the fine glass injection needle to travel through the gel without bending. Because the gel provides moisture to prevent desiccation, the animals can remain immobilized greater than 45 min with no loss of viability, compared to conventional mounting in which all animals die in <30 min (Figure S4E). We designed needles (Figure 2B, inset) with sufficiently long and thin penetrating shaft (i = 25 µm) to minimize tissue damage and rapid taper (ii = 100 µm) to prevent bending and enable rapid penetration. The needles are pulled with an open-end tip with submicrometer diameter to eliminate the need for manual needle breaking and thus reduce variability. We also used procedures to prevent clogging of needles (see Materials and Methods).

Figure 2.

Figure 2

Open in a new tab

Computer-assisted microinjection (CAMI) platform. (A) Schematic of the gel mounting procedure. All animals in the multiwell plate are simultaneously immobilized flat against the glass within a single image plane enabling rapid XY scanning and mapping of the entire multiwell plate (see Figure 3A). (B) Overview of the instrumentation for the CAMI platform. Dashed box indicates animal undergoing microinjection. Microneedle inset: (i) Indicates penetrating needle tip. (ii) Indicates rapid taper. Bar, 50 µm.

Next, the multiwell plate is transferred to a temperature-controlled microscope stage for microinjection. The CAMI hardware system consists of a high-speed camera mounted on an automated inverted microscope with objective turret, DIC optics, robotic XY stage, temperature controlled incubator, micromanipulator for needle positioning, piezoelectric unit for needle vibration pulsing, and pneumatic valves for pressurized dispensing of reagents (Figure 2B). The multiwell plate is scanned at high speed under a 2× objective to collect an image of the entire plate at ∼1-µm resolution, allowing fine anatomical features including gonad positions to be discerned automatically. Computer vision algorithms detect animal locations, define the target gonad locations, and plan an efficient travel path between targets using a nearest neighbor optimization (Arya et al. 1998) (Figure 3). Once targets are mapped at 2×, the system can then remain at 20× magnification throughout the entire set of injections while traveling from one predetermined target to another, which avoids time delays due to objective changing and refocusing (see File S2, Supplementary Video).

The computer then rotates the objective turret to 20× for microinjection, and begins to visit each gonad target. The autofocus laser system brings the gonad into view and, if necessary, the user can make a small XY alignment using only a single mouse click, which also initiates a digital zoom feature (Figure 4A). The needle is brought into the image plane enabling axial penetration of the cuticle and gonad sheath with the assistance of piezoelectric vibration, simulating the manual tapping used in a standard microinjection procedure (Figure 4B). After penetrating the gonad, a single mouse click is used to dispense the DNA mixture followed by axial removal of the needle (Figure 4C). The stage then travels to the next target where the injection procedure is repeated. After all injections are complete, the hydrogel is diluted with M9 medium and placed in a shaking incubator at 15° for 10 min to return the gel to liquid state, following which animals can be pipetted or poured onto standard growth medium for recovery.

Figure 4.

Figure 4

Open in a new tab

Computer-assisted delivery of genetic vectors via gonadal microinjection. The schematic (left) illustrates a cross-sectional view of an animal during the injection process while the images (right) demonstrate microinjection under 20× DIC objective using the CAMI platform. (A) The autofocus laser brings the gonad (green overlay) into focus, ΔF indicated set distance above glass surface. The cross indicates target for injection and the inset is magnified to show honeycomb pattern of the gonad nuclei. Bar, 50 µm; inset, 10 µm. (B) The microneedle penetrates the cuticle and gonad sheath (green overlay) from the right side via precision robotic control and piezoelectric vibration with a 5-µm overshoot (ΔO) to ensure full penetration. A constant back pressure (3 Psi) dispenses a small amount of plasmid (purple overlay) into the gonad, which indicates correct positioning and prevents biological debris from clogging the needle. The magnified inset shows needle penetration. (C) The plasmid (purple overlay) is then dispensed at high pressure (60 Psi) over a short duration (200 ms) as the needle is retracted 5 µm (ΔR) to free the needle tip from obstruction. The needle is then fully retracted above the gel along its 45° axis. The magnified inset shows the minor damage at the entry point of the needle.

CAMI operates at a rate of ∼25 sec per gonad target with minimal user fatigue, in contrast to conventional manual microinjection at ∼2–3 min per animal where, in addition, most users cannot sustain injecting for more than 1–2 hr (Table 1). Mounting and recovery of animals brings the total time per set of injections to ∼20–30 min; however, recovery can be performed in parallel with mounting for the successive set of injections (see Supporting Methods). Typically, one needs to inject ∼15 animals to ensure obtaining a transgenic line. Importantly, however, recent genome-editing techniques such as CRISPR/Cas9-mediated genome editing require larger cohorts of ∼50 animals, which are challenging using manual methods but are easily accommodated using CAMI.

