Differential CaKAN3-CaHSF8 associations underlie distinct immune and heat responses under high temperature and high humidity conditions

Abstract

High temperature and high humidity (HTHH) conditions increase plant susceptibility to a variety of diseases, including bacterial wilt in solanaceous plants. Some solanaceous plant cultivars have evolved mechanisms to activate HTHH-specific immunity to cope with bacterial wilt disease. However, the underlying mechanisms remain poorly understood. Here we find that CaKAN3 and CaHSF8 upregulate and physically interact with each other in nuclei under HTHH conditions without inoculation or early after inoculation with R. solanacearum in pepper. Consequently, CaKAN3 and CaHSF8 synergistically confer immunity against R. solanacearum via activating a subset of NLRs which initiates immune signaling upon perception of unidentified pathogen effectors. Intriguingly, when HTHH conditions are prolonged without pathogen attack or the temperature goes higher, CaHSF8 no longer interacts with CaKAN3. Instead, it directly upregulates a subset of HSP genes thus activating thermotolerance. Our findings highlight mechanisms controlling context-specific activation of high-temperature-specific pepper immunity and thermotolerance mediated by differential CaKAN3-CaHSF8 associations.

Fig. 1. CaKAN3 silencing significantly increased the susceptibility of pepper plants to R. solanacearum inoculation (RSI) under HTHH.

a The success of CaKAN3 silencing by virus-induced gene silencing (VIGS) by measuring the transcript levels of CaKAN3 in RTHH-, RSRT-, HTHH- and RSHT-challenged TRV:CaKAN3 pepper plants at 24 hours post-treatment (hpt). The transcript levels of TRV:00/RTHH were set to 1. b Effect of CaKAN3 silencing on the response of pepper plants to RSRT and RSHT treatment at 3 and 12 dpt, respectively. c The disease index of CaKAN3-silenced pepper plants challenged with RSRT or RSHT from 0 to 12 dpt. A total of 24 plants were dynamically scored for each species. In b, c, the experiment was carried out three times with similar results. d Growth of R. solanacearum in R. solanacearum-inoculated CaKAN3-silenced plants at room temperature or under HTHH, shown as colony-forming units (cfu). Data are shown as the means ± standard errors of eight replicates. Asterisks above the bars indicate significant differences between means (P < 0.01), as calculated with a t test. The center line represents the median value and the boundaries indicate the 25th percentile (upper) and the 75th percentile (lower). Whiskers extend to the largest and smallest value. e Decreased flg22-induced H₂O₂ production in CaKAN3-silenced pepper plants at HTHH. The results shown are representative of two independent experiments. Data are shown as the means ± standard errors of six replicates. f, Relative transcript levels of CaMgst3 and CaPRP1 in TRV:00 and TRV:CaKAN3 pepper plants challenged with RTHH, RSRT, HTHH or RSHT. The transcript levels of TRV:00/RTHH were set to 1. In a and f, data represent the mean ± SD of four replicates. CaActin was used as an internal control, and asterisks above the bars indicate significant differences between means (P < 0.01), as calculated with Fisher’s protected t test. All replicates were from different plants. In a–f, source data are provided as a Source Data file.

Fig. 2. CaKAN3 interacted with CaHSF8.

a Comparative assay of interacting proteins of CaKAN3 in pepper plants challenged by RSRT or RSHT, shown as Venn diagrams, the experiment was carried out once. b The top 10 specific interacting proteins of CaKAN3 with high levels of confidence in pepper plants challenged by RSHT. c BiFC analysis of the interaction between CaKAN3 and CaHSF8 in N. benthamiana leaves. NbH2B (histone H2B)-RFP was used to indicate the nucleus. Red fluorescence and yellow fluorescence, visible light and merged images were taken on a confocal microscope at 48 hpi. Bars = 25 μm. d In vitro interaction between CaHSF8 and CaKAN3 revealed using MST. CaKAN3-GST was regarded as the target, and the CaHSF8-6×His protein was used as the ligand and diluted to a range of concentrations from 1.0E-10 mM to 1.0E-3 mM. The mixtures of CaKAN3-GST/EV-6×His or CaKAN3-GST/CaHSF8-6×His were loaded into Monolith NT.115 capillaries, which were measured using 50% IR laser power and an LED excitation source with λ = 470 nm at ambient temperature. e Pull-down assay revealing the in vitro interaction between CaHSF8 and CaKAN3. CaHSF8-GST was incubated with CaKAN3-6×His and Ni Smart beads for 3 h at 4 °C under slow rotation. The bound proteins were eluted from the beads and detected using an anti-His or an anti-GST antibody. f Interaction between CaKAN3 and CaHSF8 in vivo, as determined by a coimmunoprecipitation assay. Proteins were isolated from pepper leaves transiently overexpressing CaHSF8-GFP/CaKAN3-Myc, CaHSFA1-GFP/CaKAN3-Myc, CaHSF8-GFP/CaKAN4-Myc and CaHSFA1-GFP/CaKAN4-Myc, which were immunoprecipitated with an anti-Myc antibody. The presence of the tested interacting proteins was detected using an antibody against GFP by western blotting. g Analysis of the domain of the interaction between CaKAN3 and CaHSF8 by pull-down, revealing that HR-A/B in CaHSF8 is responsible for the CaKAN3 and CaHSF8 interaction. In c, f CaHSFA1 and CaKAN4 were used as negative controls, respectively. In c to g, the experiment was carried out twice with similar results.

