CLCuMuB βC1 Subverts Ubiquitination by Interacting with NbSKP1s to Enhance Geminivirus Infection in Nicotiana benthamiana

Abstract

Viruses interfere with and usurp host machinery and circumvent defense responses to create a suitable cellular environment for successful infection. This is usually achieved through interactions between viral proteins and host factors. Geminiviruses are a group of plant-infecting DNA viruses, of which some contain a betasatellite, known as DNAβ. Here, we report that Cotton leaf curl Multan virus (CLCuMuV) uses its sole satellite-encoded protein βC1 to regulate the plant ubiquitination pathway for effective infection. We found that CLCuMu betasatellite (CLCuMuB) βC1 interacts with NbSKP1, and interrupts the interaction of NbSKP1s with NbCUL1. Silencing of either NbSKP1s or NbCUL1 enhances the accumulation of CLCuMuV genomic DNA and results in severe disease symptoms in plants. βC1 impairs the integrity of SCFCOI1 and the stabilization of GAI, a substrate of the SCFSYL1 to hinder responses to jasmonates (JA) and gibberellins (GA). Moreover, JA treatment reduces viral accumulation and symptoms. These results suggest that CLCuMuB βC1 inhibits the ubiquitination function of SCF E3 ligases through interacting with NbSKP1s to enhance CLCuMuV infection and symptom induction in plants.

Fig 1. NbSKP1s interacts with CLCuMuB βC1 in vitro and in vivo.

(A) Growth of SKY48 yeast strains containing NLS-LexA BD-CLCuMuB βC1 transformed with AD-NbSKP1s, AD-NbSKP1L1 or AD (control) on Leu-containing (Leu⁺) and Leu-deficient (Leu⁻) medium with galactose (Gal) and raffinose (Raf) at 28°C for 4 d. Yeast cells were plated at OD₆₀₀ = 1, 0.1, 0.01. (B) In vitro GST pull-down assays. His-HA-NbSKP1.1 and His-HA-NbSKP1L1 were pulled down by GST-CLCuMuB βC1 (GST-βC1), GST or GST-βC1ΔC43. βC1ΔC43 represents a βC1 mutation with the deletion of C-terminal 43 amino acids. GST beads were washed and proteins were analyzed via SDS-PAGE and western blot assays using anti-GST and anti-HA antibodies. (C) Co-immunoprecipitation (co-IP) assays show that CLCuMuB βC1 interacted with NbSKP1.1 and NbSKP1L1 in vivo. GFP-tagged CLCuMuB βC1 (GFP-βC1) was co-expressed with 2×HA-tagged NbSKP1.1 or NbSKP1L1 (HA-NbSKP1.1 or HA-NbSKP1L1) in N.benthamiana leaves by agroinfiltration. GFP co-expressed with HA-NbSKP1.1 or HA-NbSKP1L1 was introduced as a negative control. At 48 hpi, leaf lysates were immunoprecipitated (IP) with GFP-Trap agarose, then the immunopercipitates were detected by western blotting using anti-GFP and anti-HA antibodies. (D) A confocal image of BiFC shows a positive result in leaf epidermal cells. NbSKP1.1 or NbSKP1L1 fused with HA and the C-terminal fragment of YFP (HA-cYFP-NbSKP1.1 or HA-cYFP-NbSKP1L1) was transiently co-expressed in leaves of N. benthamiana with CLCuMuB βC1 or βC1ΔC43 fused with HA and N-terminal fragment of YFP (HA-βC1-nYFP or HA-βC1ΔC43-nYFP). Bar scale represents 40 μm. Photos were imaged at 48 hpi using a Zeiss LSM 710 laser scanning microscope. nLUC represents the N-terminal fragment of firefly luciferase.

Fig 2. The N-terminal domain of NbSKP1.1 interacts with CLCuMuB βC1 in yeast.

Growth of SKY48 yeast strains containing NLS-LexA BD-CLCuMuB βC1 (BD-βC1) transformed with AD fused full length, N-terminal fragment (N98aa), C-terminal fragment (C57aa) of NbSKP1.1 or AD (control) on Leu-containing (Leu⁺) and Leu-deficient (Leu⁻) medium with galactose (Gal) and raffinose (Raf) at 28°C for 6 d. Yeast cells were plated at OD₆₀₀ = 1, 0.1, 0.01.

Fig 3. CLCuMuB βC1 interferes with the interaction between NbCUL1 and NbSKP1.1 in vitro and in vivo.

