The human cytomegalovirus lytic cycle is induced by 1,25-dihydroxyvitamin D3 in peripheral blood monocytes and in the THP-1monocytic cell line

Shu-En Wu, William E. Miller n

Department of Molecular Genetics, Biochemistry, and Microbiology, University of Cincinnati College of Medicine, 231 Albert Sabin Way, Cincinnati, OH45267-0524, United States

a r t i c l e i n f o

Article history:Received 17 December 2014Returned to author for revisions12 January 2015Accepted 2 April 2015

Keywords:HCMVCytomegalovirusLatencyLytic replication1,25-dihydroxyvitamin D3Myeloid progenitorsMonocytesMacrophages

a b s t r a c t

Human cytomegalovirus (HCMV) resides in a latent form in hematopoietic progenitors and undiffer-entiated cells within the myeloid lineage. Maturation and differentiation along the myeloid lineagetriggers lytic replication. Here, we used peripheral blood monocytes and the monocytic cell line THP-1 toinvestigate the effects of 1,25-dihydroxyvitamin D3 on HCMV replication. Interestingly, 1,25-dihydrox-yvitamin D3 induces lytic replication marked by upregulation of HCMV gene expression and productionof infectious virus. Moreover, we demonstrate that the effects of 1,25-dihydroxyvitamin D3 correlatewith maturation/differentiation of the monocytes and not by directly stimulating the MIEP. These resultsare somewhat surprising as 1,25-dihydroxyvitamin D3 typically boosts immunity to bacteria and virusesrather than driving the infectious life cycle as it does for HCMV. Defining the signaling pathways kindledby 1,25-dihydroxyvitamin D3 will lead to a better understanding of the underlying molecularmechanisms that determine the fate of HCMV once it infects cells in the myeloid lineage.

& 2015 Elsevier Inc. All rights reserved.

Introduction

Human cytomegalovirus (HCMV) is a β-herpesvirus thatspreads broadly throughout the human population (Sinzgeret al., 2008). In general, about 50–70% of people are serologicallypositive for HCMV worldwide (Bate et al., 2010). Although inimmunocompetent individuals HCMV infection is typically asymp-tomatic, in the case of congenital infection, the virus can causesevere neurological sequelae such as deafness and developmentaldefects following infection of the fetus (Grosse et al., 2008;Johnson and Anderson, 2014). In immunocompromised indivi-duals, including those with HIV/AIDS or those receiving organtransplants, HCMV can cause devastating morbidity and mortalityincluding pneumonia, retinitis, and transplant rejection (Ljungmanet al., 2011; Paya et al., 2004; Yen et al., 2015). Moreover, manystudies have shown that HCMV can be associated with chronicdiseases such as atherosclerosis and hypertension, cancer, andautoimmune disease (Dziurzynski et al., 2012; Li et al., 2011;Streblow et al., 2001; Varani et al., 2009). Therefore, understand-ing the biology of HCMV infection is both clinically relevant andintensively studied with regards to potential pharmacological

intervention. Like other herpesviruses, HCMV can establish latencyin the human body, thus making the eradication of the virus frominfected individual a difficult task (Grinde, 2013; White et al.,2012). The cellular reservoirs for HCMV latency includehematopoietic stem cells, common myeloid progenitor cells, andmonocytes (Bego and St Jeor, 2006; Goodrum et al., 2002; Sinclair,2008; Taylor-Wiedeman et al., 1991). Deciphering the mechanismsthat regulate the latent/lytic switch in HCMV infected cells couldlead to the identification of novel therapeutics that could be usedto regulate latency. Previous studies have indicated that both viraland cellular factors are involved in the control of latent and lyticcycles in myeloid progenitors and monocytes (Chan et al., 2012;Goodrum et al., 2007; Kew et al., 2014; Keyes et al., 2012b;O'Connor et al., 2014; Smith et al., 2004; Stevenson et al., 2014);however, the molecular mechanisms remain unresolved, and it ishighly probable that there are numerous cellular and viral reg-ulatory factors that have yet to be identified. In light of this, furtherinvestigation of the mechanisms and factors that influence theswitch between HCMV latency and lytic replication in clinicallyrelevant myeloid cell types is needed.

It is known that the developmental maturation of monocytes intomacrophages and dendritic cells can reactivate HCMV from latencyleading to the production of new infectious virus (Chan et al., 2008;Reeves and Sinclair, 2013; Smith et al., 2004; Soderberg-Naucler et al.,1997, 2001; Stevenson et al., 2014). In addition, there are various

Contents lists available at ScienceDirect

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

Virology

http://dx.doi.org/10.1016/j.virol.2015.04.0040042-6822/& 2015 Elsevier Inc. All rights reserved.

n Corresponding author. Fax: þ1 513 558 8474.E-mail address: william.miller@uc.edu (W.E. Miller).

Virology 483 (2015) 83–95

extracellular stimuli (i.e. PMA) that can trigger monocyte to macro-phage differentiation (Greenberger et al., 1980; Hemmi and Breitman,1985; Huber et al., 2014; Naito, 2008; Nakamura et al., 1986; Netea etal., 2008) and some of these stimuli have also been shown to directlyinduce the HCMV immediate-early gene promoter, which is essentialfor induction of the HCMV lytic cycle (Ghazal et al., 1992; Kline et al.,1998; Stein et al., 1993). Since the activities of these stimuli appear tobe multi-factorial, it is difficult to determine if the major influence ofthese stimuli on lytic replication is induction of the IE promoter,promotion of cellular maturation/differentiation or a combination ofboth activities. THP-1 cells are a monocytic cell line that is commonlyused in combination with primary blood derived monocytes to studythe interaction between HCMV and myeloid cells and gain insight intothe latent/lytic switch (Saffert et al., 2010; Van Damme et al., 2014). Itis well known that HCMV enters latency or a quiescent state inundifferentiated THP-1, and the virus typically enters into the lyticcycle after it infects phorbol 12-myristate 13-acetate (PMA) treatedTHP-1 cells (Qin et al., 2013; Weinshenker et al., 1988). As aconsequence, PMA is a reagent of choice used to promote myeloiddifferentiation in studies aimed at inducing lytic replication in in vitrosystems. However, PMA is a synthetic compound resembling diacyl-glycerol (DAG) that is capable of activating a broad range of cellsignaling pathways (Castagna et al., 1982; Niedel et al., 1983; Swindle,Hunt and Coleman, 2002). In this research we sought to identifyadditional physiologically relevant compounds that could trigger bothmonocyte differentiation and HCMV lytic infection. Vitamin D3 is ahormone that is produced by the human body and acquired in asupplemental fashion through diet (Baeke et al., 2010; Holick, 2003;Lamberg-Allardt, 2006). The most well-known effects of vitamin D3and its active metabolite 1,25-dihydroxyvitamin D3 are to regulatehomeostasis of calcium and phosphorus and promote bone develop-ment through interaction with the vitamin D receptor (VDR), amember of the nuclear receptor family of transcription factors(Goltzman et al., 2014; Kannan and Lim, 2014). Interestingly, bloodleukocytes robustly express the VDR and results of studies performedin vitro in human myeloid cell lines and ex vivo in murine bonemarrow cells have demonstrated that 1,25-dihydroxyvitamin D3 hasthe ability to induce monocyte–macrophage differentiation (Gemelliet al., 2008; Hmama et al., 1999; Lagishetty et al., 2011; Liu et al., 2006;O'Kelly et al., 2002, Bhalla, 1983 #83; Provvedini et al., 1983). It istherefore not surprising that 1,25-dihydroxyvitamin D3 has beendemonstrated to exhibit antibacterial and antiviral effects (Korf et al.,2014; Luong and Nguyen, 2011; Maxwell et al., 2012; Spector, 2011).The importance of 1,25-dihydroxyvitamin D3 in regulation of immunesystem function has been further highlighted by studies which suggestthat 1,25-dihydroxyvitamin D3 or synthetic analogs of 1,25-dihydrox-yvitamin D3 could be used as potent candidates for the treatment forautoimmune diseases, infectious diseases and anticancer therapies(Salomon et al., 2014; Yuzefpolskiy et al., 2014; Zhang et al., 2013).Nonetheless, the effect of 1,25-dihydroxyvitamin D3 on HCMV repli-cation in monocytes and macrophages remains unknown. Therefore,we explored the possibility that peripheral blood monocytes and THP-1 cells could be used to determine the effect of 1,25-dihydroxyvitaminD3 on HCMV replication in myeloid cells. According to the results ofprevious studies, 1,25-dihydroxyvitamin D3 treatment induces THP-1cells to differentiate into mature monocytes, with high CD14 expres-sion (Daigneault et al., 2010; Hmama et al., 1999; Schwende et al.,1996) and therefore we also hypothesized that we also could use thismodel to study HCMV replication in 1,25-dihydroxyvitamin D3 treatedcells that are in the transition from the promonocytic to macrophagestages.

Interestingly, we found that the HCMV lytic phase can beinduced in 1,25-dihydroxyvitamin D3 treated primary monocytesand in THP-1 cells with infectious virus being produced by thesecells. In contrast to PMA treated cells, 1,25-dihydroxyvitamin D3does not have a direct effect on the HCMV immediate-early gene

promoter in reporter gene assays suggesting that the predominanteffect of 1,25-dihydroxyvitamin D3 is to drive differentiation andnot necessarily to directly stimulate IE promoter activity. When1,25-dihydroxyvitamin D3 is combined with PMA to differentiateTHP-1 cells, no additive effect on HCMV replication is observed.These results demonstrate that 1,25-dihydroxyvitamin D3 inducesa set of differentiation related signaling pathways that creates afavorable cellular milieu for HCMV lytic infection. Moreover, ourresults suggest that clinical/dietary supplementation with vitaminD3 could be problematic in patients susceptible to reactivation-based HCMV disease.

