(ADAR1) Promotes EIAV Replication and Infectivity

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Virology 476 (2015) 364–371

Contents lists available at ScienceDirect

Virology

journal homepage: www.elsevier.com/locate/yviro

Double-stranded RNA-speciﬁc adenosine deaminase 1 (ADAR1) promotes EIAV replication and infectivity

Yan-Dong Tang a,b,1, Lei Na a,1, Li-Hua Fu a, Fei Yang a, Chun-Hui Zhu a, Li Tang a, Qiang Li c, Jia-Yi Wang a, Zhan Li a, Xue-Feng Wang a, Cheng-Yao Li b,n, Xiaojun Wang a,nn, Jian-Hua Zhou a,d,nnn

a State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute of Chinese Academy of Agricultural Sciences, Harbin 150001, China b Biotechnology Institute of Southern Medical University, Guangzhou 510515, China c Harbin Weike Biotechnology Development Company, Harbin 150069, China d Harbin Pharmaceutical Group Biovaccine Company, Harbin 150069, China

article info abstract

Article history: Adenosine deaminases that act on RNA (ADARs) have been reported to be functional on various viruses. Received 2 October 2014 ADAR1 may exhibit antiviral or proviral activity depending on the type of virus. Human immunodeﬁ- Returned to author for revisions ciency virus (HIV)-1 is the most well-studied lentivirus with respect to its interaction with ADAR1, and 23 October 2014 variable results have been reported. In this study, we demonstrated that equine ADAR1 (eADAR1) was a Accepted 23 December 2014 positive regulator of equine infectious anemia virus (EIAV), another lentivirus of the Retroviridae family. Available online 9 January 2015 First, eADAR1 signiﬁcantly promoted EIAV replication, and the enhancement of viral protein expression Keywords: was associated with the long terminal repeat (LTR) and Rev response element (RRE) regions. Second, the EIAV RNA binding domain 1 of eADAR1 was essential only for enhancing LTR-mediated gene expression. Third, ADAR1 in contrast with APOBEC proteins, which have been shown to reduce lentiviral infectivity, eADAR1 Infectivity increased the EIAV infectivity. This study indicated that eADAR1 was proviral rather than antiviral Replication for EIAV. & 2014 Elsevier Inc. All rights reserved.

Introduction human T-cell leukemia virus type 2 and simian T-cell leukemia virus type 3 (Ko et al., 2012), during viral infection. Notably, ADAR1 can Adenosine deaminases that act on RNA (ADARs) are RNA- either promote or inhibit viral replication. For certain viruses, such editing enzymes that recognize double stranded RNA (dsRNA) and as human immunodeficiency virus (HIV)-1, vesicular stomatitis virus mediate the conversion of adenosine (A) to inosine (I) (Bass, 2002; (VSV) and hepatitis D virus (HDV) (Casey, 2006; Clerzius et al., 2009; Nishikura, 2010; Samuel, 2011). The base-pairing properties of inosine Doria et al., 2011; Nie et al., 2007; Phuphuakrat et al., 2008), ADAR1 are different from those of adenosine, and thus, A-to-I editing alters is proviral. However, ADAR1 inhibits the replication of other viruses, RNA structure, coding potential and splicing patterns (Samuel, 2011). such as hepatitis C virus (Taylor et al., 2005)andthemeaslesvirus Recently, ADARs have been shown to edit the genomes of many (Ward et al., 2011). For HIV-1 and VSV, ADAR1 enhances viral replica- viruses, including the measles virus (Cattaneo et al., 1988), human tion by inhibition of dsRNA activated protein kinase (PKR), which parainfluenza virus (Murphy et al., 1991), respiratory syncytial virus suppresses viruses replication by shutting off the expression of viral (Martinez et al., 1997), influenza virus (Gutierrez et al., 2013; Suspene proteins (Clerzius et al., 2009; Gelinas et al., 2011; Li et al., 2010; Nie et al., 2011), lymphocytic choriomeningitis virus (Zahn et al., 2007), et al., 2007). HDV replication requires ADAR1 editing activity for producing its large antigen by catalyzing stop codon UAG into tryptophan codon UGG in HDV mRNA (Casey, 2006). However, in

n Correspondence to: Department of Transfusion Medicine, Southern Medical general, virus replication inhibition was suspected due to ADAR1 University, Guangzhou 510515, China. Tel./fax: þ86 2061648466. massive editing activity (Ko et al., 2012; Suspene et al., 2011; Taylor nn Corresponding author. Tel.: þ86 18946066285; fax: þ86 45151997166. et al., 2005). nnn Corresponding author at: Harbin Pharmaceutical Group Biovaccine Company, Similar to HIV-1, equine infectious anemia virus (EIAV) is a Harbin 150069, China. Tel.: þ86 13796085512. member of the Lentivirus genus of the Retroviridae family. EIAV E-mail addresses: [email protected] (C.-Y. Li), [email protected] (X. Wang), [email protected] (J.-H. Zhou). causes a persistent infection characterized by recurring febrile 1 These authors contributed equally to this article. episodes associated with viremia, fever, thrombocytopenia and