(A) GFP competitive pull-down assay in vitro. His-βC1 was expressed in E. coli as inclusion body and refolded through urea-arginine dialysis. BSA (NEB, USA) was used as a control. GFP-NbCUL1 or GFP was expressed in N. benthamiana leaves and trapped through GFP-Trap agarose. After the supernatant was discarded, GFP-Trap agarose was incubated with E. coli-expressed His-HA-NbSKP1.1, then the supernatant was discarded. GFP-Trap agarose was incubated with gradient dilutions (1, 1/2, 1/4) of His-βC1. Finally, agarose was washed and proteins were analyzed via SDS-PAGE and western blot assays using anti-GFP and anti-HA antibodies. Input was analyzed by the anti-His antibody (EASYBIO, China) and supernatant was analyzed by the anti-HA antibody. Intensity was detected through Total Lab TL120. (B) A confocal image of BiFC assays show that CLCuMuB βC1 interfered with the interaction between NbCUL1 and NbSKP1.1 in vivo. Photos were taken at 48 hpi. Bar scale represents 200 μm. (C) BiFC intensity (means±SEM, n = 4) was quantified by YFP fluorescence. Relative BiFC intensity was normalized to the control. The raw data were analyzed by two-sample t-test to show the significance level at 0.01 (**). (D) The protein level of cYFP-NbCUL1 and nYFP-NbSKP1.1 were checked with the polyclonal GFP antibody (Huaxin Bochuang, China). The PVDF membrane was stained with Ponceaux to visualize the large subunit of ribulose-1,5-bisphosphate as the loading control.

Fig 4. Silencing of NbSKP1s enhances CLCuMuV DNA accumulation and results in typical disease symptoms.

(A1, A2 and A3) Six- to seven-week-old N. benthamiana plants were agroinoculated with CLCuMuV (CA) and βM2-SKP1F1 (A1), βM2-SKP1F2 (A2), βM2-SKP1F3 (A3) or βM2-βC1F (as the control). (B1, B2 and B3) Silencing of NbSKP1s enhanced CLCuMuV DNA accumulation. Each group contained 7 plants. At 14 dpi, total DNA was extracted from each plant respectively and subjected to quantitative real-time PCR (means±SEM, n = 7) to quantify viral DNA accumulation. The internal reference method was used to calculate the relative amount of viral DNA. (C1, C2 and C3) Real-time RT-PCR confirmed silencing of NbSKP1s. Total RNA was extracted from each plant respectively and subjected to quantitative RT-PCR (means±SEM, n = 4) to quantify NbSKP1s mRNA level. Actin was used as the internal reference. The raw data of (B1–B3) and (C1–C3) were analysed by two-sample t-test to show the significance level at 0.05 (*) and 0.01 (**). These experiments were repeated at least twice. (D1, D2 and D3) 50% plants infected with CA+βM2-SKP1F1 (D1), 50% plants infected with CA+βM2-SKP1F2 (D2) and 100% plants infected with CA+βM2-SKP1F3 (D3) show severe symptoms at 21 dpi. (E1, E2, E3 and E4) Apical leaves of plants infected with CA+βM2-βC1F (E1), CA+βM2-SKP1F1(E2), CA+βM2-SKP1F2 (E3) and CA+βM2-SKP1F3 (E4) at 21 dpi.

Fig 5. Silencing of NbCUL1 enhances CLCuMuV DNA accumulation and results in typical disease symptoms.

(A1 and A2) Six- to seven-week-old N. benthamiana plants were agroinoculated with CLCuMuV (CA) and βM2-CUL1F1 (A1), βM2-CUL1F2 (A2) or βM2-βC1F (as the control). (B1 and B2) Silencing of NbCUL1 enhanced CLCuMuV DNA accumulation. Each group contained 7 plants. At 14 dpi, total DNA was extracted from each plant respectively and subjected to quantitative real-time PCR (means±SEM, n = 7) to quantify viral DNA accumulation. The internal reference method was used to calculate the relative amount of viral DNA. (C1 and C2) Real-time RT-PCR confirmed silencing of NbCUL1. Total RNA was extracted from each plant respectively and subjected to quantitative RT-PCR (means±SEM, n = 4) to quantify NbCUL1 mRNA level. Actin was used as the internal reference. The raw data of (B1 and B2) and (C1 and C2) were analysed by two-sample t-test to show the significance level at 0.05 (*) and 0.01 (**). These experiments were repeated at least twice. (D1 and D2) 100% plants infected with CA+βM2-CUL1F1 (D1) or CA+βM2-CUL1F2 (D2) show severe symptoms at 21 dpi.

Fig 6. CLCuMuB βC1 represses JA responses though interfering with the integrity of SCF^COI1.

