Phosphoproteomic Analysis Reveals the Importance of Kinase Regulation During Orbivirus Infection

Abstract

Bluetongue virus (BTV) causes infections in wild and domesticated ruminants with high morbidity and mortality and is responsible for significant economic losses in both developing and developed countries. BTV serves as a model for the study of other members of the Orbivirus genus. Previously, the importance of casein kinase 2 for BTV replication was demonstrated. To identify intracellular signaling pathways and novel host-cell kinases involved during BTV infection, the phosphoproteome of BTV infected cells was analyzed. Over 1000 phosphosites were identified using mass spectrometry, which were then used to determine the corresponding kinases involved during BTV infection. This analysis yielded protein kinase A (PKA) as a novel kinase activated during BTV infection. Subsequently, the importance of PKA for BTV infection was validated using a PKA inhibitor and activator. Our data confirmed that PKA was essential for efficient viral growth. Further, we showed that PKA is also required for infection of equid cells by African horse sickness virus, another member of the Orbivirus genus. Thus, despite their preference in specific host species, orbiviruses may utilize the same host signaling pathways during their replication.

Fig. 1.

Experimental design and validation of infection with BTV. A, A schematic representation of the phosphoproteomic experimental design is shown. The use of medium and heavy media for the 12h and 18h samples respectively was observed for experiments 1 and 3, and in experiment 2 the media was switched to control for potential impacts on cell growth or infection. B, Infection conditions within the various samples submitted for LC-MS/MS analysis were confirmed by Western blotting against viral antigens indicated on the left. Cellular GAPDH was used as a loading control. C, The number (percentage of the table) and D, percentage identity of phosphorylation sites within the total phosphoproteomic dataset obtained is shown.

Fig. 2.

Identification of sequence motifs within the phosphoproteome of BTV-infected cells. The 922 high confidence phosphosites (see supplemental Table S3) were analyzed with sequence logos prepared using the weblogo software. These could be further divided into phospho(S/T/Y) sites and the phospho(S) sites divided down further still into proline-directed, acidophilic, basophilic or other sites.

Fig. 3.

Inhibition of PKA reduces BTV replication whereas further stimulation of PKA enhances BTV replication. HeLa and sheep PT cells infected with BTV1 (MOI = 5) were treated 1 h.p.i with 40 μm H89 or 1 mm Dibutyryl-cAMP and harvested 12 h.p.i (A) and 18 h.p.i (B). Mock infected and DMSO treated cells were included as controls. Samples were analyzed by Western blot and densitometry analysis of phosphorylated-PKA substrates and NS2. Results are expressed as the relative protein levels as indicated on the left. Error bars represent the S.D. values of stimulations from three independent experiments. A star (*) denotes a significant difference from control (p < 0.05).

Fig. 4.

Decreases or increases in PKA-dependent phosphorylated substrate levels prior to infection do not affect BTV replication. HeLa and sheep PT cells were treated with 40 μm H89 or 1 mm Dibutyryl-cAMP for 1 h prior to infection with BTV1 (MOI = 5) and harvested 18 h.p.i. As controls, mock infected and DMSO treated cells were included. Samples were analyzed by Western blot densitometry analysis of phosphorylated-PKA substrates and NS2 are shown. Results are expressed as the percentage of protein levels as indicated. Error bars represent the S.D. values of stimulations from three independent experiments. A star (*) denotes a significant difference from control (p < 0.05).

Fig. 5.

Changes in the levels of PKA-dependent phosphorylated substrates post or prior to infection do not further affect AHSV replication. HeLa and Equine dermal (E. Derm) cells were treated 1 h.p.i (A) or for 1 h prior (B) to infection with AHSV1 (MOI = 5) with 40 μm H89 or 1 mm Dibutyryl-cAMP and harvested 18 h.p.i. As controls, mock infected and DMSO treated cells were included. Densitometry analysis of phosphorylated-PKA substrates and NS2 are shown and the results presented as protein percentage, as indicated. Error bars represent the S.D. values of stimulations from three independent experiments. A star (*) denotes a significant difference from control (p < 0.05).

Fig. 6.

BTV1 and AHSV1 increase AKT-dependent phosphorylated substrates. HeLa and sheep PT cells infected with BTV1 (MOI = 5) (A) or HeLa and Equine dermal (E. Derm) cells infected with AHSV1 (MOI = 5) (B) were treated 1 h.p.i with 4 μm Akt Inhibitor VIII (AKT VIII) and harvested 18 h.p.i. Densitometry analysis of phosphorylated-AKT substrates and NS2 are presented as relative protein level percentage. Error bars represent the S.D. values of stimulations from three independent experiments. A star (*) denotes a significant difference from control (p < 0.05).

Fig. 7.

AKT-dependent phosphorylated substrates decrease in BTV1 infected cells but remain elevated in AHSV1 infected cells between 18 and 36 h.p.i. HeLa cells infected with BTV1 (MOI = 5) (A) or HeLa cells infected with AHSV1 (MOI = 5) (B) were harvested 18, 24, and 36 h.p.i. Densitometry analysis of phosphorylated-AKT substrates are expressed as relative protein level as indicated. Error bars represent the S.D. values of stimulations from three independent experiments. A star (*) denotes a significant difference from control (p < 0.05).

Fig. 8.

A schematic representation of pathways summary showing identified kinases in BTV infected cells. A, Protein kinase A (PKA) activity may contribute to apoptosis inhibition and arrest of cell proliferation, extending the duration of viral replication. Validation experiments confirmed the importance of PKA activity for BTV replication. B, BTV infection induces autophagy. Death-associated protein kinase 1 (DAPK1) activity was identified and may contribute to autophagy induction during BTV infection. C, PAS Domain Containing Serine/Threonine Kinase (PASK) activity increased during BTV infection, possibly increasing translation efficiency during viral protein synthesis. D, Apoptosis is known to be controlled during BTV infection. Several kinases were identified as regulated by BTV infection which may contribute to the inhibition of apoptosis, these include ribosomal protein S6 kinase beta-1 (p70S6K), ribosomal protein S6 kinase A1 and 3 (RPS6KA1 and 3), p21-activated protein kinase 2 (PAK2), protein kinase C delta type (PKCD) activation and polo-like kinase 3 (PLK3). Inset shows various symbols: phosphorylation events, pathway activation, inhibition and down-regulation. A yellow shadow shows known kinase activity and a purple shadow depicts novel kinase activity.