Evolutionary plasticity in the innate immune function of Akirin

Abstract

Eukaryotic gene expression requires the coordinated action of transcription factors, chromatin remodelling complexes and RNA polymerase. The conserved nuclear protein Akirin plays a central role in immune gene expression in insects and mammals, linking the SWI/SNF chromatin-remodelling complex with the transcription factor NFκB. Although nematodes lack NFκB, Akirin is also indispensable for the expression of defence genes in the epidermis of Caenorhabditis elegans following natural fungal infection. Through a combination of reverse genetics and biochemistry, we discovered that in C. elegans Akirin has conserved its role of bridging chromatin-remodellers and transcription factors, but that the identity of its functional partners is different since it forms a physical complex with NuRD proteins and the POU-class transcription factor CEH-18. In addition to providing a substantial step forward in our understanding of innate immune gene regulation in C. elegans, our results give insight into the molecular evolution of lineage-specific signalling pathways.

Fig 1. Akirin acts downstream of Gα to regulate the expression of nlp-29.

A. Ratio of green fluorescence (GFP) to size (time of flight; TOF) in IG274 worms carrying the integrated array frIs7 (containing nlp-29p::gfp and col-12p::DsRed) treated with RNAi against negative and positive controls (sta-1, dcar-1, respectively) or akir-1 and infected by D. coniospora (Infected), wounded (Wound), treated with a 5 mM dihydrocaffeic acid solution (DHCA) or exposed to 300mM NaCl (High salt). Here and in subsequent figures representing Biosort results, unless otherwise stated, graphs are representative of at least 3 independent experiments. The black bar represents the mean value for (from left to right), n = 189, 164, 179, 184, 183, 169; 23, 21, 30, 29, 36, 68; 97, 86, 114, 111, 104, 94; 96, 85, 100, 150, 129, 113; **** p<0.0001, ns p>0.05, Dunn’s test. B. Fluorescent images of adult worms carrying frIs7, expressing a constitutively active Gα protein, GPA-12*, in the epidermis and treated with RNAi against the indicated genes. Almost all of the residual GFP expression seen upon akir-1(RNAi), most prominent in the vulval muscle cells, comes from unc-53Bp::gfp used as a transgenesis marker.

Fig 2. Akirin regulates multiple nlp genes in the epidermis.

A. Confocal images of IG1485 transgenic worms expressing an akir-1p::gfp reporter gene showing epidermal and neuronal expression of GFP. The lateral epithelial seam cells are indicated by the arrowheads. Much of the fluorescence in the head and tail comes from neurons, seen more clearly in the right panel. Scale bar 50 μm. B. Ratio of green fluorescence (GFP) to size (TOF) in rde-1(ne219);wrt-2p::RDE-1 worms that are largely resistant to RNAi except in the epidermis, carrying the array frIs7, treated with RNAi against the indicated genes and infected by D. coniospora. The black bar represents the mean value for (from left to right), n = 135, 49, 155, 102, 130, 94; **** p<0.0001, Dunn’s test. C. Quantitative RT-PCR analysis of the expression of genes in the nlp-29 cluster in rde-1(ne219); wrt-2p::RDE-1 worms treated with RNAi against the indicated genes and infected by D. coniospora; results are presented relative to those of uninfected worms. Data (with average and SD) are from three independent experiments (S2B Fig). **, p < 0.001; *, p < 0.01; 1-tailed ratio paired t test.

Fig 3. Akirin expression in the epidermis regulates resistance to fungal infection.

Survival of rde-1(ne219);wrt-2p::RDE-1 worms treated with RNAi against sta-1 (negative control; n = 50) or akir-1 (n = 50), infected with D. coniospora and cultured at 15°C. The difference between the sta-1(RNAi) and akir-1(RNAi) animals is highly significant (p<0.0001; one-sided log rank test). Data are representative of three independent experiments.

Fig 4. LET-418 NuRD and MEC complexes regulate nlp-29 gene expression.

A. Ratio of green fluorescence (GFP) to size (TOF) in worms carrying frIs7, treated with RNAi against control (sta-1, dcar-1, akir-1), NuRD and MEC complex component, and non-NuRD chromatin remodelling component genes, and infected or not with D. coniospora. A minimum of 130 worms was used for each experiment. The black bar represents the mean value; **** p<0.0001 upon infection, relative to sta-1(RNAi), Dunn’s test; for the other conditions there is not a significance decrease. B. Fluorescent images of adult worms carrying frIs7 and expressing GPA-12* in the epidermis and treated with RNAi against the indicated genes. See legend to Fig 2 for more details.

Fig 5. AKIR-1 interactors identified by label-free quantitative immunoprecipitation.