Results

1,25-dihydroxyvitamin D3 promotes HCMV replication in primarymonocytes and THP-1 cells

Vitamin D3 is a natural hormone that is produced by humanbody and typically supplemented through diet (Baeke et al., 2010;Holick, 2003; Lamberg-Allardt, 2006). In addition to the regulationof the homeostasis of calcium and phosphorus (Garabedian andUlmann, 1979; Goltzman et al., 2014), vitamin D3 has been shownto play multiple roles in immune responses including modulatingT cell and B cell activity (Terrier et al., 2012), promoting monocyte–macrophage differentiation (Pan et al., 1997; Takahashi et al.,1997), stimulating the anti-bacterial and anti-viral effects ofmacrophages (Campbell and Spector, 2012; Verway et al., 2013),and driving lineage commitment of hematopoietic progenitor cells(Bunce et al., 1997). Vitamin D3 like other endocrine hormones iscarried by the circulatory system to various tissues (Baeke et al.,2010), and research has also shown that many cells within theimmune system have the enzyme that can convert Vitamin D3 intoits active form, 1,25-dihydroxyvitamin D3 (Ooi et al., 2014;Shahijanian et al., 2014; Stoffels et al., 2006). Therefore, vitaminD3 can execute its effects on a wide variety of cells in either anendocrine or paracrine fashion (Hewison, 2012). HCMV is a β-herpesvirus which can infect myeloid cells and establish latentand/or lytic infections within cells of this lineage (Sinclair, 2010).Although studies have shown that allogeneic stimulation orstimulation with cytokines like TNF-α can stimulate HCMV IEpromoter activity and drive lytic replication in myeloid cells(Prosch et al., 1995; Soderberg-Naucler et al., 1997; Stein et al.,1993), the effect of 1,25-dihydroxyvitamin D3 on HCMV replica-tion in myeloid cells remains unexplored. To determine if 1,25-dihydroxyvitamin D3 may influence HCMV lytic replication inmyeloid cells, we examined the effect of 1,25-dihydroxyvitamin D3on the ability of CD14 positive peripheral blood monocytes tosupport lytic replication. Peripheral blood mononuclear cells(PBMCs) from healthy anonymous donors were isolated byFicoll-paque and CD14 positive monocytes were subsequentlyisolated from PBMCs using CD14 magnetic beads (Miltenyi).Monocytes were treated with 1,25-dihydroxyvitamin D3(100 nM) for 2 days and infected with HCMV TB40E at an MOI of10. On days 4 and 6 post-infection, cells were harvested andsubsequently co-cultured with human foreskin fibroblasts (infec-tious center assays) to determine whether the infected monocyteswere producing infectious virus. Interestingly, monocytes treatedwith 1,25-dihydroxyvitamin D3 exhibited a 5–8 fold increase ininfectious centers over cells treated with the vehicle controlethanol (Fig. 1A). These data indicate that 1,25-dihydroxyvitaminD3 treatment can create a milieu in blood monocytes that moreefficiently supports HCMV virus production. HCMV replication inprimary monocytes treated with the phorbol ester phorbal 12-myristate 13-acetate (PMA) was similarly examined (Fig. 1B). PMAis a well-established inducer of HCMV replication in a number of

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–9584

systems (Qin et al., 2013; Weinshenker et al., 1988). PMA treatedcells exhibited a 15–30 fold increase in infectious centers over cellstreated with the vehicle control DMSO (Fig. 1B). Thus, for compar-ison, while it is readily apparent that 1,25-dihydroxyvitamin D3 isa robust inducer of HCMV replication in monocytes, the level ofreplication achieved is not as strong as that achieved by thephorbol ester, consistent with what might be expected for anatural compound like 1,25-dihydroxyvitamin D3.

We then sought a model system that could be utilized toprovide a more mechanistic explanation for this finding. THP-1, anestablished monocytic cell line and model system frequently usedin HCMV studies (Keyes et al., 2012a; Saffert et al., 2010), was thendeployed to further explore the effects of 1,25-dihydroxyvitaminD3 on HCMV lytic replication. Since it is well established that PMAcan drive HCMV production in THP-1 cells, we again used PMA as acontrol in these experiments (Weinshenker et al., 1988). THP-1cells were treated with vehicle (ethanol), 1,25-dihydroxyvitaminD3 (100 nM) or PMA (80 nM) for 3 days before infection. Cellswere then infected with HCMV TB40E at an MOI of 10. On day6 post-infection, THP-1 cells were co-cultured with HFFs ininfectious center assays. In THP-1 cells treated with the ethanolcontrol, only a very low number of plaques were detected ininfectious center assays, which indicate that THP-1 cells rarelysupport lytic phase replication after HCMV infection (Fig. 2). InPMA treated THP-1 cells, there was a 40-fold increase in thenumber of plaques arising in infectious center assays supportingearly studies that reported the induction of lytic phase replicationby PMA treatment. Importantly, 1,25-dihydroxyvitamin D3 treat-ment of THP-1 cells resulted in a 10-fold increase in the number ofplaques arising in infectious center assays. We repeated thisexperiment and examined infectious center production at days6 and 8 post-infection and obtained similar results, indicating thatthe difference in lytic replication observed between 1,25-dihy-droxyvitamin D3 and PMA treated cells is not simply the result of adelay in virus replication in the 1,25-dihydroxyvitamin D3 treatedcells (data not shown). Therefore, since 1,25-dihydroxyvitamin D3can promote lytic virus production in both peripheral bloodmonocytes and in the monocytic cell line THP-1, we conclude thatTHP-1 cells would provide a viable model to recapitulate andfurther explore the effects of 1,25-dihydroxyvitamin D3 on HCMVreplication in myeloid cells.

1,25-dihydroxyvitamin D3 treatment does not influence the ability ofHCMV to establish an initial infection in monocytes

Since we found that 1,25-dihydroxyvitamin D3 treatment candramatically increase the number of plaques that arise from THP-1cells in co-culture infectious center assays, we wanted to deter-mine whether this difference could be the result of increased

infectivity or entry of HCMV virions into 1,25-dihydroxyvitaminD3 treated cells. We used semi-quantitative PCR to examine viralgenome copy number in cells at various time points post-infection(Fig. 3). If 1,25-dihydroxyvitamin D3 leads to increased infectivityof the monocytes, we would expect to see increased viral DNAlevels at early time points post-infection. However, at day 1 post-infection, cells treated with 1,25-dihydroxyvitamin D3 or PMAexhibited similar viral copy numbers to that of vehicle controlcells. The results are depicted in Fig. 3A and quantitative resultsfrom six independent experiments are shown in Fig. 3B. This resultsuggested that HCMV infects control and 1,25-dihydroxyvitaminD3 treated cells with equivalent efficiency. At day 6 post-infection,the PCR signal for viral genomes in control THP-1 cells declinedwhile the signal from 1,25-dihydroxyvitamin D3- and PMA-treatedcells was maintained or increased, consistent with the conclusionthat 1,25-dihydroxyvitamin D3 and PMA treated cells are support-ing lytic HCMV replication.

1,25-dihydroxyvitamin D3 treated THP-1 cells are more likely toexhibit IE gene expression following infection

While an equivalent amount of HCMV DNA is initially presentfollowing infection of control or 1,25-dihydroxyvitamin D3 treatedcells, it is clear that the 1,25-dihydroxyvitamin D3 treated cells supporta robust increase in productive HCMV replication. Therefore, we nextchose to examine HCMV gene expression profiles in 1,25-dihydrox-yvitamin D3 and PMA treated cells. Immediate early (IE) gene expres-sion is typically repressed in cells that fail to undergo lytic phaseinduction, but is expressed rapidly after infection in cells capable ofsupporting lytic replication (Keyes et al., 2012a; Meier, 2001; Turtinen

Fig. 1. 1,25-dihydroxyvitamin D3 promotes HCMV replication in primary peripheral blood derived monocytes. Monocytes were treated with 100 nM 1,25-dihydroxyvitaminD3 (A) or 80 nM PMA (B) for 2 days and then infected with HCMV TB40E at a MOI of 10. On 4 days and 6 days post- infection, cells were harvested and co-cultured with HFFfibroblasts for 2 days. After 2 days of co-culture, fibroblast monolayers were overlayed with CMC/MEM and incubated for 8 days to allow for plaque development. The datarepresent 4–8 independent experiments performed in duplicate. VitD3, 1,25-dihydoxyvitamin D3. npo0.05, nnpo0.01.

Fig. 2. 1,25-dihydroxyvitamin D3 promotes HCMV replication in the THP-1 mono-cytic cell line. THP-1 monocytes were treated with 1,25-dihydroxyvitamin D3(100 nM) or PMA (80 nM) for 3 days and then infected with HCMV TB40E at a MOIof 10. On day 6 post-infection, cells were harvested and co-cultured with HFFfibroblasts for 2 days. After 2 days of co-culture, fibroblast monolayers wereoverlayed with CMC/MEM and incubated for 8 days to allow for plaque develop-ment. The data represent five independent experiments performed in duplicate.VitD3, 1,25-dihydoxyvitamin D3. nnpo0.01, nnnpo0.001.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–95 85

and Seufzer, 1994). We examined IE gene expression in cells treatedwith vehicle control, 1,25-dihydroxyvitamin D3 or PMA (Fig. 4). Flowcytometric staining with anti-IE-Alex488 antibodies was performed asthis enabled us to assess not only the frequency with which IE positivecells arise but also the relative level of IE antigens per cell. Infected cellswere harvested and examined at day 1 post-infection. Compared tovehicle control, 1,25-dihydroxyvitamin D3 treatment resulted in asignificantly higher percentage of IE positive cells by day 1 post-infection (Fig. 4A and B). Interestingly, while the number of IE positivecells is significantly increased with 1,25-dihydroxyvitamin D3 treat-ment, there is no difference in the relative IE expression per cell as themean fluorescent intensities are similar when comparing vehicle and1,25-dihydroxyvitamin D3 treated cells (Fig. 4C). PMA, in contrast,caused an increase in both the percentage of IE positive cells andrelative IE expression per cell, suggesting that the mechanisms utilizedby 1,25-dihydroxyvitamin D3 and PMA to promote lytic replicationmay be distinct. We did not detect any differences in the subcellularlocalization of IE1/2 when comparing control, 1,25-dihydroxyvitaminD3, and PMA treated cells indicating that changes in the compartmen-talization of IE proteins is unlikely to account for the mechanism of1,25-dihydroxyvitamin D3 induced HCMV replication (data notshown).

Taken together, while viral genome copy numbers are initiallyequivalent, the 1,25-dihydroxyvitamin D3 treated cells are more highlylikely to be capable of initiating IE protein production consistent withtheir ability to progress to the lytic phase (Figs. 2 and 4). Moreover,while 20–40% of THP-1 cells treated with 1,25-dihydroxyvitamin D3 orPMA express IE antigens, it is evident that not all cells that progressthrough the IE phase go on to produce infectious virus based oninfectious center assays, indicating that there are additional blockssubsequent to IE expression that control the progression to the lyticphase in HCMV infected myeloid cells.

HCMV early and late genes are expressed in 1,25-dihydroxyvitaminD3 stimulated cells

Although IE expression is important for initiation of lyticinfection, the expression of early and late genes are needed to

complete the lytic phase (McDonough and Spector, 1983; Wathenand Stinski, 1982). Therefore, the expression of early and lateHCMV genes was examined by western blot in vehicle control,1,25-dihydroxyvitamin D3 and PMA treated cells (Fig. 5A and B).For these experiments, we analyzed UL44, a processivity factorassociated with the viral DNA polymerase (Sinigalia et al., 2008),which is expressed with delayed-early kinetics (Hwang et al.,2000) and pp65, a tegument protein that is expressed with latekinetics (Kalejta, 2008). Vehicle control infected cells exhibitedundetectable levels of either the delayed early UL44 or late pp65proteins while cells treated with 1,25-dihydroxyvitamin D3 orPMA prior to infection showed dramatic upregulation of bothUL44 and pp65. Representative western blots are depicted inFig. 5A and the results are depicted graphically in Fig. 5B in whichthe blots were quantitated and viral protein levels are shownrelative to the cellular actin protein as a control. To furtherinvestigate viral gene expression patterns in these cells anddetermine what percentage of IE positive cells progress to theearly phase as evidenced by UL44 expression, cells at days 1 and 4post-infection were co-stained for IE1/2 and UL44 expression andanalyzed by flow cytometry (Fig. 5C). In vehicle treated cultures,about 7.5% of the cells were IE positive at day 1 post-infection,while the percentage of IE positive cells dropped to 4.4% at day 4post-infection. Of the 4.4% IE positive cells at day 4 post-infectiononly 11% of cells were UL44 positive (0.5% of the total cellularpopulation). In 1,25-dihydroxyvitamin D3 treated cultures, appro-ximately 33% of the cells were IE positive at day 1 post-infection,

Fig. 3. 1,25-dihydroxyvitamin D3 treatment does not influence the ability of HCMVto establish an initial infection. (A) THP-1 monocytes were treated 1,25-dihydrox-yvitamin D3 (100 nM) or PMA (80 nM) for 3 days and then infected with HCMVTB40E at a MOI of 10. On days 1 and 6 post-infection, DNA from THP-1 cellssubjected to the indicated treatments was amplified by PCR for HCMV genomes (IEregion) or cellular genomes (GAPDH). PCR products were visualized by agarose gelelectrophoresis. (B) The PCR signals of viral DNA were normalized to the signals ofGAPDH. Data shown are the means7SEM of six independent experiments. VitD3,1,25-dihydroxyvitamin D3. n.s. non-significant, npo0.05, nnpo0.01, d.p.i. day post-infection.

