Human Cytomegalovirus Inhibits the Parsylation Activity of Tankyrase—A Potential Strategy for Suppression of the Wnt Pathway

Article Human Cytomegalovirus Inhibits the PARsylation Activity of Tankyrase—A Potential Strategy for Suppression of the Wnt Pathway

Sujayita Roy, Fengjie Liu and Ravit Arav-Boger *

Received: 23 October 2015; Accepted: 25 December 2015; Published: 29 December 2015 Academic Editor: Eric O. Freed

Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore 21287, MD, USA; [email protected] (S.R.); [email protected] (F.L.) * Correspondence: [email protected]; Tel.: +1-410-614-3917; Fax: +1-410-614-1491

Abstract: Human cytomegalovirus (HCMV) was reported to downregulate the Wnt/β-catenin pathway. Induction of Axin1, the negative regulator of the Wnt pathway, has been reported as an important mechanism for inhibition of β-catenin. Since Tankyrase (TNKS) negatively regulates Axin1, we investigated the effect of HCMV on TNKS expression and poly-ADP ribose polymerase (PARsylation) activity, during virus replication. Starting at 24 h post infection, HCMV stabilized the expression of TNKS and reduced its PARsylation activity, resulting in accumulation of Axin1 and reduction in its PARsylation as well. General PARsylation was not changed in HCMV-infected cells, suggesting specific inhibition of TNKS PARsylation. Similarly, treatment with XAV939, a chemical inhibitor of TNKS’ activity, resulted in the accumulation of TNKS in both non-infected and HCMV-infected cell lines. Reduction of TNKS activity or knockdown of TNKS was beneficial for HCMV, evidenced by its improved growth in fibroblasts. Our results suggest that HCMV modulates the activity of TNKS to induce Axin1, resulting in inhibition of the β-catenin pathway. Since HCMV replication is facilitated by TNKS knockdown or inhibition of its activity, TNKS may serve as an important virus target for control of a variety of cellular processes.

Keywords: Human cytomegalovirus (HCMV); tankyrase; Axin1; PARsylation; ubiquitination; Wnt/β-catenin

1. Introduction Infection with human cytomegalovirus (HCMV), a β-herpesvirus, is near universal in humans and can cause signiﬁcant morbidity and mortality in immunocompromised hosts, such as organ and bone marrow transplant recipients, and patients with AIDS [1]. HCMV is also the most common infectious cause of congenital birth defects [1]. Upon infection, HCMV manipulates and controls the host cell to facilitate its successful replication by dysregulating cell-signaling events [2]. An important signaling pathway hijacked by herpesviruses is the canonical Wnt/β-catenin pathway [3]. This pathway is critical for normal embryonic development and cellular differentiation and is the subject of investigation of numerous developmental and disease models [4]. Altered modulation of the Wnt pathway is a characteristic of multiple cancers and virus infections [3,5]. The key feature of the Wnt pathway is degradation of the downstream modulator β-catenin by the so-called “destruction complex” in the cytoplasm. The destruction complex is composed of a group of proteins: adenomatous polyposis coli (APC), glycogen-synthase kinase 3 (GSK3α, β) and Axin [6–9]. Wnt activation dissociates this complex and releases β-catenin which translocates into the nucleus and regulates the transcription of Wnt-responsive genes [10]. Axin1 is the focal point of control of the destruction complex; the concentration of Axin1 is the rate-limiting step in formation of the

Viruses 2016, 8, 8; doi:10.3390/v8010008 www.mdpi.com/journal/viruses Viruses 2016, 8, 8 2 of 17 destruction complex [11]. Thus, modulation of Axin1 is a critical step in the outcome of Wnt signaling. Axin1 stabilization and degradation is subject to post-translational modifications, ubiquitination and PARsylation [12]. PARsylation of proteins is accomplished by poly-ADP ribose polymerases (PARPs), a group of enzymes that modify their target proteins (including themselves) post-translationally by catalyzing covalent addition of adenosine diphosphate (ADP) ribose units from nicotinamide adenine dinucleotide (NAD) to give rise to branched ADP-ribose chains on the acceptor protein. Axin1 is specifically PARsylated by Tankyrase 1 and 2 (also called PARP 5a and 5b) which promotes Axin1 degradation through the ubiquitin proteasome pathway [12]. Unlike the γ-herpesviruses (Epstein Barr virus, EBV and Kaposi sarcoma Herpesvirus, KSHV) and α-herpesviruses HSV1 and HSV2, we and others have reported that HCMV inhibits the Wnt/β-catenin pathway [13,14]. HCMV infection of human foreskin fibroblasts induced Axin1, the negative regulator of Wnt, resulting in decreased expression of β-catenin. To better understand strategies used by HCMV to inhibit Wnt/β-catenin signaling, we investigated whether HCMV modulated TNKS. Our results show that the upregulation of Axin1 follows an increase in TNKS stabilization in HCMV-infected cells. This apparent increase in TNKS protein level is a result of reduced PARsylation activity of TNKS, because TNKS also autoregulates its own stability through PARsylation. The inhibitory effect on the enzymatic activity of TNKS may be beneficial for HCMV, because a chemical inhibitor of TNKS (XAV939), or the knockdown of TNKS facilitated its replication. Therefore, inhibition of TNKS activity is a viral strategy for efficient replication and may contribute to inhibition of the Wnt pathway by HCMV.

2. Results

2.1. Human Cytomegalovirus (HCMV) Infection Stabilizes TNKS HCMV infection is characterized by an increase in Axin1 expression followed by degradation of β-catenin [13,14], while infection with HSV1 or HSV2 does not result in β-catenin degradation [15]. The mechanism(s) involved in Axin1 accumulation by HCMV remain to be determined. Axin1 expression is tightly regulated in the cell; one mechanism of its regulation is through PolyADP Ribose Polymerase 5 (PARP5), also known as TNKS [12]. To determine whether the reported Axin1 accumulation in HCMV is mediated by TNKS, the effect of HCMV infection on the expression of TNKS 1 and 2, Axin1 and β-catenin was measured in HFFs at 24, 48 and 72 hours post infection (hpi). This was compared to HSV1-infected HFFs which were harvested at 6, 24 and 48 hpi. TNKS has two isoforms, 1 and 2, and the antibody used in our study can identify both and, henceforth, is referred to as TNKS. At all time points, Axin1 was induced and β-catenin was decreased in HCMV-infected HFFs, in agreement with previous reports [13,14]. There was a distinct accumulation of TNKS at all time points (Figure1A and quantified in Figure1B) accompanying the Axin1 accumulation. In contrast to HCMV, HSV1 did not significantly modify the levels of TNKS, Axin1, or β-catenin, in agreement with a previous report [15]. These results suggest that the observed changes in TNKS and Axin1 might be HCMV-specific. To examine whether HCMV replication was responsible for the accumulation of TNKS, HFFs were infected with UV-inactivated HCMV and lysates prepared from these cells were compared to lysates from wild-type HCMV-infected HFFs. Ultraviolet (UV)-inactivation of HCMV was confirmed by the absence of IE1/IE2 expression (Figure1C). TNKS accumulation was not observed in cells infected with the UV-inactivated Towne (Figure1C and quantified in Figure1D), indicating that active HCMV replication was required for TNKS accumulation. Viruses 2016, 816, 8, 0000

Axin1 stabilization and degradation is subject to post-translational modifications, ubiquitination and PARsylation [12]. PARsylation of proteins is accomplished by poly-ADP ribose polymerases (PARPs), a group of enzymes that modify their target proteins (including themselves) post- translationally by catalyzing covalent addition of adenosine diphosphate (ADP) ribose units from nicotinamide adenine dinucleotide (NAD) to give rise to branched ADP-ribose chains on the acceptor protein. Axin1 is specifically PARsylated by Tankyrase 1 and 2 (also called PARP 5a and 5b) which promotes Axin1 degradation through the ubiquitin proteasome pathway [12]. Unlike the γ-herpesviruses (Epstein Barr virus, EBV and Kaposi sarcoma Herpesvirus, KSHV) and α-herpesviruses HSV1 and HSV2, we and others have reported that HCMV inhibits the Wnt/β- catenin pathway [13,14]. HCMV infection of human foreskin fibroblasts induced Axin1, the negative regulator of Wnt, resulting in decreased expression of β-catenin. To better understand strategies used by HCMV to inhibit Wnt/β-catenin signaling, we investigated whether HCMV modulated TNKS. Our results show that the upregulation of Axin1 follows an increase in TNKS stabilization in HCMV- infected cells. This apparent increase in TNKS protein level is a result of reduced PARsylation activity of TNKS, because TNKS also autoregulates its own stability through PARsylation. The inhibitory effect on the enzymatic activity of TNKS may be beneficial for HCMV, because a chemical inhibitor of TNKS (XAV939), or the knockdown of TNKS facilitated its replication. Therefore, inhibition of TNKS activity is a viral strategy for efficient replication and may contribute to inhibition of the Wnt pathway by HCMV.

