Musashi-1 and miR-147 Precursor Interaction Mediates Synergistic Oncogenicity Induced by Co-Infection of Two Avian Retroviruses

Synergism between avian leukosis virus subgroup J (ALV-J) and reticuloendotheliosis virus (REV) has been reported frequently in co-infected chicken flocks. Although significant progress has been made in understanding the tumorigenesis mechanisms of ALV and REV, how these two simple oncogenic retroviruses induce synergistic oncogenicity remains unclear. In this study, we found that ALV-J and REV synergistically promoted mutual replication, suppressed cellular senescence, and activated epithelial-mesenchymal transition (EMT) in vitro. Mechanistically, structural proteins from ALV-J and REV synergistically activated the expression of Musashi-1(MSI1), which directly targeted pri-miR-147 through its RNA binding site. This inhibited the maturation of miR-147, which relieved the inhibition of NF-κB/KIAA1199/EGFR signaling, thereby suppressing cellular senescence and activating EMT. We revealed a synergistic oncogenicity mechanism induced by ALV-J and REV in vitro. The elucidation of the synergistic oncogenicity of these two simple retroviruses could help in understanding the mechanism of tumorigenesis in ALV-J and REV co-infection and help identify promising molecular targets and key obstacles for the joint control of ALV-J and REV and the development of clinical technologies.

ALV-J induces late-onset myelocytomas, hemangiomas, and various other tumors in chickens, and REV induces chronic lymphomas in chickens, ducks, geese, pheasants, quails, and turkeys [24]. The two avian retroviruses continue to be of great interest in understanding the molecular mechanisms of tumorigenesis. In ALV-J and REV, the induction of neoplasms occurs in a minority of cases and only after several months of infection, presenting great difficulties in exploring the mechanisms of synergistic tumorigenesis. Since the suppression of cellular senescence and the activation of EMT are prerequisites for

Hematoxylin and Eosin Staining
The cancerous tissues and corresponding non-cancerous tissues of chickens were formalin-fixed, paraffin-embedded, sectioned, and stained for histopathological observation.

Illumina Small-RNA Deep Sequencing
Total RNA from infected CEF cell samples, either mock-infected or infected with ALV-J or REV alone, or co-infected with both ALV-J and REV for 72 hpi, was separated on 15% agarose gels to extract small RNA (18-30 nt). Illumina small-RNA deep sequencing was performed as previously described [19].

Quantitative Real-Time Polymerase Chain Reaction
The specific primer sequences for pri-miR-147, KIAA1199, NF-κB p65, NF-κB p50, MSI1, EGFR, and GAPDH used in this study are listed in Table S1. Total RNA from CEF cells that had been either mock-infected, mono-infected with ALV-J or REV, or co-infected with ALV-J and REV was isolated using the Tiangen RNeasy mini kit (TIANGEN) according to the manufacturer's instructions, with optional on-column DNase digestion. RNA integrity and concentrations were assessed by means of agarose gel electrophoresis and spectrophotometry, respectively. RNA (1 µg per triplicate reaction) was reverse-transcribed to cDNA using the Taqman Gold Reverse Transcription kit (Applied Biosystems). Real-time RT-PCR (qRT-PCR) was performed using SYBR ® Premix Ex Taq, and specific primers (Table S1). All values were normalized to endogenous GAPDH levels to control for variation. For qRT-PCR analysis of miR-147, we used an miRcute miRNA first-stand cDNA synthesis kit and an miRcute miRNA qPCR detection kit (SYBR Green) (TIANGEN). The reverse primer provided in the miRcute miRNA qPCR detection kit was complementary to the poly (T) adapter. Data were collected on an ABI PRISM 7500 and analyzed using Sequence Detector v1.1 software (Applied Biosystems, USA). All values were normalized to endogenous U6 to control for variations. The primers specific for U6 are listed in Table S1. Assays were performed in triplicate, after which the average threshold cycle (CT) values were used to determine relative concentration differences based on the ∆∆CT method of relative quantization described in the manufacturer's protocol.

