Proteasome inhibitor MG132 blocks viral DNA replication and ... - Core

Download PDF
3 downloads 0 Views 366KB Size Report
Comment
Jan 31, 2008 - Marion Kasparia, Nina Tavalaib, Thomas Stammingerb, Albert Zimmermannc,. Rita Schilfa, Elke Bognera,* a Institute of Virology, Charité ...
FEBS Letters 582 (2008) 666–672

Proteasome inhibitor MG132 blocks viral DNA replication and assembly of human cytomegalovirus Marion Kasparia, Nina Tavalaib, Thomas Stammingerb, Albert Zimmermannc, Rita Schilfa, Elke Bognera,* b

a Institute of Virology, Charite´ Campus Mitte, Universita¨tsmedizin Berlin, Charite´platz 1, 10117 Berlin, Germany Institut of Clinical and Molecular Virology, Friedrich-Alexander Universita¨t Erlangen-Nu¨rnberg, Erlangen, Germany c Institute of Virology, Heinrich-Heine-Universita¨t Du¨sseldorf, Germany

Received 20 December 2007; revised 14 January 2008; accepted 24 January 2008 Available online 31 January 2008 Edited by Jacomine Krijnse-Locker

Abstract This study provides evidence that proteasomal activity is required at multiple steps in human cytomegalovirus replication. Electron microscopy revealed that no viral particles were assembled in the presence of proteasome inhibitor MG132. Immunofluorescence and Western blot analyses using MG132 demonstrated that immediate early gene expression was suppressed at low but not high MOI. In contrast, expression of late proteins was completely blocked independent of MOI. Additionally, pulsed-field gel electrophoresis demonstrated that MG132 interferes with cleavage of HCMV DNA. Bromodeoxyuridine incorporation studies showed that de novo viral DNA synthesis is reduced in the presence of MG132. Furthermore, in contrast to previous hypotheses we demonstrated that neither the ND10 components PML and hDaxx nor NFjB activation represent the target for MG132.  2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Human cytomegalovirus; Viral DNA replication; Proteasome inhibitors; Antiviral target

1. Introduction Viruses have evolved multiple strategies to take over host cell pathways and promote viral replication. The ubiquitinproteasome system (UPS) seems to be a cellular pathway that viruses utilize for their own benefit. Previous studies have reported that proteasome inhibitors block replication of various viruses by targeting different steps of the replication cycle. For example, it has been demonstrated that proteasome inhibitors prevent translocation of influenza virus to the nucleus [1]. In the case of HSV-1, proteasome inhibitors decreased immediate early and late protein expression [2]. Watanabe et al. reported that paramyxovirus maturation is blocked by proteasome inhibitors [3], while the UPS affects budding events of rhabdoviruses [4] and HIV [5]. Pro¨sch et al. recently discussed the suppression of HCMV replication and virus induced immunomodulation by proteasome inhibitors [6]. HCMV has a sequential regulation of viral gene expression, leading to induction and repression cycles occurring in the immediate early (IE), early (E) and late phase (L) of viral replication. IE1 and IE2 induce expression of early proteins, mediate G1/S cell cycle arrest and host replication shut-off [7,8]. Early genes pre* Corresponding author. Fax: +49 30 450 525907. E-mail address: [email protected] (E. Bogner).

dominantly encode viral DNA replication factors, repair enzymes or immune evasion proteins, and induce the expression of L genes [9,10]. L proteins down-regulate expression of E genes and are mainly involved in virion assembly. HCMV DNA replication is thus the key step in maturation of new virions. In this report we characterize the impact of proteasome inhibition on viral protein expression. We also investigate the influence of MG132 on viral DNA replication and virion assembly.

2. Materials and methods 2.1. Cells and viruses Human embryonic lung fibroblasts (HELF) were grown in EagleÕs minimum essential medium supplemented with 2% fetal calf serum (FCS), 2 mM L -glutamine, 1 mM sodium pyruvate, 1· non-essential amino acids and gentamicin sulfate (50 lg/ml). Stably transduced human foreskin fibroblasts (HFF) were grown in DulbeccoÕs Modified EagleÕs Medium supplemented with 10% FCS, L -glutamine (350 lg/ ml), gentamicin sulfate (10 lg/ml) and puromycin (5 lg/ml). Infection of cells with HCMV AD169 was carried out as described previously [9]. Proteasome inhibitors were added after adsorption. 2.2. Proteasome inhibitors MG132 and lactacystin (Calbiochem) were dissolved in 95% (w/v) EtOH and DMSO respectively. Stock solutions were stored at 20 C before use. Cell toxicity (CC50) of MG132 and lactacystin were determined for HELF using the Cell Proliferation Kit II (Roche) according to the manufacturerÕs recommendation. 2.3. Construction of a recombinant HCMV The virus mutant HB5D NFjB lacking the four NFjB binding sites within the major immediate early (MIE) enhancer/promoter was constructed in two steps from the AD169 BACmid pHB5 [11] using procedures described previously [12]. For construction of pHB5-BACmid pHB5DIEtet a PCR fragment was generated from the contiguous primers AZ-IE-tet1 (5 0 ACGTACCGTGGCACCTTGGAGGAAGGGCCCTCGTCAGGATTATCAGGGTCCATCTTTCTCCCAGTGAATTCGAGCTCGGTAC-3 0 ) and AZ-IE-tet2 (5 0 ACGGGGTCATTAGTTCATAGCCCATATATGGAGTTCCGCGTTACATAACTTACGGTAAATGACCATGATTACGCCAAGCTCC-3 0 ). The PCR fragment containing a tetracycline resistance gene was inserted into pHB5 by homologous recombination in Escherichia coli, replacing the IE region between nucleotides 173678 and 175159 (AD169 sequence) by the tetracycline resistance gene. The 4773 bp SmaI–EcoRI fragment from pCM5801 containing the 11 kb EcoRI-J fragment of AD169 representing the IE region including the MIE enhancer/ promoter was subcloned into pUC19 generating plasmid pIESE. A SacI–BglI fragment from p18mut containing the enhancer region in which the four putative NFjB binding sites were destroyed by site directed mutagenesis was cloned into pIESE. The newly generated plasmid was designated pIED18. The insert was isolated by EcoRI–SmaI digest and recloned in pET-21d (Novagen) generating pET-IEDNFjB.

