Salivary carbonic anhydrase II in winged aphid morph facilitates plant infection by viruses

Abstract

Aphids are the most common insect vector transmitting hundreds of plant viruses. Aphid wing dimorphism (winged vs. wingless) not only showcases the phenotypic plasticity but also impacts virus transmission; however, the superiority of winged aphids in virus transmission over the wingless morph is not well understood. Here, we show that plant viruses were efficiently transmitted and highly infectious when associated with the winged morph of Myzus persicae and that a salivary protein contributed to this difference. The carbonic anhydrase II (CA-II) gene was identified by RNA-seq of salivary glands to have higher expression in the winged morph. Aphids secreted CA-II into the apoplastic region of plant cells, leading to elevated accumulation of H⁺. Apoplastic acidification further increased the activities of polygalacturonases, the cell wall homogalacturonan (HG)-modifying enzymes, promoting degradation of demethylesterified HGs. In response to apoplastic acidification, plants accelerated vesicle trafficking to enhance pectin transport and strengthen the cell wall, which also facilitated virus translocation from the endomembrane system to the apoplast. Secretion of a higher quantity of salivary CA-II by winged aphids promoted intercellular vesicle transport in the plant. The higher vesicle trafficking induced by winged aphids enhanced dispersal of virus particles from infected cells to neighboring cells, thus resulting in higher virus infection in plants relative to the wingless morph. These findings imply that the difference in the expression of salivary CA-II between winged and wingless morphs is correlated with the vector role of aphids during the posttransmission infection process, which influences the outcome of plant endurance of virus infection.

Keywords: aphid; carbonic anhydrase; cell wall; vesicle trafficking; virus transmission.

Fig. 1.

Salivary components of aphids regulated posttransmission infection of virus in the plant. (A) Schematic representation of the virus transmission process by wingless and winged aphids on local and systemic areas of aphid-infested leaves. The CMV copy numbers of (B) local area and (C) systemic area of viral wingless or winged aphid-infested leaf at 3, 4, and 5 d after infestation (n = 15, Mann–Whitney U test, *P < 0.05, **P < 0.01, ***P < 0.001). (D–G) Time spent in nonpenetration, pooled pathway phase activities in intercellular space of epidermis and mesophyll cells, salivation time, and phloem sap ingestion of wingless vs. winged M. persicae when associated with N. tabacum. Feeding activities during the 8-h feeding period were determined using electrical penetration graph (EPG) technology (n = 20, Mann–Whitney U test, *P < 0.05, ***P < 0.001). (H) The CMV copy numbers on N. tabacum when coinfiltrated with saliva of wingless and winged M. persicae (Mp_WL and Mp_W), the natural CMV vector insect, or coinfiltrated with saliva of wingless and winged A. pisum (Ap_WL and Ap_W), the nonvector that could not feed with N. tabacum. Mp_W+Prot K and Ap_W+Prot K represented the saliva of winged M. persica and A. pisum treated with proteinase K, respectively. Different lowercase letters indicate significant differences (n = 8, one-way ANOVA).

Fig. 2.

Aphid CA-II, localized in the secretory cells of salivary glands, could be delivered into plant cells. (A) Volcano plot showing the number of DEGs in salivary glands of winged vs. wingless aphids for M. persicae and A. pisum. (B) Twenty-one homologous genes had higher expressions in winged vs. wingless aphids for both aphid species. Only four DEGs (with red underline) in M. persicae and A. pisum contained signal peptides but lacked the transmembrane regions and accordingly were considered as candidate secretory proteins in salivary glands. Four biological replicates were conducted in the RNA-seq analysis. (C) qPCR results confirmed that these four genes had higher expression in winged vs. wingless M. persicae (n = 4, Student’s t test, *P < 0.05). (D and E) Knockdown of CA-II and carbonic anhydrase VII (CA-VII) in winged M. persicae reduced the gene expression in salivary glands (n = 4, Student’s t test, *P < 0.05, ***P < 0.001). (F) Winged aphids had higher CA-II than wingless aphids in salivary glands at the mRNA level (n = 4, Student’s t test, **P < 0.01). (G) RNA FISH localization of CA-II in salivary glands of wingless and winged aphids. Average intensity of CA-II in salivary glands of wingless and winged aphids (n = 4, Student’s t test, *P < 0.05). (H) Detection of CA-II in aphid-infested plant leaves by western blot. Lane 1: uninfested leaf; lane 2: aphid-infested leaf; (I) immunohistochemical localization of CA-II in the wingless and winged aphid–infested leaves. Calcofluor white was used to stain the cell wall. CA-II antibody was used to detect M. persicae CA-II in the plant cell. SG, salivary gland; WL, wingless aphid; W, winged aphid; SP, signal peptide; TM, transmembrane region; Ye, protein yellow (LOC111037988); Tre, trehalase-like protein (LOC111032474); Un, uninfested leaves; In, infested leaves; CW, Calcofluor white; CA-II, CA-II antibody; WLI, wingless aphid-infested leaves; WI, winged aphid-infested leaves.

Fig. 3.

