Neuroligin-mediated neurodevelopmental defects are induced by mitochondrial dysfunction and prevented by lutein in C. elegans

Complex-I-deficiency represents the most frequent pathogenetic cause of human mitochondriopathies. Therapeutic options for these neurodevelopmental life-threating disorders do not exist, partly due to the scarcity of appropriate model systems to study them. Caenorhabditis elegans is a genetically tractable model organism widely used to investigate neuronal pathologies. Here, we generate C. elegans models for mitochondriopathies and show that depletion of complex I subunits recapitulates biochemical, cellular and neurodevelopmental aspects of the human diseases. We exploit two models, nuo-5/NDUFS1- and lpd-5/NDUFS4-depleted animals, for a suppressor screening that identifies lutein for its ability to rescue animals’ neurodevelopmental deficits. We uncover overexpression of synaptic neuroligin as an evolutionarily conserved consequence of mitochondrial dysfunction, which we find to mediate an early cholinergic defect in C. elegans. We show lutein exerts its beneficial effects by restoring neuroligin expression independently from its antioxidant activity, thus pointing to a possible novel pathogenetic target for the human disease.


Supplementary Figure 3
Supplementary Figure 3. Strong RNAi against the mitochondrial Complex I proteins does not lead to significant signs of oxidative damage. a) Proton leak and b) Spare capacity were measured using the Seahorse XF24 Analyzer. Three biologically independent experiments n≥1000 per group. p-values in a: con vs nuo-5 p = 0.0012, in b: con vs nuo-5 p = 0.0036. c) Quantification of GFP expression in a WT strains expressing the agIs219 transgene, which is comprised of the promoter of a PMK-1-regulated gene, T24B8.5 fused to GFP and provides an in vivo sensor of PMK-1 pathway activity. Two biologically independent experiments, n≥40. d) Lipid oxidation and e) amide oxidation assessed with SR-μFTIR. Three biologically independent experiments, n≥75 per group. p-values in d: con vs nuo-5 p < 0.0001, con vs lpd-5 p < 0.0001, p-values in e: con vs nuo-5 p < 0.0001, con vs lpd-5 p < 0.0001 f) Nuclear DNA (nDNA) damage, g) Mitochondrial DNA (mtDNA) damage and h) mtDNA copy number upon strong suppression of lpd-5 and nuo-5. Three biologically independent experiments, n=18. p-values in f: con vs nuo-5 p = 0.0117, con vs lpd-5 p = 0.0468. In all figure's panels worms were fed as in Fig 2 and statistical analysis was carried out using oneway ANOVA. Box plots indicate median (middle line), 25th, 75th percentile (box) and 5th and 95th percentile (whiskers) as well as outliers (single points). Schematic experimental strategy to proof the reversibility of the strong phenotype. Animals were cultured on RNAi plates (strong treatment); then worms were moved on control plates (fed the empty vector pL4440) either as eggs or at different larval stages. The development was checked for rescue on day 4 after egg hatching (representative pictures of a partial rescue obtained returning the nematodes on pL4440 within 24 hours). b) Flowchart with representative pictures of the compounds screen in search of suppressors of the nuo-5 and lpd-5 strong (developmental arrest) phenotype.

Supplementary Tables
Supplementary Table 1 All these clones showed a strong phenotype already in the parental generation (P0, see Fig. 1) when used undiluted, and the mild effect was thus achieved by diluting the dsRNA expressing bacteria (P0, 1/10 or 1/50). a % increase normalized mean lifespan compared to control; b Kaplan-Meier survival analysis, Log-rank test against control; c Kaplan-Meier survival analysis, Log-rank test between mild and strong treatment.

Supplementary Table 3. Summary of chemotaxis assays in C. elegans models of Complex I deficiency
RNAi clone RNAi power

Lifespan assay
Survival analysis started from hatching and was carried out at 20°C. Animals were scored as dead or alive and transferred every day on fresh plates during the fertile period, and then every other day or every 3 days until death. Worms were considered dead when they stop pharyngeal pumping and responding to touch. Worms that died because of internal bagging, desiccation due to crawling on the edge of the plates, or gonad extrusion were scored as censored. These animals were included in lifespan analyses up to the point of censorship and were weighted by half in the statistical analysis. We calculated mean lifespan, standard deviation of the mean, and P value (Mantel-Cox regression analysis) from Kaplan-Meyer survival curves of pooled population of animals coming from at least two independent replicas. For statistical analysis we used the Online Application for Survival analysis OASIS 2 [16].