Fig. 4. 1,25-dihydroxyvitamin D3 treatment increases the percentage of cellssupporting HCMV IE gene expression. (A) THP-1 monocytes were treated with1,25-dihydroxyvitamin D3 (100 nM) or PMA (80 nM) for 3 days and then infectedwith HCMV TB40E at a MOI of 10. At 1 day post-infection, cells were fixed,permeabilized and stained with anti-HCMV IE antibody mAB 810-Alexa488. Cellswere analyzed by flow cytometry. (B) The percentage of IE positive cells at day1 post-infection is presented graphically. The data are derived from four indepen-dent experiments including the one depicted in panel A. (C) The mean fluorescenceintensity of IE positive cells at day 1 post-infection is presented graphically. Thedata are derived from four independent experiments including the one depicted inpanel A. VitD3, 1,25-dihydroxyvitamin D3. nnpo0.01, nnnpo0.001.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–9586

while the percentage of IE positive cells declined to 19% at day 4post-infection. However, of the 19% IE positive cells, 33% wereUL44 also positive (6.5% of the total cellular population). In PMAtreated cultures, 40% of the cells were IE positive at day 1 post-infection and the percentage decreased slightly to 38% at day 4post-infection. Of the 38% IE positive cells, 48% were also UL44positive (18% of the total cellular population). The results of theseearly and late gene expression profiling experiments are also inline with the results of infectious center assay and are allconsistent with the conclusion that 1,25-dihydroxyvitamin D3promotes HCMV lytic replication in myeloid cells.

1,25-dihydroxyvitamin D3 uses a mechanism distinct from that ofPMA to promote lytic replication

Due to the fact that both 1,25-dihydroxyvitamin D3 and PMAcan prime THP-1 cells to support lytic infection, it would beintriguing to determine if these two reagents deploy the samemechanism or if 1,25-dihydroxyvitamin D3 functions in a mannerdistinct from that of PMA. Moreover, since it is clear that IE proteinexpression is critical for the onset of lytic replication, we wished toinvestigate the effects of 1,25-dihydroxyvitamin D3 and PMA on IEgene expression in a more detailed manner. Based on publishedreports (Abraham and Kulesza, 2013), it has been demonstratedthat the HCMV IE enhancer region in THP-1 cells after infection ismarked by histone 3 lysine 27 trimethylation (H3K27me3), andthat the H3K27me3 mark at the IE enhancer is significantlydecreased after PMA treatment. H3K27me3 is associated with a

closed chromatin conformation and silenced gene expression (Fuet al., 2014), therefore it appears that decreased H3K27me3 in theIE region correlates with an open chromatin conformation andincreased MIEP activity. We wanted to determine whetherH3K27me3 associated with the IE enhancer is also decreased in1,25-dihydroxyvitamin D3 treated cells. Using chromatin immu-noprecipitation (CHIP) followed by PCR for the IE enhancer region,we find that the IE enhancer region in control cells is in factmodified by H3K27me3 as reported by others (Fig. 6A) (Abrahamand Kulesza, 2013; Rossetto et al., 2013). However, in both 1,25-dihydroxyvitamin D3 and PMA treated cells the CHIP-PCR signal is4 to 10 fold weaker indicative of decreased H3K27me3 at the IEenhancer. Thus, these results are consistent with our analyses ofIE1 protein expression and suggest that the transition of the MIEPenhancer region into an open conformation in 1,25-dihydroxyvi-tamin D3 treated cells is an important prerequisite for thetransition to the lytic phase.

The MIEP contains binding sites for a number of transcriptionfactors that are responsive to PMA such as NF-κB and CREB (Liu etal., 2010), but it is unknown if the MIEP would be directlyresponsive to 1,25-dihydroxyvitamin D3. We cloned the HMCV IEpromoter from the HCMV-FIX strain into the luciferase reporterpGL3 so that we could test whether 1,25-dihydroxyvitamin D3,like PMA would be able to directly stimulate the MIEP-luciferasereporter gene. We transfected THP-1 cells with pGL3-MIEP, sti-mulated cells with either 1,25-dihydroxyvitamin D3 or PMA, andmeasured luciferase activity. Reporter luciferase was internallycontrolled by comparison to constitutively expressed Rennila-

Fig. 5. 1,25-dihydroxyvitamin D3 promotes HCMV early and late gene expression. (A) THP-1 monocytes were treated with 1,25-dihydroxyvitamin D3 (100 nM) or PMA(80 nM) for 3 days and then infected with HCMV TB40E at a MOI of 10. At the indicated times post-infection, cells extracts were analyzed by western blot for UL44 (early) andpp65 (late) gene expression. Western blot analyses demonstrated that UL44 and pp65 are robustly expressed in 1,25-dihydroxyvitamin D3 and PMA treated cells. Cellextracts were also analyzed for actin expression as an internal control. (B) The results of four independent experiments are shown graphically as the ratio of UL44 or pp65 toactin. (C) THP-1 cells treated as described in panel A were stained for IE and UL44 proteins at days 1 and 4 post-infection and analyzed by flow cytometry. IEþ cells weregated (left panel in each pair) and then plotted as a function of UL44 expression (right panel in each pair) to determine the fraction of IEþ cells that have entered into theearly phase as assessed by UL44 expression. The results shown are representative of 4 independent experiments. VitD3, 1,25-dihydroxyvitamin D3, npo0.05, nnpo0.01.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–95 87

luciferase. Interestingly, the Luciferase assay results demonstratethat while PMA can directly stimulate MIEP promoter activity,1,25-dihydroxyvitamin D3 cannot (Fig. 6B). These data are con-sistent with our flow cytometry data in Fig. 4C, which indicatedthat 1,25-dihydroxyvitamin D3 treated cells are more likely tosupport IE protein expression but do not exhibit increased levels of

the IE protein on a per cell basis (Fig. 4C). Thus, taken together,while both 1,25-dihydroxyvitamin D3 and PMA can promote anopen conformation of the MIEP followed by IE protein productionand onset of the lytic phase, the mechanisms used by the twoinducers are not identical as the effects of PMA can at least bepartially explained by a direct effect on the MIEP promoter.

Since the two inducers appeared to not utilize totally over-lapping mechanisms we investigated whether the effects of 1,25-dihydroxyvitamin D3 would be additive regarding the onset oflytic phase. To test this hypothesis, THP-1 cells were treated withPMA or PMAþ1,25-dihydroxyvitamin D3 prior to infection. Cul-ture media was harvested at multiple time points post-infectionand viral titers were assessed by plaque assay. Compared to PMAtreated cells, 1,25-dihydroxyvitamin D3 plus PMA did not show anadditive effect as the kinetics and magnitude of the viral growth isthe same (Fig. 7). Thus, while the mechanisms are not totallyoverlapping, the convergence of the two compounds on creatingan environment more suitable for IE protein expression seems tobe a necessary prerequisite leading to the onset of the lytic virallifecycle.

The differentiation of THP-1 cells triggered by 1,25-dihydroxyvitaminD3 plays an important role in releasing the restriction on HCMV IEexpression

Based on published studies it is clear that 1,25-dihydroxyvita-min D3, like PMA, is an inducer of monocyte differentiation andmaturation (Greenberger et al., 1980; Hmama et al., 1999;Schwende et al., 1996). However, while PMA induces the THP-1cells to mature into a more highly differentiated macrophage-likestate marked by strong adherence of the cells to plastic, 1,25-dihydroxyvitamin D3 induces differentiation into a mature mono-cyte marked by a non-adherent phenotype (Schwende et al., 1996).Therefore, we wished to investigate the influence of monocytedifferentiation properties on 1,25-dihydroxyvitamin D3 inducedHCMV replication in the THP-1 model. Consistent with previousstudies (Daigneault et al., 2010; Schwende et al., 1996), weobserved that PMA but not 1,25-dihydroxyvitamin D3 inducedthe appearance of a macrophage phenotype with flattened cellstightly adhered to the flask (Fig. 8A). Immunostaining for the celldifferentiation markers CD14, CD11b, CD36, CD54, and CD68 wasalso performed and the results demonstrate that 1,25-dihydrox-yvitamin D3 consistently induces strong CD14 and moderateCD11b expression while having little to no effect on CD36, CD54,and CD68 expression (Fig. 8B and Supplementary Fig. 1). Incontrast, PMA induced moderate levels of CD14, while inducingrobust levels of CD11b, CD36, CD54, and CD68 (Fig. 8B andSupplementary Fig. 1). These results are entirely consistent withthose of published studies and are further supportive of theconclusion that 1,25-dihydroxyvitamin D3 induces an intermedi-ate differentiation phenotype typical of a mature monocyte, whilePMA causes a terminally differentiated phenotype typical of amacrophage (Kremlev and Phelps, 1997; Kunisch et al., 2004;Munoz-Pacheco et al., 2012; Schwende et al., 1996). Interestingly,while PMA induces terminal differentiation and halts cellularproliferation, 1,25-dihydroxyvitamin D3 treated cells continuedto proliferate in agreement with previous studies (data not shown)(Schwende et al., 1996).

We then asked whether there was a correlation between 1,25-dihydroxyvitamin D3 promoted differentiation and onset of lyticreplication. Since 1,25-dihydroxyvitamin D3 appeared to notdirectly stimulate MIEP promoter activity, we hypothesized thatthe effects of 1,25-dihydroxyvitamin D3 on IE expression might bedelayed and only occur as the cells differentiated into maturemonocytes. To test this postulate, THP-1 cells were treated with1,25-dihydroxyvitamin D3 for different times (6 h to 3 days) prior

Fig. 6. 1,25-dihydroxyvitamin D3 promotes K27 demethylation of histone H3associated with the HCMV IE enhancer region, but does not directly stimulate IEpromoter activity. (A) THP-1 monocytes were treated with 1,25-dihydroxyvitaminD3 (100 nM) or PMA (80 nM) for 3 days and then infected with HCMV TB40E at aMOI of 10. At 2 days post-infection, chromatin immunoprecipitation was used toexamine the K27 methylation status of histone H3 associated with the HCMV IEenhancer region. The results presented are derived from 3 independent experi-ments. (B) MIEP-luciferase activity was assessed in transient reporter gene assays inthe presence of 1,25-dihydroxyvitamin D3 or PMA. MIEP-luciferase activity wasnormalized to the internal control renilla luciferase. The results presented arederived from 3–5 independent experiments performed in duplicate. VitD3, 1,25-dihydroxyvitamin D3. npo0.05, nnpo0.01.