Determination of NF-κB p65 Nuclear Translocation
The translocation of NF-κB p65 from the cytoplasm to the nucleus was examined via immunofluorescence. CEFs were washed with PBS and fixed with 4% paraformaldehyde for 20 min. After fixation, the cells were permeabilized with 0.25% Triton X-100 in PBS for 10 min and blocked with 10% BSA in PBS for 1 h. Anti-NF-κB p65 antibody (1:100, Bioss) was incubated overnight at 4 • C, followed the next day by a one-hour incubation at room temperature with anti-rabbit IgG antibody labeled with FITC (1:1000, Engibody). Finally, the cells were washed with PBS, incubated with DAPI for 5 min, and then observed under a fluorescence microscope after washing with PBS.

Senescence-Associated β-Gal Staining
A senescence β-galactosidase staining kit was purchased from Beyotime (Shanghai, China) and was used to evaluate cellular senescence according to the manufacturer's instructions.

ELISA for NF-κB p65 and P-IκBα Assays
Chicken NF-κB p65 and chicken NF-κB P-IκBα ELISA kits were purchased from Senbeijia (Nanjing, China) and were used to determine the expression levels of NF-κB p65 according to the manufacturer's instructions.

RNA ChIP Assay
The RNA ChIP kit was purchased from Active Motif (Shanghai, China) and was used to assay RNA-protein interactions according to the manufacturer's instructions.

Statistical Analysis
Data are presented as the mean ± standard deviation(s). The t-test and one-way ANOVA tests were performed using SPSS v. 13.0 statistical software (SPSS, Chicago, IL, USA). Statistical significance was set at p ≤ 0.05.

ALV-J and REV Synergistically Suppress Cellular Senescence and Activate Epithelial-Mesenchymal Transition (EMT) In Vitro
The suppression of cellular senescence or activation of EMT is a prerequisite for tumorigenesis and metastasis. To understand whether ALV-J and REV synergistically affected oncogenicity, cellular senescence and EMT were measured in ALV-J and REV co-infected cells. The RNA levels and protein levels of both ALV-J and REV were increased significantly in co-infected cells compared to those in mono-infected cells ( Figure 1A,B,E-G), confirming that there was synergistic replication between ALV-J and REV. Senescence-associated (SA)β-Gal staining revealed that ALV-J and REV synergistically inhibited cellular senescence ( Figure 1C,D). The expression levels of EMT-associated proteins, assessed via Western blotting (WB) analysis, suggested that ALV-J and REV synergistically activated the EMT process ( Figure 1E-G). These findings confirm that ALV-J and REV synergistically suppress cellular senescence and activate the EMT process in vitro.

Identification of Key Host Molecules Responsible for Synergistic Oncogenicity Induced by ALV-J and REV
Chick embryo fibroblasts (CEFs) co-infected with ALV-J and REV, mono-infected with ALV-J or REV, and mocks were analyzed using tandem mass tag (TMT)-based proteomics combined with miRNA whole-genome sequencing analysis. Among the 33 differentially expressed proteins and 17 differentially expressed miRNAs (Figure 2A,B), only miR-147 (or pri-miR-147) exhibited potential interactions with Musashi-1 (MSI1), NF-κB p50, and KIAA1199, which are associated with the cancer signaling pathway. The decreased miR-147 expression was verified using qPCR ( Figure 2C), and the increased expression of MSI1, KIAA1199, and NF-κB p50 were verified using WB in co-infected CEFs ( Figure 2D). RNAprotein interaction analysis showed interactions between miR-147 and MSI1, KIAA1199, and NF-κB p50 in ALV-J and REV co-infected cells. These results suggest that the interaction between miR-147 and activated MSI1, KIAA1199, and NF-κB p50 may be important in the synergistic oncogenicity induced by ALV-J and REV.

