Aberrant Activation of p38 MAP Kinase-Dependent Innate Immune Responses Is Toxic to Caenorhabditis elegans

Abstract

Inappropriate activation of innate immune responses in intestinal epithelial cells underlies the pathophysiology of inflammatory disorders of the intestine. Here we examine the physiological effects of immune hyperactivation in the intestine of the nematode Caenorhabditis elegans. We previously identified an immunostimulatory xenobiotic that protects C. elegans from bacterial infection by inducing immune effector expression via the conserved p38 MAP kinase pathway, but was toxic to nematodes developing in the absence of pathogen. To investigate a possible connection between the toxicity and immunostimulatory properties of this xenobiotic, we conducted a forward genetic screen for C. elegans mutants that are resistant to the deleterious effects of the compound, and identified five toxicity suppressors. These strains contained hypomorphic mutations in each of the known components of the p38 MAP kinase cassette (tir-1, nsy-1, sek-1, and pmk-1), demonstrating that hyperstimulation of the p38 MAPK pathway is toxic to animals. To explore mechanisms of immune pathway regulation in C. elegans, we conducted another genetic screen for dominant activators of the p38 MAPK pathway, and identified a single allele that had a gain-of-function (gf) mutation in nsy-1, the MAP kinase kinase kinase that acts upstream of p38 MAPK pmk-1. The nsy-1(gf) allele caused hyperinduction of p38 MAPK PMK-1-dependent immune effectors, had greater levels of phosphorylated p38 MAPK, and was more resistant to killing by the bacterial pathogen Pseudomonas aeruginosa compared to wild-type controls. In addition, the nsy-1(gf) mutation was toxic to developing animals. Together, these data suggest that the activity of the MAPKKK NSY-1 is tightly regulated as part of a physiological mechanism to control p38 MAPK-mediated innate immune hyperactivation, and ensure cellular homeostasis in C. elegans.

Keywords: C. elegans genetics; genetics of immunity; host-pathogen interactions; immune regulation; innate immunity.

Figure 1

p38 MAPK mutants have a Xenobiotic toxicity suppressor (Xts) phenotype. (A) Representative images of C. elegans mutants with an Xts phenotype photographed after 3 d of development at 20°C in the presence (+) or absence (–) of 140 μM R24. C. elegans N2 animals were used as the wild-type control strain. (B) Quantification of the percentage of animals that grew from the L1 to young adult (YA) stage for the experiment shown in (A). Data are the average of two technical replicates, with error bars giving the standard deviation between plates. The sample sizes for this experiment are: N2 (196), nsy-1(ums1) (171), tir-1(ums2) (193), tir-1(ums3) (220), pmk-1(ums4) (145), sek-1(ums5) (196), pmk-1(km25) (155), tir-1(qd4) (269) and nsy-1(ag3) (193). * P < 0.05, ** P = 0.07. The data are representative of multiple biological replicate experiments, which were conducted during the backcrossing of the ums mutants to wild-type animals. (C) A P. aeruginosa pathogenesis assay with C. elegans wild-type N2 and mutants with an Xts phenotype is shown. The difference in P. aeruginosa susceptibility between all mutants with an Xts phenotype and wild-type animals is significant (P < 0.001), except nsy-1(ums1) (P = n.s.). Sample sizes are: wild-type N2 (152), nsy-1(ums1) (140), tir-1(ums2) (151), tir-1(ums3) (150), pmk-1(ums4) (124), and sek-1(ums5) (139). (D) Immunoblot analysis of lysates from L4 larval stage animals of the indicated genotype using antibodies that recognize the doubly phosphorylated TGY motif of PMK-1 (p-PMK-1) and actin.

