Tomato yellow leaf curl virus infection mitigates the heat stress response of plants grown at high temperatures

Abstract

Cultured tomatoes are often exposed to a combination of extreme heat and infection with Tomato yellow leaf curl virus (TYLCV). This stress combination leads to intense disease symptoms and yield losses. The response of TYLCV-susceptible and resistant tomatoes to heat stress together with viral infection was compared. The plant heat-stress response was undermined in TYLCV infected plants. The decline correlated with the down-regulation of heat shock transcription factors (HSFs) HSFA2 and HSFB1, and consequently, of HSF-regulated genes Hsp17, Apx1, Apx2 and Hsp90. We proposed that the weakened heat stress response was due to the decreased capacity of HSFA2 to translocate into the nuclei of infected cells. All the six TYLCV proteins were able to interact with tomato HSFA2 in vitro, moreover, coat protein developed complexes with HSFA2 in nuclei. Capturing of HSFA2 by viral proteins could suppress the transcriptional activation of heat stress response genes. Application of both heat and TYLCV stresses was accompanied by the development of intracellular large protein aggregates containing TYLCV proteins and DNA. The maintenance of cellular chaperones in the aggregated state, even after recovery from heat stress, prevents the circulation of free soluble chaperones, causing an additional decrease in stress response efficiency.

Figure 1. Infected TYLCV-susceptible (967) and resistant (GF967) tomatoes grown at different temperature regimen.

(A) 1. (A) tomato plants of line 967 inoculated with TYLCV and maintained at high temperatures (regime 2: 40–45 ^oC/20–25 ^oC, day/night) and (B) at normal temperatures (regime 1: 22–25 ^oC/18–20 ^oC, day/night). 2: tomato plants of lines GF967 (A) and 967 (B) inoculated with TYLCV and maintained at high temperatures (40–45 ^oC/20–25 ^oC, day/night). Pictures were taken at 28 dpi. (B) qPCR estimation of relative TYLCV amounts in untreated (C) or heat stress (HS) treated TYLCV-susceptible (line 967) and TYLCV-resistant (line GF967) genotypes. The results were normalized using the tomato expressed gene as an internal marker. Results were analyzed using student’s t test. One asterisk denotes p < 0.05. Two asterisks denotes p < 0.01. Bars represent the average and standard deviation of the relative expression from five independent biological repeats. (C) Western blot analyses of TYLCV CP in tomatoes grown at normal temperatures (−, regime 1) and high temperatures (+, regime 2). Samples were taken every seven days after the onset of viral infection (0, 7, 14, 21, 28 dpi). The OE33 chloroplast protein (OE33) was used as unrelated control.

Figure 2. Changes in the patterns of TYLCV CP under constant HS.

TYLCV-infected 967 tomatoes were grown at normal (regime, 1: 22–25 ^oC/18–20 ^oC, day/night) and at high temperatures (regime 2: 40–45 ^oC/20–25 ^oC, day/night). Native proteins, extracted from those tomatoes at 21 and 28 dpi, were subjected to ultracentrifugation on sucrose gradients, which were subsequently divided in 10 fractions; aliquots were analyzed by western blots with antibodies against CP and HSP70 (used as a plant protein marker).

Figure 3. Differences in protein amounts and expression of HS-inducible genes in TYLCV-susceptible (967) and resistant (GF967) tomatoes grown at normal and high temperatures.

(A) Western blot analyses of plant transcription factor HSFA2 and HSPs in tomatoes grown at normal temperature (−, regime 1) and at high temperatures (+, regime 2). The OE33 chloroplast protein was used as protein loading control. (B) qRT-PCR analyses of Hsf, Hsp and Apx gene transcripts of the two lines grown for 21 days at normal temperature (−, regime 1) and at high temperatures (+, regime 2) at 14 dpi. The expression level of each gene was calculated in relation to leaves of susceptible line grown at high temperature. The results were normalized using the β-actin gene as an internal marker. Results were analyzed using student’s t test. One asterisk denotes p < 0.05. Two asterisks denotes p < 0.01. Bars represent the average and standard deviation of the relative expression from five independent biological repeats; pooled leaves of three different plants were taken for each sample.

Figure 4. Analyses of proteins and expression profiles of HS-depended genes upon TYLCV infection.

Detached leaves from uninfected and TYLCV-infected (14, 21 and 49 dpi) tomato plants of line 967 were incubated at ambient (23–25 ^oC) and high (42–43 ^oC) temperatures for 2 h followed by 2 h recovery at ambient temperature. (A) Immunodetection of HSFA2 and HSP90/HSP70 in leaf samples. The OE33 chloroplast protein (OE33) was used as unrelated control. (B) qPCR transcription profile in uninfected (c) and infected (14 and 49 dpi) leaf samples. The expression level of each gene was calculated in relation to uninfected leaves after 2 h heat shock. The results were normalized using the β-actin gene as an internal marker. Results were analyzed using analysis of variance (ANOVA). One asterisk denotes p < 0.05. Two asterisks denotes p < 0.01. Bars represent the average and standard deviation of the relative expression from five independent biological repeats. Bars represent the average and standard deviation of the relative expression from five independent biological repeats. Pooled leaves of three different plants were taken for each sample.

Figure 5. HSFA2 changes in TYLCV-infected tomatoes.

(A) Immunodetection of HSFA2 in cytoplasmic (Cyt) and nuclear (N) protein extracts of heat shock-treated (1 h) tomato leaves of uninfected vs. infected (14 and 21 dpi) 967 tomatoes. Cytoplasmic HSP70 and nuclear Histone 3 were used as internal markers to assess the purity of the cellular fractions. (B) Pull-down assay with purified TYLCV proteins (CP, V2, C1, C2, C3, C4) used as baits for binding with HSFA2 present in leaf total protein extracts. HSFA2 immunodetection revealed potential complexes in plant elutes 1 and/or 2 (E1, E2), but not in E. coli protein extracts (Ec). The immunodetection in crude extracts was used as direct control (D). (C) Immunodetection of HSFA2 in pull down assays, where TYLCV CP preferentially binds HSFA2 in the nuclear protein fraction of uninfected and infected leaves. (D) Co-localization of CP and HSFA2 in cytoplasm and nuclei of infected (14 dpi) tomato leaf (cross leaf section and leaf midrib). Fluorescent microscopy using primary anti-CP antisera and Cy2-labeled secondary antibody, primary anti-HSFA2 antisera and Cy3-labeled secondary antibody; nuclei were DAPI stained. Viral CP appears as green, cellular HSFA2 as red, nuclei as blue; CP co-localizing with HSFA2 as yellow (yellow arrow), CP co-localizing with HSFA2 in nuclei as pink (pink arrow). Bar: 50 μm.

Figure 6. Influence of TYLCV infection on the pattern of HSP90 and HSP70 upon heat shock and recovery.

Infected and uninfected tomato leaves were incubated at room temperature (control), and at 42–43 ^oC for 2 h (heat shock), followed by 2 h recovery at room temperature (recovery). Extracts of native proteins were subjected to ultracentrifugation on sucrose gradients, which were subsequently divided in 10 fractions; aliquots were analyzed by western blots with anti-HSP90 and HSP70 antibodies.