2. Results

2.1. Human Cytomegalovirus (HCMV) Infection Stabilizes TNKS HCMV infection is characterized by an increase in Axin1 expression followed by degradation of β-catenin [13,14], while infection with HSV1 or HSV2 does not result in β-catenin degradation [15]. The mechanism(s) involved in Axin1 accumulation by HCMV remain to be determined. Axin1 expression is tightly regulated in the cell; one mechanism of its regulation is through PolyADP Ribose Polymerase 5 (PARP5), also known as TNKS [12]. To determine whether the reported Axin1 accumulation in HCMV is mediated by TNKS, the effect of HCMV infection on the expression of TNKS 1 and 2, Axin1 and β-catenin was measured in HFFs at 24, 48 and 72 hours post infection (hpi). This was compared to HSV1-infected HFFs which were harvested at 6, 24 and 48 hpi. TNKS has two isoforms, 1 and 2, and the antibody used in our study can identify both and, henceforth, is referred to as TNKS. At all time points, Axin1 was induced and β-catenin was decreased in HCMV-infected HFFs, in agreement with previous reports [13,14]. There was a distinct accumulation of TNKS at all time points (Figure 1A and quantified in Figure 1B) accompanying the Axin1 accumulation. In contrast to HCMV, HSV1 did not significantly modify the levels of TNKS, Axin1, or β-catenin, in agreement with a previous report [15]. These results suggest that the observed changes in TNKS and Axin1 might be HCMV-specific. To examine whether HCMV replication was responsible for the accumulation of TNKS, HFFs were infected with UV-inactivated HCMV and lysates prepared from these cells were compared to lysates from wild-type HCMV-infected HFFs. Ultraviolet (UV)-inactivation of HCMV was confirmed Viruses 2016by, 8 ,the 8 absence of IE1/IE2 expression (Figure 1C). TNKS accumulation was not observed in cells 3 of 17 infected with the UV-inactivated Towne (Figure 1C and quantified in Figure 1D), indicating that active HCMV replication was required for TNKS accumulation.

Viruses 2016, 816, 8, 0000

Figure 1. Cont.

Figure 1.FigureTNKS 1. isTNKS induced is induced by human by human cytomegalovirus cytomegalovirus (HCMV) (HCMV) but butnot herpesvirus not herpesvirus (HSV) infection. (HSV) infection. (A) Western(A) blotWestern showing blot showing protein protein profile profile in response in response to to HCMV HCMV and HSV HSV infection infection at the at theindicated indicated hpi. hpi. HFFs were infected with HCMV Towne or HSV1 at multiplicity of infection (MOI = 1) and HFFs wereharvested infected at withindicated HCMV times; Towne (B) Shows or HSV1 the quantification at multiplicity of ofthe infectionsame from (MOI three= independent 1) and harvested at indicatedexperiments times; (B) with Shows standard the quantificationerror of mean (C) of Western the same blot fromindicating three requirement independent of active experiments HCMV with standardreplication error of mean for TNKS (C) Westernaccumulation. blot indicatingHFFs were requirementinfected with HCMV of active Towne HCMV or UV-inactivated replication for TNKS accumulation.HCMV HFFs Towne were at MOI infected = 1 and with harvested HCMV at 72 Towne hpi; (D or) Shows UV-inactivated the quantificati HCMVon of Townethe same at from MOI = 1 and harvestedthree at 72 independent hpi; (D) Shows experiments the quantification with standard error of the of samemean ( fromE) qRT-PCR three independentshowing mRNA experiments levels of with TNKS 1 and 2, Axin 1 and β-catenin at 24 hpi in infected HFFs. Quantification is averaged from four β standardindependent error of mean experiments (E) qRT-PCR with standa showingrd error mRNA of means. levels * indicates of TNKS p value 1 < and 0.05, 2, ** Axinindicates 1 and p < 0.01,-catenin at 24 hpi inand infected *** indicates HFFs. p Quantification< 0.001 and ns indicates is averaged non-significance. from four independent experiments with standard error of means. * indicates p value < 0.05, ** indicates p < 0.01, and *** indicates p < 0.001 and ns indicates2.2. Effect non-significance. of HCMV Infection on TNK Accumulation is Post-Translational HCMV upregulates multiple host proteins by inducing mRNA synthesis through viral 2.2. Effecttranscription of HCMV Infectionfactors, and on TNKS TNK might Accumulation be upregulated is Post-Translational in a similar way. However, qRT-PCR at 24 hpi revealed that mRNA levels of TNKS 1 and 2 were not significantly changed during infection, HCMVindicating upregulates the upregulation multiple of TNKS host was proteins a post-transcriptional by inducing effect mRNA (Figuresynthesis 1E). Additionally, through viral transcriptionAxin1 factors,or β-catenin and were TNKS not induced might at be the upregulated mRNA level, indicating in a similar that the way. Wnt pathway However, is likely qRT-PCR at 24 hpi revealed that mRNA levels of TNKS 1 and 2 were not significantly changed during infection, Viruses 2016, 8, 8 4 of 17 indicating the upregulation of TNKS was a post-transcriptional effect (Figure1E). Additionally, Axin1 or β-catenin were not induced at the mRNA level, indicating that the Wnt pathway is likely modulated at the protein level, in agreement with [13]. Accumulation of TNKS could result from decreasedViruses protein 2016, 8 ubiquitination16, 8, 0000 and subsequent reduction in its proteasomal degradation [12,16]. TNKS ubiquitination was measured by immunoprecipitation (IP) using anti-ubiquitin antibody and immunoblotmodulated using TNKSat the protein antibody level, (Figure in agreement2A). Awith significant [13]. Accumulation and reproducible of TNKS could difference result from in the level decreased protein ubiquitination and subsequent reduction in its proteasomal degradation [12,16]. of ubiquitinatedTNKS ubiquitination TNKS between was measured non-infected by immunoprecipitation and HCMV-infected (IP) using human anti-ubiquitin foreskin antibody fibroblasts and (HFFs) was observed,immunoblot suggesting using TNKS that theantibody ubiquitination (Figure 2A). A status significant of TNKS and reproducible was increased, difference rather in the than level decreased upon infectionof ubiquitinated (Figure2 A).TNKS These between results non-infected were alsoand HCMV-infected supported by human the degradation foreskin fibroblasts of TNKS, (HFFs) measured in the presencewas observed, of cycloheximide, suggesting that a translational the ubiquitination inhibitor, status inof non-infectedTNKS was increased, and HCMV-infected rather than HFFs decreased upon infection (Figure 2A). These results were also supported by the degradation of TNKS, after 48 hpimeasured (Figure in2B the and presence quantified of cycloheximide, in Figure2 aC). translational TNKS degradation inhibitor, in non-infected occurred in and both HCMV- non-infected and HCMV-infectedinfected HFFs lysates, after 48 hpi indicating (Figure 2B that and proteasomalquantified in Figure degradation 2C). TNKS ofdegradation TNKS was occurred not inhibitedin by HCMV. Takenboth non-infected together, TNKS and HCMV-infected accumulation lysates, in HCMV-infected indicating that proteasomal cells was degradation not due toof TNKS an upregulation was of transcriptionnot inhibited or post-translational by HCMV. Taken changes together, TNKS resulting accumulation in altered in HCMV-infected degradation. cells was not due to an upregulation of transcription or post-translational changes resulting in altered degradation.

Figure 2. TNKS degradation is not inhibited upon HCMV infection. (A) Western blot showing Figure 2. ubiquitinationTNKS degradation of TNKS immunoprecipitated is not inhibited from upon non-infected HCMV and infection. HCMV-infected (A) Western HFFs using blot an showing ubiquitinationanti-ubiquitin of TNKS antibody. immunoprecipitated HFFs were infected with from HCMV non-infected Towne at MOI and = HCMV-infected1 and harvested at 72 HFFs hpi; using an anti-ubiquitin(B) Western antibody. blot showing HFFs werestability infected of TNKS with after 48 HCMV h of infection Towne at atMOI MOI = 3 followed = 1 and by harvested treatment at 72 hpi; (B) Westernwith blot cycloheximide showing stabilityto inhibit de of novo TNKS translation after and 48 hharvested of infection at indicated at MOI time = points. 3 followed M indicates by treatment treatment with MG132 as a proteasome inhibitor control; (C) Quantification of TNKS’ degradation with cycloheximide to inhibit de novo translation and harvested at indicated time points. M indicates normalized to β-actin from (B) Cycloheximide was added at 48 hpi for the times indicated in the X treatmentaxis with (h of MG132 CHX). Data as a shown proteasome are the average inhibitor of three control; experiments (C) Quantiﬁcationwith standard error of of TNKS’the means. degradation normalized to β-actin from (B) Cycloheximide was added at 48 hpi for the times indicated in the X axis (h of CHX). Data shown are the average of three experiments with standard error of the means.

