A myeloid differentiation-like protein in partnership with Toll5 from the pest insect Spodoptera litura senses baculovirus infection

Abstract

Many types of viruses infect insects and other arthropods. In contrast, little is known about how arthropods sense viruses, although several innate immune pathways including Toll have antiviral functions. Large DNA viruses in the family Baculoviridae are used to control a number of pest insects. Here, we studied Spodoptera litura and Autographa californica multiple nucleopolyhedrovirus (AcMNPV) to test the hypothesis that one or more myeloid differentiation-like (ML) proteins and Toll family members sense baculoviruses. We identified 11 ML and 12 Toll genes in the S. litura genome. A series of experiments indicated that S. litura ML protein 11 (SlML-11) binds the budded form of AcMNPV and partners with S. litura Toll5 (SlToll5). SlML-11 also bound sphingomyelin (SPM), which is a component of the virion envelope. Disabling SlML-11 and SlToll5 increased susceptibility to infection, while priming larvae with SPM reduced susceptibility as measured by increased survival to the adult stage and clearance of AcMNPV from individuals that emerged as adults. We conclude that SPM is a pathogen-associated molecular pattern molecule while SlML-11 and SlToll5 interact to function as a pattern recognition receptor that senses AcMNPV.

Keywords: immunity; insect; pattern recognition receptor; virus.

Fig. 1.

Recombinant SlML-11 slows AcMNPV infection of Sl221 cells. (A) Cells were infected at an MOI of 3.6 in medium containing SlML-11 or RFP, and examined by light and epifluorescence microscopy at 24, 36, and 48 hpi. Infected cells (green) are visible in the epifluorescent images while the number of infected cells per field of view from five biological replicates is shown in the graphs to the right. Scale bars in the light microscopy images at 24 hpi equal 200 μm. (B) Infection at 24 to 48 hpi as measured by AcMNPV titer. (C) Infection as measured by vp39 transcript abundance. (D) Infection as measured by VP39 abundance on immunoblots using Tubulin as a loading control. (E) Quantification of VP39 band intensity from three independent biological replicates that were immunoblotted as in D. Bars indicate mean intensity ± SEM. Treatments at each time point postinfection in A–D were compared to the RFP control by the t test. Asterisks: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. (F) Viral DNA per Sl221 cell when infected at 4 °C in medium containing SlML-11, RFP, or no recombinant protein (Control). Three independent assays were conducted per treatment followed by comparing cells in SlML-11 or RFP to cells in medium with no recombinant protein by the t test. ***P ≤ 0.001.

Fig. 2.

SlML-11 binds AcMNPV and SPM. (A) Uninfected Sl221 cells (Control) or Sl221 cells transiently transfected with pIEx-4-RFP-Flag (RFP), pIEx-4-SlML-9-Flag (SlML-9), or pIEx-4-SlML-11-Flag (SlML-11) and cultured for 8 h before infection with AcMNPV at an MOI of 3.6. Medium was collected 48 hpi and used in IP assays. Anti-Flag antibody plus protein G/A beads immunoprecipitated RFP-Flag, SlML-9-Flag, or SlML-11-Flag as detected by probing the immunoblot (Upper) with anti-Flag. Probing the immunoblot (Lower) with anti-VP39 showed that AcMNPV coimmunoprecipitated with only SlML-11-Flag. Molecular mass markers and their size (kDa) are indicated to the left of each blot. (B) Molecular docking of SlML-11 with PS, SPM, or LPS. The smaller value in free energy (Kcal/mol) indicates stronger predicted affinity. (C–F) SlML-11 binds SPM but not PS or LPS. SPR sensorgrams show SPM binding to surface-immobilized SlML-11 (C) with a K_D of 70.2 ± 0.15 μM (D), but no binding to PS (E) or LPS (F). Assays shown in A and C–F were performed in triplicate using independent biological samples but only a single representative result is presented.

Fig. 3.