Fig. 7. 1,25-dihydroxyvitamin D3 and PMA do not function in a cooperative manner toincrease lytic replication. THP-1 cells were differentiated with PMA (80 nM) or PMA(80 nM) and 1,25-dihydroxyvitamin D3 (100 nM) for one day, and then infected withHCMV at an MOI of 7. Cell supernatants were harvested at the indicated times post-infection and viral titers were determined by plaque assay on human foreskin fibroblasts.The results indicate that 1,25-dihydroxyvitamin D3 and PMA do not function coopera-tively to enhance HCMV replication in THP-1 cells. The results are derived from4 independent experiments. VitD3, 1,25-dihydroxyvitamin D3.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–9588

to infection with HCMV (experimental conditions are depictedgraphically in Fig. 9C). Expression of CD14 was monitoredthroughout the time course to assess cellular differentiation andIE positivity was assessed as a measure of the relative ability of thecells to promote lytic replication. IE expression was measured at24 h post-infection in all cases. This experimental system enablesus to assess the expression of the differentiation marker CD14 andthe HCMV IE protein in a series of cells in different stages ofmaturation. CD14 expression as measured by flow cytometrypeaked at between 1 and 2 days post-stimulation with 1,25-dihydroxyvitamin D3 (Fig. 9A). Interestingly, we did not observesignificant effects of 1,25-dihydroxyvitamin D3 on IE positivityuntil 2 days post-stimulation with 1,25-dihydroxyvitamin D3(Fig. 9B). This finding is consistent with the conclusion that 1,25-dihydroxyvitamin D3-induction of lytic HCMV replication corre-lates with 1,25-dihydroxyvitamin D3 induced cellular differentia-tion. Moreover, the lack of a significant effect of 1,25-dihydroxyvitamin D3 at early time points argues against a directeffect of 1,25-dihydroxyvitamin D3 on the MIEP similar to what weobserved in reporter assays in Fig. 6.

Discussion

In this study, we found that 1,25-dihydroxyvitamin D3, ahormone present in the circulatory system and in many tissues(Prietl et al., 2013), can promote HCMV replication in primaryperipheral blood monocytes and in THP-1 cells. Our data are

consistent with a mechanism whereby 1,25-dihydroxyvitamin D3induced HCMV replication involves the induction of monocytedifferentiation. Based on our data, we propose that monocytematuration/differentiation induced by 1,25-dihydroxyvitamin D3leads to a modification of histone 3 K27 methylation in the HCMVIE enhancer region, which results in a conversion to an openchromatin conformation and induction of IE gene expression.Moreover, since 1,25-dihydroxyvitamin D3 does not promote cellcycle arrest and terminal differentiation of monocyte cell linesin vitro (Schwende et al., 1996), this system represents an inter-esting paradigm that could be utilized to study reactivation ofvirus in model systems. In particular, while previous studies haveshown that PMA differentiated THP-1 macrophages can supportpermissive infection, in this study we found that monocytes in thetransition between mature monocytes and macrophage stages,like the state induced by 1,25-dihydroxyvitamin D3, can alsosupport permissive infection.

Our work is the first to explore the effect of 1,25-dihydrox-yvitamin D3 on HCMV in myeloid cells. Although the concentra-tion of 1,25-dihydroxyvitamin D3 used in our study is somewhathigher than the concentrations found in vivo (Grande et al., 2002),published studies have demonstrated that in hematopoietic pro-genitor cells derived from umbilical cord blood, a similar dose of1,25-dihydroxyvitamin D3 added once a week has the same effectson monocyte–macrophage differentiation as does physiologicalconcentrations of 1,25-dihydroxyvitamin D3 supplemented daily(Grande et al., 2002). In this case the difference between the twoconditions is that high dose treatment causes more rapid

Fig. 8. 1,25-dihydroxyvitamin D3 and PMA induced THP-1 differentiation is phenotypically distinct. (A) THP-1 cells were treated with 1,25-dihydroxyvitamin D3 (100 nM) orPMA (80 nM) for 3 days and photographed with a Olympus Q Color 5 camera equipped with QCapture Pro Software. (B) Cells treated as in panel A were stained with theindicated antibodies and analyzed by flow cytometry. Data from four independent experiments is depicted graphically. VitD3, 1,25-dihydroxyvitamin D3. nnpo0.01.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–95 89

differentiation and lineage commitment than does physiologicalconcentrations and thus the high doses like those used in ourstudy simply facilitate experiments performed in an in vitrosetting.

Although in this research, our main focus is to define the effectof 1,25-dihydroxyvitamin D3 on HCMV lytic replication in periph-eral blood monocytes, an equally attractive question to beaddressed in the future is to ask what effect 1,25-dihydroxyvita-min D3 has on HCMV replication in hematopoietic progenitor cells(HPCs), a cell type well-known as a reservoir for HCMV latency(Goodrum et al., 2002; Maciejewski and St Jeor, 1999). In particulardoes 1,25-dihydroxyvitamin D3 treatment of HPCs affect theability of HCMV to choose a latent or lytic path and would thepresence of high concentrations of 1,25-dihydroxyvitamin D3promote lytic reactivation? It is known in the HCMV field thatmacrophages and mature dendritic cells can support lytic infectionbut hematopoietic progenitor cells, myeloid progenitor cells andmonocytes are cell types known to typically establish a latentinfection (Sinclair, 2008; Sinclair, 2010). THP-1 cells treated with1,25-dihydroxyvitamin D3 do not show a phenotype characteristicof mature macrophages and therefore, 1,25-dihydroxyvitamin D3appears to induce a differentiation state in between monocytesand macrophages. Previously, studies of HCMV in myeloid cells

have focused on hematopoietic progenitor cells (Goodrum et al.,2007; Maciejewski et al., 1992), myeloid progenitor cells (Cheunget al., 2006; Kondo et al., 1994), monocytes (Chan et al., 2012;Keyes et al., 2012b; Stevenson et al., 2014) or macrophages(Sanchez et al., 2012), and thus our research offers some additionalinsight into HCMV replication in cells that are transitioningbetween the monocyte and macrophage stages. Although previousstudies suggest that HCMV typically enters the lytic phase inmature macrophages (Smith et al., 2004; Turtinen and Seufzer,1994; Weinshenker et al., 1988) our results indicated that cellstransitioning between the monocyte and macrophage stages canalso support lytic infection. HCMV infection itself has been shownto promote monocyte differentiation but the differentiation pat-terns triggered by infection alone are not typically capable ofefficiently driving lytic infection (Chan et al., 2012; Stevensonet al., 2014). It is highly possible that there is a differentiationthreshold that is needed to be passed in order to appropriatelykindle a lytic infection. 1,25-dihydroxyvitamin D3 treatment mayprime the cells in the differentiation process, and when infectionprovides the appropriate additional differentiation signals, thethreshold is surpassed.

In these studies we examined whether 1,25-dihydroxyvitaminD3 differentiated monocytes can support lytic infection, but asmentioned above an important question to ask in the future iswhether 1,25-dihydroxyvitamin D3 may also be involved in thereactivation of HCMV from latency. Recently, glucocorticoids havebeen shown to trigger reactivation of HCMV in primary monocytesthrough direct activation of the IE promoter (Van Damme et al.,2014). Since there are studies showing crosstalk between gluco-corticoids and 1,25-dihydroxyvitamin D3 (Hidalgo et al., 2011), andsince 1,25-dihydroxyvitamin D3 can enhance glucocorticoid actionin human monocytes (Zhang et al., 2013, 2014), it is reasonable tospeculate that 1,25-dihydroxyvitamin D3 may also be involved inthe regulation of HCMV reactivation or work in concert withglucocorticoids in this process. Perhaps since the mechanisms ofactions appear to be disparate (glucocorticoids directly on theMIEP and 1,25-dihydroxyvitamin D3 on cellular differentiation)these two hormones may work synergistically to affect HCMVreplication and/or reactivation.

Differentiation of monocytes by 1,25-dihydroxyvitamin D3appears to robustly induce the HCMV lytic phase, but the precisemolecular mechanism(s) that regulates this activity remainunknown. Based on our studies, we postulate that signalingactivity and gene expression patterns typically triggered by 1,25-dihydroxyvitamin D3 to drive monocyte maturation/differentia-tion are required for HCMV lytic infection to proceed. Globalcomparison of downstream gene expression induced by 1,25-dihydroxyvitamin D3, PMA and/or glucocorticoids in monocytesby microarray or RNA-seq will be helpful to narrow down the listof candidates, and increase the likelihood of identifying theessential molecules that regulate this switch.

From a clinical perspective, there are many ongoing studiesshowing that 1,25-dihydroxyvitamin D3 or its synthetic analogscan have anti-cancer effects, and can have beneficial effects oncardiovascular or autoimmune disease (Delvin et al., 2014; Jameset al., 1998; Menezes et al., 2014). Since there is some evidencesupporting a role for HCMV infection in the progression of cancers(Michaelis et al., 2011; Soroceanu and Cobbs, 2011), cardiovascularand autoimmune diseases (Caposio et al., 2011; Varani et al.,2009), our study could prompt questions regarding whether ornot 1,25-dihydroxyvitamin D3 and its synthetic analogs should beused therapeutically. In addition, the effect of 1,25-dihydroxyvita-min D3 on HCMV replication in other cell types includingcancer cells is still unknown. Moreover, while 1,25-dihydroxyvi-tamin D3 can have genomic and non-genomic effects (Normanet al., 1992), whether those synthetic analogs have the same effect

Fig. 9. The timing and magnitude of 1,25-dihydroxyvitamin D3 induced differ-entiation of THP-1 monocytes is important for supporting HCMV lytic infection.(A) THP-1 cells were treated with 1,25-dihydroxyvitamin D3 (100 nM) for indicatedtimes and CD14 expression was analyzed by flow cytometry. (B) Cells were treatedwith 1,25-dihydroxyvitamin D3 for the indicated times prior to infection withHCMV TB40E. At 24 h post-infection cells were fixed, permeabilized and stainedwith HCMV IE antibody mAb810-Alexa488. The results represent flow cytometricanalysis of IE positive cells and are derived from three independent experiments.(C) Experimental set up for the experiments depicted here. Cells were treated forvariable lengths prior to infection with HCMV TB40E. In all cases IE gene expressionwas analyzed at 24 h post-infection. VitD3, 1,25-dihydroxyvitamin D3. nnpo0.01,nnnpo0.001.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–9590

of 1,25-dihydroxyvitamin D3 on HCMV replication is anotherintriguing question to answer and which may provide an importanttool to tease out the specific pathways involved in lytic induction.

Materials and methods

General reagents

1,25-dihydroxyvitamin D3 and phorbol 12-myristate 13-acetate(PMA) were purchased from Sigma-Aldrich. APC conjugated anti-human CD14, anti-human CD11b, anti-human CD54, anti-humanCD36 antibodies and PE conjugated anti-mouse IgG1 antibodieswere obtained from eBioscience. Anti-CMV IE1/IE2 antibody(mAb810) and an Alexa Fluors 488 conjugated version ofmAB810 were purchased from Millipore. Anti-CMV UL44 antibody(mAb 25G11, IgG1 isotype) was a kind gift of John Shanley, andanti-CMV pp65 antibody was obtained from Virusys Corporation.