Table 1. Microinjection timing.

Step Mean time ± SD (sec)a
1. Stage XY travel to target 2 ± 0.5
2. Z-focus/fine alignment 5 ± 1
3. Needle insertion 10 ± 1
4. Plasmid dispense 5 ± 1
5. Needle exit 3 ± 0.5
Total 25 ± 4
Open in a new tab
a

Timing rounded to nearest half second.

To quantify the transgenesis efficiency using CAMI, we replicated a classic experiment that was used to establish the conventional transgenesis protocol (Mello et al. 1991). In this experiment, a 100 µg/ml solution of the plasmid pRF4 bearing a semidominant allele of the rol-6 gene was injected into the gonads of young adult animals. The F1 progeny that bear the transgene exhibit an easily scored “roller” (Rol) phenotype, in which animals crawl in circles. Most of these transgenic F1 progeny are transient, and only a small fraction transmit the transgene to the F2 generation. In three independent trials, we injected 10–15 gonads each and obtained 19.0 ± 1.4 transgenic F1 animals per gonad, comparable to a published rate of 12.5 F1 animals per gonad by the conventional method (Table 2). We found that 6.0 ± 2.5% of these F1 animals transmitted the transgene to the F2 generation, comparable to a published rate of 9.4% by the conventional method (Table 2). Thus, transgenesis by CAMI occurs at an efficiency well in line with conventional techniques.

Table 2. Transformation efficiency.

Trial Single gonad injections F1 Rollers Heritable expression F2 lines/total F1 (%)
Total Average per injected gonad Maximum per injected gonad
1 14 266 19 46 15/266 (5.6%)
2 15 272 18 79 10/272 (3.7%)
3 10 208 21 46 18/208 (8.7%)
Total 39 746 19 ± 1.4a 79 43/746 (6 ± 2.5%)a
Open in a new tab

A standard assay of rol-6 plasmid expression efficiency was conducted to determine the efficiency of the injection process. In each trial 100 µg/ml of rol-6 was injected into single gonads. The resulting F1 transformants were placed on individual agar plates and heritable expression was scored in the F2 generation.

a

Mean ± SD.

To test if this method can be used effectively to combine a vector and strain library as illustrated in Figure 1A, we performed a small screen focused on sensory neuron development. This screen was motivated by the following rationale: first, in a previous genetic screen we found that the zona pellucida (ZP) domain protein DYF-7 is required for sensory neurons to extend their dendrites to the nose tip (Heiman and Shaham 2009); second, ZP domain proteins are secreted factors that often heteromultimerize into a matrix; third, the C. elegans genome encodes dozens of predicted ZP domain proteins, raising the possibility that other ZP domain proteins may cooperate with DYF-7 to promote sensory dendrite extension. Therefore, we wished to determine if mutants in other ZP domain proteins might exhibit defects in sensory dendrite extension similar to those seen in the absence of DYF-7.

As a vector library, we generated a collection of plasmids bearing cell-specific markers to individually label most of the head sensory neurons. These include glial-ensheathed neurons—namely, 3 of the 12 amphid neurons (ADF, AWC, and AFD) and all of the outer and inner labial and cephalic neurons (OLL, OLQ, IL1, IL2, and CEP)—as well as non-glial-ensheathed neurons that also extend dendrites to the nose tip (BAG, FLP, URX, and URY). Together, these markers allow us to assay morphology of 38 individual neurons in 12 classes. We divided these markers into four plasmid mixes, each containing CFP, YFP, and mCherry-based plasmids (Figure 1B and Table S2). As a mutant library, we selected a small panel of mutants that represent the three major C. elegans strain collections: publicly available mutants generated by individual laboratories (ram-5) (Yu et al. 2000), deletion mutants generated by the North American and Japanese knockout consortium (cut-1, cut-5, and cut-6), and predicted null mutants identified through the Million Mutation Project (cutl-14 and T23F1.5) (see Table S4).

We used CAMI to introduce each of the four plasmid mixes into each of the six mutant strains, generating 24 lines, allowing us to look for phenotypes in a total of 78 mutant neuron classes (4 transgenic lines × 6 mutants × 3 neuron classes per line) (Figure 5 and Table S4). Although we observed very low penetrance defects in CEP neurons of the cut-5 mutant strain (not shown), in most cases these mutants had no effect on neuronal morphology, suggesting that ZP domain proteins do not broadly contribute to sensory neuron dendrite extension and thus that the phenotypes we observe with DYF-7 are likely to be specific.