0014-5793/$34.00  2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2008.01.040

M. Kaspari et al. / FEBS Letters 582 (2008) 666–672 From this plasmid a BamHI–BglI fragment was isolated and cloned into the BamHI cleaved shuttle mutagenesis vector pST76KSR. The mutated MIE region from plasmid pST76KSR-IEDNFjB was inserted into pHB5DIEtet by two-step homologous recombination [13] resulting in BAC pHB5DNFjB. Recombinant clones were isolated after selection on tetracycline containing agar plates. Mutagenesis of BAC pHB5DNFjB was validated by restriction analysis and sequencing of the MIE region, confirming the deletion of NFjB binding sites. Recombinant virus was reconstituted by transfection of MRC-5 using Superfect reagent (Qiagen). 2.4. Immunofluorescence HELF (1 · 105) were grown on cover slips, infected at MOI 0.1 or 1 and treated with 0.5 lM MG132 or left untreated. At 72 h p.i. cells were fixed with 3% paraformaldehyde [14]. Staining for IE1, pUL44 and MCP was performed using mAbs p63-27, BS510 and 28-4, respectively, for 45 min at RT prior to incubation with Cy3-labeled goat antimouse F(ab 0 )2 fragments for 45 min. Nuclei were stained using DAPI (1 lg/ml). Samples were mounted in Fluoroprep (bioMerieux) containing 2.5% (w/v) 1,4-diazabicyclo[2.2.2]octan and examined using an Olympus BX 50 fluorescence microscope. Images were captured with an Olympus Colorview II camera in conjunction with the Cell D software program (Olympus). 2.5. BrdU labeling and immunostaining HELF (1 · 105) were grown on cover slips, infected at MOI 1 and treated with 0.05–0.5 lM MG132 or left untreated. BrdU (10 lM) was added at 45 h p.i. Cells were fixed at 72 h p.i. and DNA was denatured in 2 N HCl for 30 min at 37 C followed by neutralization with 1· TBE. Immunostaining was performed using mAb a-BrdU (Dianova) for 45 min at RT prior to incubation with Cy3-labeled goat antimouse F(ab 0 )2 fragments. Nuclei were stained using DAPI (1 lg/ml). Samples were mounted as described above and analyzed using a Zeiss Laser Scanning Microscope LSM 510/ConfoCor2. Images were processed using the Axioplan software program (Zeiss). 2.6. Proteasome activity assay Cell extracts of HELF (1 · 106) were prepared using lysis buffer (20 mM Tris-HCl pH 7.5, 1 mM EDTA, 1 mM dithiothreitol, 50 mM NaCl, 0.1% NP40) followed by sedimentation of cell fragments. Supernatant was treated with protease inhibitor cocktail (Roche). Chymotrypsin-like proteasome activity was determined using the substrate Succ-LLVY-AMC (Bachem). Cell extracts were incubated for 30 min with assay buffer (20 mM Tris–HCl pH 7.5, 1 mM EDTA) containing 50 lM substrate. Released AMC was detected using a fluorometer (BioTek Instruments) in conjunction with the KC4 software program. Proteasome activity was calculated as fluorescence units/30 min/lg protein. 2.7. Plaque reduction assay HELF (5 · 105) were seeded in 12-well plates and infected at MOI 0.001. After adsorption the inoculum was replaced with semisolid medium containing 0.05–0.5 lM MG132. At day 7 p.i. cells were stained with mAb p63-27 against IE1 for 1 h at 37 C prior to incubation with HRP-conjugated anti-mouse F(ab 0 )2 fragments for 2 h. Nuclei of infected cells were visualized using the AEC Staining Kit (Sigma). Plaques were counted using a light microscope. 2.8. SDS–PAGE and Western blot Extracts from mock-infected or HCMV infected HELF (1 · 106; different MOI) treated with 0.5 lM MG132 or left untreated were separated on 10% polyacrylamide gels, transferred onto nitrocellulose membranes and subjected to Western blot analysis. Membranes were incubated with mAbs E13, BS510, 27-156, 28-4 and 58-15 against IE1/IE2, pUL44, gB, MCP and pp65, respectively, prior to incubation with HRP-conjugated anti-mouse F(ab 0 )2 fragments. Detection of protein bands was performed using ECL reagent (Pierce). Membranes were re-probed for b-actin to verify equal loading. 2.9. Pulsed-field gel electrophoresis (PFGE) HELF (1 · 106) were seeded in 25 cm2 flasks, infected at MOI 1, treated with 0.05–0.5 lM MG132 or left untreated and prepared for PFGE as described previously [15]. Samples and Lambda Ladder PFG Marker

