The barley powdery mildew candidate secreted effector protein CSEP0105 inhibits the chaperone activity of a small heat shock protein

Abstract

Pathogens secrete effector proteins to establish a successful interaction with their host. Here, we describe two barley (Hordeum vulgare) powdery mildew candidate secreted effector proteins, CSEP0105 and CSEP0162, which contribute to pathogen success and appear to be required during or after haustorial formation. Silencing of either CSEP using host-induced gene silencing significantly reduced the fungal haustorial formation rate. Interestingly, both CSEPs interact with the barley small heat shock proteins, Hsp16.9 and Hsp17.5, in a yeast two-hybrid assay. Small heat shock proteins are known to stabilize several intracellular proteins, including defense-related signaling components, through their chaperone activity. CSEP0105 and CSEP0162 localized to the cytosol and the nucleus of barley epidermal cells, whereas Hsp16.9 and Hsp17.5 are cytosolic. Intriguingly, only those specific CSEPs changed localization and became restricted to the cytosol when coexpressed with Hsp16.9 and Hsp17.5, confirming the CSEP-small heat shock protein interaction. As predicted, Hsp16.9 showed chaperone activity, as it could prevent the aggregation of Escherichia coli proteins during thermal stress. Remarkably, CSEP0105 compromised this activity. These data suggest that CSEP0105 promotes virulence by interfering with the chaperone activity of a barley small heat shock protein essential for defense and stress responses.

Figure 1.

Silencing of CSEP0105 and CSEP0162 by HIGS reduces the Bgh haustorial formation rate. One-week-old barley leaves were bombarded with an RNAi construct and a GUS reporter construct. Two days later, leaves were infected with Bgh, and at 3 dpi, they were stained with a 5-bromo-4-chloro-3-indoxyl-β-d-glucuronic acid, cyclohexylammonium salt solution and scored for fungal haustorial formation rate, which was calculated as the number of GUS-expressing cells with haustoria divided by the total number of GUS-expressing cells. The relative haustorial formation rate of each construct was obtained by comparing with the empty vector control of each experiment, which was set to 100%. Data represent means ± se of five independent experiments, except for Mlo RNAi (n = 3). A total of 1,821, 2,102, 1,779, and 617 cells were assessed for the empty vector, CSEP0105, CSEP0162, and Mlo RNAi, respectively. Bars marked with different letters are significantly different at P < 0.01.

Figure 2.

Expression patterns of CSEP0105 and CSEP0162 at different stages of Bgh development. Total RNA was isolated from Bgh-infected (isolate DH14) barley leaves (cv Golden Promise) at 0, 3, 6, 12, 24, and 48 hpi. H and E denote haustorial and epiphytic expression. Expression of Bgh glyceraldehyde 3-phosphate dehydrogenase was used to normalize the CSEP expression in each sample. Relative expression was determined compared with time point 0 hpi, arbitrarily set to 1. Three biological and two technical repetitions were included for each time point. Data shown are means ± se of three independent biological repetitions.

Figure 3.

CSEP0105 and CSEP0162 interact with the barley Hsp16.9 and Hsp17.5 proteins in a Y2H assay. Yeast was transformed with bait and prey constructs. Interactions were selected on dropout (DO) medium lacking Leu (L), His (H), adenine (A), and Trp (W), supplemented with 5 mm 3-AT. The β-galactosidase assay (5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside [X-gal]) was performed on a filter paper print of DO-Leu-Trp-grown yeast. SNF4/SNF1 interaction was used as a positive control, and CSEP0081 and CSEP0254 were used as negative controls.

Figure 4.

Amino acid-based phylogenetic tree of all 37 identified full-length barley sHsps, tomato RSI2, Ntshsp17, and Nbshsp17. The lower left branches mainly contain cytosolic (black) sHsps with high similarity to each other. Those with noncytosolic and unknown localizations are shown in purple and green, respectively. Letters following the molecular weight indicate their localizations: C, cytosolic; ER, endoplasmic reticulum; MI, mitochondria; P, chloroplast; and Px, peroxisome. The two sHsps identified in the Y2H screen, RSI2, Nthsp17, Nbhsp17, and Mds1 are indicated by arrows. The naming follows Reddy et al. (2014) for those starting with HvHsp. The MLOC names are from the annotation of the barley genome. There are four sequences solely based on cDNA clones, with GenBank accession numbers given.