Identification of Key Host Molecules Responsible for Synergistic Oncogenicity Induced by ALV-J and REV
Chick embryo fibroblasts (CEFs) co-infected with ALV-J and REV, mono-infected with ALV-J or REV, and mocks were analyzed using tandem mass tag (TMT)-based proteomics combined with miRNA whole-genome sequencing analysis. Among the 33 differentially expressed proteins and 17 differentially expressed miRNAs (Figure 2A,B), only miR-147 (or pri-miR-147) exhibited potential interactions with Musashi-1 (MSI1), NF-κB p50, and KIAA1199, which are associated with the cancer signaling pathway. The decreased miR-147 expression was verified using qPCR ( Figure 2C), and the increased expression of MSI1, KIAA1199, and NF-κB p50 were verified using WB in co-infected CEFs ( Figure 2D). RNA-protein interaction analysis showed interactions between miR-147 and MSI1, KIAA1199, and NF-κB p50 in ALV-J and REV co-infected cells. These results suggest that the interaction between miR-147 and activated MSI1, KIAA1199, and NF-κB p50 may be important in the synergistic oncogenicity induced by ALV-J and REV.

MSI1 Directly Targeted pri-miR-147 to Inhibit miR-147 Maturation
We intended to detect the expression of the miR-147 precursor (pri-miR-147) to determine whether mature miR-147 was inhibited before or after pri-miR-147 transcription. In contrast to mature miR-147, ALV-J and REV synergistically activated the expression of pri-miR-147 rather than inhibiting it. This showed that miR-147 was inhibited after pri-mir147 transcription ( Figure 4A). The relationship between MSI1 and pri-miR-147/miR-147 was further analyzed. MiR-147 levels were detected when MSI1 was over-or under-expressed in CEFs. When MSI1 was overexpressed, the levels of mature miR-147 were suppressed more than 3.77-fold ( Figure 4B). Upon MSI1 knockdown, the levels of mature miR-147 were elevated more than 5.58-fold ( Figure 4C). Furthermore, RNA chromatin immunoprecipitation (ChIP) revealed over 15.6-fold enrichment of pri-miR-147, which is associated with MSI1 ( Figure 4D,E), indicating that MSI1 directly targeted pri-miR-147 RNA. To identify the domain in MSI1 that binds pri-miR-147, we constructed four MSI1 mutants based on its RNA-binding sites [41,42], and transfected CEFs to detect mature miR-147 levels. The four MSI1 mutants are shown in Figure 4F. WB analysis confirmed that all MSI mutants were successfully transfected into the CEF cell line DF-1 ( Figure 4G). The miR-147 expression level showed that only MSI1 mut1 relieved the inhibition of miR-147 maturation ( Figure 4H), implying that the RNA-binding site (amino acid sequences 33 and 35 to 39) was the key domain for inhibiting miR-147 maturation. These data suggest that MSI1 directly targets pri-miR-147 through its RNA binding site (amino acid sequences 33 and 35 to 39), leading to the inhibition of miR-147 maturation.

Ectopic Expression of miR-147, MSI1, KIAA1199, and NF-κB p50 Is Associated with Oncogenicity
To verify the association of miR-147, MSI1, KIAA1199, and NF-κB p50 with oncogenicity, we measured cellular senescence and EMT through the construction and transfection of miR-147 mimics, miR-147 inhibitors, FLAG-MSI1, MSI1 Cas9/gRNA, FLAG-KIAA1199, KIAA1199 Cas9/gRNA, KIAA1199 shRNAs, FLAG-NF-κB p50, NF-κB p50 Cas9/gRNA, and NF-κB p50 shRNAs in CEF cells. SA-β-Gal staining and WB assays re-   We intended to detect the expression of the miR-147 precursor (pri-miR-147) to determine whether mature miR-147 was inhibited before or after pri-miR-147 transcription. In contrast to mature miR-147, ALV-J and REV synergistically activated the expression of pri-miR-147 rather than inhibiting it. This showed that miR-147 was inhibited after pri- that all MSI mutants were successfully transfected into the CEF cell line DF-1 ( Figure 4G). The miR-147 expression level showed that only MSI1 mut1 relieved the inhibition of miR-147 maturation ( Figure 4H), implying that the RNA-binding site (amino acid sequences 33 and 35 to 39) was the key domain for inhibiting miR-147 maturation. These data suggest that MSI1 directly targets pri-miR-147 through its RNA binding site (amino acid sequences 33 and 35 to 39), leading to the inhibition of miR-147 maturation.  Data are presented as mean ± SEM for n = 3, with each experiment being performed in triplicate. ** p ≤ 0.01 determined using Student s t-test versus the Neg. Ctrl. group. * p ≤ 0.05 determined using Student s t-test versus the Neg. Ctrl. group. n.s., not significant.