http://dx.doi.org/10.1016/j.virol.2014.12.038 0042-6822/& 2014 Elsevier Inc. All rights reserved. Y.-D. Tang et al. / Virology 476 (2015) 364–371 365 wasting symptoms in equids (Leroux et al., 2004). However, with downstream luciferase activity. To determine whether eADAR1 affects respect to HIV-1 studies, different results have been found EIAV LTR-mediated gene expression, fetal equine dermal (FED) cells regarding the interaction of ADAR1 and HIV-1. Many groups have were co-transfected with the LTR-luciferase plasmid and either the found that ADAR1 positively regulates HIV-1 (Clerzius et al., 2009; eADAR1 expression plasmid or an empty vector. The results, presented Doria et al., 2011; Phuphuakrat et al., 2008), and one group has in Fig. 2A, showed that eADAR1 overexpression increased luciferase reported that ADAR1 inhibits HIV-1 replication (Biswas et al., activity by a factor of 4.22 compared with the vector control, indicating 2012). Consequently, investigating the interaction between EIAV that eADAR1 expression enhances LTR-mediated gene expression. The and ADAR1 will increase our understanding of the ADAR1 function result indicated that the eADAR1-mediated increase in luciferase in lentiviruses. expression correlated with the LTR region. Next, we examined the role of the RRE region in eADAR1- promoting activity. We constructed two different expression Results plasmids for EIAV env expression that expressed either the wild- type env (pEIAV-Env) with a functional RRE region or a codon- Equine ADAR1 positively regulated EIAV, HIV-1 and SIV replication optimized env (pEIAV-Env-opti), in which the RRE secondary structure was destroyed. HEK293T cells were then transfected To investigate the role of eADAR1 in EIAV replication, HEK293T with one of these plasmids and the ADAR1 expression vector. The cells were co-transfected with the eADAR1 expression plasmid and level of viral envelope protein (Env) expressed in these cells was EIAV-CMV-3-8, an EIAV proviral clone derived from the live attenu- examined by western blot. As presented in Fig. 2B and C, eADAR1 ated vaccine strain EIAVFDDV13 (Tang et al., 2014; Wang et al., 2011). The overexpression increased the Env expression in cells expressing amounts of EIAV-CMV-3-8 in the cell lysates and culture supernatants the wild-type env gene, but not in cells expressing the env gene were estimated by measuring the levels of EIAV Gag (p55) and capsid that lacked the RRE secondary structure. This finding indicates that (CA or p26, cleaved from p55) protein by western blot at 48 h post an intact RRE secondary structure is required for eADAR1-mediated transfection (hpt). As shown in Fig. 1A, the levels of the viral capsid promotion of Env expression. protein and its precursor in the cell lysate were significantly greater in We also investigated the effects of ADAR1 on two non- cells co-transfected with ADAR1, and this increase was dose depen- structural EIAV proteins: transactivator of transcription (Tat) and dent. Furthermore, the levels of p26 in viral particles released into the regulator of virion expression (Rev), which regulate viral replica- culture supernatant also increased substantially in response to ADAR1 tion (Dorn and Derse, 1988; Leroux et al., 2004; Martarano et al., overexpression. 1994). EIAV Tat or Rev was expressed in HEK293T cells, together These results indicate that EIAV production increased in response with increasing doses of eADAR1. The levels of Tat and Rev protein to eADAR1 overexpression. To determine whether eADAR1 promotes in cell lysates were examined by western blot. We did not observe the replication of viruses other than EIAV, we examined the effects of any noticeable changes in the densities of the relevant bands eADAR1 overexpression on human lentivirus HIV-1 and simian (Fig. 2D and E), indicating that eADAR1 overexpression did not lentivirus simian immunodeficiency virus (SIV) replication. HEK293T affect the expression of these regulatory proteins. The Tat coding cells were co-transfected with the plasmids expressing either HIV- region does not encode a double-stranded RNA secondary struc- 1NL4–3 or SIVmac239, together with increasing doses of the eADAR1 ture, and Rev contains only a partial RRE sequence (Lee et al., expression vector. The levels of viral Gag protein in the cell lysates and 2008). Our results further confirm that an intact RRE is essential culture supernatants were examined by western blot. As shown in for eADAR1 to promote viral protein expression. Fig. 1B and C, the amounts of both HIV-1 and SIV structural proteins in both cell lysates and supernatants increased with increasing levels of The eADAR1 RNA binding domain 1 was required to promote EIAV eADAR1 expression, indicating that eADAR1 activity upregulates the expression replication of multiple lentiviral species. To examine whether endogenous eADAR1 also promotes EIAV When compared with the sequence of human ADAR1, eADAR1 replication, we targeted ADAR1 in eMDMs with specificsiRNAs, was predicted to consist of three functional regions. The N- resulting in efficient knockdown of ADAR1 expression at the mRNA terminal region contains two copies of a Z-DNA binding domain, level (Fig. 1D). Twelve hours post siRNA transfection, 2.5 ng of Zα and Zβ. The central region of eADAR1 contains a double- EIAVDLV34 (as determined by RT activity) was used to infect the cells stranded RNA-binding domain, designated as dsRBD or R. The in 96-well plates. Viruses were collected from the supernatant and C-terminal region of eADAR1 is the deaminase catalytic domain titrated by measuring the viral RT activity 72 h post infection (hpi). (Cat). As shown in Fig. 1A and 2A, eADAR1 enhanced EIAV protein Our results show that the knockdown of eADAR1 significantly expression and was associated with the LTR region. To determine inhibited the production of EIAV (Fig. 1E) compared with the mock which domain or domains of eADAR1 were required to promote and non-specific siRNA controls. EIAV gene expression, we generated constructs in which the DNA binding domain, the RNA binding domain or the Cat domain of Equine ADAR-1-mediated promotion of EIAV expression is associated eADAR1 was deleted (Fig. 3A). We then examined the ability of with the LTR and RRE regions these deletion mutants to promote LTR-mediated gene expression by co-transfecting plasmids expressing the eADAR1 deletion ADAR1 binding on RNA depends on the double-stranded RNA mutants and the LTR-luciferase reporter plasmid into FED cells secondary structure (Bass, 2002; Nishikura, 2010; Samuel, 2011). It is and measuring the resulting luciferase activity. Western blot logical to investigate whether this structure is also essential for the analysis showed that these eADAR1 mutants were expressed at activity of ADAR1 in promoting EIAV expression. In the single-stranded the predicted molecular weights (Fig. 3B). As shown in Fig. 3C, positive-sense RNA genome of EIAV, two regions contain such deletion of the RNA-binding domain, but not the DNA-binding secondary structures: the long terminal repeat (LTR)andtheRev domain or Cat domain, significantly decreased the luciferase response element (RRE)regionsintheenvelope (env) gene. We first activity, indicating that RNA binding was required for ADAR1 up- investigated whether the LTR region is necessary for eADAR1- regulation of EIAV protein expression. mediated promotion of EIAV expression. The EIAV LTR region functions We further examined the roles of these three domains sepa- as the promoter for EIAV transcription. We subcloned this fragment rately with respect to the activity of a LTR-luciferase reporter into a pGL3-Basic plasmid to measure the activity of LTR by detecting (Fig. 3D and E). Only the RNA-binding region was sufficient to 366 Y.-D. Tang et al. / Virology 476 (2015) 364–371