A. Experimental design. Protein extracts from mixed-stage worms expressing AKIR-1::GFP were incubated with anti-GFP conjugated or control resins before proteolytic release of peptides from the immunoprecipitated proteins. The relative abundance of co-precipitated proteins was assessed by mass spectrometry. B. Volcano plot showing specific interaction partners (in red) of AKIR-1::GFP. The mean values for fold change from 3 independent experiments are shown. The SAM (significance analysis of microarrays) algorithm was used to evaluate the enrichment of the detected proteins. Proteins that met the combined enrichment threshold (hyperbolic curves, t₀ = 1.2) are colored in red. Proteins with the gene ontology annotation “DNA-binding” (GO:0003677) are depicted as triangles. Known members of the NuRD complex are shown in blue. C. NuRD complex and/or DNA-binding proteins among the 53 high confidence AKIR-1::GFP interaction partners.

Fig 6. Validation of AKIR-1 interactors by Western blotting.

A. Complexes immunopurified using an anti-GFP antibody from control or infected worms with a single copy AKIR-1::GFP insertion (wt; frSi12[pNP157(akir-1p::AKIR-1::GFP)] II) were probed with specific antibodies. The results for two independent pull-downs are shown. The presence of HDA-1 and LET-418 (NuRD complex components) could be confirmed. Anti-ACT-1 was used to control the total input for each immunoprecipitation. B. Complexes immunopurified using an anti-FLAG antibody, from a strain co-expressing AKIR-1::GFP and FLAG-tagged CEH-18 (wt; frSi12[pNP157(akir-1p::AKIR-1::GFP)] II; wgIs533[CEH-18::TY1::GFP::3xFLAG + unc-119(+)]), were probed with anti-FLAG (top panel) and anti-GFP (bottom) antibodies. In addition to the immunopurified complex (IP), the extract before immunopurification (Input), the unbound fraction (flow-through: FT) and proteins immunopurified using an unrelated antibody (Mock) were also analysed.

Fig 7. CEH-18 plays a role in host defence.

A. Ratio of green fluorescence (GFP) to size (TOF) in wild-type (IG274) or ceh-18(mg57) mutant (IG1714) worms carrying frIs7, infected or not with D. coniospora for 16 h (yellow and blue, respectively; data for IG274 is as Fig 3B in [53]), and IG274 worms treated with RNAi against sta-1 (control) or ceh-18 and, from left to right, exposed to high salt (purple; cpsf-2(RNAi) is a positive control, sta-2(RNAi) a negative control [11]), in worms also expressing GPA-12* in the epidermis, and in the rde-1(ne219);wrt-2p::RDE-1 background and infected by D. coniospora. For the latter 2 panels, sta-2(RNAi) is a positive control. A minimum of 45 worms was used for each condition. The black bar represents the mean value; *** p<0.001, **** p<0.0001, relative to sta-1(RNAi), Dunn’s test; for the other conditions there is not a significance decrease. The results of the 3 right panels are unpublished results from [11], representative of at least 4 independent experiments. B. Quantitative RT-PCR analysis of the expression of several genes in the nlp-29 cluster in wild-type worms, sta-2 and ceh-18 mutants infected by D. coniospora; results are presented relative to those of uninfected worms. Data (with average and SD) are from three independent experiments. **, p < 0.001; *, p < 0.01; 1-tailed ratio paired t test. C. Results of 2 independent tests of survival of rde-1(ne219);wrt-2p::RDE-1 worms treated with RNAi against sta-1, sta-2, akir-1 or ceh-18, infected with D. coniospora and cultured at 25°C (n>65 for all tests). The difference between the sta-1(RNAi) and ceh-18(RNAi) animals is highly significant in both trials (p<0.0001; one-sided log rank test).

Fig 8. AKIR-1 binds preferentially to nlp gene promoters in the absence of infection.

A. Specific binding of AKIR-1::GFP onto promoters (left panels) or 3’ UTR (right panel) of act-1 (left panel; 2 different PCR amplicons, A and B), and nlp-29, nlp-31 and nlp-34, represented as the fold enrichment of the specific ChIP signal obtained using an anti-GFP antibody for immunoprecipitation relative to that when blocked beads were used, measured by quantitative PCR. Data is normalised to input; the average (and standard error) from three independent experiments is shown. ***, p < 0.0001; **, p < 0.001; ns, p > 0.1; paired 2-tail Student’s t test. B. Model for the role of AKIR-1 in the regulation of nlp AMP gene expression upon infection. Under normal conditions (left), the AKIR-1/NuRD complex is recruited to the nlp-29 locus, leading to modification (red stars) of histones (ovoids), and formation of an open chromatin structure. Upon infection, STA-2 is activated and, following removal of the AKIR-1/NuRD complex, is responsible for expression of the nlp genes. Infection could impact chromatin structure, but here we assume that it does not. When AKIR-1 is absent (right), an open chromatin structure cannot be formed, precluding STA-2-dependent expression of the nlp genes following infection, but not affecting the low basal STA-2-independent gene expression. The images are adapted, with permission, from https://www.activemotif.com.