Figure 5.

CSEP0105, CSEP0162, and CSEP0254 have cytosolic and nuclear localization in barley epidermal cells, whereas Hsp16.9 and Hsp17.5 are cytosolic. Ubiquitin promoter-driven expression constructs were generated for all. A to C, Constructs encoding signal peptide-lacking CSEPs fused to the C terminus of YFP were cotransformed with a free mCherry construct into barley leaf epidermal cells using particle bombardment. Two to 4 d later, cells were analyzed using a Leica SP5 confocal laser scanning microscope. All the YFP-CSEP fusion signals are in the cytosol (C) and the nucleus (N). D and E, Constructs encoding full-length Hsps fused to the N terminus of mCherry were cotransformed with a free YFP construct. Both Hsp-mCherry signals are in the cytosol. The free mCherry and YFP markers localize to the cytosol and nucleus. Bars = 20 µm.

Figure 6.

CSEP0105 and CSEP0162 accumulate exclusively in the cytosol when coexpressed with Hsp16.9 and Hsp17.5. The same YFP-CSEP and Hsp-mCherry fusion constructs were used as described in Figure 5. The YFP-CSEP constructs were cotransformed with each Hsp-mCherry construct into barley leaf epidermal cells using particle bombardment. Two to 4 d later, individual cells were visualized using a Leica SP5 confocal laser scanning microscope. CSEP0254 was included as a negative control. The YFP-CSEP0105, YFP-CSEP0162, Hsp16.9-mCherry, and Hsp17.5-mCherry signals are in the cytosol (C), whereas YFP-CSEP0254 is in the cytosol and the nucleus (N). Bars = 20 µm.

Figure 7.

Hsp16.9 protects the heat denaturation of E. coli proteins in vitro. Thermostability is shown for total protein from E. coli expressing GUS or Hsp16.9. Cell-free total protein extracts (250 µg mL⁻¹) were heated for 15 min. Heat-denatured proteins were removed by centrifugation, and the protein content of the supernatant fractions was determined. Values are relative to the unheated controls. Data shown are means of three replicates ± se. Asterisks show significant differences (P < 0.05) between the GUS control and Hsp16.9 at each heat treatment analyzed by Duncan’s multiple range test. The experiment was repeated once with a similar result.

Figure 8.

CSEP0105 inhibits the chaperone activity of Hsp16.9. Thermostability is shown at 60°C for total protein from E. coli expressing GUS or Hsp16.9 with and without CSEPs. Purified CSEPs were added in a 1:2 protein molar ratio with Hsp16.9 in the cellular lysate (250 µg mL⁻¹). CSEP0254 was added as a negative control. Values are relative to the unheated controls. Data shown are means of three replicates ± se. Means with different letters are significantly different (P < 0.05). Duncan’s multiple range test was used to compare all the means. The experiment was repeated once with a similar result.

Figure 9.

Localization of CSEP0105, CSEP0162, Hsp16.9, and Hsp17.5 proteins in Bgh-infected barley cells. Confocal images show cells transiently cotransformed by particle bombardment with the respective constructs as described in Figure 5. The Hsp16.9-mCherry and Hsp17.5-mCherry fluorescent signals are in the cytosol (C). YFP-CSEP0105 fluorescent signal is in the cytosol when coexpressed with Hsp16.9-mCherry and in the cytosol and nucleus (N) when coexpressed with the free mCherry construct. YFP-CSEP0162, as well as the free YFP and mCherry marker proteins, localized to the cytosol, nucleus, and EHMx. The fluorescent signals labeled EHMx also include the cytosol surrounding the haustoria (H). Since the nucleus and haustoria were not found in a single confocal plane, images showing nuclei of the same cells are presented in Supplemental Figure S3. Bars = 20 µm.