HIV-1 EIAV SIV

-p24 Supernatant Supernatant -p27 Supernatant -p26

Control Control Control

-eADAR1 -eADAR1 -eADAR1

-p55 -p55 -p55

Cell lysate Cell lysate Cell lysate

-p24 -p27

-p26

- -actin - -actin - -actin

eADAR1: Control eADAR1: Control eADAR1: Control

siADAR1

Mock siScr 1 2 3

-eADAR1

- -actin

Fig. 1. Equine ADAR1 enhances lentivirus production. (A) Overexpression of eADAR1 increased EIAV production, as observed both in cell lysates and culture supernatants. An infectious EIAV clone, EIAV-CMV3-8 (50 ng of DNA), was transfected into HEK293T cells along with different amounts of the eADAR1 expression vector (0, 0.5, 1 or 2 μg). Virus production was quantified by western blotting using an anti-EIAV p26 monoclonal antibody. EIAV p26 enhanced by eADAR1 was quantified by densitometry scanning of western blot images of three independent experiments. *Po0.05. (B) Overexpression of eADAR1 increased HIV-1 production. HEK293T cells were transfected with 500 ng of HIV-1NL4–3 proviral DNA along with different amounts of the eADAR1 expression vector (0, 0.5, 1 or 2 μg). The levels of HIV-1NL4–3 and eADAR1 expression were detected by western blot. HIV-1 p24 expression enhanced by ADAR1 was quantified by densitometry scanning of western blot images of three independent experiments. *Po0.05.

(C) Overexpression of eADAR1 increased SIV production. HEK293T cells were transfected with 500 ng of SIVmac239 proviral DNA along with different amounts of the ADAR1 expression vector (0, 0.5, 1 or 2 μg). The levels of SIVmac239 and eADAR1 expression were detected by western blot. SIV p27 expression enhanced by ADAR1 was quantified by densitometry scanning of western blot images of three independent experiments. *Po0.05. (D) eADAR1 transcription was knocked down by siRNA. eMDMs were transfected with either 50 nM of one of three ADAR1-specific siRNA species (siADAR1 1-3) or 50 nM of a scrambled siRNA control (siScr) or mock transfected (Mock). At 24 hpt, eADAR1 (upper panel) and β-actin (lower panel) mRNA levels were quantified by semi-quantitative PCR. (E) Knockdown of eADAR1 transcription increased EIAV replication in equine monocyte-derived macrophages (eMDMs). eMDMs were transfected with 50 nM siADAR1 or siScr for 12 h and then infected with 2.5 ng of EIAVDLV34. The viral titers in the supernatants were assessed at 72 hpi by measuring the RT activity. P values o0.05 were considered statistically significant. The images shown in A, B and C are representative of three independent experiments. *Po0.05, from three independent experiments.

