Equine ANP32 proteins support influenza A virus RNA polymerase activity

Abstract

Host ANP32 family proteins are crucial for maintaining the activity of influenza RNA polymerase and play an important role in the cross-species transmission of influenza viruses. To date, the molecular properties of equine ANP32 (eqANP32) protein are poorly understood, particularly the mechanisms that affect equine influenza virus (EIV) RNA polymerase activity. Here, we found that there are six alternative splicing variants of equine ANP32A (eqANP32A) with different levels of expression. Further studies showed that these six splicing variants of eqANP32A supported the activity of EIV RNA polymerase to varying degrees, with the variant eqANP32A_X2 having the highest expression abundance and exhibiting the highest support of polymerase activity. Sequence analysis demonstrated that the differences in the N-Cap regions of the six splicing variants significantly affected their N-terminal conformation, but did not affect their ability to bind RNA polymerase. We also demonstrated that there is only one transcript of eqANP32B, and that this transcript showed only very low support to the EIV RNA polymerase. This functional defect in eqANP32B is caused by the sequence of the 110-259 amino acids at its C-terminus. Our results indicated that it is the eqANP32A_X2 protein that mainly determines the efficiency of the EIV replication in horses. In conclusion, our study parsed the molecular properties of eqANP32 family proteins and revealed the sequence features of eqANP32A and eqANP32B, suggesting for the first time that the N-cap region of ANP32A protein also plays an important role in supporting the activity of the influenza virus polymerase.

Keywords: Equine influenza virus (EIV); N-Cap domain; RNA polymerase activity; equine ANP32A; equine ANP32B.

Fig. 1

Expression of eqANP32 family genes and their distribution in horse tissues. A Schematic diagram illustrating the genomic analysis of equine ANP32A and ANP32B within the chromosomes. B–G Expression of mRNA from eqANP32 family genes in different equine tissue samples, including lung (B), liver (C), spleen (D), kidney (E), funicle (F), and placenta (G). The comparative Ct method was used to determine the relative mRNA expression of genes normalized by the GAPDH. H Cloning of eqANP32 family genes in equine lung tissue. (I) Stable expression of 1 ​μg of each FLAG-tagged ANP32A in wild-type 293T cells. At 48 ​h post-transfection, cells were lysed in radio immunoprecipitation assay (RIPA) lysis buffer, and proteins were separated with 4%–12% SDS-PAGE, followed by immunoblotting with anti-β-actin antibody and anti-FLAG antibody to detect the FLAG-tagged ANP32 proteins. hu, human; eq, equine. Statistical differences between samples are indicated, according to a one-way ANOVA, followed by a Dunnett's test (ns, not significant; ∗∗, P ​< ​0.01; ∗∗∗, P ​< ​0.001; ∗∗∗∗, P ​< ​0.0001). Error bars represent the SEM within one representative experiment. The results represent at least three independent experiments.

Fig. 2

Support of equine influenza viral replication by equine ANP32A or ANP32B, and species-dependent support of equine ANP32A or B for influenza A viral replication. Vectors carrying 20 ​ng of ANP32A or ANP32B genes, or empty vectors, were co-transfected into DKO cells (a 293T knockout cell line), together with a minigenome reporter, a Renilla expression control, and influenza virus polymerases from either equine influenza H3N8_XJ07 (A); human influenza H7N9_AH13 (B); canine influenza H3N2_GD11 (C); swine influenza H1N1_NC08 (D); avian influenza H9N2_ZJ12 (E). Luciferase activity was measured 20 ​h later, and data indicate the firefly luciferase gene activity normalized to the Renilla luciferase gene activity. Statistical differences between samples are indicated, according to a one-way ANOVA, followed by a Dunnett's test (ns, not significant; ∗, P ​< ​0.05; ∗∗, P ​< ​0.01; ∗∗∗, P ​< ​0.001; ∗∗∗∗, P ​< ​0.0001). Error bars represent the SEM within one representative experiment. The results represent at least three independent experiments. Western blotting to detect the ANP32A and viral proteins used specific antibodies: PA antibody and PB2 antibody (from our lab), PB1 antibody (NBP2-42877, NOVUS), Anti-Flag antibody (F1804, SIGMA) and Anti-β-actin antibody (F1804, SIGMA). ch, chicken; sw, swine; hu, human; eq, equine.