Figure 5.

Figure 5

Open in a new tab

Deep phenotyping of sensory neuron morphology. A total of 24 transgenic strains were generated, each consisting of a plasmid mix bearing fluorescent markers of three sensory neuron classes, indicated at left, introduced by microinjection into a different mutant strain, as indicated at top (see Table S4). Dendrite length, morphology, and position were analyzed in at least 25 individuals per strain. Representative images are shown.

Discussion

Microinjection is a powerful method for manipulating model organisms. For example, several groups including ours have developed automated microinjection of zebrafish embryos (Wang et al. 2007; Hogg et al. 2008; Spaink et al. 2013; Chang et al. 2014). However, the small size of C. elegans and its special handling requirements have rendered it previously inaccessible to such higher-throughput microinjection. Here, we have demonstrated how CAMI can be used to simplify and accelerate microinjection, and we used it to screen for novel phenotypes in a panel of C. elegans mutants and thus assign new functions to genes. This method can similarly be used to deliver small molecules in a screen for novel therapeutics, to deliver RNAi constructs in cases where traditional feeding RNAi is not potent enough, to perform large-scale genome editing using CRISPR/Cas9 or TALENs, or to assay the pathogenicity of human gene variants.

The hardware set-up cost of the system as presented (∼$100,000, File S1, Table S1) is suitable for a core facility, and roughly comparable to employing an injection technician for 1–2 years, but will be prohibitive for smaller labs. Future improvements can certainly allow for less expensive microscope systems and robotics and can also be integrated with more streamlined software. Additional modifications can increase the speed of the system; for example, the ability to flush the needle would allow plasmid mixes to be changed automatically as we demonstrated in another previous study (Steinmeyer and Yanik 2012). Combining this system with recently reported antibiotic selection vectors for C. elegans can speed the isolation of transgenic lines following microinjection (Cornes et al. 2014). Portions of the CAMI protocol can be adapted to existing microinjection systems; for example, the hydrogel mounting methods can, in principle, be used with manual injection systems, provided that the micromanipulator offers “on axis” movement in addition to more conventional XYZ movement. It could also be combined with recently reported manual optical immobilization of C. elegans in hydrogel (Hwang et al. 2014). Bombardment is another technique that can potentially be used for large-scale transgenesis (Semple and Lehner 2014); however, bombardment requires the transgenes and selection markers to be cloned into a single plasmid, and thus is not compatible with existing vector libraries and lacks the versatility of plasmid mixtures (Wilm et al. 1999). Bombardment also produces low-copy number transgenes, which can provide more physiological expression but are not compatible with visualizing weaker fluorescent markers. The facile transgenesis offered by CAMI brings the field one step closer to Brenner’s pioneering goal of “microbiologizing” an animal system to be as manipulable as bacteria (Wood 1988).

Supplementary Material

Supporting Information
supp_201_1_39__index.html (1.3KB, html)

Acknowledgments

This work was supported by a National Institutes of Health (NIH) New Innovator Award (1-DP2-OD002989), NIH Transformative Research R01 Award (R01 NS073127), Packard Award in Science and Engineering, Sloan Award in Neuroscience, National Science Foundation (NSF) Career Award, NIH Director's Pioneer Award (DP1 OD006782), and an NIH R01 Award (R01 GM108754). C.L.G. was supported by an NSF graduate research fellowship and an NIH biotechnology training grant. We thank Christoph Engert of the Horvitz lab and Joseph Steinmeyer of the Yanik lab for helpful discussions.

Author contributions: C.L.G. designed the microinjection platform, developed the hydrogel immobilization technique and wrote the supplementary information. C.L.G., A.T.F., and J.N. developed algorithms for precision gonad targeting of injections. C.L.G. and A.T.F. performed all microinjection screening experiments and designed microneedle parameters. M.G.H. designed the biological assay, prepared all plasmids, and imaged resulting transgenic strains. C.L.G., M.G.H., and M.F.Y. wrote the manuscript. M.G.H. and M.F.Y. supervised the research at all times.

Footnotes

Communicating editor: O. Hobert

Supporting information is available online at www.genetics.org/lookup/suppl/doi:10.1534/genetics.115.179648/-/DC1.

