A high-throughput screen for the identification of compounds that inhibit nematode gene expression by targeting spliced leader trans-splicing

Abstract

Infections with parasitic nematodes are among the most significant of the neglected tropical diseases affecting about a billion people living mainly in tropical regions with low economic activity. The most effective current strategy to control nematode infections involves large scale treatment programs with anthelmintic drugs. This strategy is at risk from the emergence of drug resistant parasites. Parasitic nematodes also affect livestock, which are treated with the same limited group of anthelmintic drugs. Livestock parasites resistant to single drugs, and even multi-drug resistant parasites, are appearing in many areas. There is therefore a pressing need for new anthelmintic drugs. Here we use the nematode Caenorhabditis elegans as a model for parasitic nematodes and demonstrate that sinefungin, a competitive inhibitor of methyltransferases, causes a delay in development and reduced fecundity, and inhibits spliced leader trans-splicing. Spliced leader trans-splicing is an essential step in gene expression that does not occur in the hosts of parasitic nematodes, and is therefore a potential target for new anthelmintic drugs. We have exploited the ability of sinefungin to inhibit spliced leader trans-splicing to adapt a green fluorescent protein based reporter gene assay that monitors spliced leader trans-splicing for high-throughput screening for new anthelmintic compounds. We have established a protocol for robust high-throughput screening, combining mechanical dispensing of living C. elegans into 384- or 1536- well plates with addition of compounds using an acoustic liquid dispenser, and the detection of the inhibition of SL trans-splicing using a microplate reader. We have tested this protocol in a first pilot screen and envisage that this assay will be a valuable tool in the search for new anthelmintic drugs.

Keywords: Anthelminitics; Caenorhabditis elegans; HTS assay; SL1 trans-splicing; Sinefungin.

Graphical abstract

Fig. 1

A. SL1 trans-splicing mechanism. SL1 spliced leader trans-splicing is the addition of a specialised exon known as the spliced leader 1 (SL1) to the 5′ ends of pre-mRNA transcripts at the SL1 splice acceptor site, resulting in the removal of the 5′ untranslated region referred to as the “outron”. The spliced leader SL1 is donated by a separate RNA, the SL1 RNA. B. A GFP based in vivo assay that monitors SL1 trans-splicing. Schematic representation of the gfp reporter gene with the outron sequence and the SL1 splice acceptor site 3′ of the ATG initiation codon (schematic adapted from (Philippe et al., 2017)). During SL1 trans-splicing, the outron sequence is removed from gfp mRNA and replaced by SL1, resulting in the removal of the initiation codon and thus preventing synthesis of functional GFP. Inhibition of SL1 trans-splicing results in GFP expression, because the start codon is retained in the gfp reporter mRNA. Presence and absence of SL1 trans-splicing are indicated with GO and STOP signs, respectively.

Fig. 2

Sinefungin inhibits SL trans-splicing. A. Exposure to sinefungin activates gfp reporter gene expression. PE796 animals in S complete medium supplemented with OP50 bacteria were either left untreated, treated with DMSO (2%) or with 20 μM, 100 μM or 400 μM sinefungin in 2% DMSO for 21 h. gfp expression was then analysed by fluorescence microscopy. In the micrographs shown the exposure time was 5 ms and the scale bar represents 100 μm. The graph plots gfp expression standardised with respect to mCherry in arbitrary units. 10 animals were analysed for each treatment, shown are the median and first and third quartile. (“ns” indicates values not significantly higher than in untreated and DMSO treated animals (P > 0.05; ANOVA), ** indicates values significantly higher than in untreated and DMSO treated animals (P ≤ 0.01; ANOVA)). B. Treatment with sinefungin inhibits SL trans-splicing. PE796 animals were treated with sinefungin for 21 h as described under (A) and then harvested for cDNA synthesis as described in Materials and Methods. SL1 trans-splicing of gfp reporter transcripts and of endogenous rps-3 transcripts was analysed by reverse transcription followed by qPCR. The schematic diagrams indicate the location of primers used for PCR on the gfp and rps-3 RNAs. “outron” indicates the primer pair used to measure the efficiency of SL1 spliced leader trans-splicing, and “internal” indicates the primer pair used to measure RNA levels. Outron sequences are shown in blue and the SL1 acceptor sites in pink. gfp and rps-3 open reading frames are shown in green and brown, respectively. The levels of non-trans-spliced outron-gfp reporter transcripts and of outron-rps-3 transcripts were standardised with respect to RNA levels measured using the internal primer pairs, and the levels in untreated animals were defined as 1. Note that an increase in outron-gfp RNA levels reflects the inhibition of trans-splicing. The graphs show the average of three biological replicates, error bars indicate the standard deviation. Inhibition of SL1 trans-splicing, as measured by outron retention, increases significantly between treatments with 20 μM, 100 μM and 400 μM sinefungin (P ≤ 0.05; t-test). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Culture volume, read mode, and dispensing affect high-throughput screen robustness. A. Culture volume and read mode. 20–25 animals were manually transferred into wells of a 384-well plate in either 10 μl or 20 μl of S-complete medium containing OP50 bacteria. Animals were then exposed to 400 μM sinefungin/2% DMSO (32 wells) or with 2% DMSO only (32 wells), using an ECHO 550 acoustic liquid handler to dispense the compounds. Fluorescence was measured after 21 h using either single read or orbital averaging mode of a PHERAstar FS microplate reader. The graphs show fluorescence measurements taken in either single read or orbital averaging mode for 20 μl cultures for the treatments with sinefungin/DMSO or with DMSO only. The table shows signal/background ratios (S/B) with standard deviations and Z′-factors from the analysis using orbital averaging mode and single read mode for both culture volumes, calculated as described in Materials and Methods. B. Dispensing method and read mode. 20–25 animals were transferred into wells of a 384-well plate in 20 μl of S-complete medium containing OP50 bacteria using the methods indicated. Animals were then treated with 400 μM sinefungin/2% DMSO (32 wells) or with 2% DMSO only (32 wells). Fluorescence was measured after 21 h using a PHERAstar FS microplate reader using either single read or orbital averaging mode. The graph shows fluorescence measurements of the treatments with sinefungin/DMSO or with DMSO only taken in the orbital averaging mode. The table shows signal/background ratios (S/B) with standard deviations and Z′-factors for measurements taken in single read and orbital averaging mode.