Cell culture and differentiation of THP-1 cells

THP-1 cells were maintained in RPMI-1640 (Roswell ParkMemorial Institute Institute-1640) supplemented with 10% FBS,100 IU/ml penicillin and 100 μg/ml streptomycin at 37 1C in 5%CO2. THP-1 cells were passaged every 3 days to maintain the celldensity between 0.2�106 and 1�106 cells/ml. Human foreskinfibroblasts (HFFs) were maintained in DMEM (Dulbecco ModifiedEagle's Medium) supplemented with 10% Fetal Clone III serum,100 IU/ml penicillin and 100 μg/ml streptomycin at 37 1C in 5%CO2. THP-1 cells were treated with 80 nM PMA or 100 nM 1,25-dihydroxyvitamin D3 for three days to induce cellular maturation/differentiation.

Propagation and purification of virus

The HCMV TB40E-mCherry(3XFLAGUS28) virus was generouslyprovided by Dr. Christine O' Connor from the Cleveland Clinic(Miller et al., 2012; O'Connor and Shenk, 2011). This virus wascharacterized and demonstrated to grow with similar kinetics andto similar titers as does HCMV TB40E. To propagate virus, HFFswere infected with TB40E viruses at an m.o.i. of 0.04. Viralsupernatant was harvested at days 9, 11, and 13 post-infection.Cell Culture supernatant containing virus was centrifuged at 1800gfor 3 min at 21 1C to remove cellular debris. The clarified super-natant was overlayed on a 20% D-sorbitol/1 mM MgCl2 cushionand subjected to ultracentrifugation at 24,000 rpm for 1 h at 21 1C.Supernatant was decanted, and the viral pellet was resuspended inRPMI-1640 culture media. Viral supernatant was aliquoted andstored at –80 1C.

Isolation of monocytes from peripheral blood of normal donors

Blood was diluted 2 fold with Dulbecco's Phosphate-BufferedSaline (DPBS) containing 2 mM EDTA. Diluted blood was carefullylayered over 15 ml of Ficoll-paques PLUS in a 50 ml conical tube.Conical tubes were centrifuged at 400g for 30–40 min at 20 1C in aswinging bucket rotor without brake. After centrifuge, the upperlayer was aspirated, leaving buffy coat containing the mononuclearcell layer undisturbed at interphase. The buffy coat was transfer toclean tube, and fresh DPBS with 2 mM EDTA was added to fill thetubes. The cells were centrifuged at 300g for 10 min at 20 1C. Thensupernatant was carefully removed. In order to remove platelets,cells were resuspended in 50 ml of DPBS with 2 mM EDTA, andcentrifuged at 200g for 10–15 min at 20 1C. The supernatant wasremoved afterward. The wash step was repeated once to furtherdeplete platelets. Cells were resuspended in 80 ml of buffer (DPBS

containing 0.5% BSA and 2 mM EDTA) per 107 cells. 20 ml of CD14MicroBeads (Miltenyi Biotec) was added per 107 cells, and cellswere incubated in the cold room (2–8 1C) for 30 min. Afterincubation, cells were washed by adding 1–2 ml of buffer (DPBScontaining 0.5% BSA and 2 mM EDTA) per 107 cells, and centri-fuged at 300g for 10 min. 108 cells were resuspended in 500 ml ofbuffer, and passed through LS Column (Miltenyi Biotec). Thecolumnwas washed 3 times with buffer. The column was removedfrom MACS Separator (Miltenyi Biotec), and cells were flushed outby firmly pushing the plunger into the column.

Culture and infection of primary peripheral blood derived monocytes

CD14þ monocytes isolated as described above were resus-pended in RPMI-1640 supplemented with 10% FBS, 100 IU/mlpenicillin and 100 μg/ml streptomycin. Cells were then culturedin 1,25-dihydroxyvitamin D3 (100 nM), PMA (80 nM) or theappropriate vehicle control (EtOH or DMSO) for 2 days at 37 1Cin 5% CO2. After 2 days of culture in 1,25-dihydroxyvitamin D3,PMA or vehicle, viral supernatant was added at a MOI of 10, andcells were incubated overnight. After overnight incubation withvirus, cells treated with vehicle or 1,25-dihydroxyvitamin D3 werepelleted by centrifugation, viral inocula were removed, and cellswere treated with 1X trypsin for 5 min to remove attached but un-internalized virions (O'Connor and Murphy, 2012). Cells were thenresuspended in fresh RPMI supplemented as described above andcultured for 4–6 days. In the case of the PMA treated cells thatwere adhered to the culture plates, viral inocula were aspirated,cells were washed thoroughly with DPBS, fed with fresh RPMI andsupplements and cultured for 4–6 days.

HCMV infection of THP-1 cells

THP-1 cells were infected with HCMV TB40E-mCherry (3XFLA-GUS28) at MOIs as indicated in the figure legends. After the viralsupernatant was added, cells were centrifuged at 21 1C and 1000gfor 30 min to enhance infectivity. After overnight culture, cellswere spun down and the inoculums were removed. Cells wereincubated in 1X trypsin for 5 min to remove attached but un-internalized virions (O'Connor and Murphy, 2012). The trypsinreactions were neutralized by adding equal volumes of freshculture media. The supernatant is aspirated, and cells wereresuspended in fresh culture media. Because the PMA-differentiated cells firmly adhere to the plates, the centrifuge stepis omitted from infection protocol. After overnight culture, culturemedia were removed and cells were washed with 1X DPBS, andfresh media were added to the cells.

Western blot analysis of HCMV gene expression

1�106 infected THP-1 cells were lysed in NP-40 cell lysis buffer(50 mM HEPES pH7.4, 0.5% NP-40, 250 mM NaCl, 20% glycerol,2 mM EDTA, 100 μM Sodium Orthovanadate, 1 mM Sodium Fluor-ide, 1X complete Mini protease inhibitor). Cell lysates weresonicated for 20 s on level 1 using a Sonic Dismembrator Model100 (Fisher Scientific). Protein concentrations for each lysatesample were determined using the Bio-Rad protein assay reagent.Lysates were mixed with Laemmli sample buffer, and heated at1001 for 10 min. 30 μg of protein was loaded into each lane forelectrophoresis. Proteins were transferred to nitrocellulose mem-branes using a semi-dry transfer apparatus (Continental LabProducts). Membranes were blocked with 5% non-fat dried milkfor 1 h and membranes were incubated with primary antibody at41 overnight. Membranes were washed 3 times with Tris bufferedsaline with Tween-20 (TBST) and then incubated with secondaryantibody for 2 h. Membranes were washed with TBST 3 times and

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–95 91

subjected to antibody detection using the SuperSignal West Picochemiluminescent substrate (Thermo Scientific). Luminescenceemitted from the membranes was detected by classic blue auto-radiography film BX. Films were developed by Kodak min-Rmammography processor.

MIEP reporter gene construction

The HCMV immediate-early gene promoter (bp 52586–53162in GenBank entry AC146907.1) was PCR amplified from FIX-BACbacmid DNA using the following primers. HCMV MIEP promoterforward primer: 50-TAACCCGGGTAGTAATCAATTACGGGG-30, HCMVMIEP promoter reverse primer: 50-TCGAGATCTCTGACGGTTCAC-TAAACG-30. PCR amplification was performed for 30 cycles, con-sisting of denaturation at 94 1C for 30 s, annealing at 55 1C for 30 s,and extension at 72 1C for 30 s. The HCMV MIEP promoterfragment was cloned into the XmaI and BglII sites of pGL3 andsequenced to confirm the identity of the MIEP fragment(Genewiz, Inc).

Luciferase assays

2�105 THP-1 cells were plated per well in 24-well plates andcultured overnight. For each transfection, 410 ng pcDNA3, 60 ngpGL3-HCMV MIEP, 30 ng phRGTK-renilla and 1.5 μl TransIT-2020was diluted into 50 μl RPMI-1640 and incubated for 15 min. Theincubated transfection reagent was then added into each well. 4 hafter transfection, vehicle (ethanol or DMSO), 1,25-dihydroxyvita-min D3 (100 nM), or PMA (80 nM) was added to the designatedwells and the cells were incubated for a further 24 h. Cells werelysed in 200 μl of 1X passive lysis buffer and 10 μl of the cell lysatewas used in luciferase assay reactions. 50 μl of Firefly-luciferasesubstrate was added to cell lysate and luciferase activity wasmeasured on a Glomax 20/20 luminometer (Promega). 50 μl Stopand Glow solution was added to each reaction, and luminescenceof the control reporter Renilla-Luciferase was measured. TheFirefly-luciferase reading of vehicle control (ethanol) was dividedby the Renilla-Luciferase reading of vehicle control and that valuewas defined as 1. The fold changes were then determined bydividing the luciferase to renilla ratios of the experimental condi-tions to the ratio of vehicle control.

Infectious center assays

1�105 HFF cells were plated into wells of 12-well plates andcultured overnight. Infected THP-1 cells or primary monocyteswere harvested at six days post-infection and co-cultured withHFFs for 2 days. Culture media was then removed and the plateswere washed twice with 1X PBS. Cell monolayers were coveredwith overlay media (a 1:1 mixture of 1.5% carboxymethyl cellulose(Sigma), and 2X MEM supplemented with 20% FCIII serum,nonessential amino acids, and penicillin-streptomycin) and incu-bated for another 8 days to allow for plaque development. Cellswere fixed with methanol, stained with 10% Geimsa (Sigma) andplaques were counted using a dissecting microscope.

Plaque assays

After infection, culture media from each cell sample washarvested on the indicated days, and 10 μl of culture media wasadded to HFF monolayers plated in 12-well plates the previousday. Virus was adsorbed to HFF monolayers for 3 h, culture mediawere removed, and covered with overlay media (a 1:1 mixture of1.5% carboxymethyl cellulose (Sigma), and 2X MEM supplementedwith 20% FCIII serum, nonessential amino acids, and penicillin-streptomycin) and incubated for another 8 days to allow for plaque

development. Cells were fixed with methanol, stained with 10%Geimsa (Sigma) and plaques were counted using a dissectingmicroscope.

Flow cytometry

For cell surface marker analyses, cells were harvested andresuspended in 50 μl of a 0.5% BSA/DPBS solution containing1:200 dilution of the appropriate APC-conjugated antibody(CD11b, CD14, etc). Cells were incubated at room temperaturefor 1 h. Cells were washed with 0.5% BSA/DPBS solution, resus-pended in fresh DPBS and analyzed by flow cytometry on a BDFACSCalibur. For HCMV IE protein staining, cells were harvested,resuspended in 100 μl DPBS and 1 ml of 70% ice cold EtOH wasadded to fix the cells. Fixed cells were washed and permeabilizedwith 0.5% BSA/DPBS solution containing 0.5% tween-20. Cells werethen resuspended in 0.5% BSA/DPBS solution containing 1:200dilution of Alexa488 conjugated anti-HCMV IE antibody, andincubated at room temperature for 1 h. Stained cells were washedwith 500 μl 0.5% BSA/DPBS solution containing 0.5% tween-20. ForUL44 and IE co-staining, cells were fixed and permeabilized asabove and then resuspended in 0.5% BSA/DPBS solution containing1:10 dilution anti-HCMV UL44 antibody. Cells were incubated for1 h at room temperature, washed as described above and thenincubated for 1 h in 0.5% BSA/DPBS solution containing 1:250dilution of PE-conjugated anti-mouse IgG1 to label the UL44primary antibody. UL44 stained cells were then washed andstained for IE proteins as described above. After staining, cellswere resuspended in DPBS and analyzed by flow cytometry.