RNAi knockdown of SlToll5 increases AcMNPV infection of Sl221 cells. (A) Relative transcript abundance of SlToll family members in AcMNPV-infected and age-matched uninfected (Control) Sl221 cells. Cells were infected at an MOI of 3.6 and analyzed at 48 hpi. Transcript abundances for each toll family member were measured in three independent samples, which were then compared by the t test. **P ≤ 0.01, ***P ≤ 0.001, ns (nonsignificant). (B) Pretreatment of Sl221 cells with ds-toll5 reduces transcript abundance of Sltoll5 when compared to cells treated with ds-rfp. Three independent biological samples for each treatment were measured at 36 and 48 h posttreatment. (C) Pretreatment of Sl221 cells with ds-toll5 followed by AcMNPV at an MOI of 3.6 increases the number of cells that are infected at 24 and 36 hpi when compared to ds-rfp pretreated controls. Cells were examined by light and epifluorescence microscopy. Infected cells (green) are visible in the epifluorescent images while the number of infected cells per field of view from five biological replicates is shown in the graph to the right. Scale bars in the light microscopy images at 24 hpi equal 200 μm. (D) AcMNPV titer at 24 and 48 hpi as measured by TCID50 assay. (E) Infection as measured by VP39 protein abundance on immunoblots using Tubulin as a loading control. (F) Quantification of VP39 band intensity from three independent biological replicates that were immunoblotted as shown in E. Bars indicate mean intensity ± SEM. Treatments in B, C, and F were compared using a t test. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Fig. 4.

SlML-11 binds AcMNPV and SlToll5. (A and B) Control S2 cells (nontransfected) or S2 cells transiently transfected with pIEx-4-RFP-Flag (RFP), pIEx-4-SlML-11-Flag (ML-11), or SlToll5^ecto-V5 (Toll5^ecto) were cultured for 48 h. Medium was then collected and used separately, after combining 1:1, or after combining 1:1 and adding SPM in IP assays. (A) Anti-V5 antibody plus protein G/A beads immunoprecipitated SlToll5^ecto-V5 (Upper), while probing the immunoblot (Lower) with anti-Flag showed that SlML-11-Flag but not RFP-Flag was coimmunoprecipitated when SlToll5^ecto-V5 was present. (B) Anti-Flag antibody plus protein G/A beads immunoprecipitated SlML-11-Flag and RFP-Flag (Upper), while probing the immunoblot (Lower) with anti-V5 showed that SlML-11-Flag but not RFP-Flag coimmunoprecipitated SlToll5^ecto-V5. Adding SPM did not alter the amount of SlToll5^ecto-V5 that was coimmunoprecipitated by SlML-11-Flag. (C) Control Sl221 cells (nontransfected) or Sl221 cells transiently transfected with pIEx-4-SlToll5^ecto-V5 (Toll5^ecto) alone or cotransfected with SlToll5^ecto-V5, RFP-Flag (RFP), or SlML-11-Flag (ML-11) for 48 h. Medium was used separately, after combining 1:1, or after adding the same amount of AcMNPV as used in infection assays. Anti-V5 plus protein G/A beads immunoprecipitated SlToll5^ecto-V5 (Upper), while probing the immunoblot with anti-VP39 (Middle) or anti-Flag (Lower) showed that anti-V5 coimmunoprecipitated AcMNPV when SlToll5^ecto-V5 and SlML-11-Flag were present. Molecular mass markers (kDa) are indicated to the left of each immunoblot shown in A–C. (D) Confocal microscopy images of uninfected (Control) and AcMNPV-infected Sl221 cells at 36 hpi. Control cells were transiently cotransfected with pIEx-4-SlToll5^ecto-TM-V5, pIEx-4-SlML-11-Flag, and pIEx-4-GFP while treatment cells were transiently cotransfected with pIEx-4-SlToll5^ecto-TM-V5, and pIEx-4-SlML-11-Flag followed by infection with AcMNPV. The top row shows light images. Below this are nuclei stained with 4′,6-diamidino-2-phenylindole (DAPI) (blue), GFP (green), SlToll5^ecto-TM—V5 (purple), SlML-11-Flag (red), or images showing merged signals. Intracellular GFP is detected in Control and AcMNPV-infected cells. SlToll5^ecto-TM—V5 and SlML-11-FLAG are detected on the surface of cells. (E) Fluorescence intensity for SlToll5^ecto-TM—V5 and SlML-11-FLAG is higher in infected than control cells with bars showing mean intensity ± SE, while circles or triangles show values for each cell that was measured. Treatments were compared by the t test: **P ≤ 0.01.