(A) Total root length of HA-βC1 transgenic (#2 HA-βC1 and #3 HA-βC1) and wild-type (#2 WT and #3 WT) N. benthamiana seedlings was measured every 24 h beginning at the 6th day after sowing (n ≥11). Bars represent SEM. #2 HA-βC1 and #2 WT were presented on same plates, while #3 HA-βC1 and #3 WT were presented on same plates. These experiments were repeated 3 times. (B) Jasmonate sensitivity was measured as root growth inhibition. Six-day-old seedlings (n ≥10) were grown on MS contained with 50 μM MeJA for additional 4 days. Bars represent SEM. The raw data were analysed by Mann-Whitney rank sum test to show the significance level at 0.05 (*). (C) Relative expression level of marker genes of jasmonate responses in mock- or MeJA-treated HA-βC1 transgenic and wild-type N. benthamiana seedlings determined by quantitative real-time PCR. #2 HA-βC1 and #2 WT were presented on same plates, while #3 HA-βC1 and #3 WT were presented on same plates. HA-βC1-expressing lines are compared with their corresponding control in each condition. Actin was used as the internal control. Bars represent SEM. The raw data were analysed by two-sample t-test to show the significance level at 0.05 (*), 0.01 (**) and 0.001 (***). These experiments were repeated at least twice. (D) CLCuMB βC1 enhanced degradation of COI1 in vitro. The purified Myc-COI1 protein was added to total protein extracts from N.benthamiana which expressed HA-nLUC or HA-βC1, incubated at 25°C for the indicated time periods, and subjected to immunoblot analysis with the anti-Myc antibody. Intensity was detected through Total Lab TL120. The PVDF membrane was stained with Ponceaux to visualize the large subunit of ribulose-1,5-bisphosphate as the loading control. (E) Quantitative analysis of the relative abundance of COI1 in the presence of HA-nLUC or HA-βC1 for the time periods indicated. The abundance of COI1 at the start point (0-h) was set to 100% as a reference for calculating its relative abundance after different incubation periods. Error bars represent SD. The experiment was repeated three times.

Fig 7. CLCuMuB βC1 hinders the degradation of YFP-GAI in vivo.

(A) CLCuMuB βC1 attenued degradation of YFP-GAI in vivo. YFP-GAI expression construct was coinfiltrated with constructs expressing HA-nLUC or HA-βC1 into seven to eight-week-old N. benthamiana plant leaves. Around 48 hpi, agroinfiltrated leaves were sprayed with 100 μM GA₃ or mock solution (ethonal) and visualized via a Zeiss LSM 710 laser scanning microscope. Bar scales represents 200 μm. DMSO and MG132 (50 μM) were applied into plant leaves 12 h before observation. Protein level was analyzed via SDS-PAGE and western blot analysis with the anti-GFP antibody, which also recognizes YFP. The PVDF membrane was stained with Ponceaux to visualize the large subunit of ribulose-1,5-bisphosphate as a loading control. (B) Real-time RT-PCR detected the mRNA level of YFP-GAI. Total RNA was extracted from each N. benthamiana leaves and then subjected to quantitative RT-PCR (means±SEM, n = 3) to quantify YFP-GAI mRNA level. eIF4a was used as the internal reference. (C) CLCuMuB βC1 didn’t affect stability of GFP in vivo. Detection of GFP (as an internal control) in N. benthamiana leaves coinfiltrated with the construct expressing GFP together with constructs expressing HA-nLUC or HA-βC1 and treated with 100 μM GA₃ or mock (ethanol) solution and visualized via a Zeiss LSM 710 laser scanning microscope. Bar scale represents 200 μm. Protein level was analyzed via SDS-PAGE and immunoblot analysis with anti-GFP. The PVDF membrane was stained with Ponceaux to visualize the large subunit of ribulose-1,5-bisphosphate as a loading control.

Fig 8. JA treatment enhances plant defense against CLCuMuV.

(A) Exogenous MeJA treatment delayed the incidence of symptom appearance. Six- to seven-week-old N. benthamiana (10–12 plants per treatment) were agroinoculated with CA+β, treated every other day with 50 μM MeJA or mock solution (ethonal), and recorded for the symptom appearance until 14 dpi. Means came from three independent experiments, Error bars represent SEM. (B-D) Different levels of symptoms (B: the 4^th level, C: the 2^nd level, and D: level 0 showing no symptom). (E) Exogenous MeJA treatment attenued disease symptoms level. Plants were scored for the appearance of symptoms at 14 dpi. (F) Total DNA was extracted from each plant respectively and subjected to real-time PCR (means±SEM, n = 7) to quantify viral DNA accumulation at 14 dpi. (G) Six- to seven-week-old N. benthamiana (10 plants per treatment) were agroinoculated with CA+βM1, treated every other day with 50 μM MeJA or mock solution (ethonal), Total DNA was extracted from each plant respectively and subjected to real-time PCR (means±SEM, n = 7) to quantify viral DNA accumulation at 14 dpi. The internal reference method was used to calculate the relative amount of viral DNA. The raw data of (E, F and G) were analysed by two-sample t-test to show significance level at 0.05 (*) and 0.01 (**). These experiments were repeated at least twice.