Semi-quantitative PCR for viral DNA copy number

2�105 THP-1 cells were harvested on days 1 and 6 post-infection, and lysed in 100 μl DNA lysis buffer (Kondo et al., 1994)containing 20 μg of Proteinase K at 55 1C overnight. Proteinase Kactivity was stopped by incubating DNA lysates at 100 1C for15 min and DNAs were used for semi-quantitative PCR. Primersfor HCMV IE amplification: HCMV IE forward 50-ATG-GAGTCCTCTGCCAAGAGAAAGATGGAC-30, HCMV IE reverse 50-CAA-TACACTTCATCTCCTCGAAAGG-30 (Bego et al., 2005). Primers usedfor GAPDH amplification: GAPDH forward 50-GAGCCAAAAGGGT-CATC-30, GAPDH reverse primer 50-GTGGTCATGAGTCCTTC-30

(Juckem et al., 2008). PCR amplification was performed for 30cycles, consisting of denaturation at 94 1C for 30 s, annealing at55 1C for 30 s, and extension at 721 for 30 s. Band intensities weremeasured by NIH Image software and calculated as a ratio ofHCMV IE DNA over cellular GAPDH DNA.

Chromatin immunoprecipitation

5�106 cells were harvested at 2 days post-infection and fixedin DPBS containing 1% formaldehyde for 10 min at room tempera-ture. 2.5 M glycine was added to stop the fixation reaction, andcells were washed in ice cold DPBS. Cells were lysed in cell lysisbuffer (0.5 mM PIPES Ph8, 85 mM KCl, 0.5% NP40, 1X proteininhibitor cocktail). Nuclei were pelleted by centrifuge and resus-pended in 500 μl nuclear lysis buffer (50 mM Tris–HCl Ph8, 10 mMEDTA,1% SDS,1X protein inhibitor cocktail). The lysates weresonicated for six cycles (30 s on/30 s off) using program 3 on aSonic Dismembrator Model 100 (Fisher Scientific). Lysates werethen diluted 5 fold in IP dilution buffer (0.01% SDS, 1.1% Triton-X100, 1.2 mM EDTA, 16.7 mM Tris–HCl pH8, 167 mM NaCl, 1Xprotein inhibitor cocktail) and pre-cleared with 30 μl sepharosebeads for 2 h at 4 1C. 3 μg of anti-tri-methylated histone 3 antibody(Millipore) was added into each reaction and rotated overnight at41C. Protein A/G beads were added to each reaction and incubated

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–9592

at 41C for 3 h to capture primary antibody. Protein A/G beads waspelleted by centrifugation and washed with 1X dialysis buffer(2 mM EDTA, 50 mM Tris–HCl pH8, 0.2% Sarkosyl, 1X proteininhibitor cocktail) 3 times. Bound chromatin fragments wereeluted using elution buffer (50 mM NaHCO3, 1% SDS). After elution,RNase A and NaCl were added to make final concentration of0.083 mg/ml and 0.2 M respectively and the solution was incu-bated at 65 1C overnight to reverse cross-links. 34 μg of ProteinaseK was added to digest proteins at 45 1C for 2 h. Primers foramplification of MIEP enhancer: Forward primer: 50-TTGGGCA-TACGCGATATCTG-30. Reverse primer: 50-GCCTCATATCGTCTGT-CACC-30 (Abraham and Kulesza, 2013). The DNA fragments wererecovered using a Fermentus gel extraction kit and 20 ng ofimmunoprecipitated DNA was used for PCR amplification. PCRreaction conditions are the same as mentioned above except wereperformed for 36 cycles. The signal from chromatin immunopre-cipitation samples were normalized to signal from respectiveinput samples.

Acknowledgments

We would like to thank Christine O'Connor for providing therecombinant TB40E virus expressing mCherry and J. Shanley forUL44 antibody. We thank the Cell Processing and ManipulationCore in the Translational Cores, and Physicians and Nurses atCCHMC for obtaining and processing peripheral blood samples formonocyte purification. We also thank the CCHMC TranslationalResearch Trials Office for providing the regulatory and adminis-trative support for this endeavor. S.E. Wu was supported by aTeaching Assistantship at the University of Cincinnati. This workwas supported by National Institutes of Health Grants R01-AI058159 and R56-AI095442 awarded to W.E.M.

Appendix A. Supporting information

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.virol.2015.04.004.

References

Abraham, C.G., Kulesza, C.A., 2013. Polycomb repressive complex 2 silences humancytomegalovirus transcription in quiescent infection models. J. Virol. 87 (24),13193–13205.

Baeke, F., Takiishi, T., Korf, H., Gysemans, C., Mathieu, C., 2010. Vitamin D:modulator of the immune system. Curr. Opin. Pharmacol. 10 (4), 482–496.

Bate, S.L., Dollard, S.C., Cannon, M.J., 2010. Cytomegalovirus seroprevalence in theUnited States: the national health and nutrition examination surveys, 1988–2004. Clin. Infect. Dis. 50 (11), 1439–1447.

Bego, M., Maciejewski, J., Khaiboullina, S., Pari, G., St Jeor, S., 2005. Characterizationof an antisense transcript spanning the UL81-82 locus of human cytomegalo-virus. J. Virol. 79 (17), 11022–11034.

Bego, M.G., St Jeor, S., 2006. Human cytomegalovirus infection of cells ofhematopoietic origin: HCMV-induced immunosuppression, immune evasion,and latency. Exp. Hematol. 34 (5), 555–570.

Bunce, C.M., Brown, G., Hewison, M., 1997. Vitamin D and hematopoiesis. TrendsEndocrinol. Metab. 8 (6), 245–251.

Campbell, G.R., Spector, S.A., 2012. Toll-like receptor 8 ligands activate a vitamin Dmediated autophagic response that inhibits human immunodeficiency virustype 1. PLoS Pathog. 8 (11), e1003017.

Caposio, P., Orloff, S.L., Streblow, D.N., 2011. The role of cytomegalovirus inangiogenesis. Virus Res. 157 (2), 204–211.

Castagna, M., Takai, Y., Kaibuchi, K., Sano, K., Kikkawa, U., Nishizuka, Y., 1982. Directactivation of calcium-activated, phospholipid-dependent protein kinase bytumor-promoting phorbol esters. J. Biol. Chem. 257 (13), 7847–7851.

Chan, G., Bivins-Smith, E.R., Smith, M.S., Smith, P.M., Yurochko, A.D., 2008.Transcriptome analysis reveals human cytomegalovirus reprograms monocytedifferentiation toward an M1 macrophage. J. Immunol. 181 (1), 698–711.

Chan, G., Nogalski, M.T., Yurochko, A.D., 2012. Human cytomegalovirus stimulatesmonocyte-to-macrophage differentiation via the temporal regulation of cas-pase 3. J. Virol. 86 (19), 10714–10723.

Cheung, A.K., Abendroth, A., Cunningham, A.L., Slobedman, B., 2006. Viral geneexpression during the establishment of human cytomegalovirus latent infec-tion in myeloid progenitor cells. Blood 108 (12), 3691–3699.

Daigneault, M., Preston, J.A., Marriott, H.M., Whyte, M.K., Dockrell, D.H., 2010. Theidentification of markers of macrophage differentiation in PMA-stimulatedTHP-1 cells and monocyte-derived macrophages. PLoS One 5 (1), e8668.

Delvin, E., Souberbielle, J.C., Viard, J.P., Salle, B., 2014. Role of vitamin D in acquiredimmune and autoimmune diseases. Crit. Rev. Clin. Lab. Sci. 51 (4), 232–247.

Dziurzynski, K., Chang, S.M., Heimberger, A.B., Kalejta, R.F., McGregor Dallas, S.R.,Smit, M., Soroceanu, L., Cobbs, C.S., 2012. Consensus on the role of humancytomegalovirus in glioblastoma. Neuro Oncol. 14 (3), 246–255.

Fu, Y., Chen, J., Pang, B., Li, C., Zhao, J., Shen, K., 2014. EZH2-Induced H3K27me3 isassociated with epigenetic repression of the ARHI tumor-suppressor gene inovarian cancer. Cell Biochem. Biophys..

Garabedian, M., Ulmann, A., 1979. Metabolic and cellular activity of vitamin D. Adv.Nephrol. Necker Hosp. 8, 165–177.

Gemelli, C., Orlandi, C., Zanocco Marani, T., Martello, A., Vignudelli, T., Ferrari, F.,Montanari, M., Parenti, S., Testa, A., Grande, A., Ferrari, S., 2008. The vitamin D3/Hox-A10 pathway supports MafB function during the monocyte differentiationof human CD34þ hemopoietic progenitors. J. Immunol. 181 (8), 5660–5672.

Ghazal, P., DeMattei, C., Giulietti, E., Kliewer, S.A., Umesono, K., Evans, R.M., 1992.Retinoic acid receptors initiate induction of the cytomegalovirus enhancer inembryonal cells. Proc. Natl. Acad. Sci. USA 89 (16), 7630–7634.

Goltzman, D., Hendy, G.N., White, J.H., 2014. Vitamin D and its receptor during latedevelopment. Biochim. Biophys. Acta.

Goodrum, F., Reeves, M., Sinclair, J., High, K., Shenk, T., 2007. Human cytomegalo-virus sequences expressed in latently infected individuals promote a latentinfection in vitro. Blood 110 (3), 937–945.

Goodrum, F.D., Jordan, C.T., High, K., Shenk, T., 2002. Human cytomegalovirus geneexpression during infection of primary hematopoietic progenitor cells: a modelfor latency. Proc. Natl. Acad. Sci. USA 99 (25), 16255–16260.

Grande, A., Montanari, M., Tagliafico, E., Manfredini, R., Zanocco Marani, T., Siena,M., Tenedini, E., Gallinelli, A., Ferrari, S., 2002. Physiological levels of 1alpha, 25dihydroxyvitamin D3 induce the monocytic commitment of CD34þhematopoietic progenitors. J. Leukoc. Biol. 71 (4), 641–651.

Greenberger, J.S., Newburger, P.E., Sakakeeny, M., 1980. Phorbol myristate acetatestimulates macrophage differentiation and replication and alters granulopoiesisand leukemogenesis in long-term bone marrow cultures. Blood 56 (3),368–379.

Grinde, B., 2013. Herpesviruses: latency and reactivation – viral strategies and hostresponse. J. Oral Microbiol., 5.

Grosse, S.D., Ross, D.S., Dollard, S.C., 2008. Congenital cytomegalovirus (CMV)infection as a cause of permanent bilateral hearing loss: a quantitativeassessment. J. Clin. Virol. 41 (2), 57–62.

Hemmi, H., Breitman, T.R., 1985. Induction of functional differentiation of a humanmonocytic leukemia cell line (THP-1) by retinoic acid and cholera toxin. Jpn. J.Cancer Res. 76 (5), 345–351.

Hewison, M., 2012. Vitamin D and immune function: autocrine, paracrine orendocrine? Scand. J. Clin. Lab. Investig. Suppl. 243, 92–102.

Hidalgo, A.A., Deeb, K.K., Pike, J.W., Johnson, C.S., Trump, D.L., 2011. Dexamethasoneenhances 1alpha,25-dihydroxyvitamin D3 effects by increasing vitamin Dreceptor transcription. J. Biol. Chem. 286 (42), 36228–36237.