2.3. HCMV Infection Reduces PARsylation Activity of TNKS Since transcriptional change or protein degradation did not seem to play a role in TNKS accumulation, we hypothesized that HCMV infection could inﬂuence the enzymatic activity of TNKS. Viruses 2016, 8, 8 5 of 17

Viruses 2016, 816, 8, 0000 Inhibition of TNKS activity can lead to its accumulation, and inhibition of auto-PARsylation of TNKS2.3. by HCMV XAV939 Infection (a chemical Reduces PARsylation inhibitor Activity of TNKS) of TNKS results in TNKS accumulation in multiple cell systems [12Since,17 ,18transcriptional]. The contribution change ofor auto-PARsylationprotein degradation activity did not to seem TNK accumulationto play a role wasin TNKS therefore testedaccumulation, in infected cells we usinghypothesized an in vitro thatsemi-quantitative HCMV infection could biochemical influence assay. the Followingenzymatic IPactivity of TNKS, of its auto-PARsylationTNKS. Inhibition activity of TNKS was activity determined can lead by to measuring its accumulation, its ability and to inhibiti use biotin-labeledon of auto-PARsylation nicotinamide adenineof TNKS dinucleotide by XAV939 (NAD (a chemical+) as a substrate inhibitor of and TNKS) catalyze results the in addition TNKS accumulation of biotin-labeled in multiple ADP cell to itself. systems [12,17,18]. The contribution of auto-PARsylation activity to TNK accumulation was therefore The biotin label on TNKS was measured by Western blot. TNKS from non-infected HFFs exhibited tested in infected cells using an in vitro semi-quantitative biochemical assay. Following IP of TNKS, auto-PARsylationits auto-PARsylation activity, activity evident was by determined detection ofby biotin measuring bound its to ability TNKS, to butusein biotin-labeled HCMV-infected HFFs,nicotinamide there was aadenin consistente dinucleotide and reproducible (NAD+) as a reduction substrate inand the catalyze biotin-ADP the addition signal of detected biotin-labeled on TNKS (FigureADP3A). to Asitself. expected, The biotin XAV939 label on inhibited TNKS was the measured auto-PARsylation by Western activity blot. TNKS of TNKS from when non-infected included in the PARsylationHFFs exhibited reaction auto-PARsylation (Figure3A). Theactivity, lysates evident from by non-infected detection of and biotin infected bound cells to usedTNKS, in but the in TNKS assayHCMV-infected exhibited the expectedHFFs, there activity was a whenconsistent assayed and forreproducible proteasome-mediated reduction in the degradation biotin-ADP (Figuresignal 3B), indicatingdetected that on the TNKS quality (Figure of extracts3A). As expected, was not compromised.XAV939 inhibited In the addition, auto-PARsylation the general activity proteasomal of activityTNKS was when unchanged included in the extracts PARsylation treated reaction with XAV939, (Figure 3A). as expected.The lysates from The TNKSnon-infected activity and assay indicatedinfected that cells HCMV used in directly the TNKS reduced assay exhibited the auto-PARsylation the expected activity activity when of assayed TNKS1/2. for proteasome- The differences mediated degradation (Figure 3B), indicating that the quality of extracts was not compromised. In observed in auto-PARsylation activity were expected to result in changes in the abundance of PAR addition, the general proteasomal activity was unchanged in extracts treated with XAV939, as residuesexpected. on TNKS The TNKS in these activity lysates. assay An indicated antibody that directed HCMV againstdirectly reduced PAR residues the auto-PARsylation detected a weaker signalactivity on TNKS of TNKS1/2. immunoprecipitated The differences fromobserved HCMV-infected in auto-PARsylation cells asactivity compared were expected to non-infected to result cells (Figurein changes3C). Thus, in the HCMV-infected abundance of PAR cells residues had reduced on TNKS TNKS in these activity lysates.that An antibody resulted directed in reduced against TNKS auto-PARsylation.PAR residues detected The reduction a weaker insignal TNKS-mediated on TNKS immunoprecipitated PARsylation was from likely HCMV-infected specific, since cells generalas PARsylationcompared was to non-infected not reduced cells when (Figure whole-cell 3C). Thus, lysates HCMV-infected were probed cells had with reduced the anti-PAR TNKS activity antibody (Figurethat3D). resulted Thus, inHCMV reduced infectionTNKS auto-PARsylation. does not reduce The protein reduction PARsylation in TNKS-mediated in general, PARsylation but specifically was inhibitslikely the specific, PARsylation since general activity PARsylation of TNKS. was not reduced when whole-cell lysates were probed with Axin1the anti-PAR is a target antibody protein (Figure for 3D). TNKS-mediated Thus, HCMV infection PARsylation. does not The reduce reduction protein in PARsylation TNKS activity in in general, but specifically inhibits the PARsylation activity of TNKS. HCMV-infected cells was accompanied by a decrease in PARsylation of Axin1 immunoprecipitated Axin1 is a target protein for TNKS-mediated PARsylation. The reduction in TNKS activity in fromHCMV-infected infected cells (Figurecells was3E) accompanied when compared by a decrease to non-infected in PARsylation cells. of Axin1 TNKS immunoprecipitated has only two binding motifsfrom on Axin1infected [ 19cells]; therefore, (Figure 3E) detection when compared of PARsylation to non-infected on Axin1 cells. is moreTNKS difficulthas only thantwo binding detection of TNKSmotifs auto-PARsylation. on Axin1 [19]; therefore, Thus, HCMV-mediated detection of PARsyl reductionation on Axin1 in TNKS is more activity difficult translated than detection into reduced of PARsylationTNKS auto-PARsylation. of both TNKS and Thus, Axin1, HCMV which-mediated may reduction contribute in TNKS to increased activity stabilitytranslated and into accumulation reduced of bothPARsylation proteins inof infectedboth TNKS cells. and Axin1, which may contribute to increased stability and accumulation of both proteins in infected cells.

FigureFigure 3. Cont.Cont . Viruses 2016, 8, 8 6 of 17 Viruses 2016, 816, 8, 0000

Figure 3. HCMV inactivates TNKS TNKS activity. activity. ( (AA)) Western Western blot showing auto auto-PARsylation-PARsylation activity of immunoprecipitatedimmunoprecipitated TNKS at 72 hpi with HCMV To Townewne at MOI = 3. Immunoprecipitated Immunoprecipitated TNKS was used for the PARsylation assay as described in Materials and Methods. XAV939 XAV939 was included as a positive control for TNKS inhibition; ( B)) Proteasomal degradation degradation assay assay from from lysates lysates assayed assayed in in ( (AA)) to indicate lysate lysate quality quality and and activity; activity; ( (CC)) Western Western blot showing showing detection detection of of PAR PAR residues on TNKS immunoprecipitatedimmunoprecipitated fromfrom non-infectednon-infected and and HCMV-infected HCMV-infected HFFs HFFs (MOI (MOI = 3= at3 at 72 72 hpi); hpi); (D ()D Detection) Detection of ofPAR PAR residues residues by by Western Western blot blot in non-infectedin non-infected and and HCMV HCMV Towne-infected Towne-infected HFF HFF (MOI (MOI = 3, = 72 3, hpi) 72 hpi) cell celllysates lysates to show to show no loss no ofloss PARsylation of PARsylation activity activity from other from PARPs other PARPs in general; in general; (E) Western (E) Western blot showing blot showingdetection detection of PAR residues of PAR on residues Axin1 immunoprecipitatedon Axin1 immunoprecipitated from non-infected from non-infected and HFFs infectedand HFFs by infectedHCMV Towneby HCMV at MOI Towne = 3 atat 72MOI hpi. = 3 at 72 hpi.

2.4. Inhibition Inhibition of TNKS Activity Promotes HCMV Replication Since HCMVHCMV inhibitedinhibited the the PARsylation PARsylation activity activity of TNKS,of TNKS, the effectthe effect of chemical of chemical inhibition inhibition of TNKS’ of TNKS’activity activity on HCMV on replicationHCMV replication was studied. was Compoundsstudied. Compounds that inhibit that the inhibit Wnt pathway the Wnt were pathway reported were as reportedHCMV inhibitors as HCMV [14 ];inhibitors the Wnt inhibitor,[14]; the Salinomycin,Wnt inhibitor, reduced Salinomycin, Wnt signaling reduced and Wnt suppressed signaling HCMV and suppressedreplication (6).HCMV XAV939, replication an enzymatic (6). XAV939, inhibitor an of TNKS,enzymatic inhibits inhibitor the Wnt of pathwayTNKS, inhibits by upregulating the Wnt pathwayAxin1 [12 by]. Weupregulating used XAV939 Axin1 in HCMV-infected[12]. We used XAV HFFs939 at in concentrations HCMV-infected that HFFs inhibit at concentrations TNKS activity that(0.1 andinhibit 1 µ TNKSM) [20 activity,21]. At (0.1 these and concentrations, 1 μM) [20,21]. XAV939At these didconcentr not changeations, PARP1XAV939 or did PARP2 not change at the PARP1mRNA or level PARP2 (Figure at the S1). mRNA Unlike level the reported(Figure S1). Wnt Unlike inhibitor, the Salinomycin,reported Wnt XAV939inhibitor, did Salinomycin, not inhibit XAV939HCMV replication did not inhibit (Figure HCMV4A,B). replication A plaque (Figure reduction 4A,B). assay A performedplaque reduction with the assay laboratory-adapted performed with thestrain laboratory-adapted of HCMV, Towne, strain and of the HCMV, TB40endotheliotropic Towne, and the TB40 strain endotheliotropic showed a slight strain increase showed in both a slight the increasenumber in and both size the of number plaques and (Figure size 4ofA,C plaques and Figure(Figure S2) 4A,C with and XAV939 Figure S2) treatment, with XAV939 suggesting treatment, that suggestingTNKS inactivation that TNKS may conferinactivation a favorable may environmentconfer a favorable for HCMV environment replication for (Figure HCMV4A). Unlikereplication the (Figurestrong inhibition 4A). Unlike of HCMV-encoded the strong inhibition pp65 expression of HCMV-encoded by Salinomycin, pp65 noexpression inhibition by was Salinomycin, observed with no inhibitionXAV939 (Figure was observed4B). To test with whether XAV939 inhibition (Figure of 4B). TNKS To (PARP5) test whether reflects inhibition general PARPof TNKS inhibition (PARP5) by reflectsHCMV, general the PARP1and PARP inhibition 2 inhibitor, by ABT-888, HCMV, wasthe PA alsoRP1and used. There2 inhibitor, was a ABT- significant888, was reduction also used. in plaque There wassize witha significant ABT-888 reduction (Figure4 Cin and plaque Figure size S2), with but ABT-888 the number (Figure of infectious4C and Figure foci was S2), not but significantly the number ofreduced infectious (Figure foci4 A).was A not growth significantly assay (Figure reduced4D) (Figure similarly 4A). showed A growth that XAV939assay (Figure improved 4D) similarly HCMV showedgrowth, whilethat XAV939 Salinomycin improved fully inhibitedHCMV growth, HCMV. while ABT888 Salinomycin did not affect fully the inhibited number HCMV. of virions. ABT888 These didresults not suggestaffect the that number HCMV-mediated of virions. These inhibition results of suggest PARP activitythat HCMV-mediated could be specific inhibition to PARP5. of PARP Other activityPARPs maycould still be playspecific a role to inPARP5. virus propagation.Other PARPs Takenmay still together, play a inhibitionrole in virus of TNKSpropagation. activity Taken alone together,creates a favorableinhibition environmentof TNKS activity for HCMV alone crea replicationtes a favorable and inhibition environment of TNKS for activityHCMV replication might be a andstrategy inhibition that HCMV of TNKS exploits activity for amight variety be of a cellularstrategy effects. that HCMV exploits for a variety of cellular effects. Viruses 2016, 8, 8 7 of 17 Viruses 2016, 816, 8, 0000