Literature Cited

  1. Arya S., Mount D. M., Netanyahu N. S., Silverman R., Wu A. Y., 1998.  An optimal algorithm for approximate nearest neighbor searching fixed dimensions. J. Assoc. Comput. Mach. 45: 891–923. [Google Scholar]
  2. Brenner S., 1974.  The genetics of Caenorhabditis elegans. Genetics 77: 71–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. C. elegans Deletion Mutant Consortium, 2012.  Large-scale screening for targeted knockouts in the Caenorhabditis elegans genome. G3 (Bethesda) 2: 1415–1425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Tsung-Yao C., Shi P., Steinmeyer J. D., Chatnuntawech I., Tillberg P., et al. , 2014.  Organ-targeted high-throughput in vivo biologics screen identifies materials for RNA delivery. Integr. Biol. 6: 926–934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cornes E., C. A. Quéré, R. Giordano-Santini, and D. Dupuy, 2014.  Applying antibiotic selection markers for nematode genetics. Methods 68: 403–408. [DOI] [PubMed] [Google Scholar]
  6. Dupuy D., Qian R. L., Deplancke B., Boxem M., Hao T., et al. , 2004.  A first version of the Caenorhabditis elegans promoterome. Genome Res. 14: 2169–2175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Fire A., Xu S., Montgomery M. K., Kostas S. A., Driver S. E., et al. , 1998.  Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391: 806–811. [DOI] [PubMed] [Google Scholar]
  8. Friedland A. E., Y. B. Tzur, K. M. Esvelt, M. P. Colaiácovo, G. M. Church et al, 2013.  Heritable genome editing in C. elegans via a CRISPR-Cas9 system. Nat. Methods 10: 741–743. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Heiman M. G., Shaham S., 2009.  DEX-1 and DYF-7 establish sensory dendrite length by anchoring dendritic tips during cell migration. Cell 137: 344–355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Hogg R. C., et al. , 2008.  An automated system for intracellular and intranuclear injection. J. Neurosci. Methods 169: 65–75. [DOI] [PubMed] [Google Scholar]
  11. Hwang, HKrajniak J., Matsunaga Y., G. M. Benian, and H. Lu, 2014.  On-demand optical immobilization of Caenorhabditis elegans for high-resolution imaging and microinjection. Lab Chip 14: 3498–3501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ishii N., Wadsworth W. G., Stern B. D., Culotti J. G., Hedgecock E. M., 1992.  UNC-6, a laminin-related protein, guides cell and pioneer axon migrations in C. elegans. Neuron 9: 873–881. [DOI] [PubMed] [Google Scholar]
  13. Kamath R. S., Ahringer J., 2003.  Genome-wide RNAi screening in Caenorhabditis elegans. Methods 30: 313–321. [DOI] [PubMed] [Google Scholar]
  14. Kenyon C., Chang J., Gensch E., Rudner A., Tabtiang R., 1993.  A C. elegans mutant that lives twice as long as wild type. Nature 366: 461–464. [DOI] [PubMed] [Google Scholar]
  15. Kerr R., Lev-Ram V., Baird G., Vincent P., Tsien R. Y., et al. , 2000.  Optical imaging of calcium transients in neurons and pharyngeal muscle of C. elegans. Neuron 26: 583–594. [DOI] [PubMed] [Google Scholar]
  16. Kimura K. D., Tissenbaum H. A., Liu Y., Ruvkun G., 1997.  daf-2, an insulin receptor-like gene that regulates longevity and diapause in Caenorhabditis elegans. Science 277: 942–946. [DOI] [PubMed] [Google Scholar]
  17. Ko M. P., Van Gundy S. D., 1988.  An alternative gelling agent for culture and studies of nematodes, bacteria, fungi, and plant tissues. J. Nematol. 20: 478–485. [PMC free article] [PubMed] [Google Scholar]
  18. Lamesch P., Milstein S., Hao T., Rosenberg J., Li N., et al. , 2004.  C. elegans ORFeome version 3.1: increasing the coverage of ORFeome resources with improved gene predictions. Genome Res. 14: 2064–2069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lee R. C., Feinbaum R. L., Ambros V., 1993.  The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell 75: 843–854. [DOI] [PubMed] [Google Scholar]
  20. Mello C. C., Kramer J. M., Stinchcomb D., Ambros V., 1991.  Efficient gene transfer in C. elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J. 10: 3959–3970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Nagel G., Brauner M., Liewald J. F., Adeishvili N., Bamberg E., et al. , 2005.  Light activation of Channelrhodopsin-2 in excitable cells of Caenorhabditis elegans triggers rapid behavioral responses. Curr. Biol. 15: 2279–2284. [DOI] [PubMed] [Google Scholar]