Hmama, Z., Nandan, D., Sly, L., Knutson, K.L., Herrera-Velit, P., Reiner, N.E., 1999.1alpha,25-dihydroxyvitamin D(3)-induced myeloid cell differentiation is regu-lated by a vitamin D receptor-phosphatidylinositol 3-kinase signaling complex.J. Exp. Med. 190 (11), 1583–1594.

Holick, M.F., 2003. Vitamin D: a millenium perspective. J. Cell. Biochem. 88 (2),296–307.

Huber, R., Pietsch, D., Gunther, J., Welz, B., Vogt, N., Brand, K., 2014. Regulation ofmonocyte differentiation by specific signaling modules and associated tran-scription factor networks. Cell. Mol. Life Sci. 71 (1), 63–92.

Hwang, E.S., Kim, J., Jong, H.S., Park, J.W., Park, C.G., Cha, C.Y., 2000. Characteristicsof DNA-binding activity of human cytomegalovirus ppUL44. Microbiol. Immu-nol. 44 (10), 827–832.

James, S.Y., Mercer, E., Brady, M., Binderup, L., Colston, K.W., 1998. EB1089, asynthetic analogue of vitamin D, induces apoptosis in breast cancer cells in vivoand in vitro. Br. J. Pharmacol. 125 (5), 953–962.

Johnson, J., Anderson, B., 2014. Screening, prevention, and treatment of congenitalcytomegalovirus. Obstet. Gynecol. Clin. N. Am. 41 (4), 593–599.

Juckem, L.K., Boehme, K.W., Feire, A.L., Compton, T., 2008. Differential initiation ofinnate immune responses induced by human cytomegalovirus entry intofibroblast cells. J. Immunol. 180 (7), 4965–4977.

Kalejta, R.F., 2008. Tegument proteins of human cytomegalovirus. Microbiol. Mol.Biol. Rev. 72 (2), 249–265, table of contents.

Kannan, S., Lim, H.W., 2014. Photoprotection and vitamin D: a review. Photoder-matol. Photoimmunol. Photomed. 30 (2–3), 137–145.

Kew, V.G., Yuan, J., Meier, J., Reeves, M.B., 2014. Mitogen and stress activatedkinases act co-operatively with CREB during the induction of human cytome-galovirus immediate-early gene expression from latency. PLoS Pathog. 10 (6),e1004195.

Keyes, L.R., Bego, M.G., Soland, M., St Jeor, S., 2012a. Cyclophilin A is required forefficient human cytomegalovirus DNA replication and reactivation. J. Gen. Virol.93 (Pt 4), 722–732.

Keyes, L.R., Hargett, D., Soland, M., Bego, M.G., Rossetto, C.C., Almeida-Porada, G.,St Jeor, S., 2012b. HCMV protein LUNA is required for viral reactivation fromlatently infected primary CD14(þ) cells. PLoS One 7 (12), e52827.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–95 93

Kline, J.N., Hunninghake, G.M., He, B., Monick, M.M., Hunninghake, G.W., 1998.Synergistic activation of the human cytomegalovirus major immediate earlypromoter by prostaglandin E2 and cytokines. Exp. Lung Res. 24 (1), 3–14.

Kondo, K., Kaneshima, H., Mocarski, E.S., 1994. Human cytomegalovirus latentinfection of granulocyte-macrophage progenitors. Proc. Natl. Acad. Sci. USA 91(25), 11879–11883.

Korf, H., Decallonne, B., Mathieu, C., 2014. Vitamin D for infections. Curr. Opin.Endocrinol. Diabetes Obes. 21 (6), 431–436.

Kremlev, S.G., Phelps, D.S., 1997. Effect of SP-A and surfactant lipids on expressionof cell surface markers in the THP-1 monocytic cell line. Am. J. Physiol. 272(6 Pt 1), L1070–L1077.

Kunisch, E., Fuhrmann, R., Roth, A., Winter, R., Lungershausen, W., Kinne, R.W.,2004. Macrophage specificity of three anti-CD68 monoclonal antibodies (KP1,EBM11, and PGM1) widely used for immunohistochemistry and flow cytome-try. Ann. Rheum. Dis. 63 (7), 774–784.

Lagishetty, V., Liu, N.Q., Hewison, M., 2011. Vitamin D metabolism and innateimmunity. Mol. Cell. Endocrinol. 347 (1-2), 97–105.

Lamberg-Allardt, C., 2006. Vitamin D in foods and as supplements. Prog. Biophys.Mol. Biol. 92 (1), 33–38.

Li, S., Zhu, J., Zhang, W., Chen, Y., Zhang, K., Popescu, L.M., Ma, X., Lau, W.B., Rong, R.,Yu, X., Wang, B., Li, Y., Xiao, C., Zhang, M., Wang, S., Yu, L., Chen, A.F., Yang, X.,Cai, J., 2011. Signature microRNA expression profile of essential hypertensionand its novel link to human cytomegalovirus infection. Circulation 124 (2),175–184.

Liu, P.T., Stenger, S., Li, H., Wenzel, L., Tan, B.H., Krutzik, S.R., Ochoa, M.T., Schauber,J., Wu, K., Meinken, C., Kamen, D.L., Wagner, M., Bals, R., Steinmeyer, A., Zugel,U., Gallo, R.L., Eisenberg, D., Hewison, M., Hollis, B.W., Adams, J.S., Bloom, B.R.,Modlin, R.L., 2006. Toll-like receptor triggering of a vitamin D-mediated humanantimicrobial response. Science 311 (5768), 1770–1773.

Liu, X., Yuan, J., Wu, A.W., McGonagill, P.W., Galle, C.S., Meier, J.L., 2010. Phorbolester-induced human cytomegalovirus major immediate-early (MIE) enhanceractivation through PKC-delta, CREB, and NF-kappaB desilences MIE geneexpression in quiescently infected human pluripotent NTera2 cells. J. Virol. 84(17), 8495–8508.

Ljungman, P., Hakki, M., Boeckh, M., 2011. Cytomegalovirus in hematopoietic stemcell transplant recipients. Hematol. Oncol. Clin. N. Am. 25 (1), 151–169.

Luong, K., Nguyen, L.T., 2011. Impact of vitamin D in the treatment of tuberculosis.Am. J. Med. Sci. 341 (6), 493–498.

Maciejewski, J.P., Bruening, E.E., Donahue, R.E., Mocarski, E.S., Young, N.S.,St Jeor, S.C., 1992. Infection of hematopoietic progenitor cells by humancytomegalovirus. Blood 80 (1), 170–178.

Maciejewski, J.P., St Jeor, S.C., 1999. Human cytomegalovirus infection of humanhematopoietic progenitor cells. Leuk. Lymphoma 33 (1–2), 1–13.

Maxwell, C.S., Carbone, E.T., Wood, R.J., 2012. Better newborn vitamin D statuslowers RSV-associated bronchiolitis in infants. Nutr. Rev. 70 (9), 548–552.

McDonough, S.H., Spector, D.H., 1983. Transcription in human fibroblasts permis-sively infected by human cytomegalovirus strain AD169. Virology 125 (1),31–46.

Meier, J.L., 2001. Reactivation of the human cytomegalovirus major immediate-early regulatory region and viral replication in embryonal NTera2 cells: role oftrichostatin A, retinoic acid, and deletion of the 21-base-pair repeats andmodulator. J. Virol. 75 (4), 1581–1593.

Menezes, A.R., Lamb, M.C., Lavie, C.J., DiNicolantonio, J.J., 2014. Vitamin D andatherosclerosis. Curr. Opin. Cardiol. 29 (6), 571–577.

Michaelis, M., Baumgarten, P., Mittelbronn, M., Driever, P.H., Doerr, H.W., Cinatl Jr.,J., 2011. Oncomodulation by human cytomegalovirus: novel clinical findingsopen new roads. Med. Microbiol. Immunol. 200 (1), 1–5.

Miller, W.E., Zagorski, W.A., Brenneman, J.D., Avery, D., Miller, J.L., O'Connor, C.M.,2012. US28 is a potent activator of phospholipase C during HCMV infection ofclinically relevant target cells. PLoS One 7 (11), e50524.

Munoz-Pacheco, P., Ortega-Hernandez, A., Miana, M., Cachofeiro, V., Fernandez-Cruz, A., Gomez-Garre, D., 2012. Ezetimibe inhibits PMA-induced monocyte/macrophage differentiation by altering microRNA expression: a novel anti-atherosclerotic mechanism. Pharmacol. Res. 66 (6), 536–543.

Naito, M., 2008. Macrophage differentiation and function in health and disease.Pathol. Int. 58 (3), 143–155.

Nakamura, T., Hemmi, H., Aso, H., Ishida, N., 1986. Variants of a human monocyticleukemia cell line (THP-1): induction of differentiation by retinoic acid,interferon-gamma, and T-lymphocyte-derived differentiation-inducing activity.J. Natl. Cancer Inst. 77 (1), 21–27.

Netea, M.G., Lewis, E.C., Azam, T., Joosten, L.A., Jaekal, J., Bae, S.Y., Dinarello, C.A.,Kim, S.H., 2008. Interleukin-32 induces the differentiation of monocytes intomacrophage-like cells. Proc. Natl. Acad. Sci. USA 105 (9), 3515–3520.

Niedel, J.E., Kuhn, L.J., Vandenbark, G.R., 1983. Phorbol diester receptor copurifieswith protein kinase C. Proc. Natl. Acad. Sci. USA 80 (1), 36–40.

Norman, A.W., Nemere, I., Zhou, L.X., Bishop, J.E., Lowe, K.E., Maiyar, A.C.,Collins, E.D., Taoka, T., Sergeev, I., Farach-Carson, M.C., 1992. 1,25(OH)2-vitaminD3, a steroid hormone that produces biologic effects via both genomic andnongenomic pathways. J. Steroid Biochem. Mol. Biol. 41 (3–8), 231–240.

O'Connor, C.M., Murphy, E.A., 2012. A myeloid progenitor cell line capable ofsupporting human cytomegalovirus latency and reactivation, resulting ininfectious progeny. J. Virol. 86 (18), 9854–9865.

O'Connor, C.M., Shenk, T., 2011. Human cytomegalovirus pUS27 G protein-coupledreceptor homologue is required for efficient spread by the extracellular routebut not for direct cell-to-cell spread. J. Virol. 85 (8), 3700–3707.

O'Connor, C.M., Vanicek, J., Murphy, E.A., 2014. Host microRNA regulation of humancytomegalovirus immediate early protein translation promotes viral latency.J. Virol. 88 (10), 5524–5532.

O'Kelly, J., Hisatake, J., Hisatake, Y., Bishop, J., Norman, A., Koeffler, H.P., 2002.Normal myelopoiesis but abnormal T lymphocyte responses in vitamin Dreceptor knockout mice. J. Clin. Investig. 109 (8), 1091–1099.

Ooi, J.H., McDaniel, K.L., Weaver, V., Cantorna, M.T., 2014. Murine CD8þ T cells butnot macrophages express the vitamin D 1alpha-hydroxylase. J. Nutr. Biochem.25 (1), 58–65.