positively regulate LTR-luciferase expression (Fig. 3F). Because the RI alone eliminated the enhancement of LTR-luciferase activity by RNA-binding region contains three RNA-binding domains (RI, RII eADAR1 (Fig. 4C), and expression of only RI was sufﬁcient to and RIII), each of these domains was further either deleted or enhance the LTR activity to an extent similar to that observed with expressed separately to investigate its role in enhancing LTR- intact eADAR1 (Figs. 4F and 3F). These results indicate that the RI luciferase expression (Fig. 4A, B and D, E). To our surprise, deleting domain is essential for the eADAR1-mediated promotion of LTR- Y.-D. Tang et al. / Virology 476 (2015) 364–371 367

GL3-Basic-LTR Fold change (Luc)

Vector eADAR1

-Env-opti -Env

Control Control

-eADAR1 -eADAR1

- -actin - -actin

-Rev -Tat

Control Control

-eADAR1 -eADAR1

- -actin - -actin

Fig. 2. Equine ADAR1-mediated promotion of EIAV expression is correlated with the LTR and RRE regions. (A) eADAR1 activity is correlated with the LTR region. Either the eADAR1 expression vector (pADAR1) or the empty control vector (100 ng) was co-transfected with pEIAV-LTR-luc (10 ng) into fetal equine dermal (FED) cells. Luciferase activity was detected at 24 hpt. The data shown are from three independent experiments. *Po0.05. (B) Overexpression of pADAR1 promoted Env expression. HEK293T cells were co-transfected with 0.5 μg of the non-codon-optimized Env expression vector pEIAV-Env and different amounts of pADAR1 (0, 3, or 6 μg). At 48 hpt, the cells were lysed, and the culture supernatants were collected. Env expression was then analyzed by western blot using an equine infectious anemia (EIA)-positive serum. (C) Overexpression of pADAR1 did not increase the expression of codon-optimized Env. HEK293T cells were co-transfected with 0.1 μg of the codon-optimized Env expression vector pEIAV-Env-opti and different amounts of pADAR1 (0, 3, or 6 μg). At 48 hpt, the cells were lysed, and the culture supernatants were collected. Env expression was then analyzed by western blot using an EIA-positive serum. (D) and (E) pADAR1 overexpression did not promote Tat and Rev expression. HEK293T cells were co-transfected with 0.1 μg of either a Tat or a Rev expression plasmid and increasing doses of pADAR1 (0, 3 or 6 μg). The expression levels of Tat and Rev were examined by western blot at 48 hpt using an HA antibody to the HA tag fused to Tat and Rev. The images in B, C, D and E show representative results from three independent experiments. *Po0.05, from three independent experiments. mediated gene expression. This characteristic differs from human Overexpression of equine ADAR1 increased EIAV infectivity ADAR for HIV-1, for which integrity of the three RNA binding domains is necessary for HIV-1 LTR activation (Clerzius et al., The APOBEC3 family of cytidine deaminases plays an important 2009). role in restricting retroviruses by introducing G-to-A hypermutation 368 Y.-D. Tang et al. / Virology 476 (2015) 364–371

Z Z RI R II RIII Cat eADAR1

eADAR-∆DNA

eADAR-∆Cat

eADAR-∆RNA

A NA DNAD Cat RN ∆ ∆ ∆ - - - R1 R or ADAR ADAR Vect eADAeADAR1 eADAR eADAR eADAe

-ADAR/HA Fold change (Luc)

- -actin

at RNA DNA C ∆ ∆ ∆ - Vector - - eADAR1 R R A DAR D eA eADA eA

Z Z RI R II RIII Cat eADAR1

eADAR-DNA

eADAR-RNA

eADAR-Cat

DNA RNA Cat - - -

AR D A Vector eADAR1 e eADAR eADAR

-HA Fold change (Luc)

t - -actin a C AR1 DNA RNA - D - - Vector AR eA AR D D A ADAR eA e e

Fig. 3. The equine ADAR1 RNA-binding domain is required to promote EIAV LTR activity. (A) Schematic representation of the eADAR1 protein showing the sites and domains that were deleted for the deletion mutant experiments. (B) The deletion mutants were correctly expressed. A western blot of HEK293T cells at 48 hpt is shown. (C) Analysis of LTR transcriptional activity in response to wild-type eADAR1 or the eADAR1 deletion mutants. The expression levels of the mutants were veriﬁed by western blot. The eADAR1 expression plasmid and eADAR1 deletion mutant expression plasmids (100 ng each) were co-transfected with pEIAV-LTR-luc (10 ng) into FED cells. Luciferase activity was detected at 24 hpt. (D) Plasmids expressing each of the three functional domains of eADAR1 were constructed. (E) The expression levels of these mutants were examined by western blot using an anti-HA antibody to the HA tag fused to these eADAR1 fragments. (F) eADAR1 enhanced LTR transcriptional activity when the RNA binding domain was present. Expression plasmids for either eADAR1 or each of the three mutants (100 ng) were co-transfected with pEIAV-LTR-luc (10 ng) into FED cells. Luciferase activity was detected at 24 hpt. The images in B and E show representative results of three independent experiments. *Po0.05, from three independent experiments. of viral genomes, resulting in decreased viral infectivity. APOBEC3 Mangeatetal.,2003;Zhangetal.,2003).Here,wehaveshownthat proteins are packaged in HIV-1 particles and restrict replication of eADAR1 also upregulates the expression of the EIAV capsid protein the virus in its subsequent cycle of infection (Lecossier et al., 2003; and enhances LTR promoter activity. We therefore examined the Y.-D. Tang et al. / Virology 476 (2015) 364–371 369