Fig. 3

Differences in the N-Cap region of eqANP32A lead to changes in the N-terminal conformation of eqANP32A proteins. A Amino acid sequences from differential splicing variants of equine ANP32A were aligned using the DNAMAN software. Gaps are marked with dashes. B The amino acid sequences of the N-Cap domain of eqANP32A and ANP32A proteins from other species were aligned in the Geneious Prime software. C–H Structure prediction of differential splicing of equine ANP32A (eqANP32A_X1, eqANP32A_X2, eqANP32A_X3, eqANP32A_X4, eqANP32A_X5 and eqANP32A_X6) using Phyre2. The analysis was performed using the Swiss-PBD Viewer software. ch, chicken; sw, swine; hu, human; eq, equine; ca, canine.

Fig. 4

The N-Cap domain does not affect the interaction of eqANP32A with the trimeric equine influenza virus polymerase complexes. A DKO cells (a 293T knockout cell line) were transfected with different ANP32A (0.6 ​μg) and polymerase plasmids (0.6 ​μg ​PA, 1 ​μg PB1, and 1 ​μg PB2) from equine influenza virus H3N8_XJ07. The cells were lysed at 36 ​h post-transfection. Co-IP was performed using anti-FLAG M2 magnetic beads, followed by Western blotting to detect the ANP32A and viral proteins using specific antibodies: PA antibody and PB2 antibody (from our lab), PB1 antibody (NBP2-42877, NOVUS), Anti-Flag antibody (F1804, SIGMA) and Anti-β-actin antibody (F1804, SIGMA). To compare the binding ability of eqANP32 with each of the polymerase protein, the intensity of each protein band was measured using the ImageJ software. The intensity bands ratio was the value of the intensity of each polymerase protein divided by that of eqANP32 (B and C). P values were determined using one-way ANOVA followed by a Dunnett's multiple comparisons test. ns, not significant; ∗, P ​< ​0.05; ∗∗, P ​< ​0.01. eq, equine.

Fig. 5

Differences in ANP32B support the activity of viral polymerases from influenza viruses infecting different species. DKO cells were co-transfected with expression plasmids carrying PB1 (40 ​ng), PB2 (40 ​ng), PA (20 ​ng) and NP (80 ​ng) from equine influenza H3N8_XJ07 (A); canine influenza H3N2_GD11 (B), together with 40 ​ng minigenome reporter and 1 ​ng Renilla luciferase expression plasmids (pRL-TK, as an internal control) in the presence of chANP32B, swANP32B, eqANP32B, huANP32B or caANP32B protein or empty vector. Cells were then lysed using passive lysis buffer, and luciferase activity was measured at 20 ​h post transfection. Statistical differences between samples are indicated, according to a one-way ANOVA, followed by a Dunnett's test (ns, not significant; ∗, P ​< ​0.05; ∗∗, P ​< ​0.01; ∗∗∗, P ​< ​0.001). Error bars represent the SEM within one representative experiment. The results represent at least three independent experiments. ch, chicken; sw, swine; hu, human; ca, canine.

Fig. 6

The key region responsible for the functional deficiency of eqANP32B is 111–259aa. (A) Amino acid sequences from swine, eqANP32B and huANP32B were aligned using the DNAMAN software. Gaps are marked with dashes. (B) Schematic diagram of chimeric clones between equine and human ANP32B, constructed according to the known domains. The bars indicate the origins of the genes by color as follows: pink, eqANP32B; and orange, huANP32B. (C) Stable expression of 1 ​μg of FLAG-tagged eq110-huANP32B and hu110-eqANP32B in wild-type 293T cells. At 48 ​h post-transfection, cells were lysed and analyzed by Western blotting. (D) Plasmids (20 ​ng) carrying ANP32B or mutants, or empty vectors were co-transfected with plasmids carrying polymerase from H3N8_XJ07. Luciferase activity was assayed at 20 ​h after transfection. (Data are firefly activity normalized to Renilla. Statistical differences between cells are labeled according to a one-way ANOVA followed by a Dunnett's test; ∗∗, P ​< 0.01; ∗∗∗P ​< ​0.001, ∗∗∗∗P ​< ​0.0001). sw, swine; hu, human; eq, equine.