Fig. 4

Treatment with sinefungin produces a robust signal that can be detected for at least 19 h and scaled up for high-throughput screening. A. Time course. PE796 animals were grown, dispensed and treated with 800 μM sinefungin/2% DMSO (24 wells) or 2%DMSO (24 wells) as described. The graph shows fluorescence measurements at the indicated times after compound addition for the treatment with sinefungin/DMSO and for DMSO only. B. Assays in 384- and 1536-well plates. Animals in 192 or 796 wells were either treated with 800 μM sinefungin/2% DMSO or with DMSO only and GFP fluorescence was measured after 4 h and 19 h of incubation. The graphs show the measurements taken after 4 h and 19 h. Signal to background (S/B) ratios with standard deviations and Z′-factors were calculated as described. C. Pilot screen of drug-like compounds. PE796 animals were dispensed into 384-well plates in the presence of OP50 bacteria. On each plate, 32 wells were treated with 2% DMSO/800 μM sinefungin, and 32 wells with 2% DMSO. Animals in the remaining wells were treated with 50 μM compound/2% DMSO, and fluorescence was analysed after 4 h exposure. The graph shows fluorescence caused by control treatments with sinefungin/DMSO and with DMSO only, and with compounds from the BioAscent Compound Cloud from a typical 384-well plate. BCC0052390, BCC0054675 and BCC0115265 were further investigated.

Fig. 5

Sinefungin-induced reporter gene activation is similar in the presence and absence of E. coli OP50. PE796 animals were dispensed with or without OP50 E. coli and treated with compounds as described. GFP fluorescence was measured at the indicated times after sinefungin addition. The graph shows fluorescence measurements for treatments with 800 μM sinefungin/2% DMSO (16 replicates each) and with 2% DMSO only (32 replicates) at the indicated times. The table summarises signal/background ratios (S/B) with standard deviations and Z′-factors for treatments with 400 μM and 800 μM sinefungin compared to the DMSO treatment control.

Fig. 6

The bus-8 mutation affecting cuticle integrity increases high-throughput screen robustness.bus-8⁽⁺⁾ and bus-8⁽⁻⁾ animals (PE796 and PE797) animals were grown, dispensed and treated with sinefungin/DMSO or DMSO only in the absence of OP50 bacteria. GFP fluorescence was measured at the indicated times after the addition of compounds. The graph shows fluorescence measurements 4 h after addition of compounds. Normally 12 or 15 wells were measured for incubations with PE797 and 32 for incubations with PE796 except for the treatments of PE796 with 400 μM sinefungin, where 8 wells were measured. The table shows signal/background ratios (S/B) with standard deviations and Z′-factors for 4 h, 6.5 h and 21.5 h treatment with compounds. The values for PE796 animals are as in Fig. 5 without OP50 bacteria. Note that the assay is more robust with bus-8⁽⁻⁾ PE797 animals. The reduced absolute levels of GFP fluorescence in these animals may be linked to the different genetic background of this strain.