Chemotaxis assays
Sodium azide (NaN3) was used to anesthetize worms. It was placed on buffered agar 180 degrees opposite on a 10 cm dish; the attractant (or repellent) was then placed on one NaN3 spot, and ethanol (neutral odor for the worms in which chemical was diluted) on the other spot; a population of 80-100 age-synchronized animals was placed in the center of the testing plate and the number of worms at attractant and control was counted every 15 minutes for four hours to calculate the Chemotaxis Index (CI). CI=(A-B)/(A+B+C), where A is the number of worms at attractant, B is the number of worms at control and C is the number of animals which didn't reach any of the two spots at the end of the two hours. For a population of 3 days old, wild-type animals, a good CI is around 0.8 for attractants and -0.8 for repellents after two hours (CI=0 means no attraction, while CI=1 or -1 represent maximum attraction or repulsion, but there are always some animals that, also for a wild-type strain, remain randomly dispersed in the assay plate or reach the control spot instead of the attractant) [17].

ATP levels
Worms at the desired developmental stage (L2/L3 larvae) were transferred to 15mL conical tube, spun at 2200 RCF and the supernatant was removed. Nematodes were resuspended in k-medium to a concentration of 1.0±0.2 nematodes/μl. 50-100 animals were pipetted into each well of a white 96-well plate (4-5 wells per treatment), and brought to a final volume 100uL with k-medium. Kmedium alone was used for blank measurements. A plate reader was used to read first the GFP fluorescence (Emissions filter: 502nm; excitation filter: 485nm) and then the luminescence after injection of a luminescence buffer. The ATP production was then calculated dividing the luminescence signal by the GFP intensity.
Mitochondrial DNA copy number, mitochondrial DNA and nuclear DNA damage Synchronized wild-type (N2) worms were grown at 20°C for two consecutive generations on bacteria expressing dsRNA against the genes of interest: nuo-5 and, lpd-5. nDNA and mtDNA from these animals was compared with stage-matched animals fed bacteria expressing the empty vector pL4440. Six worms were picked and pooled in a single tube per biological replicate, and three biological replicates were taken per treatment in three experiments separated in time.
This assay defines the control samples as undamaged and determines a lesion frequency in experimental samples based on any decrease in amplification efficiency relative to the control samples [19]. One nuclear genome target (9,3 kb) and one mitochondrial genome target (10,9 kb) were amplified. The DNA damage assay is able to quantitatively measure the number of polymerase-stalling lesions based on the amount of amplification obtained from the QPCR [20][21][22]. To calculate DNA copy number, we utilized a real-time PCR assay in which, by using a standard, the actual DNA content can be calculate [23,24]. This method is particularly advantageous because the standard curve is run along with the samples and the actual number of copies can be calculated.

Quantification of Gene Expression through fluorescent transgene reporters
For the SJ4100 and CL2166 strains the GFP intensity of the total body was quantified ( Supplementary Fig. 4a). Concerning the JIN1679 strain, for each animal the number of nuclei showing HLH-30 nuclear translocation (between the pharynx and the vulva) (Supplementary Fig.   4b) were counted with a macro specifically customized with ImageJ. With the KP3928 strain the intensity of the fluorescence was quantified in the pharynx, specifically, two circular area were designed with ImageJ corresponding to anterior and posterior pharyngeal bulbs and the GFP intensity was quantified in the two areas (Supplementary Fig. 11b). For Pglb-10::GFP the intensity of the neuronal nuclei located in the pharyngeal area was quantified (Supplementary Fig. 11c).
For the Pdct-1::GFP a circular region in the middle body was selected to quantify the GFP signal ( Supplementary Fig. 8g). For the selection of the area and the quantification of the signal the software ImageJ (http://imagej.nih.gov/ij/) was used. For the LX929 the intensity of the second proximal head neuron was quantified (Supplementary Fig. 11d). With the vjIs47, vjIs105 and CRR104 strains the intensity of the fluorescence was quantified in the ventral nerve cord (VNC), specifically, a rectangular area was designed in the posterior half of the nematode, between the vulva and the tale, and the GFP intensity was quantified in the selected areas ( Fig. 7e and Fig. 8f).

RNA extraction for Microarray
Worms at the desired larval stage (L3) were washed off three big agar plates spotted with the RNAi bacteria under study (nuo-5, lpd-5 and pL4440 for control). The plates were supplemented either with 1 µM Lutein (Sigma Aldrich PHR1699) or 0.25 % DMSO for control. Individual C. elegans were collected into nuclease-free water and frozen at −80 °C until needed. Frozen worm pellets were re-suspended in lysis buffer and the same volume of stainless-steel beads (6.0 mm RNase free) was added. A TissueLyser II (Qiagen) was used with 2 shakes for 2 minutes at 30 Hz. Total RNA was prepared from each sample using the RNeasy Mini Kit and QIAshredder according to the manufacturer's instructions (Qiagen-74104 and 79654). Total RNA from 5 different biological replicas was extracted, on column purified to remove small RNA fractions, and treated with DNAse (RNeasy kit, Qiagen-74004).