RI R II RIII eADAR-RNA

eADAR-RNA-∆RI

eADAR-RNA-∆RII

eADAR-RNA-∆RIII

I III RI RIR R ∆ ∆ ∆ - - -

RNA RNA -RNA - -RNA - R AR A D A AD Vector eADAR eADAR e e

-eADAR/HA Foldc hange (Luc)

I I I R II ctor ∆ RI R RNA - ∆ ∆ - - - Ve A A AR N N R NA - -R R - -actin AD R - e A D DAR eA eADAR eA

RI R II R III eADAR-RNA

eADAR-RNA-RI

eADAR-RNA-RII

eADAR-RNA-RIII

I II II RI - -RIIR -RRIII A RNA RNA RNRNA -RNA - - - R R A A D DDA ADAR A Vector eADAR eAeADAR e e

-eADAR1/HAeADAR1/HA Foldc hange (Luc)

II RI I tor NA - RII R R - - A A - -actin Vec N N -RNA R - -R ADAR- e AR DAR D eADAR A eA e

Fig. 4. The RNA binding domain 1 is required for eADAR1-mediated enhancement of LTR activity. (A) Deletion mutants of the RNA-binding motifs 1–3 of the eADAR1 RNA- binding (R) domain were generated. (B) The deletion mutants were expressed correctly in HEK293T cells, as shown by western blot using an anti-HA antibody to the HA tag fused to these R domain fragments. (C) LTR transcriptional activity was enhanced by eADAR1 when RNA binding domain 1 was present. Plasmids expressing either eADAR1 or one of the deletion mutants missing one of the RNA binding motifs 1–3 (100 ng) were co-transfected with pEIAV-LTR-luc (10 ng) into FED cells. Luciferase activity was detected at 24 hpt. (D) Plasmids expressing each one of the RNA binding motifs 1–3 were constructed. (E) The expression levels of the three RNA binding motifs 1–3 were examined by western blot using an anti-HA antibody to the HA tags fused to these motif fragments. (F) LTR transcriptional activity was enhanced by motif 1 of the RNA binding domain. Plasmids expressing one of the RNA binding motifs (1–3) were co-transfected with pEIAV-LTR-luc (10 ng) into FED cells. Luciferase activity was detected at 24 hpt. The images in B and E show representative results from three independent experiments. *Po0.05, from three independent experiments. effect of this equine adenosine deaminase on EIAV infectivity. We eADAR1 (ADAR1) or donkey APOBEC A3Z3 (doA3Z3), a donkey prepared two one-life cycle EIAV-luciferase reporter viruses using homolog of human APOBEC3s that restricts EIAV infectivity (unpub- HEK293T cells co-transfected with a plasmid expressing either lished data). These two pseudoviruses, EIAV-ADAR1 and EIAV- 370 Y.-D. Tang et al. / Virology 476 (2015) 364–371