miR-147 Targets NF-κB p50 and KIAA1199
To confirm that miR-147 directly targeted KIAA1199 and NF-κB p50, a dual-luciferase assay was performed in CEFs. The KIAA1199 3 untranslated region (UTR) luciferase reporter assay revealed that miR-147 significantly inhibited the activity of the KIAA1199 3 UTR reporter and that of the NF-κB p50 3 UTR reporter, but not that of control reporters ( Figure 5A,B). miR-147 inhibited the activities of the KIAA1199 3 UTR reporter and the NF-κB p50 3 UTR reporter and suppressed the endogenous expression levels of KIAA1199 and NF-κB p50 in CEF cells in a dose-dependent manner ( Figure 5C-E). In contrast, the miR-147 inhibitor upregulated the expression of endogenous KIAA1199 and NF-κB p50 in a dosedependent manner ( Figure 5F). Bioinformatics analysis identified one putative miR-147 binding site at the KIAA1199 3 UTR and the NF-κB p50 3 UTR, respectively ( Figure 4G,H). Mutations in the putative miR-147 binding site eliminated the inhibitory effect of miR-147 on the reporter activities of the KIAA1199 3 UTR and NF-κB p50 3 UTR (Figure 5I,J). These findings suggested that miR-147 directly targeted KIAA1199 and NF-κB p50.  Ctrl.) or miR-147 into 293T cells, whereas mutant KIAA1199 3 UTR construct (mut KIAA1199) was co-transfected with Neg. Ctrl. or miR-147. (J) NF-κB p50 3 UTR wild type (WT NF-κB p50) was co-transfected with a negative control (Neg. Ctrl.) or miR-147 into 293T cells, whereas mutant NF-κB p50 3 UTR construct (mut NF-κB p50) was co-transfected with Neg. Ctrl. or miR-147. The above luciferase assays were performed 48 h later; data are presented as the mean ± SEM for n = 3, with each experiment being performed in triplicate. ** p ≤ 0.01 determined using Student s t-test versus the Neg. Ctrl. group. n.s., not significant.

ALV-J and REV Synergistically Activated the NF-κB/KIAA1199/EGFR Signaling Pathway
Previous studies have shown that KIAA1199, an oncogene that is transcriptionally induced by NF-κB proteins, promotes EGFR stability and contributes to the activation of NF-κB/EGFR signaling pathway crosstalk in breast cancer [43,44]. To determine whether EGFR is involved in NF-κB/KIAA1199 signaling that is synergistically activated by ALV-J and REV, expression levels of NF-κB p65, phosphorylated IκBα, and EGFR were detected using qPCR, ELISA, and WB. Compared to mono-infection, NF-κB p65 RNA levels in coinfected cells were elevated 1.78-and 1.91-fold, respectively ( Figure 6A). ELISA was used to confirm these results ( Figure 6B). Compared with single infection, translocation of NF-κB to the nucleus was significantly observed in CEF-co-infected ALV-J and REV ( Figure S1). The increase in phosphorylated IκBα levels indicated that ALV-J and REV synergistically activated the NF-κB signaling pathway ( Figure 6C). Furthermore, we found that ALV-J and REV synergistically enhanced EGFR expression levels ( Figure 6D,E). To determine the correlation between NF-κB, KIAA1199, and EGFR in ALV-J and REV co-infected cells, RNA interference was carried out via the construction and transfection of NF-κB p65 or KIAA1199 shRNAs into DF-1 cells. Upon NF-κB p65 or KIAA1199 knockdown, the expression levels of KIAA1199 and EGFR ( Figure 6F-H) or EGFR and NF-κB p65 were suppressed ( Figure 6I-K). Taken together, these findings suggested that ALV-J and REV synergistically activate the NF-κB/KIAA1199/EGFR signaling pathway.