Figure 4. TNKS inhibition favors HCMV replication. ( A)) Plaque Plaque assay assay for HCMV Towne and TB40 strains with XAV939 (PARP5 inhibitor),inhibitor), andand ABT-888ABT-888 (PARP1and(PARP1and 22 inhibitor).inhibitor). HFFs were infected to yield approximately 50-100 plaques/well,plaques/well, stained stained an andd quantified quantified at 10 days after plating; ( B)) Western Western blot to detect viral proteins at 72 hpi in HFFs infectedinfected with HCMV Towne atat MOIMOI == 1. Fold change in protein expression normalized to β-actin are quantified quantified under the blot; (C) Plaques from (A) imaged after staining at 1010×ˆ magnification;magnification; ( (DD)) Viral Viral growth growth assay assay for for Towne Towne and and TB40 TB40 strains strains with with XAV939 XAV939 (PARP5 inhibitor),inhibitor), SalinomycinSalinomycin (Wnt(Wnt inhibitor) and ABT-888ABT-888 (PARP1and(PARP1and 22 inhibitor).inhibitor). HFFs were infected andand supernatantssupernatants werewere collected collected at at 24-h 24-h intervals intervals till till 120 120 hpi. hpi. Released Released virions virions were were titered titered by aby standard a standard plaque plaque assay. assay. Average Average pfu/mL pfu/mL calculated calculated from from quadruplicate quadruplicate wells wells of one of one experiment experiment are shown.are shown. * indicates * indicatesp value p value < 0.05, < 0.05, ** indicates ** indicatesp < p 0.01. < 0.01.

To further confirm the role of TNKS during HCMV infection, TNKS1 and TNKS2 were knocked down in HFF using shRNAs. TNKS knockdown reduced levels of both TNKS1 and TNKS2 mRNA (Figure 5A) and protein (Figure 5B). Reduction of TNKS levels increased endogenous Axin1 and Viruses 2016, 8, 8 8 of 17

To further conﬁrm the role of TNKS during HCMV infection, TNKS1 and TNKS2 were knocked down in HFF using shRNAs. TNKS knockdown reduced levels of both TNKS1 and TNKS2 mRNA (FigureViruses5A) 2016 and, 816, protein 8, 0000 (Figure5B). Reduction of TNKS levels increased endogenous Axin1 and reduced β-catenin in HFFs (Figure5B) as expected [ 12]. Knockdown of TNKS did not alter cell viability reduced β-catenin in HFFs (Figure 5B) as expected [12]. Knockdown of TNKS did not alter cell or change cell growth when compared to the control cells (Figure5C). Growth of a pp28-luciferase viability or change cell growth when compared to the control cells (Figure 5C). Growth of a pp28- Towne strain was improved in the TNKS knockdown cell line when compared to the control line, both luciferase Towne strain was improved in the TNKS knockdown cell line when compared to the by luciferasecontrol line, counts both (Figureby luciferase5D) andcounts pp65 (Figure expression 5D) and pp65 (Figure expression5F). Ganciclovir (Figure 5F). (GCV) Ganciclovir inhibited (GCV) virus replicationinhibited in bothvirus cellreplication lines. Virion in both release cell lines. from Virion the tworelease lines from was the also two measured lines was inalso second measured cycle in assay usingsecond supernatants cycle assay from using Figure supernatants5D, showing from moreFigurevirions 5D, showing were more released virions from were the released TNKS from knockdown the cell lineTNKS as knockdown compared tocell the line control as compared line by to pp28-luciferasethe control line by activity pp28-luciferase (Figure activity5E) and (Figure pp65 5E) expression and (Figurepp655G). expression Thus, knockdown (Figure 5G). of Thus, TNKS knockdown conferred of a TNKS growth conferred advantage a growth for HCMV. advantage for HCMV.

FigureFigure 5. TNKS 5. TNKS knockdown knockdown improves improves HCMV HCMV replication replication inin HFFs.HFFs. ( A)) qRT-PCR qRT-PCR showing showing reduction reduction in TNKS1in andTNKS1 TNKS2 and TNKS2 mRNA mRNA level in level double in double knockdown knockdown cells comparedcells compared to control to control (TRC); (TRC); (B) Western(B) blot showingWestern blot effect showing of TNKS effect knockdown of TNKS knockdown on expression on expression of Wnt proteins of Wnt prot (TNKS,eins (TNKS, Axin1 andAxin1β -catenin);and (C) Cellβ-catenin); viability (C assay) Cell viability for control assay and for TNKS control knockdown and TNKS knockdown lines; (D) lines; pp28-luciferase (D) pp28-luciferase assay to assay measure to measure viral growth in control and TNKS knockdown cell lines; (E) Luciferase and Western blot viral growth in control and TNKS knockdown cell lines; (E) Luciferase and Western blot (F) of second (F) of second cycle infection from D; (G) Western blot to detect viral protein pp65 in lysates from (E) cycle infection from D; (G) Western blot to detect viral protein pp65 in lysates from (E) * indicates * indicates p value < 0.05, ** indicates p < 0.01, and ns indicates non-significance. p value < 0.05, ** indicates p < 0.01, and ns indicates non-signiﬁcance.

Viruses 2016, 8, 8 9 of 17

2.5. Accumulation of TNKS during Infection Correlates with Wnt Inhibition Viruses 2016, 816, 8, 0000 Since downregulation of TNKS activity was an HCMV-driven mechanism, we investigated the effects2.5. of Accumulation TNKS inhibition of TNKS on theduring expression Infection ofCorrelates Wnt pathway with Wnt proteins Inhibition during HCMV infection. Changes in TNKS,Since Axin1 downregulation and β-catenin of expression TNKS activity were was measured an HCMV-driven in HFFs andmechanism, the placental we investigated cytotrophoblastic the cell line,effects IST-1 of TNKS (Figure inhibition6A,B, quantiﬁed on the expression in Figure 6ofC,D), Wnttreated pathway with proteins XAV939 during or Salinomycin. HCMV infection. In both cell lines,Changes the inductionin TNKS, Axin1 of Axin1 and by β-catenin HCMV expression resulted in were reduced measured expression in HFFs of andβ-catenin the placental (Figure 6A, lanescytotrophoblastic 1 and 5, and Figure cell line,6B, IST-1 lanes (Figure 1 and 4).6A,B, Inactivation quantified in of Figure TNKS 6C,D), by XAV939 treated iswith expected XAV939to or lead to itsSalinomycin. cellular accumulation. In both cell lines, This the was induction clearly of observed Axin1 by byHCMV the resulted dose-dependent in reduced increase expression of of TNKS levelsβ-catenin with XAV939 (Figure (Figure 6A, lanes6A 1 lanesand 5, 3–4 and in Figure HFFs, 6B, Figure lanes 61B, and lanes 4). Inactivation 2–3 in IST-1). of TNKS Whenever by XAV939 XAV939 inducedis expected accumulation to lead to of its inactive cellular TNKS,accumulation HCMV. This was was able clearly to drive observed even by further the dose-dependent accumulation of increase of TNKS levels with XAV939 (Figure 6A lanes 3–4 in HFFs, Figure 6B, lanes 2–3 in IST-1). TNKS (Figure6A, lanes 7–8, compared to lanes 3–4, and Figure6B, lanes 5–6, compared to lanes 2–3). Whenever XAV939 induced accumulation of inactive TNKS, HCMV was able to drive even further An increaseaccumulation of TNKS of TNKS levels (Figure was accompanied 6A, lanes 7–8, by comp accumulationared to lanes of 3–4, Axin1 and (Figure Figure6 A,6B, laneslanes 3–4,5–6, and Figurecompared6B, lanes to 2–3), lanes in 2–3). agreement An increase with theof TNKS expected levels effect was ofaccompanied TNKS inhibition by accumulation on Axin1 accumulation. of Axin1 In all(Figure instances, 6A, lanes infection 3–4, withand Figure HCMV 6B, resulted lanes 2–3), in reducedin agreementβ-catenin with the levels, expected compared effect of to TNKS the levels observedinhibition in non-infected on Axin1 accumulation. cells (Figure In6 A,all lanesinstances, 7–8, infection compared with to HCMV lanes 3–4,resulted and in Figure reduced6B, β lanes- 5–6, comparedcatenin levels, to compared lanes 2–3). to the Infection levels observed resulted inin non-infected reduced levels cells (Figure of β-catenin, 6A, lanes irrespective7–8, compared of the amountto lanes of Axin 3–4, orand TNKS. Figure Data 6B, lanes in Figure 5–6, compared6C,D indicate to lanes that 2–3). HCMV Infection always resulted increases in reduced TNKS levels and of Axin1 whileβ lowering-catenin, irrespectiveβ-catenin, of irrespective the amount of of the Axin starting or TNKS. basal Data level in of Figure these 6C,D proteins. indicate Therefore, that HCMV a robust HCMV-drivenalways increases mechanism TNKS and is inAxin1 place while for lowering TNKS inactivation, β-catenin, irrespective which plays of the a starting role in basal inhibition level of of the these proteins. Therefore, a robust HCMV-driven mechanism is in place for TNKS inactivation, which Wnt/β-catenin pathway. However, HFFs and intermediate trophoblastic cell line, ISTs were not as plays a role in inhibition of the Wnt/β-catenin pathway. However, HFFs and intermediate β sensitivetrophoblastic to XAV939-mediated cell line, ISTs were suppression not as sensitive of -catenin to XAV939-mediated (Figure6A, lanes suppression 3–4 and Figureof β-catenin6B, lanes 2–3) as(Figure has been6A, lanes reported 3–4 and in Figure colon cancer6B, lanes cell 2–3) lines as ha [12s been]. This reported “non-responsiveness” in colon cancer cell to lines XAV939 [12]. has also beenThis “non-responsiveness” reported in other cancer to XAV939 cell-lines has [also22]. been In the reported colon cancerin other cell cancer line cell-lines SW480, [22]. we alsoIn the found that βcolon-catenin cancer was cell reduced line SW480, in thewe presencealso found of that XAV939 β-catenin (Figure was reduced S3). Thus, in the although presence useof XAV939 of XAV939 sheds(Figure light on S3). the Thus, requirement although use for of inactivating XAV939 sheds TNKS light for on HCMV the requirement infection, for HCMV-mediated inactivating TNKS control for of β-cateninHCMV may infection, involve HCMV-mediated additional Wnt control proteins of β-catenin that are may not involve directly additional controlled Wnt by proteins TNKS. that are not directly controlled by TNKS.