et al., 2003; Mangeat et al., 2003; Zhang et al., 2003), the A-to-G mutation does not generate stop codons that cause the premature termination of protein translation. Additionally, in contrast with APOBEC3G, which is packaged into virions and induces G-to-A mutations in the genomes of viral progeny, ADAR1 introduces A-to- G mutations to the genomes of viral progeny, and these mutations improve viral replication in the next replication cycle. Thus, ADAR1 enhances the ability of a virus to infect a host. ADAR1 has been identified as a positive regulator of HIV-1 in Luciferase activity several studies (Clerzius et al., 2009; Doria et al., 2011; Phuphuakrat et al., 2008). However, a recent publication reported that ADAR1 inhibits HIV-1 replication (Biswas et al., 2012). It is unclear why these studies obtained contrary results. These inconsistencies may result from ADAR1 editing activities. Indeed, ADAR1 has dual roles in virus replication. If ADAR1 edits viral genomes to an appropriate extent, viral replication is favored. In contrast, when ADAR1 edits viral genomes massively, virus replication can be hindered. In the -doA3Z3 present study, we showed that eADAR1 did not massively edit EIAV Supernatant genomes (unpublished data), which may be one of the mechanisms by which eADAR1 enhances EIAV replication. -p26 In conclusion, we demonstrated that eADAR1 positively regulates EIAV replication, as evidenced by its ability to enhance LTR-mediated Fig. 5. Equine ADAR1 enhances EIAV infectivity. (A) HEK293T cells were transfected gene expression, EIAV capsid expression and infectivity. However, with 3.5 μg of plasmids expressing either eADAR1 or donkey APOBEC A3Z3 further investigation is necessary to determine whether additional together with DNA from a three-plasmid lentivirus package system (1.5 μgof genomic mutations are required to enhance eADAR1-induced virus μ pONY8.1-LUC transfer plasmid, 1.5 g of pEIAV-GagPol packaging plasmid and production. 0.5 μg of pMD2.0G envelope plasmid), using the calcium phosphate method. The resulting pseudoviruses were collected and quantified by measuring the viral RT activity. HEK293T cells were infected with identical doses of EIAV. Luciferase activity was detected at 24 hpi. (B) eADAR1 was not packaged into EIAV virions. Materials and methods The pseudovirus in the supernatant of transfected HEK293T cells was collected, and the virion quantities were adjusted based on RT activity before being examined by Plasmids western blot. An EIA-positive serum was used to detect EIAV p26, and an HA antibody was used to detect the HA tag fused to eADAR1 and donkey A3Z3. Data were collected from three independent experiments. The image shown is from a Equine ADAR1 (eADAR1) cDNA was cloned from mRNA extracted representative experiment. *Po0.05. from eMDMs and expressed using the adenoviral expression vector pDC315 (Microbix Biosystems, Canada) to generate a protein fused to doA3Z3, were used to infect HEK293T cells at identical doses, which an HA tag at the C-terminus (pADAR1). A set of expression vectors were standardized using a reverse transcriptase (RT) activity assay with deletion mutations in the eADAR1 cDNA were constructed using (Fig. 5B). Intracellular luciferase activity was quantified at 24 hpi. As the primers listed in Table S1.Alloftheconstructedmutantswere shown in Fig. 5A, EIAV-doA3Z3 was poorly infectious, whereas EIAV- confirmed by sequencing. The plasmid pEIAV-Env expressing the ADAR1 was highly infectious, suggesting that ADAR1 plays a different codon-optimized EIAV envelope gene (env)wasderivedfromthe role in viral infections than the APOBEC3 proteins. EIAV strain EIAVFDDV3-8 (FDDV3-8) and was kindly provided by Dr. Yi- Ming Shao (Meng et al., 2011). EIAV-CMV-3-8 is an infectious EIAV clone derived from FDDV3-8 by substituting the U3 region of the 50 Discussion LTR with the CMV promoter (Wang et al., 2011). The HIV-1 NL4–3and SIVmac239 infectious clones were generously provided by Dr. Yong- The ability of this adenosine deaminase to promote EIAV Hui Zheng. The EIAV Tat fragment was amplified from FDDV3-8 and replication can be explained by the following eADAR1 activities. subcloned into the pEGFP-N1 mammalian expression vector (Clontech, First, the eADAR1 promotion of EIAV replication is dependent on USA) with an HA tag (Tang et al., 2014). The expression plasmids for RNA binding activity (Figs. 3 and 4). This property may be result the EIAV non-optimized envelope protein (Env) and the nonstructural from ADAR1 binding to PKR, which inhibits PKR kinase activity, protein Rev were constructed as described previously (Yin et al., 2014). resulting in reduced eIF-2a phosphorylation and protein synthesis (Clerzius et al., 2009, 2011; Pfaller et al., 2011). Indeed, our results Cells and virus stocks show that the eADAR1-mediated promotion of EIAV gene expression is associated with the LTR and RRE regions (Fig. 2), which have Primary eMDMs were prepared from equine peripheral blood been reported to contain RNA secondary structures. The secondary mononuclear cells (PBMCs) as described previously (Lin et al., structure is necessary for PKR activation (Clerzius et al., 2011; Cole, 2011) and maintained in RPMI 1640 medium containing 60% fetal 2007; Sadler and Williams, 2007). However, further studies are bovine serum (FBS). HEK293T and HEK293 cells were maintained necessary to determine whether the eADAR1 stimulation of EIAV in Dulbecco's Modified Eagle's Medium containing 10% FBS. A expression is mediated by the binding of ADAR1 to RNA secondary donkey MDM (dMDM)-adapted pathogenic EIAV strain, EIAVDLV34 structures or by other editing-independent pathways, for instance, (DLV34), was titrated using a reverse transcriptase (RT) assay kit through competitive inhibition of RNA binding to toll-like recep- (Roche, Switzerland) according to the manufacturer's instructions. tors (TLRs) and retinoic acid inducible-gene I (RIG-I), which are innate sensors of foreign RNAs and induce antiviral responses (Wu EIAV infectivity assay using a luciferase-expressing reporter virus and Chen, 2014). Second, eADAR1 induces an A-to-G mutation, which facilitates the adaptation of the virus to a new host. Unlike HEK293T cells were co-transfected with 3.5 μg of either pADAR1 the G-to-A hypermutation induced by APOBEC3 proteins (Lecossier or donkey APOBEC A3Z3 and 1.5 μgofpONY8.1-LUC,1.5μgpEIAV- Y.-D. Tang et al. / Virology 476 (2015) 364–371 371