CA-II decreased feeding efficiency of winged M. persicae but promoted virus infection in plants. (A) Individuals associated with dsCA-II-injected and dsGFP-injected winged aphids were moved to new leaves or new plants for 2 h’ feeding (n = 100). (B) Movement trajectories and (C) movement distances of dsCA-II-injected and dsGFP-injected winged aphids in petri dishes during 30 min (n = 20, Student’s t test, *P < 0.05). (D–G) The feeding activities including time spent in nonpenetration, pooled pathway phase activities in intercellular space of the epidermis and mesophyll cells, salivation time, phloem sap ingestion, and (H) first time to phloem of dsCA-II-injected and dsGFP-injected winged aphids (n = 20, Mann–Whitney U test, *P < 0.05, **P < 0.01, ***P < 0.001). (I–L) The feeding activities including time spent in nonpenetration, pooled pathway phase activities in intercellular space of the epidermis and mesophyll cells, salivation time, phloem sap ingestion, and (M) first time to phloem of wingless aphids when fed on EV-infiltrated and CA-II-overexpressing plants (n = 20, Mann–Whitney U test, *P < 0.05, **P < 0.01, ***P < 0.001). (N–Q) The relative expression of coat protein (CP) of CMV and TuMV in virus-inoculated leaves or in systemic leaves when artificially inoculated on EV-infiltrated and CA-II-overexpressing plants with viruses (n = 20, Mann–Whitney U test). (R) Schematic representation of virus transmission by wingless and winged aphids when associated with EV-infiltrated and CA-II-overexpressing plants. The relative expression of coat protein of (S) CMV and (T) TuMV in EV- and CA-II-overexpressing plants 3 d after infested by viral wingless and winged aphids (n = 20, Kruskal–Wallis test). Means with different lowercase letters were significantly different. *P < 0.05, **P < 0.01, and ***P < 0.001. ns, no significant difference.

Fig. 4.

Aphid CA-II led to apoplastic acidification and degradation of the cell wall pectin. Apoplastic pH values of (A) dsGFP and dsCA-II aphid-infested leaves and (B) EV and CA-II-overexpressing leaves visualized by HPTS staining (n = 10, Student’s t test). (C) The degradation process of cell wall pectin. HGs, the main component of pectin, were transported to the cell wall in a highly methylesterified form and were demethylesterified in the cell wall by PMEs. PGs cleaved demethylesterified HGs into oligogalacturonides, enhancing the cell wall loosening. Total PG activity of (D) dsGFP and dsCA-II aphid-infested leaves, (E) EV- and CA-II-overexpressing leaves, and (F) leaves with pH value of 5.0 and pH value of 5.5 (n = 10, Student’s t test). Demethylesterified HG (green) immunolabeled with LM20 monoclonal antibodies in (G) dsGFP and dsCA-II aphid-infested leaves, (H) EV- and CA-II-overexpressing leaves, and (I) leaves with pH value of 5.0 and pH value of 5.5. The cell wall (blue) was stained by Calcofluor white. (Scale bar, 50 μm.) The relative fluorescence signal intensity of LM20 was normalized with Calcofluor white (n = 4, Student’s t test, *P < 0.05, **P < 0.01, and ***P < 0.001). Pro, fluorescent signals for the protonated HPTS form; Depro, fluorescent signals for the deprotonated HPTS form; CW, Calcofluor white; OE_CA-II, CA-II-overexpressing plants.

Fig. 5.

Acceleration of vesicle trafficking facilitated virus spread. Shown are representative images of AtRab A5d-mCherry-labeled vesicles and their movement distances in (A) dsGFP vs. dsCA-II aphid-infested leaves, (B) EV vs. CA-II-overexpressing leaves, and (C) leaves with pH value of 5.0 vs. pH value of 5.5. Two vesicles were randomly selected and marked with white arrows (T1 and T2). The tracking trajectories of two marked vesicles within 100 s were shown on the right side. The movement distance of all vesicles was calculated (n = 4, Student’s t test, ***P < 0.001). Subcellular transmission electron microscope images of vesicles of (D) dsGFP and dsCA-II infested leaves, (E) EV- and CA-II-overexpressing leaves, and (F) leaves with pH values of 5.0 and 5.5. (Scale bar, 10 μm.) (G and H) The relative expression of CP of CMV and TuMV in leaves with higher and lower apoplastic pH values (pH = 5.5 vs. 5.0) (n = 20, Student’s t test, **P < 0.01, ***P < 0.001). OE_CA-II, CA-II-overexpressing plants. (I) Colocalization between AtRab A5d-mCherry and S6K-GFP in plant cells under higher and lower apoplastic pH values (pH = 5.5 vs. 5.0) (Pearson’s coefficient, mean ± SE, n = 4).

Fig. 6.

Model of the aphid-derived effector CA-II in promotion of infection of a plant by viruses. Carbonic anhydrase CA-II was more abundant in salivary glands of winged M. persicae than wingless morph, and it could be secreted into intracellular and apoplastic regions of plant cells accompanied with viruses. It produced excessive H⁺ and led to apoplastic acidification. The activity of PGs was then increased to promote degradation of demethylesterified HGs, loosening the cell wall. The secretory vesicle trafficking was accelerated to facilitate the virus translocation from the endomembrane system to apoplast, by which virus particles dispersed from infected cells to neighboring cells. Difference in aphid CA-II quantity resulted in higher virus infection efficiency by the winged aphid morph than by the wingless morph during the posttransmission infection.