Figure 6. Cont. Viruses 2016, 816, 8, 0000 Viruses 2016, 8, 8 10 of 17

Figure 6. Cont.

FigureFigure 6. HCMV-induced 6. HCMV-induced TNKS TNKS accumulation accumulation correlates with with decreased decreased β-catenin.β-catenin. Western Western blot blot showingshowing levels levels of Wntof Wnt proteins proteins (TNKS, (TNKS, Axin1,Axin1, β-catenin)-catenin) in in infected infected HFFs HFFs (A) (andA) andIST-1 IST-1 (B) cells. (B) cells. XAV939XAV939 (0.1 and(0.1 1andµM) 1 and μM) Salinomycin and Salinomycin (0.1 µ M)(0.1 were μM) used were as used modulators as modulators of Wnt. of TNKS Wnt. accumulates TNKS in responseaccumulates to HCMV in response infection to HCMV and infectionβ-catenin and is β reduced-catenin is upon reduced infection; upon infection; (C) and (C ()D and) show (D) the quantificationshow the quantification of (A) and ( Bof) respectively(A) and (B) respectively from three from representative three representative experiments. experiments. Data plottedData is meanplotted˘ standard is mean error, ± standard SE. Lane error, numbers SE. Lane in numb Westerners in blot Western figures blot correspond figures correspond to lane numbersto lane in numbers in quantification plots. * indicates p value < 0.05, ** indicates p < 0.01, and *** indicates p < quantification plots. * indicates p value < 0.05, ** indicates p < 0.01, and *** indicates p < 0.001, for 0.001, for changes in β-catenin. changes in β-catenin. 3. Discussion 3. Discussion HCMV often manipulates host cell proteins to promote its replication. There are multiple HCMVinstances often of HCMV manipulates co-opting host critical cell proteinscellular pathwa to promoteys to itsassist replication. its replication, There including are multiple cell cycle instances of HCMVregulators co-opting and other critical signaling cellular pathways. pathways HCMV to assist recruits its replication,pRb, Cyclin Dependent including cellKinases cycle (CDKs) regulators and other[23], signalingp53 [24,25], pathways. Pro Myelocytic HCMV Leukemia recruitspRb, (PML) Cyclin [25] Dependentand other cellular Kinases proteins (CDKs) for [23 ],efficient p53 [24 ,25], Pro Myelocyticreplication and Leukemia at various (PML) points [25 during] and otherits replication cellular [26]. proteins for efficient replication and at various We reported previously that HCMV upregulated Axin1 and reduced β-catenin [14]. Inhibitors points during its replication [26]. of the Wnt pathway such as Salinomycin and Monensin efficiently suppressed HCMV replication, β Wesuggesting reported a fine previously balance of that the HCMV Wnt pathway upregulated is important Axin1 for and HCMV reduced replication.-catenin In the [14 ].current Inhibitors of thestudy, Wnt we pathway show that such HCMV as Salinomycin inactivates TNKS, and Monensin the negative efficiently regulator suppressed of Axin1. Infection HCMV leads replication, to suggestingaccumulation a fine and balance stabilization of the of Wnt TNKS pathway in infected is important cells resulting for from HCMV its enzymatic replication. inhibition In the and current study,reduced we show PARsylation that HCMV activity. inactivates Reduced PARsylation TNKS, the negativemay also explain regulator the changes of Axin1. in Axin1 Infection stability. leads to accumulationAlthough the and direct stabilization mechanism of by TNKS which in HCMV infected inac cellstivates resulting TNKS activity from its remains enzymatic to be elucidated, inhibition and reducedit is PARsylationpossible that viral activity. or even Reduced cellular PARsylation proteins induced may by also infection explain interact the changes with TNKS in Axin1 and slow stability. Althoughdown theits directenzymatic mechanism activity. A by protein which array HCMV target inactivatesing cellular TNKS proteins activity that remainsinteract with to be HCMV elucidated, it is possibleImmediate that Early viral 1 (IE1) or even identified cellular Tankyrase1 proteins Binding induced Protein by infection 1 (TNKS1BP1) interact as withan IE1-interacting TNKS and slow protein based on hybridization and IP [27], supporting the requirement for active virus replication down its enzymatic activity. A protein array targeting cellular proteins that interact with HCMV and de novo synthesis of HCMV proteins (particularly IE1) for TNKS accumulation. Thus, it is possible Immediatethat viral Early components 1 (IE1) identifiedtarget TNKS Tankyrase1 and regulate Binding its activity. Protein In a 1recent (TNKS1BP1) study [28] asusing an IE1-interactinga genome- proteinwide based deep on sequencing hybridization and ribosome and IP [ 27profiling], supporting approach, the no requirement significant increase for active in TNKS virus replicationtranslation and de novowithsynthesis HCMV infection of HCMV at 5–72 proteins h of infection (particularly was seen. IE1) Thus, for TNKS TNKS accumulation. stabilization may Thus, be controlled it is possible by that viralpost-translational components target modification. TNKS and In regulateaddition, itsrate activity.s of protein In a production recent study and [28 turnover] using may a genome-wide change deepunder sequencing different and cellular ribosome conditions profiling [29]. approach, no significant increase in TNKS translation with HCMV infectionUsing a semi-quantitative at 5–72 h of infection assay, we was observed seen. Thus,a decrease TNKS in stabilizationthe auto-PARsylation may be activity controlled of by post-translationalTNKS upon infection, modification. which should In addition, be expected rates to of reduce protein its production degradation. and However, turnover TNKS may was change underdegraded different efficiently cellular in conditions both non-infected [29]. and HCMV-infected HFFs. This apparent conflict probably is due to residual PARsylation activity in HCMV-infected HFFs. Thus, unPARsylated TNKS might Using a semi-quantitative assay, we observed a decrease in the auto-PARsylation activity of TNKS accumulate slowly, but eventually TNKS will get PARsylated, ubiquitinated and degraded even in uponinfected infection, HFFs. which HCMV should slows be down expected the toPARsylation reduce its activity degradation. of TNKS, However, which is TNKS reflected was in degraded its efficientlyaccumulation in both during non-infected infection. and The HCMV-infected accumulated TNKS HFFs. plays This an apparent as yet undefined conflict role probably in HCMV is due to residual PARsylation activity in HCMV-infected HFFs. Thus, unPARsylated TNKS might accumulate slowly, but eventually TNKS will get PARsylated, ubiquitinated and degraded even in infected HFFs. HCMV slows down the PARsylation activity of TNKS, which is reflected in its accumulation during infection. The accumulated TNKS plays an as yet undefined role in HCMV replication. Reduction in Viruses 2016, 8, 8 11 of 17