GagPol and 0.5 μg of pMD2.0 G, using the calcium phosphate Gelinas, J.F., Clerzius, G., Shaw, E., Gatignol, A., 2011. Enhancement of replication of method (Tang et al., 2014; Yin et al., 2014). The VSV-G-packaged RNA viruses by ADAR1 via RNA editing and inhibition of RNA-activated protein kinase. J. Virol. 85, 8460–8466. pseudovirus was collected 48 hpt and used to infect HEK293T cells Gutierrez, R.A., Viari, A., Godelle, B., Frutos, R., Buchy, P., 2013. Biased mutational seeded in 48-well plates. These cells were washed and subjected to pattern and quasispecies hypothesis in H5N1 virus. Infect., Genet. Evol.: J. Mol. luciferase analysis at 24 hpi. Epidemiol. Evol. Genet. Infect. Dis. 15, 69–76. Ko, N.L., Birlouez, E., Wain-Hobson, S., Mahieux, R., Vartanian, J.P., 2012. Hyperedit- ing of human T-cell leukemia virus type 2 and simian T-cell leukemia virus type siRNA knockdown of eADAR1 mRNA in eMDMs 3 by the dsRNA adenosine deaminase ADAR-1. J. Gen. Virol. 93, 2646–2651. Lecossier, D., Bouchonnet, F., Clavel, F., Hance, A.J., 2003. Hypermutation of HIV-1 Three ADAR1-specific siRNAs (siADAR 1-3; see Table S1 for the DNA in the absence of the Vif protein. Science 300, 1112. Lee, J.H., Culver, G., Carpenter, S., Dobbs, D., 2008. Analysis of the EIAV Rev- sequences) and a scrambled siRNA negative control (siScr) were responsive element (RRE) reveals a conserved RNA motif required for high synthesized by RiboBio, China. Equine MDMs were seeded in 96- affinity Rev binding in both HIV-1 and EIAV. PLoS One 3, e2272. well plates and cultivated for three days. The cells were then Leroux, C., Cadore, J.L., Montelaro, R.C., 2004. Equine infectious anemia virus (EIAV): what has HIV's country cousin got to tell us? Vet. Res. 35, 485–512. washed twice with PBS before being transfected with either fi fi Li, Z., Wolff, K.C., Samuel, C.E., 2010. RNA adenosine deaminase ADAR1 de ciency eADAR1-speci c or negative control siRNA (diluted to 50 nM in leads to increased activation of protein kinase PKR and reduced vesicular serum-free media). Knockdown of eADAR1 mRNA was verified by stomatitis virus growth following interferon treatment. Virology 396, 316–322. reverse transcription PCR. Lin, Y.Z., Cao, X.Z., Li, L., Jiang, C.G., Wang, X.F., Ma, J., Zhou, J.H., 2011. The pathogenic and vaccine strains of equine infectious anemia virus differentially induce cytokine and chemokine expression and apoptosis in macrophages. Transfections and western blotting Virus Res. 160, 274–282. Mangeat, B., Turelli, P., Caron, G., Friedli, M., Perrin, L., Trono, D., 2003. Broad Cells were transiently transfected with the plasmids described antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature 424, 99–103. above using Lipofectamine 2000 (Invitrogen, USA) according to the Martarano, L., Stephens, R., Rice, N., Derse, D., 1994. Equine infectious anemia virus manufacturer's instructions. At 48 hpt, the cells and culture super- trans-regulatory protein Rev controls viral mRNA stability, accumulation, and natants were collected separately. The cells were washed once alternative splicing. J. Virol. 68, 3102–3111. Martinez, I., Dopazo, J., Melero, J.A., 1997. Antigenic structure of the human with PBS and then lysed in RIPA Lysis Buffer containing a protease respiratory syncytial virus G glycoprotein and relevance of hypermutation inhibitor cocktail (Roche, Switzerland). The supernatants were events for the generation of antigenic variants. J. Gen. Virol. 78 (Pt 10), centrifuged at 12,000g for 10 min at 4 1C to remove cell debris 2419–2429. and centrifuged again at 20,000g for 2 h at 4 1C to precipitate the Meng, Q., Lin, Y., Ma, J., Ma, Y., Zhao, L., Li, S., Liang, H., Zhou, J., Shen, R., Zhang, X., Shao, Y., 2011. A pilot study on an attenuated Chinese EIAV vaccine inducing EIAV particles. The proteins in the cell lysates and supernatant broadly neutralizing antibodies. Arch. Virol. 156, 1455–1462. precipitates were separated by SDS-PAGE, transferred to PVDF Murphy, D.G., Dimock, K., Kang, C.Y., 1991. Numerous transitions in human membranes (Millipore, Germany) and probed with antibodies. parainfluenza virus 3 RNA recovered from persistently infected cells. Virology 181, 760–763. Nie, Y., Hammond, G.L., Yang, J.H., 2007. Double-stranded RNA deaminase ADAR1 increases host susceptibility to virus infection. J. Virol. 81, 917–923. Acknowledgments Nishikura, K., 2010. Functions and regulation of RNA editing by ADAR deaminases. Annu. Rev. Biochem. 