Structural Proteins, Especially Gags from ALV-J and REV, Synergistically Activate MSI1
The genome of simple retroviruses is composed of gag, pol, and env, which encode core proteins, proteases, and envelope proteins, respectively [45]. To determine whether the synergistic activation of MSI1 by ALV-J and REV is caused by its structural proteins, we detected the expression of MSI1 in cells infected with REV and transfected with ALV-J structural proteins (gag, pol, and env), or infected with ALV-J and transfected with REV structural proteins (gag, pol, and env). All viral structural proteins showed the synergistic promotion of MSI1 expression; however, ALV-J gag and REV gag showed the most significant promotion effect on MSI1 expression ( Figure 7A,B). Furthermore, cotransfection of the two gags from ALV-J and REV demonstrated synergism in regard to MSI1 activation ( Figure 7C). These findings confirmed that all structural proteins of ALV-J and REV were involved in the synergistic activation of MSI1. EGFR antibodies. Data are presented as mean ± SEM for n = 3, with each experiment being performed in triplicate. **p ≤ 0.01 determined using Student′s t-test versus the Neg. Ctrl. group. *p ≤ 0.05 determined using Student′s t-test versus the Neg. Ctrl. group.

Structural Proteins, Especially Gags from ALV-J and REV, Synergistically Activate MSI1
The genome of simple retroviruses is composed of gag, pol, and env, which encode core proteins, proteases, and envelope proteins, respectively [45]. To determine whether the synergistic activation of MSI1 by ALV-J and REV is caused by its structural proteins, we detected the expression of MSI1 in cells infected with REV and transfected with ALV-J structural proteins (gag, pol, and env), or infected with ALV-J and transfected with REV structural proteins (gag, pol, and env). All viral structural proteins showed the synergistic promotion of MSI1 expression; however, ALV-J gag and REV gag showed the most significant promotion effect on MSI1 expression ( Figure 7A,B). Furthermore, co-transfection of the two gags from ALV-J and REV demonstrated synergism in regard to MSI1 activation ( Figure 7C). These findings confirmed that all structural proteins of ALV-J and REV were involved in the synergistic activation of MSI1.

MSI1-miR-147 Regulated NF-κB/KIAA1199/EGFR Pathways Present in Tumors Induced by ALV-J and REV
To verify the synergistic oncogenicity occurring in ALV-J and REV co-infected tumor-bearing chickens, we established a tumor model induced by ALV-J and REV and measured the key molecules and the NF-κB/KIAA1199/EGFR pathway in three type tumors. All chickens were euthanized at 18 weeks of age; 4 of 10 (40%) ALV-J infected chickens, 0 of 13 (0%) REV-infected chickens, and 7/8 (87.5%) co-infected chickens bore tumors. Histopathological examination showed that the tumors induced by ALV-J were myelocytomas; however, ALV-J and REV induced myelocytomas, lymphomas, and endocardial fibromas ( Figure 8A). These findings suggested that ALV-J and REV synergistically promoted tumorigenesis in chickens. To validate the results of the in vitro experiments, we detected the RNA levels of miR-147, MSI1, KIAA1199, NF-κB p50, and EGFR in the cancerous tissues of chickens in different infection groups, including the bone marrow, liver, and heart. Compared to the mono-infection group, the RNA levels of miR-147 were significantly downregulated, whereas MSI1, KIAA1199, NF-κB p50, and EGFR were significantly upregulated in the bone marrow, livers, and hearts of chickens ( Figure 8B-F). Next, we determined the RNA expression levels of miR-147 in five cases of myelocytomas, lymphomas, and endocardial fibromas. Compared to non-cancerous tissues in co-infected