TNKS activity in infected cells is also reflected in some reduction in Axin1 PARsylation in infected cells. However, TNKS-mediated regulation may not be the only mechanism leading to Axin1 stabilization in HCMV infection. Axin1 levels are tightly controlled by the “destruction complex” (DC) present in the cytoplasm. KSHV employs a mechanism targeting the critical DC protein GSK-3β that involves LANA-mediated redistribution of GSK-3β [30]. Similarly, the involvement of APC and casein kinase 1 (CKI) in HCMV-mediated Wnt modulation remains to be explored, and might still play a significant role in regulating Axin1 stability and degradation during HCMV infection. Studies of the interdependence of TNKS and Axin and how they mutually regulate each other identified RNF146 (RING Finger Protein), an ubiquitin E3 ligase which serves as a positive regulator of Wnt signaling [17,31]. RNF146, TNKS, and Axin formed a protein complex, and RNF146 was responsible for ubiquitination of the trimer and targeting it for proteasomal degradation. A Trp–Trp–Glu (WWE) motif in RNF146 recognized PARsylated proteins for degradation through a direct binding between the WWE domain and PAR moiety. In addition to TNKS, other target proteins are modified by the same mechanism. TNKS activity reduced RNF146 protein levels. The mechanism of Wnt regulation via RNF146 is entirely unexplored in HCMV and it is possible that HCMV might modify or change the equilibrium of the proteins in the trimer, thus modulating changes in Wnt. The WWE domain of RNF146 shown to be a PAR-binding domain is also found in other E3 ligases, namely, HUWE1 (HECT, UBA and WWE domain containing 1), DTX1–4 (deltex homologues 1–4), TRIP12 (thyroid hormone receptor interactor 12), reviewed in [32]. These E3 ligases may also regulate protein turnover of TNKS and Axin in a PARsylation-dependent manner. It is possible that HCMV co-opts this finely-tuned mechanism for protein degradation in the cell. The accumulation of TNKS during HCMV infection and resulting Axin1 stabilization and decrease in β-catenin expression is a unique feature of HCMV infection. The effect of XAV939 on β-catenin expression may be both time- and cell-dependent, since in HFFs, despite the accumulation of TNKS and Axin1, the expression of β-catenin was determined largely by infection and not by XAV939. Initial studies of XAV939 in the colon carcinoma in SW480 cells revealed that exposure to XAV939 resulted in reduced levels of β-catenin. Similarly, we found the same effect of XAV939 on β-catenin in SW480 cells, which differed from its effects in HFFs (Figure S3). Reports suggest that the outcome of XAV939 treatment may be cell-type specific, and, in some cells, prolonged exposure (>6 h) to XAV939 results in a reduced effect on β-catenin inhibition [22]. XAV939 insensitivity was conferred by β-catenin’s association with Lymphoid-enhancer factor 1 (LEF1) and B cell lymphoma 9 (BCL9-2/B9L), which normally accumulate during Wnt stimulation, suggesting that β-catenin inhibition requires the targeting of its interaction with LEF1 and/or BCL9/B9L. Alternatively, it is possible that mechanisms other than the TNKS-Axin axis are involved in β-catenin inhibition by HCMV.

4. Materials and Methods

4.1. Compounds XAV939, an inhibitor of TNKS activity, Salinomycin, an inhibitor of the Wnt pathway that decreases the phosphorylation of low-density lipoprotein receptor-related protein 6, and ABT-888, PARP1 and PARP2 inhibitor, were obtained from Sigma-Aldrich (St. Louis, MO, USA) and dissolved ˝ in water or DMSO. A stock of 10 mM was stored at ´80 C. The 50% inhibitory concentration (IC50) of XAV939 for TNKS1 and 2 is 10 and 4 nM, respectively and for PARP1and 2–2.2 and 0.1 µM. Therefore, XAV939 was used at concentrations of 0.1 and 1 µM, which are speciﬁc for inhibition of TNKS1 and 2. The IC50 of ABT88 towards PARP1 and TNK1 is 5.2 nM, and 15 µM; therefore, it was used at 0.1 µM. Cycloheximide (Sigma-Aldrich) was dissolved in DMSO and used at 100 µg/mL. Ganciclovir (Sigma-Aldrich) was dissolved in PBS and used at 5 µM. Viruses 2016, 8, 8 12 of 17

4.2. Viruses The HCMV strains, Towne (American Type Culture Collection, ATCC, Manassas, VA, USA, VR-977) and TB40-GFP (ATCC VR-1578) were used as speciﬁed for each experiment. Towne HCMV was UV-inactivated by spreading a thin layer of stock suspension in an uncovered six-well tissue culture plate and exposing to a total dose of 720 mJ/cm2 in a UV crosslinker (Spectrolinker XL-1000, Spectronics, Westbury, NY, USA) [33]. A clinical isolate of HSV1 was obtained from the clinical microbiology laboratory at Johns Hopkins, with no identiﬁers that could be linked to a patient. The pp28-luciferase Towne strain was constructed as previously described [34]. This recombinant virus expresses luciferase under the control of the UL99 (pp28) late promoter 48–72 h post infection (hpi) and luciferase activity highly correlates with plaque reduction assay.

4.3. Cell Culture, Virus Infection and Anti-Viral Assays Human foreskin fibroblasts (HFFs) passage 12–16 (ATCC, CRL-2088) were grown in Dulbecco’s Modified Eagle Medium (DMEM) (Life Technologies, Grand Island, NE, USA) containing 10% fetal bovine serum (FBS) (Gibco, Carlsbad, CA, USA) in a 5% CO2 incubator at 37 ˝C and used for infection with HCMV. The cytotrophoblastic cell line, IST-1, was provided by Ie Ming-Shih, Johns Hopkins University School of Medicine, and was maintained in RPMI 1640 (Life Technologies, Grand Island, NY, USA) containing 10% FBS in a 5% CO2 incubator at 37 ˝C[35]. Colorectal adenocarcinoma, SW480 cells (ATCC CCL-228) were grown in L-15 medium containing 10% FBS (Gibco) under atmospheric conditions at 37 ˝C. When used, MG132 (10 µM) was included in the culture medium for 8–10 h before harvesting. Cell viability was tested using the CellTiter-Glo luminescent assay kit (Promega, Madison, WI, USA) according to manufacturer’s instructions. Infection was carried out at multiplicity of 1 plaque forming unit (PFU)/cell (MOI = 1), unless otherwise noted. Following 90 min adsorption, media containing virus was removed and replaced by DMEM with 4% FBS (Gibco) with or without compounds. Infected or infected-treated HFFs were collected at specific time points depending on the assay used. For plaque reduction assay, HFFs were seeded at 1 ˆ 106 cells/ 24-well plate one day prior to infection. HCMV (Towne or TB40) was diluted in DMEM to a desired titer which gave approximately 50 plaques/well and added to each well in quadruplicates. Plates were incubated for 90 min with shaking every 10 min; thereafter overlaid with carboxy methyl cellulose supplemented with XAV939 or ABT-888. After 10 days of incubation, cells were stained with crystal violet and plaques were counted at 40ˆ magnification and photographed using a Nikon Eclipse E-800 fluorescence microscope (Nikon, Melville, NY, USA). The pp28-luciferase Towne virus was used for estimation of viral growth. Infected HFFs were collected at 96 hpi, and pp28 activity was measured by a luciferase assay kit (Promega) on the GloMax-Multi detection system (Promega). In second-cycle replication assays, supernatants were collected from the tested conditions of the first cycle and used for infection of fresh HFFs in 96-well plates. Luciferase activity was measured 120 h after second-cycle infection. Viral growth assay was carried out as reported [34,36]. Briefly, HFFs were infected with HCMV Towne or TB40 (MOI 0.1) and treated with drugs. Culture supernatants were collected every 24 h until 120 hpi and frozen at ´80 ˝C. Collected samples were thawed and used for titration of infectious virus by plaque assay.

4.4. SDS-PAGE and Immunoblot Analysis Cell lysates were quantified for total protein content using Pierce bicinchoninic acid (BCA) protein assay kit (Thermo-Fisher Scientific, Waltham, MA, USA). Equivalent amount of proteins were mixed with an equal volume of sample buffer (125 mM Tris-HCl (pH 6.8), 4% SDS, 20% glycerol, 5% β-mercaptoethanol) and boiled at 100 ˝C for 10 min. Denatured proteins were resolved in Tris-glycine polyacrylamide gels (10%–12%) and transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories, Hercules, CA, USA) by electroblotting. Membranes were incubated in blocking solution Viruses 2016, 8, 8 13 of 17

(5% nonfat dry milk/BSA and 0.1% Tween 20 in PBS (PBST)) for 1 h, washed three times with PBST, and incubated with primary antibodies diluted in 5% milk at 4 ˝C overnight. The membranes were washed with PBST, followed by incubation with horseradish peroxidase (HRP)-conjugated secondary antibodies in 5% milk/BSA for 1 h at room temperature. After three washing steps with PBST, protein bands were visualized by chemiluminescence using SuperSignal West Pico reagent (Pierce Chemical, Rockford, IL, USA). The following antibodies were used: mouse monoclonal anti-pp65 antibody (Vector Laboratories, Burlingame, CA, USA), mouse monoclonal anti-PAR antibody (Trevigen, Gaithersburg, MD, USA), rabbit monoclonal anti-Axin1 (Cell Signaling Technologies, Danvers, MA, USA), rabbit polyclonal anti-β-catenin, anti-TNKS 1 and 2 (Santa Cruz Biotechnologies, Santa Cruz, CA, USA), mouse monoclonal anti-ubiquitin and anti-β-actin antibody (Santa Cruz Biotechnology), HRP-conjugated goat anti-rabbit IgG antibody (Cell Signaling Technology), and HRP-conjugated sheep anti-mouse IgG (GE Healthcare, Waukesha, WI, USA). Protein A-HRP was obtained from ThermoFisher Scientiﬁc (Grand Island, NY, USA). Proteins levels were quantiﬁed using ImageJ 1.48v (NIH, Bethesda, MD, USA).

4.5. Protein Immunoprecipitation (IP) Approximately 107 HFFs were lysed with RIPA buffer and total lysate was precleared with 50 µL protein A/G agarose bead slurry (Santa Cruz) for 1 h. The precleared lysates were quantiﬁed as described above and 500 µg to 1 mg of total lysate was incubated overnight with anti-ubiquitin (1 µg), anti-TNKS (2 µg), or anti-Axin1 (2 µg) antibody. The antibody–lysate complexes were incubated with protein A/G beads (Santa Cruz) at 4 ˝C and washed three times with RIPA buffer. Samples were boiled in SDS sample buffer, and the supernatant was analyzed on SDS-PAGE gel after immunoblotting as described previously. Then, 1%–5% of the cell lysate used for IP was loaded on gels as “Input”.