Figure 2

The nsy-1(ums8) allele encodes a gain-of-function mutation in nsy-1. (A) The putative domain architecture of NSY-1, based on homology to mammalian ASK1, is presented. This schematic was adapted from Bunkoczi et al. (2007). The putative location of the central serine-threonine kinase domain, and two coiled-coil domains in the N and C termini are shown in red and blue, respectively. The boundary of the N-terminal negative regulatory domain relative to the NSY-1 protein is presented above the diagram. The Arg that was mutated to Gln in the ums8 strain (R246Q) is highlighted in red. Amino acid sequence alignment of human ASK1, Drosophila Pk92B (the ASK1 homolog) and C. elegans NSY-1 demonstrates that the ums8 mutation is located in a strongly conserved region. Dark blue shading indicates identical amino acids, with progressive lighter blue shading indicating the level of similarity in amino acid class as determined by the software ClustalW2. (B) Wild-type (WT) and nsy-1(ums8) animals were exposed to the feeding RNAi bacteria strain transformed with the control vector (L4440), or with two separate RNAi constructs (pHC1 and pHC2) that target different areas of coding region in the nsy-1 gene and photographed. Green is F08G5.6::GFP induction and red is myo-2::mCherry, which was used as the co-injection marker. (C) The expression of the indicated genes was determined using qRT-PCR in wild-type (+/+), nsy-1(ums8)/+ heterozygotes, and nsy-1(ums8)/nsy-1(ums8) homozygotes. F08G5.6::GFP was used as the wild-type strain. Data are the average of three replicates, each normalized to a control gene with error bars representing SEM, and are presented as the value relative to the average expression of the indicated gene in wild-type animals. * equals P < 0.05. There was no statistical difference in the levels of induction between the heterozygous and homozygous samples for either gene. For additional genes tested, see Figure S1A. (D) Immunoblot analysis of lysates from L4 larval stage animals of the indicated genotype using antibodies that recognize the doubly phosphorylated TGY motif of PMK-1 (p-PMK-1) and actin. Thirty μg of nsy-1(ums8), pmk-1(km25) and nsy-1(ag3) total protein were loaded on the gel alongside a dilution series of wild-type template [15 μg, 20 μg, and 30 μg of total protein (left to right)] to control for the ability of the p-PMK-1 antibody to detect different concentrations of substrate. The arrow on the right highlights the PMK-1 band, which is absent in the pmk-1(km25) and nsy-1(ag3) mutants. The arrowheads on the left point to nonspecific bands. Data are representative of two biological replicates.

Figure 3

p38 MAPK-dependent putative immune effectors are constitutively activated in the nsy-1(ums8) mutant. (A) A scatter plot compares the expression of 118 C. elegans genes, which were analyzed using nanoString nCounter gene expression system in wild-type, nsy-1(ums8) and nsy-1(ag3) animals. Data are the average of two replicates for nsy-1(ums8), and nsy-1(ag3), and are from one sample for wild-type. The expression of each gene was normalized to the geometric mean of the expression of three control genes. Genes that are outside the two parallel lines on each graph are differentially regulated more than fivefold from the expression in wild-type animals. Genes that are previously characterized targets of the p38 MAPK PMK-1, and were strongly differentially regulated in this experiment, are highlighted. See Table S1 for the expression levels of all the genes. (B) qRT-PCR was used to study the expression of six putative immune effectors in RNAi-treated, mixed-stage animals of the indicated genotypes. All animals were grown on the RNAi bacteria feeder strain HT115 expressing the empty vector L4440, except for the two indicated samples that were exposed to bacteria expressing the nsy-1(RNAi) construct. The location on the scatter plots of these genes is indicated in (A) with red (left) and blue (right) dots. Data are the average of three replicates, each normalized to a control gene with error bars representing SEM, and are presented as the value relative to the average expression of the indicated gene in wild-type animals. * equals P < 0.05.

Figure 4

Endogenous hyperactivation of p38 MAPK innate immune responses protects nematodes from bacterial infection and delays the development of wild-type nematodes. In both (A) and (B), wild-type animals were grown on RNAi bacteria expressing the empty vector L4440 (Wild-type), or a construct designed to knockdown nsy-1 [nsy-1(RNAi)]. nsy-1(ums8) animals, were grown in parallel on RNAi bacteria expressing the empty vector L4440 [nsy-1(ums8)], or a construct designed to knockdown nsy-1 [nsy-1(ums8) + nsy-1(RNAi)]. (A) P. aeruginosa pathogenesis assays were performed on RNAi-treated animals of the indicated genotypes. The difference in P. aeruginosa susceptibility between wild-type and nsy-1(ums8) animals is significant, as is the survival difference between nsy-1(ums8) and nsy-1(ums8) + nsy-1(RNAi) (P < 0.001). Data are representative of two biological replicates. The sample sizes for this experiment are: wild-type (115), nsy-1(RNAi) (122), nsy-1(ums8) (111), and nsy-1(ums8) + nsy-1(RNAi) (133) (B) The development of RNAi-treated animals of the indicated genotypes to the L4 larval stage or older was recorded. The data are the average of three plates, with error bars showing the standard deviation between plates. The sample sizes for this experiment are: wild-type + L4440 (1,200), wild-type + nsy-1(RNAi) (1179), nsy-1(ums8) + L4440 (956), and nsy-1(ums8) + nsy-1(RNAi) (1010). Data are representative of two biological replicates. * P < 0.001.