MSI1-miR-147 Regulated NF-κB/KIAA1199/EGFR Pathways Present in Tumors Induced by ALV-J and REV
To verify the synergistic oncogenicity occurring in ALV-J and REV co-infected tumorbearing chickens, we established a tumor model induced by ALV-J and REV and measured the key molecules and the NF-κB/KIAA1199/EGFR pathway in three type tumors. All chickens were euthanized at 18 weeks of age; 4 of 10 (40%) ALV-J infected chickens, 0 of 13 (0%) REV-infected chickens, and 7/8 (87.5%) co-infected chickens bore tumors. Histopathological examination showed that the tumors induced by ALV-J were myelocytomas; however, ALV-J and REV induced myelocytomas, lymphomas, and endocardial fibromas ( Figure 8A). These findings suggested that ALV-J and REV synergistically promoted tumorigenesis in chickens. To validate the results of the in vitro experiments, we detected the RNA levels of miR-147, MSI1, KIAA1199, NF-κB p50, and EGFR in the cancerous tissues of chickens in different infection groups, including the bone marrow, liver, and heart. Compared to the mono-infection group, the RNA levels of miR-147 were significantly downregulated, whereas MSI1, KIAA1199, NF-κB p50, and EGFR were significantly upregulated in the bone marrow, livers, and hearts of chickens ( Figure 8B-F). Next, we determined the RNA expression levels of miR-147 in five cases of myelocytomas, lymphomas, and endocardial fibromas. Compared to non-cancerous tissues in co-infected chickens, miR-147 levels in myelocytomas, fibromas, and lymphomas in four, five, and five cases, respectively, were suppressed ( Figure 9A). Compared to non-cancerous tissues, ELISA revealed that four, five, and four cases of phosphorylated IκBα levels were elevated in myelocytomas, fibromas, and lymphomas, respectively ( Figure 9B). Simultaneously, elevated expression levels of EGFR, KIAA1199, NF-κB p50, and MSI1 were also confirmed in myelocytomas, fibromas, and lymphomas using WB ( Figure 9C-E). Compared to the corresponding non-cancerous tissues, MSI1 and NF-κB/KIAA1199/EGFR pathway crosstalk was upregulated in 15 of 15 and 13 of 15 tumors, respectively, whereas miR-147 was downregulated in 14 of 15 tumors ( Figure 9F). These data suggest that a synergistic tumorigenesis mechanism occurred in chickens co-infected with ALV-J and REV. chickens, miR-147 levels in myelocytomas, fibromas, and lymphomas in four, five, and five cases, respectively, were suppressed ( Figure 9A). Compared to non-cancerous tissues, ELISA revealed that four, five, and four cases of phosphorylated IκBα levels were elevated in myelocytomas, fibromas, and lymphomas, respectively ( Figure 9B). Simultaneously, elevated expression levels of EGFR, KIAA1199, NF-κB p50, and MSI1 were also confirmed in myelocytomas, fibromas, and lymphomas using WB ( Figure 9C-E). Compared to the corresponding non-cancerous tissues, MSI1 and NF-κB/KIAA1199/EGFR pathway crosstalk was upregulated in 15 of 15 and 13 of 15 tumors, respectively, whereas miR-147 was downregulated in 14 of 15 tumors ( Figure 9F). These data suggest that a synergistic tumorigenesis mechanism occurred in chickens co-infected with ALV-J and REV.  1 Figure 9. miR-147 was downregulated in ALV-J and REV coinfection-induced tumors and was negatively correlated with MSI1 and NF-κB/KIAA1199/EGFR pathway crosstalk. (A) Expression levels of miR-147 in 14 of 15 tumors were lower than those of corresponding non-cancerous tissues, as detected using qPCR. Data are presented as mean ± SEM for n = 3, with each experiment being performed in triplicate. (B) Phosphorylated IκBα expression levels in 13 of 15 tumors were elevated when compared to corresponding non-cancerous tissues, as detected using an NF-κB P-IκBα ELISA kit. Data are presented as mean ± SEM for n = 3, with each experiment being performed in triplicate. (C) Expression levels of EGFR, KIAA1199, NF-κB p50, and MSI1 in myelocytomas were detected using WB with anti-EGFR, anti-KIAA1199, anti-NF-κB p50, and anti-MSI antibodies, respectively. (D) Expression levels of EGFR, KIAA1199, NF-κB p50, and MSI1 in fibromas were detected using WB with anti-EGFR, anti-KIAA1199, anti-NF-κB p50, and anti-MSI antibodies, respectively. (E) Expression levels of EGFR, KIAA1199, NF-κB p50, and MSI1 in lymphomas were detected using WB with anti-EGFR, anti-KIAA1199, anti-NF-κB p50, and anti-MSI antibodies, respectively. (F) Comparisons of MIS1 elevation, miR-147 elevation, and NF-κB/KIAA1199/EGFR pathway crosstalk elevation in tumors. Central overlapping region represents all cases of MIS1, miR-147, and NF-κB/KIAA1199/EGFR pathway crosstalk elevations. ** p ≤ 0.01 determined using Student s t-test versus the Neg. Ctrl. group. * p ≤ 0.05 determined using Student s t-test versus the Neg. Ctrl. group.