4.6. IP of TNKS and In Vitro PARsylation Assay The protocol was modified from [12] and [37]. Briefly, 107 HFFs were pelleted and resuspended in 1 mL TNE buffer (10 mM Tris (pH 8.0), 1% Nonidet P-40, 0.15 M NaCl, 1 mM EDTA) for 1 h at 4 ˝C. Cells were lysed by passing 6 times through 25 5/8 gauge syringe, and supernatant was collected after centrifugation at 14,000 rpm for 10 min. The lysate was pre-cleared with 50 µL Protein A/G beads (Santa Cruz) for 2 h at 4 ˝C, and approximately half of the lysate was used per reaction. TNKS was immunoprecipitated using 1 µg TNKS1/2 antibody per mL TNE extract for 5 h at 4 ˝C. The antibody-lysate mix (500 µL) was loaded on 25 µL protein A/G beads per reaction (pre-washed with TNE buffer (0.5–1mL/wash, 2 washes)) for an overnight incubation at 4 ˝C. Beads were washed thrice with TNE and twice with 50 mM Tris-HCl (8.0). TNKS assay buffer ((50 mM Tris-HCl (pH 8.0), + 4 mM MgCl2, 0.2 mM dithiothreitol (DTT)) was then added with Biotin NAD (5 µM, Trevigen) with/without XAV939 (1–5 µM, final concentration). The reaction was incubated at 30 ˝C for 2.5 h, the beads were boiled after adding 5 µL 5ˆ SDS gel loading buffer (Bromophenol blue (0.25%), DTT (dithiothreitol; 0.5 M), Glycerol (50%), SDS (sodium dodecyl sulfate; 10%), Tris-HCl (0.25 M, pH 6.8) to the reaction. The samples were subjected to PAGE on 8% gel and blocked with 5% BSA, and incubated with streptavidin HRP (Pierce) at 1:20,000 to 5% BSA for overnight at 4 ˝C. The blot was washed with PBST and detected as described above.

4.7. Determination of TNKS Degradation For estimation of TNKS degradation, non-infected and HCMV-infected HFFs (48 hpi) were treated with cycloheximide (100 µg/mL) diluted in DMEM with 4% FBS and harvested at 3 h intervals for 3 h through 24 h for SDS-PAGE analysis as described above.

4.8. Proteasome Assay Proteasome activity was determined as reported [38]. Brieﬂy, non-infected and infected HFFs (MOI = 3, 72 hpi) were harvested and lysed in Proteasome Lysis/Assay Buffer containing 50 mM HEPES Viruses 2016, 8, 8 14 of 17

(pH 7.8), 10 mM NaCl, 1.5 mM MgCl2, 1 mM EDTA, 1 mM EGTA, 250 mM sucrose, and 5 mM DTT. 50 µL of cell lysate was incubated with 100 µM Suc-LLVY-AMC (chymotrypsin-like activity substrate) (Enzo Life Sciences, Farmingdale, NY, USA) diluted in 200 µL of proteasome assay buffer (lysis buffer supplemented with 2 mM ATP) and incubated at 37 ˝C for 60 min. The released ﬂuorogenic AMC was measured at 360 nm excitation and 460 nm emission using a ﬂuorometric plate reader (GloMax, Promega) and normalized to protein amount determined by standard BCA-based protein estimation assay (Pierce). MG132 (1 µM) was included in samples as a control for proteasome activity.

4.9. RNA Isolation and Real-Time Quantitative Reverse Transcriptase (qRT) PCR Total RNA was isolated from non-infected and HCMV-infected HFFs (with and without drug treatment) using RNeasy Mini kit (Qiagen, Georgetown, MD, USA) according to manufacturer’s instructions. RevertAid first strand cDNA synthesis kit (Life Technologies, Carlsbad, CA, USA) was used to synthesize first strand cDNA from 2 µg total RNA using oligo-dT primers. Negative reverse-transcriptase (-RT) reactions were included to ensure the specificity of qRT-PCR reactions. Synthesis of first strand cDNA from mRNA template was carried out at 42 ˝C for 1 h. cDNAs were diluted by 5 fold and quantitative RT-PCR (qRT-PCR) was performed using 2 µL of diluted cDNA, specific primers and SYBR green (Life Technologies) with two-step cycling protocol (95 ˝C for 15 s, 55 ˝C for 1 min). Reactions were performed in triplicates and GAPDH was used as internal control. mRNA levels in HCMV-infected cells were normalized to the mRNA produced in non-infected HFFs in addition to the internal normalization of each sample to GAPDH. Primers used to detect the mRNAs were: 51-GTCCCTGACAGCCTAGAAATAAG-31 and 51-GGGAACAGTAGCAGTTGAGTATG-31 for TNKS1, 51-TCCGGACAACAAGGTCTTAAC-31 and 51-CTCTTCCTCCACAGACTGAAAC-31 for TNKS2, 51-CTTCACCTGACAGATCCAAGTC-31 and 51-CCTTCCATCCCTTCCTGTTTAG-31 for β-catenin, 51-GAGGTATGTGCAGGAGGTTATG-31 and 51-TCCTCTGCGATCTTGTCTCT-31 for Axin1, 51-GCCGAGATCATCAGGAAGTATG-31 and 51-ATTCGCCTTCACGCTCTATC-31 for PARP1, 51-GTGGAGAAGGATGGTGAGAAAG-31 and 51-CTCAAGATTCCCACCCAGTTAC-31 for PARP2 and 51- TTGGTATCGTGGAAGGACTC-31 and 51-ACAGTCTTCTGGGTGGCAGT-31 for GAPDH.

4.10. Knockdown of TNKS ShRNA clones targeting TNKS2 (TRCN0000053239, TRCN0000053240, TRCN0000053241, TRCN0000053242, TRCN0000053238) and TNKS1 (TRCN0000040187 and TRCN0000040185) were obtained from Sigma. To package lentivirus, 21 µg of gag/pol, 7 µg of vesicular stomatitis virus glycoprotein, and 7 µg of shRNA plasmids were transfected into HEK293 cells using Lipofectamine 2000 (ThermoFisher Scientific). After 48 h, the packaged lentivirus particles were concentrated from the medium and the concentrated viruses were stored at ´80 ˝C. Lentivirus particles containing shRNA were transduced into HFFs. Then, 1.5 ˆ 106 cells were plated onto 75 cm2 flask and 15 mL of concentrated virus and polybrene (final concentration, 8 µg/mL) were added to the cells, and incubated for 24 h. Following transduction, puromycin (2 µg/mL) was added to select for stably transduced cells. TRC control HFFs and shTNKS1+2 double knockdown HFFs were counted and an equal number of cells was plated into each well prior to infection.

4.11. Statistical Analysis Statistical analyses were done using GraphPad Prism v6 (GraphPad Software, Inc. La Jolla, CA, USA) for Mac OSX. For the growth analysis, sample groups were compared to the control infected group using a one-way ANOVA. P-values were adjusted for multiple comparisons. For Western blot quantification, an unpaired two-tailed t-test with Welch’s correction was used. In all the figures, the following convention has been followed to indicate significance: * indicates p value < 0.05, ** indicates p < 0.01, and *** indicates p < 0.001. Viruses 2016, 8, 8 15 of 17

5. Conclusions Based on our data, we conclude that: (1) Decreased β-catenin expression in HCMV-infected cells is the final product of virus strategy to inhibit Wnt. (2) The inhibition of PARsylation activity is specific to TNKS and. (3) Chemical inactivation or knockdown of TNKS is beneficial for HCMV replication and, (4) regulation of TNKS may play an important role in HCMV replication and propagation.

Supplementary Materials: The following are available online at http://www.mdpi.com/1999-4915/8/1/8/s1, Figure S1: XAV939 does not induce PARP1 or PARP2 transcription, Figure S2: Growth of TB40 is enhanced in the presence of XAV939, Figure S3: Sensitivity to XAV939 is cell-type specific. Acknowledgments: We thank Ie-Ming Shih for providing the IST-1 cells. Author Contributions: Sujayita Roy and Ravit Arav-Boger conceived and designed the experiments; Sujayita Roy and Fengjie Liu performed the experiments; Sujayita Roy and Ravit Arav-Boger analyzed the data; Sujayita Roy and Ravit Arav-Boger wrote the paper. Conflicts of Interest: The authors declare no conflicts of interest.

