Neuron type-specific miRNA represses two broadly expressed genes to modulate an avoidance behavior in C. elegans

In this study, Drexel et al. research miRNA-mediated repression of broadly transcribed genes as a strategy for cellular specialization. They show that mir-791, expressed exclusively in the CO2-sensing neurons in C. elegans, represses two otherwise broadly expressed genes, which are needed for normal neuronal function and behavior of the animals toward CO2.


2014).
From plates where we detected editing events, F2 progeny was either singled out to regular OP50 plates (in case only a few F2 were left on the plate) or pooled (2-3 worms, in case of many survivors) and allowed to self. As soon as the F3 progeny is visible, F2 worms can be screened for the required homology directed repair event, which typically consisted on PCR followed by restriction digestion.
All mutant alleles were confirmed by Sanger sequencing and the sequences of the obtained 3'UTR alleles are provided below.

Behavioral Assays
Behavioral assays were performed using a similar device previously used for O 2 sensory responses (Zimmer et al. 2009) with some modifications. One-day old, well-fed adults were transferred from growth plates with OP50 bacteria to 10 cm NGM plate without food. They were then subsequently transferred to a 14 cm NGM assay plate containing an arena (56 x 56 mm) delimited by Whatman paper soaked in 20 mM CuCl 2, which prevents the animals from leaving the assay area. 50 to 100 animals were assayed per experiment. Animals were starved for 1 hour prior to recording onset. For each round of experiments, a separate WT and mir-791(0) dataset was recorded as controls in parallel with the additional tested genotypes.
Gas flow was applied through a custom made Plexiglas device (60 x 60 x 0.7 mm), which was placed onto the assay arena and connected via a static mixing element to mass flow controllers (Vögtlin Instruments) operated by customized LabView software to control gas concentrations (v/v). Gas flow was maintained at 100 ml/min. For the first 240 sec, animals were exposed to 21% oxygen. At constant 21% oxygen, CO 2 was increased to 5% at a rate of 0.5% / 5 sec. CO 2 concentration was then kept constant for 500 seconds and eventually decreased ! 3! back to 0% at the same rate. All gas mixtures were balanced with N 2 . Movies were made at 3 frames per second using a 4 Mpixel camera (Jai) streaming to Streampix software. Movies were analyzed with a modified MatLab-based tracking software (Ramot et al. 2008), available at http://med.stanford.edu/wormsense/tracker. For speed calculation only periods of continuous forward movement were used and data were binned by taking the mean of 3 consecutive frames. Reversals and turns were detected by measuring characteristic changes in angular velocity and data were binned by taking the sum of events in consecutive 15 frames intervals.
For traces in Fig. 1B  We would like to highlight that the response we observe for BAG has somewhat different kinetics than others reported in previous work (Bretscher et al. 2011, Hallem et al. 2011, Carrillo et al. 2013

Fosmid Recombineering
Fosmid-based reporters were generated as previously described (Tursun et al. 2009). Primer sequences used to build all fosmids and resulting construct sequences described in this work can be provided upon request. Briefly, the mir-791 fosmid was generated by replacing the precursor miRNA hairpin by gfp. The that was co-injected with the akap-1 reporters, but had visibly fewer GFP-labeled cells around them, and scored animals even when rotated as long as at least one BAG neuron was unobstructed. This meant that most times we scored a single BAG neuron per animal and the RME reference neurons were not visible in every animal. Therefore, we used as a reference one of the hyp5 nuclei in the anterior part of the animal. Hyp5 is the most anterior hypodermal cell and is very reliably identified.

Scoring cell type specific markers
Neurons were scored for presence or absence of the respective neuron specific markers and images were taken under a widefield microscope, Axio Imager.Z2 with sCMOS camera running under Metamorph. In bold are the 5 candidates selected for validation ( Fig. 5 and Supplemental Fig. S7) based on the presence of multiple binding sites as well as a potential connection with CO 2 sensing or neuronal signaling.

SUPPLEMENTAL REFERENCES
For akap-1, the two more conserved sites are conserved in C. briggsae, C. remanei, C. japonica, C. tropicalis and C. sp11 (based on the seed-matching sequence), but they seem to be absent in C. brenneri, C. angaria and P. pacificus. The less conserved site is only present in C. elegans.
For cah-3, the two sites are conserved in C. briggsae, C. remanei, C. brenneri, C. japonica, C. tropicalis and C. sp11 (based on the seed-matching sequence), but they seem to be absent in C. angaria and P. pacificus. Table S2. Strains used in this study  Experiments used for this analysis are the same as those described in Figure 1B.
Statistical analysis was performed on the average speed of each experiment during the 50 sec of stimulation with CO 2 (stim) and the preceeding 50 sec (pre).
Mann-Whitney tests were all non-significant (ns).