79, 321–349. This study was supported by grants from the Chinese National Pfaller, C.K., Li, Z., George, C.X., Samuel, C.E., 2011. Protein kinase PKR and RNA adenosine deaminase ADAR1: new roles for old players as modulators of the Key Programs for Infectious Diseases (2012ZX10001-008) and the interferon response. Curr. Opin. Immunol. 23, 573–582. National Natural Science Foundation of China (31070809) to J-H Z. Phuphuakrat, A., Kraiwong, R., Boonarkart, C., Lauhakirti, D., Lee, T.H., Auewarakul, We thank Dr. Yi-Ming Shao of the National Center for AIDS/STD P., 2008. Double-stranded RNA adenosine deaminases enhance expression of fi – Control and Prevention, China CDC, for the kind donation of the human immunode ciency virus type 1 proteins. J. Virol. 82, 10864 10872. Sadler, A.J., Williams, B.R., 2007. Structure and function of the protein kinase R. codon-optimized EIAV Env expression plasmids. Curr. Top. Microbiol. Immunol. 316, 253–292. Samuel, C.E., 2011. Adenosine deaminases acting on RNA (ADARs) are both antiviral and proviral. Virology 411, 180–193. Appendix A. Supporting information Suspene, R., Petit, V., Puyraimond-Zemmour, D., Aynaud, M.M., Henry, M., Guetard, D., Rusniok, C., Wain-Hobson, S., Vartanian, J.P., 2011. Double-stranded RNA adenosine deaminase ADAR-1-induced hypermutated genomes among inactivated seasonal Supplementary data associated with this article can be found in influenza and live attenuated measles virus vaccines. J. Virol. 85, 2458–2462. the online version at http://dx.doi.org/10.1016/j.virol.2014.12.038. Tang, Y.D., Na, L., Zhu, C.H., Shen, N., Yang, F., Fu, X.Q., Wang, Y.H., Fu, L.H., Wang, J.Y., Lin, Y.Z., Wang, X.F., Wang, X., Zhou, J.H., Li, C.Y., 2014. Equine viperin restricts equine infectious anemia virus replication by inhibiting the production and/or References release of viral Gag, Env and receptor by distorting the endoplasmic reticulum. J. Virol. Taylor, D.R., Puig, M., Darnell, M.E., Mihalik, K., Feinstone, S.M., 2005. New antiviral Bass, B.L., 2002. RNA editing by adenosine deaminases that act on RNA. Annu. Rev. pathway that mediates hepatitis C virus replicon interferon sensitivity through Biochem. 71, 817–846. – Biswas, N., Wang, T., Ding, M., Tumne, A., Chen, Y., Wang, Q., Gupta, P., 2012. ADAR1 ADAR1. J. Virol. 79, 6291 6298. is a novel multi targeted anti-HIV-1 cellular protein. Virology 422, 265–277. Wang, X., Wang, S., Lin, Y., Jiang, C., Ma, J., Zhao, L., Lv, X., Wang, F., Shen, R., Kong, X., Casey, J.L., 2006. RNA editing in hepatitis delta virus. Curr. Top. Microbiol. Immunol. Zhou, J., 2011. Genomic comparison between attenuated Chinese equine 307, 67–89. infectious anemia virus vaccine strains and their parental virulent strains. – Cattaneo, R., Schmid, A., Eschle, D., Baczko, K., ter Meulen, V., Billeter, M.A., 1988. Arch. Virol. 156, 353 357. Biased hypermutation and other genetic changes in defective measles viruses Ward, S.V., George, C.X., Welch, M.J., Liou, L.Y., Hahm, B., Lewicki, H., de la Torre, J.C., in human brain infections. Cell 55, 255–265. Samuel, C.E., Oldstone, M.B., 2011. RNA editing enzyme adenosine deaminase is Clerzius, G., Gelinas, J.F., Daher, A., Bonnet, M., Meurs, E.F., Gatignol, A., 2009. a restriction factor for controlling measles virus replication that also is required ADAR1 interacts with PKR during human immunodeficiency virus infection of for embryogenesis. Proc. Natl. Acad. Sci. U. S. A. 108, 331–336. lymphocytes and contributes to viral replication. J. Virol. 83, 10119–10128. Wu, J., Chen, Z.J., 2014. Innate immune sensing and signaling of cytosolic nucleic Clerzius, G., Gelinas, J.F., Gatignol, A., 2011. Multiple levels of PKR inhibition during acids. Annu. Rev. Immunol. 32, 461–488. HIV-1 replication. Rev. Med. Virol. 21, 42–53. Yin, X., Hu, Z., Gu, Q., Wu, X., Zheng, Y.H., Wei, P., Wang, X., 2014. Equine tetherin Cole, J.L., 2007. Activation of PKR: an open and shut case? Trends Biochem. Sci. 32, blocks retrovirus release and its activity is antagonized by equine infectious 57–62. anemia virus envelope protein. J. Virol. 88, 1259–1270. Doria, M., Tomaselli, S., Neri, F., Ciafre, S.A., Farace, M.G., Michienzi, A., Gallo, A., Zahn, R.C., Schelp, I., Utermohlen, O., von Laer, D., 2007. A-to-G hypermutation in 2011. ADAR2 editing enzyme is a novel human immunodeficiency virus-1 the genome of lymphocytic choriomeningitis virus. J. Virol. 81, 457–464. proviral factor. J. Gen. Virol. 92, 1228–1232. Zhang, H., Yang, B., Pomerantz, R.J., Zhang, C., Arunachalam, S.C., Gao, L., 2003. The Dorn, P.L., Derse, D., 1988. cis- and trans-acting regulation of gene expression of cytidine deaminase CEM15 induces hypermutation in newly synthesized HIV-1 equine infectious anemia virus. J. Virol. 62, 3522–3526. DNA. Nature 424, 94–98.