References

1. Knipe, D.M.; Howley, P.M. Fields virology., 6th ed.; Wolters Kluwer/Lippincott Williams and Wilkins Health: Philadelphia, PA, USA, 2013. 2. Roy, S.; Arav-Boger, R. New cell-signaling pathways for controlling cytomegalovirus replication. Am. J. Transplant. 2014, 14, 1249–1258. [CrossRef][PubMed] 3. Hayward, S.D.; Liu, J.; Fujimuro, M. Notch and wnt signaling: Mimicry and manipulation by γ herpesviruses. Sci. STKE 2006, 2006.[CrossRef][PubMed] 4. Clevers, H.; Nusse, R. Wnt/β-catenin signaling and disease. Cell 2012, 149, 1192–1205. [CrossRef][PubMed] 5. Polakis, P. Wnt signaling and cancer. Genes Dev. 2000, 14, 1837–1851. [CrossRef][PubMed] 6. Behrens, J.; Jerchow, B.A.; Wurtele, M.; Grimm, J.; Asbrand, C.; Wirtz, R.; Kuhl, M.; Wedlich, D.; Birchmeier, W. Functional interaction of an axin homolog, conductin, with β-catenin, APC, and GSK3β. Science 1998, 280, 596–599. [CrossRef][PubMed] 7. Kishida, M.; Koyama, S.; Kishida, S.; Matsubara, K.; Nakashima, S.; Higano, K.; Takada, R.; Takada, S.; Kikuchi, A. Axin prevents Wnt-3a-induced accumulation of β-catenin. Oncogene 1999, 18, 979–985. [CrossRef] [PubMed] 8. Lee, E.; Salic, A.; Kruger, R.; Heinrich, R.; Kirschner, M.W. The roles of APC and axin derived from experimental and theoretical analysis of the Wnt pathway. PLoS Biol. 2003, 1, E10. [CrossRef][PubMed] 9. Salic, A.; Lee, E.; Mayer, L.; Kirschner, M.W. Control of β-catenin stability: Reconstitution of the cytoplasmic steps of the wnt pathway in xenopus egg extracts. Mol. Cell 2000, 5, 523–532. [CrossRef] 10. Clevers, H. Wnt/β-catenin signaling in development and disease. Cell 2006, 127, 469–480. [CrossRef] [PubMed] 11. MacDonald, B.T.; Tamai, K.; He, X. Wnt/β-catenin signaling: Components, mechanisms, and diseases. Dev. Cell 2009, 17, 9–26. [CrossRef][PubMed] 12. Huang, S.M.; Mishina, Y.M.; Liu, S.; Cheung, A.; Stegmeier, F.; Michaud, G.A.; Charlat, O.; Wiellette, E.; Zhang, Y.; Wiessner, S.; et al. Tankyrase inhibition stabilizes axin and antagonizes wnt signalling. Nature 2009, 461, 614–620. [CrossRef][PubMed] 13. Angelova, M.; Zwezdaryk, K.; Ferris, M.; Shan, B.; Morris, C.A.; Sullivan, D.E. Human cytomegalovirus infection dysregulates the canonical Wnt/β-catenin signaling pathway. PLoS Pathog. 2012, 8, e1002959. [CrossRef][PubMed] 14. Kapoor, A.; He, R.; Venkatadri, R.; Forman, M.; Arav-Boger, R. Wnt modulating agents inhibit human cytomegalovirus replication. Antimicrob. Agents Chemother. 2013, 57, 2761–2767. [CrossRef][PubMed] 15. Li, Z.; Yamauchi, Y.; Kamakura, M.; Murayama, T.; Goshima, F.; Kimura, H.; Nishiyama, Y. Herpes simplex virus requires poly(ADP-ribose) polymerase activity for efﬁcient replication and induces extracellular signal-related kinase-dependent phosphorylation and ICP0-dependent nuclear localization of tankyrase 1. J. Virol. 2012, 86, 492–503. [CrossRef][PubMed] Viruses 2016, 8, 8 16 of 17

16. Yeh, T.Y.; Meyer, T.N.; Schwesinger, C.; Tsun, Z.Y.; Lee, R.M.; Chi, N.W. Tankyrase recruitment to the lateral membrane in polarized epithelial cells: Regulation by cell-cell contact and protein poly(ADP-ribosyl)ation. Biochem. J. 2006, 399, 415–425. [CrossRef][PubMed] 17. Zhang, Y.; Liu, S.; Mickanin, C.; Feng, Y.; Charlat, O.; Michaud, G.A.; Schirle, M.; Shi, X.; Hild, M.; Bauer, A.; et al. RNF146 is a poly(ADP-ribose)-directed E3 ligase that regulates axin degradation and Wnt signalling. Nat. Cell Biol. 2011, 13, 623–629. [CrossRef][PubMed] 18. Bao, R.; Christova, T.; Song, S.; Angers, S.; Yan, X.; Attisano, L. Inhibition of tankyrases induces axin stabilization and blocks wnt signalling in breast cancer cells. PLoS ONE 2012, 7, e48670. [CrossRef][PubMed] 19. Morrone, S.; Cheng, Z.; Moon, R.T.; Cong, F.; Xu, W. Crystal structure of a tankyrase-axin complex and its implications for axin turnover and tankyrase substrate recruitment. Proc. Nat. Acad. Sci. USA 2012, 109, 1500–1505. [CrossRef][PubMed] 20. Lehtio, L.; Chi, N.W.; Krauss, S. Tankyrases as drug targets. FEBS J. 2013, 280, 3576–3593. [CrossRef] [PubMed] 21. Riffell, J.L.; Lord, C.J.; Ashworth, A. Tankyrase-targeted therapeutics: Expanding opportunities in the PARP family. Nat. Rev. Drug Discov. 2012, 11, 923–936. [CrossRef][PubMed] 22. De la Roche, M.; Ibrahim, A.E.; Mieszczanek, J.; Bienz, M. LEF1 and B9l shield β-catenin from inactivation by axin, desensitizing colorectal cancer cells to tankyrase inhibitors. Cancer Res. 2014, 74, 1495–1505. [CrossRef] [PubMed] 23. Salvant, B.S.; Fortunato, E.A.; Spector, D.H. Cell cycle dysregulation by human cytomegalovirus: Influence of the cell cycle phase at the time of infection and effects on cyclin transcription. J. Virol. 1998, 72, 3729–3741. [PubMed] 24. Casavant, N.C.; Luo, M.H.; Rosenke, K.; Winegardner, T.; Zurawska, A.; Fortunato, E.A. Potential role for p53 in the permissive life cycle of human cytomegalovirus. J. Virol. 2006, 80, 8390–8401. [CrossRef][PubMed] 25. Lee, H.R.; Kim, D.J.; Lee, J.M.; Choi, C.Y.; Ahn, B.Y.; Hayward, G.S.; Ahn, J.H. Ability of the human cytomegalovirus IE1 protein to modulate sumoylation of PML correlates with its functional activities in transcriptional regulation and infectivity in cultured fibroblast cells. J. Virol. 2004, 78, 6527–6542. [CrossRef] [PubMed] 26. Michaelis, M.; Doerr, H.W.; Cinatl, J. The story of human cytomegalovirus and cancer: Increasing evidence and open questions. Neoplasia 2009, 11, 1–9. [CrossRef][PubMed] 27. Martinez, F.P.; Tang, Q. Identification of cellular proteins that interact with human cytomegalovirus immediate-early protein 1 by protein array assay. Viruses 2014, 6, 89–105. [CrossRef][PubMed] 28. Tirosh, O.; Cohen, Y.; Shitrit, A.; Shani, O.; Le-Trilling, V.T.; Trilling, M.; Friedlander, G.; Tanenbaum, M.; Stern-Ginossar, N. The transcription and translation landscapes during human cytomegalovirus infection reveal novel host-pathogen interactions. PLoS Pathog. 2015, 11, e1005288. [CrossRef][PubMed] 29. Vogel, C.; Marcotte, E.M. Insights into the regulation of protein abundance from proteomic and transcriptomic analyses. Nat. Rev. Genet. 2012, 13, 227–232. [CrossRef][PubMed] 30. Fujimuro, M.; Hayward, S.D. The latency-associated nuclear antigen of Kaposi’s sarcoma-associated herpesvirus manipulates the activity of glycogen synthase kinase-3β. J. Virol. 2003, 77, 8019–8030. [CrossRef] [PubMed] 31. Callow, M.G.; Tran, H.; Phu, L.; Lau, T.; Lee, J.; Sandoval, W.N.; Liu, P.S.; Bheddah, S.; Tao, J.; Lill, J.R.; et al. Ubiquitin ligase RNF146 regulates tankyrase and axin to promote Wnt signaling. PLoS ONE 2011, 6, e22595. [CrossRef][PubMed] 32. Aravind, L. The WWE domain: A common interaction module in protein ubiquitination and ADP ribosylation. Trends Biochem. Sci. 2001, 26, 273–275. [CrossRef] 33. Child, S.J.; Jarrahian, S.; Harper, V.M.; Geballe, A.P. Complementation of vaccinia virus lacking the double-stranded RNA-binding protein gene E3L by human cytomegalovirus. J. Virol. 2002, 76, 4912–4918. [CrossRef][PubMed] 34. He, R.; Sandford, G.; Hayward, G.S.; Burns, W.H.; Posner, G.H.; Forman, M.; Arav-Boger, R. Recombinant luciferase-expressing human cytomegalovirus (CMV) for evaluation of CMV inhibitors. Virol. J. 2011, 8. [CrossRef][PubMed] 35. Shih, I.; Wang, T.; Wu, T.; Kurman, R.J.; Gearhart, J.D. Expression of Mel-CAM in implantation site intermediate trophoblastic cell line, IST-1, limits its migration on uterine smooth muscle cells. J. Cell Sci. 1998, 111, 2655–2664. [PubMed] Viruses 2016, 8, 8 17 of 17

36. E, X.; Pickering, M.T.; Debatis, M.; Castillo, J.; Lagadinos, A.; Wang, S.; Lu, S.; Kowalik, T.F. An E2F1-mediated DNA damage response contributes to the replication of human cytomegalovirus. PLoS Pathog. 2011, 7, e1001342. [CrossRef][PubMed] 37. Smith, S.; Giriat, I.; Schmitt, A.; de Lange, T. Tankyrase, a poly(ADP-ribose) polymerase at human telomeres. Science 1998, 282, 1484–1487. [CrossRef][PubMed] 38. Valera, E.; Dargusch, R.; Maher, P.A.; Schubert, D. Modulation of 5-lipoxygenase in proteotoxicity and Alzheimer’s disease. J. Neurosci. 2013, 33, 10512–10525. [CrossRef][PubMed]

© 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).