Fig. 3. Determination of DNA-binding sites and target genes of CaKAN3/CaHSF8 by ChIP-seq.

a Genome-wide distribution of DNA-binding peaks of CaKAN3/CaHSF8. b The reads per million (RPM) of CaHSF8, CaKAN3 and input are shown as a heatmap, and 199 cotargeted genes by CaHSF8/CaKAN3 are shown as Venn diagrams. c The genes cotargeted by CaHSF8/CaKAN3 were enriched in different KEGG signaling pathways. d Integrative Genomics Viewer (IGV) images of ChIP-seq data and the locations of HSE and AACAA motifs within the promoters of the CaR1B23, CaR1B11, CaR1A, CaR1B12, CaR1A6 and CaR1B-16 genes that are cotargeted by CaHSF8/CaKAN3. e Both CaKAN3 and CaHSF8 exhibited higher levels of enrichment on the promoters of the 6 tested NLR genes. GV3101 cells containing 35 S:CaKAN3-HA and 35 S:CaHSF8-HA were infiltrated into pepper leaves, which were harvested at 48 hpi for ChIP‒qPCR analysis using specific primer pairs. IP using IgG beads was used as the control. The enrichment levels of the tested genes were compared with those in the control, and the relative enrichment of IP using anti-HA was set to a value of 1 after normalization by input. Data are shown as the means ± standard errors of three replicates. Asterisks above the bars indicate significant differences between means (P < 0.01), as calculated with Fisher’s protected t test. CaWRKY40 and CaWRKY58 were used as negative controls. The ratio of IP:anti-HA to IP:IgG is indicated on the error line of IP:IgG. All replicates were from different plants. Source data are provided as a Source Data file. f CaKAN3-GST and CaHSF8-GST bound the promoters of the 6 NLR genes in an AACAA-element- and HSE-dependent manner, as shown by EMSA. The experiment was carried out twice with similar results.

Fig. 4. Relationship between CaKAN3 and CaHSF8 in the regulation of the six tested NLRs at 1 hpt of HTHH treatment.

a, b The 6 tested NLR genes, CaR1B23, CaR1B11, CaR1A, CaR1B12, CaR1A6 and CaR1B-16, were upregulated by CaHSF8 transient overexpression, but this upregulation was blocked by CaKAN3 silencing in pepper plants at 1 hpt of HTHH, and vice versa. The 6 NLR genes were downregulated by CaKAN3 transient overexpression at 0 hpt and by CaHSF8 silencing, and CaKAN3 and CaHSF8 functioned independently at 0 hpt of HTHH. c The co-transient overexpression of CaKAN3 and CaHSF8 induced higher expression levels of the 6 tested NLR genes than transient overexpression of CaKAN3 and CaHSF8 individually at 28 °C. d By ChIP‒qPCR, CaKAN3 silencing significantly reduced the enrichment of CaHSF8 on the promoters of the tested NLR genes, but CaHSF8 silencing did not reduce that of CaKAN3 on these promoters. The enrichment of IP:anti-HA was set to 1 after normalization by input. The ratio of IP:anti-HA to IP:IgG is indicated on the error line of IP:IgG. In a–c CaActin was used as an internal control. Data are shown as the means ± standard errors of four replicates. Different uppercase letters above the bars indicate significant differences (P < 0.01) by Fisher’s protected LSD test. All replicates were from different plants. In a–d, source data are provided as a Source Data file.

Fig. 5. CaHSF8 was differentially associated with CaKAN3 in a temperature-dependent manner to coordinately activate NLR genes and HSPs.

a The data from BiFC showed that the interaction between CaKAN3 and CaHSF8 weakened as the temperature rose from 28 to 45 °C, and this interaction also weakened at 37 °C with time from 0 to 6 hpt. The experiment was carried out once. b Analysis of the interaction intensity between CaKAN3 and CaHSF8 under treatment at 28, 34, or 45 °C at 1 hpt and 1, 3, or 6 hpt at 37 °C. CaKAN3 fused with the C-terminus of Luc (LUC^c) was coexpressed with CaHSF8 fused with the N-terminus of Luc (LUC^N) in N. benthamiana leaves by GV3101 cells carrying different plasmids (using CaHSFA1-LUC^N and CaKAN4-LUC^c as negative controls). c CaKAN3-GST exhibited a reduced binding affinity to CaHSF8 at 37 °C/1 hpt compared to 37 °C/0 hpt, and CaKAN3-GST did not bind CaHSF8 at 37 °C/6 hpt or 45 °C/1 hpt by MST assay. d, e Analysis of the interaction intensity between CaKAN3 and CaHSF8 in vivo under treatment at 28, 34, 37, or 45 °C at 1 hpt and under treatment at 37 °C at 1, 3, and 6 hpt, as determined by co-IP assay. The same amounts of proteins isolated from pepper leaves transiently overexpressing CaHSF8-Myc and CaKAN3-HA were used, and the interacting partners of CaHSF8 were immunoprecipitated with an antibody against Myc. The presence of CaKAN3 in the protein complex was assayed by western blotting using an antibody against HA. f The transcript levels of the 6 tested NLR genes under transient overexpression of CaHSF8 at 1, 3 and 6 hpt of 28 or 37 °C treatment and under treatment at 28, 37 or 45 °C at 1 hpt. g Transcript levels of HSPs under transient overexpression of CaHSF8 under 28 or 37 °C treatment at 1, 3 or 6 hpt and under 28, 37 or 45 °C treatment at 1 hpt. h The mechanisms controlling the coordinative and context-specific activation of high-temperature-specific pepper immunity against RSI and thermotolerance mediated by differential CaKAN3-CaHSF8 association. In f and g, data are shown as the means ± standard errors of four replicates. Different uppercase letters or asterisks above the bars indicate significant differences (P < 0.01) by Fisher’s protected LSD or t test. Source data are provided as a Source Data file. In d and e, the experiment was carried out twice with similar results. All replicates were from different plants.