Discussion
Synergistic interactions between two retroviruses in co-infected hosts have been well documented [20][21][22][23]. Co-infection with two or more oncogenic retroviruses is known to ac-celerate cancer development [32]. Recent studies have shown that co-infection of ALV-J and REV causes higher mortality, more serious growth retardation, and immunosuppression, facilitating viral replication and changing the miRNA expression profile [17][18][19]. However, the question of whether synergism promotes oncogenicity and the underlying synergistic mechanism remains unclear. In this study, we found that ALV-J and REV synergistically suppressed cellular senescence and activated EMT in vitro, indicating that these two viruses have developed strategies to synergistically promote oncogenic potential in vitro.
The suppression of cellular senescence and the activation of EMT are prerequisites for neoplasm and metastasis, which have been commonly considered as essential indicators of cellular oncogenic potential in vitro [25][26][27][28][29][30]. To identify the key molecules responsible for the synergistic oncogenicity induced by ALV-J and REV in host cells, TMT-based proteomics, combined with miRNA whole-genome sequencing, was used to screen and identify the key molecules in the co-infected/mono-infected/mock cells. Interestingly, an miRNA molecule, miR-147, known as a tumor suppressor [46][47][48][49] showed ectopic expression, which increased in mono-infected cells and decreased in co-infected cells. However, its precursor, pri-miR-147, showed a synergistic increase in ALV-J and REV-co-infected cells. These data suggested that a certain molecule blocks miR-147 maturation, releasing its target signals for tumorigenesis. Thus, target analysis indicated that miR-147 or pri-miR-147 exhibited potential interactions with MSI1, NF-κB p50, and KIAA1199, which are associated with the cancer signaling pathway. Next, we demonstrated that miR-147, MSI1, NF-κB p50, and KIAA1199 play a critical role in cellular senescence suppression and EMT activation, indicating that these molecules are involved in the synergistic oncogenicity induced by ALV-J and REV.
MSI1, an RNA-binding protein, has been found to regulate multiple critical biological processes that are relevant to cancer initiation and progression [50]. KIAA1199, a novel proto-oncogene, has been associated with tumor progression and metastasis in numerous cancers [51]. NF-κB activation is key to the early development of some cancers [52]. Based on the functions of these molecules, we speculated that MSI1 blocks the maturation of miR-147, which relieves the inhibition of NF-κB p50 and KIAA1199. The experimental results support our speculation. MSI1 directly targeted pri-miR-147 through its RNAbinding site, inhibiting miR-147 maturation. Because miR-147 directly targets NF-κB p50 and KIAA1199, downregulation of miR-147 promoted the upregulation of NF-κB p50 and KIAA1199. Recent studies have demonstrated that NF-κB and EGFR are partners in cancer, and NF-κB-induced KIAA1199 promotes EGFR stability, contributing to the activation of the NF-κB/EGFR signaling pathway [43,44]. Because EGFR was absent in the TMT-based proteomics results, we wanted to know whether EGFR is involved in the NF-κB/KIAA1199 signaling that is synergistically activated by ALV-J and REV. The results showed that EGFR participated in the NF-κB/KIAA1199 pathway, namely, NF-κB/KIAA1199/EGFR, which was synergistically activated by ALV-J and REV.
We observed that ALV-J and REV synergistically activate MSI1, which binds pri-miR-147, blocking miR-147 maturation and thereby relieving the inhibition of the NF-κB/KIAA1199/EGFR signaling pathway. We then investigated how ALV-J and REV synergistically activate MSI1. The results showed that all structural proteins, especially gags from ALV-J and REV, synergistically activated the expression of MSI1. Generally, gag proteins from complex retroviruses or acutely transforming retroviruses (carrying gag-onc fusion genes) are involved in oncogenesis [53,54]. Here, we observed for the first time that gags from two simple retroviruses synergistically promoted oncogenicity. Finally, the key molecules and signaling pathways involved in synergistic tumorigenesis were verified in tumor-bearing chickens infected with ALV-J and REV.