  22. Reboul J., Vaglio P., Rual J.-F., Lamesch P., Martinez M., et al, 2003.  C. elegans ORFeome version 1.1: experimental verification of the genome annotation and resource for proteome-scale protein expression. Nat. Genet. 34: 35–41. [DOI] [PubMed] [Google Scholar]
  23. Sarov M., Murray J. I., Schanze K., Pozniakovski A., Niu W., et al. , 2012.  A genome-scale resource for in vivo tag-based protein function exploration in C. elegans. Cell 150: 855–866. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Semple, J. I., and B. Lehner, 2014.  Single and dual drug selection for transgenes following bombardment of Caenorhabditis species. Methods 68(3): 409–416. [DOI] [PubMed] [Google Scholar]
  25. Shaye D. D., Greenwald I., 2011.  Ortholist: a compendium of C. elegans genes with human orthologs. PLoS One 6: e20085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Spaink H. P., C. Cui, M. I. Wiweger, H. J. Jansen, W. J. Venemanet al. , 2013.  Robotic injection of zebrafish embryos for high-throughput screening in disease models. Methods 62: 246–254. [DOI] [PubMed] [Google Scholar]
  27. Steinmeyer J. D., Yanik M. F., 2012.  High-throughput single-cell manipulation in brain tissue. PLoS One 7: e35603. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Stiernagle T., 2006.  Maintenance of C. elegans. WormBook : The Online Review of C. elegans. Biology (Basel) 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Thompson O., Edgley M., Strasbourger P., Flibotte S., Ewing B., et al. , 2013.  The Million Mutation Project: a new approach to genetics in Caenorhabditis elegans. Genome Res. 23: 1749–1762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Tian L., S. A. Hires, T. Mao, D. Huber, M. E. Chiappe et al, 2009.  Imaging neural activity in worms, flies and mice with improved GCaMP calcium indicators. Nat. Methods 6: 875–881. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Waaijers S., Portegijs V., Kerver J., B. B. Lemmens, M. Tijsterman et al., 2013  CRISPR/Cas9-targeted mutagenesis in Caenorhabditis elegans. Genetics 195: 1187–1191. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Wang W., et al. , 2007.  A fully automated robotic system for microinjection of zebrafish embryos. PLoS One 2: e862. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Wightman B., Ha I., Ruvkun G., 1993.  Posttranscriptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in C. elegans. Cell 75: 855–862. [DOI] [PubMed] [Google Scholar]
  34. Wilm T., P. Demel, H. U. Koop, H. Schnabel, and R. Schnabel, 1999.  Ballistic transformation of Caenorhabditis elegans. Gene 229: 31–35. [DOI] [PubMed] [Google Scholar]
  35. Wood, W. B., 1988 The Nematode Caenorhabditis elegans, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. [Google Scholar]
  36. Yu R. Y., Nguyen C. Q., Hall D. H., Chow K. L., 2000.  Expression of Ram-5 in the structural cell is required for sensory ray morphogenesis in Caenorhabditis elegans male tail. EMBO J. 19: 3542–3555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Yuan J., Shaham S., Ledoux S., Ellis H. M., Horvitz H. R., 1993.  The C. elegans cell death gene ced-3 encodes a protein similar to mammalian interleukin-1 Beta-converting enzyme. Cell 75: 641–652. [DOI] [PubMed] [Google Scholar]
  38. Yuan J. Y., Horvitz H. R., 1990.  The Caenorhabditis elegans genes ced-3 and ced-4 act cell autonomously to cause programmed cell death. Dev. Biol. 138: 33–41. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
supp_201_1_39__index.html (1.3KB, html)
10.1534_115.179648_genetics.115.179648-1.pdf (4.4MB, pdf)
10.1534_115.179648_genetics.115.179648-6.pdf (4.3MB, pdf)
Download video file (748KB, mov)
10.1534_115.179648_genetics.115.179648-3.pdf (78.4KB, pdf)
10.1534_115.179648_genetics.115.179648-2.pdf (78KB, pdf)
10.1534_115.179648_genetics.115.179648-5.pdf (78.2KB, pdf)

Data Availability Statement

Plasmids and strains are available upon request. File S1 contains instructions for assembling and using the system. Table S1 contains a purchasing list for all hardware components. Software is available through GitHub (https://github.com/CodyLGilleland/CAMI_Gilleland_2015_GENETICS.git). Plasmids, transgenes, and strains used in this study are listed in Tables S2, S3, and S4 respectively.

Articles from Genetics are provided here courtesy of Oxford University Press

RESOURCES