Fig. 5.

Neutralization of SlML-11 and SlToll5 in S. litura larvae accelerates AcMNPV-induced mortality. (A) Immunoblot showing increased detection of SlML-11 in plasma 0 to 104 hpi by AcMNPV. (B) Immunoblot showing increased detection of SlToll5 in hemocytes relative to Tubulin (loading control) 0 to 104 hpi by AcMNPV. Molecular mass markers are indicated to the Left of the blots. Graphs to the Right of the blots in A and B show quantification of SlML-11 and SlToll5 band intensity from three independent biological replicates. Bars indicate mean intensity ± SE, with different letters indicating treatments differed as determined by ANOVA and a post hoc Tukey multiple comparison test (P > 0.05). (C) Day 3 fifth instar S. litura larvae were injected with affinity-purified control (preimmune) IgG, anti-SlML-11 IgG, anti-SlToll5 IgG, or anti-SlML-11 and anti-SlToll5 IgG. After 2 h, AcMNPV was injected into antibody-injected and untreated larvae followed by measuring survival in hours postinjection. Kaplan–Meier survival curves were analyzed by Mantel–Cox tests which indicated treatment differences. ****P ≤ 0.0001; nonsignificant (ns) P > 0.05. (D) AcMNPV genomic DNA in the fat body 0 to 120 hpi for the treatments shown in C. (E) AcMNPV genomic DNA in hemocytes 0 to 120 hpi for the treatments shown in C. Data in D and E show mean ± SE from three biological replicates. Treatments were compared at 120 hpi by the t test: *P ≤ 0.05, ns (nonsignificant) P > 0.05). (F) Relative transcript abundance for four antimicrobial peptide (AMP) genes (moricin, gloverin, lebocin, and attacin) and vago 2 in the fat body (fb) at 114 hpi for the treatments shown in C. (G) Transcript abundance for moricin, gloverin, lebocin, attacin, and vago 2 in hemocytes at 114 hpi for the treatments shown in C. Data in F and G show mean relative expression ± SE for three biological replicates. Different letters above bars indicate treatments significantly differed as determined by ANOVA and a post hoc Tukey multiple comparison test (P ≤ 0.05).

Fig. 6.

Priming S. litura larvae with SPM before AcMNPV infection increases survival. (A) Immunoblots showing increased detection of SIML-11 after injection of SPM but not PS or PBS. Molecular mass markers (kDa) are indicated to the left. The graph to the right of the blots shows SlML-11 mean band intensity ± SE for immunoblots from independent samples. Different letters above a bar at each time point indicate treatments differed as determined by ANOVA and a post hoc Tukey multiple comparison test (P ≤ 0.05). (B and C) Increased transcript abundance of cecropin A (Cec), lebocin 2 (Leb 2), and moricin (Mor) in fat body and hemocytes after priming with PS or SPM. Day 3 fifth instar S. litura were injected with PS or SPM followed by collection of fat body and hemocytes 24 h postinjection. Data show mean ± SE fold changes. Treatments were compared by t test: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ns (nonsignificant) P > 0.05. (D) Increased survival of larvae primed with SP-M versus PBS or PS before AcMNPV infection as determined by Kaplan–Meier curves and Mantel–Cox tests. Treatment differences **** P ≤ 0.0001; nonsignificant (ns) P > 0.05. (E and F) AcMNPV genomic DNA detected in fat body and hemocyte samples 0 to 120 h postinfection (hpi). Significance was determined by comparing larvae primed with SPM, PS or PBS followed by infection with AcMNPV. Treatments were compared by t test: ***P ≤ 0.001; nonsignificant (ns) P > 0.05. (G). qPCR analysis of AcMNPV genomic DNA in the exuvia of pupae and whole-body samples from adults that were primed as larvae with SPM before infection. Bars show mean ± SE for five pupal exuvia and five adults. Values for each are below the limit of detection (LOD) for the assay.