Conclusions
In conclusion, the current study revealed a synergistic oncogenicity mechanism induced by two simple retroviruses, ALV-J and REV. ALV-J and REV synergistically suppressed cellular senescence and activated EMT in vitro, and synergistically induced tumori-genesis and tumor spectrum extension in vivo. Mechanistically, as shown in Figure 10, after co-infection with ALV-J and REV, the released or expressed structural proteins from ALV-J and REV in the cytoplasm synergistically activated MSI1 expression, which directly targeted pri-miR-147 through its RNA binding site, causing the inhibition of miR-147 maturation. This relieved the inhibition of the NF-κB/KIAA1199/EGFR signaling pathway, thereby suppressing cellular senescence and activating EMT. The synergistic oncogenicity mechanism of ALV-J and REV sheds light on the identification of promising molecular targets and key barriers to the joint control of ALV-J and REV and the development of clinical technologies. In conclusion, the current study revealed a synergistic oncogenicity mechanism induced by two simple retroviruses, ALV-J and REV. ALV-J and REV synergistically suppressed cellular senescence and activated EMT in vitro, and synergistically induced tumorigenesis and tumor spectrum extension in vivo. Mechanistically, as shown in Figure  10, after co-infection with ALV-J and REV, the released or expressed structural proteins from ALV-J and REV in the cytoplasm synergistically activated MSI1 expression, which directly targeted pri-miR-147 through its RNA binding site, causing the inhibition of miR-147 maturation. This relieved the inhibition of the NF-κB/KIAA1199/EGFR signaling pathway, thereby suppressing cellular senescence and activating EMT. The synergistic oncogenicity mechanism of ALV-J and REV sheds light on the identification of promising molecular targets and key barriers to the joint control of ALV-J and REV and the development of clinical technologies. Figure 10. Schematic of the molecular mechanisms of synergistic oncogenicity induced by ALV-J and REV. After co-infection of ALV-J and REV, released or expressed structural proteins from ALV-J and REV in the cytoplasm synergistically elevate the expression levels of MSI1, which directly targets pri-miR-147, causing the inhibition of miR-147 maturation, which relieves the inhibition of the NF-κB/KIAA1199/EGFR signaling pathway and thereby suppresses cellular senescence and activates the EMT process, promoting oncogenicity.

Supplementary Materials:
The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Table S1: Primer sequences used in qPCR analysis. Figure S1: Compared with single infection, translocation of NF-κB to the nucleus was observed at significant levels in CEFs co-infected with ALV-J and REV at 72 hpi.
Author Contributions: Z.C. conceived and designed the research and wrote the manuscript. D.Z. performed the experiments and wrote the manuscript. L.D., M.X., and X.L. contributed to data extraction. X.Z., J.X., X.D., J.Z., and X.C. contributed reagents and materials. All authors discussed and analyzed the results. All authors have read and agreed to the published version of the manuscript. Figure 10. Schematic of the molecular mechanisms of synergistic oncogenicity induced by ALV-J and REV. After co-infection of ALV-J and REV, released or expressed structural proteins from ALV-J and REV in the cytoplasm synergistically elevate the expression levels of MSI1, which directly targets pri-miR-147, causing the inhibition of miR-147 maturation, which relieves the inhibition of the NF-κB/KIAA1199/EGFR signaling pathway and thereby suppresses cellular senescence and activates the EMT process, promoting oncogenicity.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/cells11203312/s1, Table S1: Primer sequences used in qPCR analysis. Figure S1: Compared with single infection, translocation of NF-κB to the nucleus was observed at significant levels in CEFs co-infected with ALV-J and REV at 72 hpi.
Author Contributions: Z.C. conceived and designed the research and wrote the manuscript. D.Z. performed the experiments and wrote the manuscript. L.D., M.X. and X.L. contributed to data extraction. X.Z., J.X., X.D., J.Z. and X.C. contributed reagents and materials. All authors discussed and analyzed the results. All authors have read and agreed to the published version of the manuscript.