Pan, Q., Granger, J., O'Connell, T.D., Somerman, M.J., Simpson, R.U., 1997. Promotionof HL-60 cell differentiation by 1,25-dihydroxyvitamin D3 regulation of proteinkinase C levels and activity. Biochem. Pharmacol. 54 (8), 909–915.

Paya, C., Humar, A., Dominguez, E., Washburn, K., Blumberg, E., Alexander, B.,Freeman, R., Heaton, N., Pescovitz, M.D., 2004. Efficacy and safety of valganci-clovir vs. oral ganciclovir for prevention of cytomegalovirus disease in solidorgan transplant recipients. Am. J. Transplant. 4 (4), 611–620.

Prietl, B., Treiber, G., Pieber, T.R., Amrein, K., 2013. Vitamin D and immune function.Nutrients 5 (7), 2502–2521.

Prosch, S., Staak, K., Stein, J., Liebenthal, C., Stamminger, T., Volk, H.D., Kruger, D.H.,1995. Stimulation of the human cytomegalovirus IE enhancer/promoter in HL-60 cells by TNFalpha is mediated via induction of NF-kappaB. Virology 208 (1),197–206.

Provvedini, D.M., Tsoukas, C.D., Deftos, L.J., Manolagas, S.C., 1983. 1,25-dihydrox-yvitamin D3 receptors in human leukocytes. Science 221 (4616), 1181–1183.

Qin, Q., Penkert, R.R., Kalejta, R.F., 2013. Heterologous viral promoters incorporatedinto the human cytomegalovirus genome are silenced during experimentallatency. J. Virol. 87 (17), 9886–9894.

Reeves, M.B., Sinclair, J.H., 2013. Circulating dendritic cells isolated from healthyseropositive donors are sites of human cytomegalovirus reactivation in vivo.J. Virol. 87 (19), 10660–10667.

Rossetto, C.C., Tarrant-Elorza, M., Pari, G.S., 2013. Cis and trans acting factorsinvolved in human cytomegalovirus experimental and natural latent infectionof CD14 (þ) monocytes and CD34 (þ) cells. PLoS Pathog. 9 (5), e1003366.

Saffert, R.T., Penkert, R.R., Kalejta, R.F., 2010. Cellular and viral control over theinitial events of human cytomegalovirus experimental latency in CD34þ cells.J. Virol. 84 (11), 5594–5604.

Salomon, D.G., Mascaro, E., Grioli, S.M., Ferronato, M.J., Vitale, C.A., Radivoy, G.E.,Curino, A.C., Facchinetti, M.M., 2014. Phosphonate analogues of 1alpha,25dihydroxyvitamin D3 are promising candidates for antitumoural therapies.Curr. Top. Med. Chem..

Sanchez, V., Dong, J.J., Battley, J., Jackson, K.N., Dykes, B.C., 2012. Human cytome-galovirus infection of THP-1 derived macrophages reveals strain-specificregulation of actin dynamics. Virology 433 (1), 64–72.

Schwende, H., Fitzke, E., Ambs, P., Dieter, P., 1996. Differences in the state ofdifferentiation of THP-1 cells induced by phorbol ester and 1,25-dihydroxyvi-tamin D3. J. Leukoc. Biol. 59 (4), 555–561.

Shahijanian, F., Parnell, G.P., McKay, F.C., Gatt, P.N., Shojoei, M., O'Connor, K.S.,Schibeci, S.D., Brilot, F., Liddle, C., Batten, M., Stewart, G.J., Booth, D.R., 2014. TheCYP27B1 variant associated with an increased risk of autoimmune disease isunderexpressed in tolerizing dendritic cells. Hum. Mol. Genet. 23 (6),1425–1434.

Sinclair, J., 2008. Human cytomegalovirus: latency and reactivation in the myeloidlineage. J. Clin. Virol. 41 (3), 180–185.

Sinclair, J., 2010. Chromatin structure regulates human cytomegalovirus geneexpression during latency, reactivation and lytic infection. Biochim. Biophys.Acta 1799 (3-4), 286–295.

Sinigalia, E., Alvisi, G., Mercorelli, B., Coen, D.M., Pari, G.S., Jans, D.A., Ripalti, A.,Palu, G., Loregian, A., 2008. Role of homodimerization of human cytomegalo-virus DNA polymerase accessory protein UL44 in origin-dependent DNAreplication in cells. J. Virol. 82 (24), 12574–12579.

Sinzger, C., Digel, M., Jahn, G., 2008. Cytomegalovirus cell tropism. Curr. Top.Microbiol. Immunol. 325, 63–83.

Smith, M.S., Bentz, G.L., Alexander, J.S., Yurochko, A.D., 2004. Human cytomegalo-virus induces monocyte differentiation and migration as a strategy fordissemination and persistence. J. Virol. 78 (9), 4444–4453.

Soderberg-Naucler, C., Fish, K.N., Nelson, J.A., 1997. Reactivation of latent humancytomegalovirus by allogeneic stimulation of blood cells from healthy donors.Cell 91 (1), 119–126.

Soderberg-Naucler, C., Streblow, D.N., Fish, K.N., Allan-Yorke, J., Smith, P.P.,Nelson, J.A., 2001. Reactivation of latent human cytomegalovirus in CD14(þ)monocytes is differentiation dependent. J. Virol. 75 (16), 7543–7554.

Soroceanu, L., Cobbs, C.S., 2011. Is HCMV a tumor promoter? Virus Res. 157 (2),193–203.

Spector, S.A., 2011. Vitamin D and HIV: letting the sun shine in. Top. Antivir. Med.19 (1), 6–10.

Stein, J., Volk, H.D., Liebenthal, C., Kruger, D.H., Prosch, S., 1993. Tumour necrosisfactor alpha stimulates the activity of the human cytomegalovirus majorimmediate early enhancer/promoter in immature monocytic cells. J. Gen. Virol.74 (Pt 11), 2333–2338.

Stevenson, E.V., Collins-McMillen, D., Kim, J.H., Cieply, S.J., Bentz, G.L., Yurochko, A.D.,2014. HCMV reprogramming of infected monocyte survival and differentiation: aGoldilocks phenomenon. Viruses 6 (2), 782–807.

Stoffels, K., Overbergh, L., Giulietti, A., Verlinden, L., Bouillon, R., Mathieu, C., 2006.Immune regulation of 25-hydroxyvitamin-D3-1alpha-hydroxylase in humanmonocytes. J. Bone Miner. Res. 21 (1), 37–47.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–9594

Streblow, D.N., Orloff, S.L., Nelson, J.A., 2001. Do pathogens accelerate athero-sclerosis? J. Nutr. 131 (10), 2798S–2804S.

Swindle, E.J., Hunt, J.A., Coleman, J.W., 2002. A comparison of reactive oxygenspecies generation by rat peritoneal macrophages and mast cells using thehighly sensitive real-time chemiluminescent probe pholasin: inhibition ofantigen-induced mast cell degranulation by macrophage-derived hydrogenperoxide. J. Immunol. 169 (10), 5866–5873.

Takahashi, T., Nakamura, K., Iho, S., 1997. Differentiation of myeloid cells and 1,25-dihydroxyvitamin D3. Leuk. Lymphoma 27 (1–2), 25–33.

Taylor-Wiedeman, J., Sissons, J.G., Borysiewicz, L.K., Sinclair, J.H., 1991. Monocytesare a major site of persistence of human cytomegalovirus in peripheral bloodmononuclear cells. J. Gen. Virol. 72 (Pt 9), 2059–2064.

Terrier, B., Derian, N., Schoindre, Y., Chaara, W., Geri, G., Zahr, N., Mariampillai, K.,Rosenzwajg, M., Carpentier, W., Musset, L., Piette, J.C., Six, A., Klatzmann, D.,Saadoun, D., Patrice, C., Costedoat-Chalumeau, N., 2012. Restoration of regula-tory and effector T cell balance and B cell homeostasis in systemic lupuserythematosus patients through vitamin D supplementation. Arthritis Res.Ther. 14 (5), R221.

Turtinen, L.W., Seufzer, B.J., 1994. Selective permissiveness of TPA differentiatedTHP-1 myelomonocytic cells for human cytomegalovirus strains AD169 andTowne. Microb. Pathog. 16 (5), 373–378.

Van Damme, E., Sauviller, S., Lau, B., Kesteleyn, B., Griffiths, P., Burroughs, A.,Emery, V., Sinclair, J., Van Loock, M., 2014. Glucocorticosteroids trigger reactivationof human cytomegalovirus from latently infected myeloid cells and increase therisk for HCMV infection in DþRþ liver transplant patients. J. Gen. Virol..

Varani, S., Frascaroli, G., Landini, M.P., Soderberg-Naucler, C., 2009. Humancytomegalovirus targets different subsets of antigen-presenting cells withpathological consequences for host immunity: implications for immunosup-pression, chronic inflammation and autoimmunity. Rev. Med. Virol. 19 (3),131–145.

Verway, M., Bouttier, M., Wang, T.T., Carrier, M., Calderon, M., An, B.S., Devemy, E.,McIntosh, F., Divangahi, M., Behr, M.A., White, J.H., 2013. Vitamin D inducesinterleukin-1beta expression: paracrine macrophage epithelial signaling con-trols M. tuberculosis infection. PLoS Pathog. 9 (6), e1003407.

Wathen, M.W., Stinski, M.F., 1982. Temporal patterns of human cytomegalovirustranscription: mapping the viral RNAs synthesized at immediate early, early,and late times after infection. J. Virol. 41 (2), 462–477.

Weinshenker, B.G., Wilton, S., Rice, G.P., 1988. Phorbol ester-induced differentiationpermits productive human cytomegalovirus infection in a monocytic cell line.J. Immunol. 140 (5), 1625–1631.

White, D.W., Suzanne Beard, R., Barton, E.S., 2012. Immune modulation duringlatent herpesvirus infection. Immunol. Rev. 245 (1), 189–208.

Yen, M., Chen, J., Ausayakhun, S., Kunavisarut, P., Vichitvejpaisal, P., Jirawison, C.,Shantha, J., Holland, G.N., Heiden, D., Margolis, T.P., Keenan, J.D., 2015. Retinaldetachment associated with AIDS-related cytomegalovirus retinitis: risk factorsin a resource-limited setting. Am. J. Ophthalmol. 159 (1), 185–192.

Yuzefpolskiy, Y., Baumann, F.M., Penny, L.A., Studzinski, G.P., Kalia, V., Sarkar, S.,2014. Vitamin d receptor signals regulate effector and memory CD8 T cellresponses to infections in mice. J. Nutr. 144 (12), 2073–2082.

Zhang, C.F., Wan, R.Z., Liu, Z.P., 2013. Recent developments of 19-nor-1,25-dihy-droxyvitamin D3 analogues. ChemMedChem 8 (8), 1249–1260.

Zhang, Y., Leung, D.Y., Goleva, E., 2013. Vitamin D enhances glucocorticoid action inhuman monocytes: involvement of granulocyte-macrophage colony-stimulat-ing factor and mediator complex subunit 14. J. Biol. Chem. 288 (20),14544–14553.

Zhang, Y., Leung, D.Y., Goleva, E., 2014. Anti-inflammatory and corticosteroid-enhancing actions of vitamin D in monocytes of patients with steroid-resistantand those with steroid-sensitive asthma. J. Allergy Clin. Immunol. 133 (6),1744–1752.e1.

S.-E. Wu, W.E. Miller / Virology 483 (2015) 83–95 95