667 (New England Biolabs) were loaded into a 1% LMP agarose gel for electrophoresis (15–110 s, 150 V, 27 h). Upon completion of electrophoresis DNA was stained with 1 lg/ml ethidium bromide and photographed. 2.10. Thin sectioning and electron microscopy HELF (1 · 106) were seeded in 25 cm2 flasks and infected at MOI 1 in the presence or absence of 0.5 lM MG132. Cells were fixed at 72 h p.i. and embedded as described before [13]. Sections were analyzed using a Tecnai G2 (FEI Company, Eindhoven) electron microscope operated at 120 kV.

3. Results 3.1. Increased 20 S proteasome activity in HCMV infected HELF To characterize 20 S proteasome activity in HCMV infected cells, HELF were infected at MOI 0.05–2 and treated with 0.5 lM MG132 for 72 h or left untreated before measuring proteasome activity. An increased MOI was linked to increased proteasome activity (Fig. 1A). In the presence of MG132 proteasome activity was reduced by 70–80% at all MOIÕs examined (Fig. 1A). This experiment demonstrated that in HCMV infected cells, activity of the 20 S proteasome increases in an MOI-dependent fashion. 3.2. 20 S proteasome activity is required for HCMV replication For determination of IC50 values plaque assays were carried out. The mean IC50 values were 0.165 lM for MG132 and 13.530 lM for lactacystin (Table 1), demonstrating that HCMV replication is strongly inhibited by MG132 and less severely suppressed by lactacystin (Fig. 1B). Both compounds have a CC50 value 4- to 10-fold higher than the IC50 value (Table 1), suggesting that suppression of viral replication was not due to cytotoxic side effects of the drugs. To assure that the observed anti-HCMV effect of MG132 is indeed due to proteasome inhibition and not the result of inhibition of other cellular proteases, we carried out plaque assays using calpain inhibitor I or II (both inhibit calpain I and II, cathepsin B and L). Fig. 1C shows that HCMV replication was not suppressed by both calpain inhibitors, indicating that blocking these cellular proteases does not contribute to the antiviral effect of proteasome inhibitors. 3.3. 20 S proteasome activity is required at multiple stages of the replication cycle To determine the stage of viral replication that is sensitive to proteasome inhibition we analyzed the effect of time-delayed drug addition. HELF were infected at MOI 0.001 and MG132 (0.5 lM) was added at different times p.i. At day 7 p.i. viral replication was examined using plaque assays. If MG132 was added after adsorption (0 h p.i.) viral replication was inhibited by 95% (Fig. 1D). Addition of MG132 at 24 h or 48 h p.i. resulted in suppression of viral replication by 82% or 91%, respectively. If the drug was added at 72 h p.i. viral replication was reduced by 52%. Addition of MG132 at 96 h p.i. had no effect on HCMV replication. This experiment suggests that all stages of the viral replication cycle are sensitive to proteasome inhibition to different extents. 3.4. Inhibition of infectious particle formation by proteasome inhibitors To investigate the formation of viral particles under proteasome inhibition, electron microscopy of infected HELF (MOI

668

M. Kaspari et al. / FEBS Letters 582 (2008) 666–672

B

A 500

120

- MG132 + MG132

400 350

plaques (% control)

proteasome activity (%)

450

300 250 200 150

MG132 Lactacystin

100 80 60 40

100 20 50 0

0 w/o

MOI 0.01 MOI 0.05 MOI 0.1

MOI 1

0.0 0.1 0.2 0.3 0.4 0.5

MOI 2

5

10 15 20 25

proteasome inhibitor (µM)

C

D plaques (% control)

120 100 80

calpain inhibitor I

60 calpain inhibitor II

40

plaques (% control)

120 140

100 80 60 40 20

20 0

0 0

1

2

3

4

5

0

calpain inhibitor (µM)

20

40

60

80

100

MG132 addition (hours post infection)

Fig. 1. Analysis of 20 S proteasome inhibitors. (A) Analysis of 20 S proteasome activity in HCMV infected HELF. HELF were infected at MOI 0.05–2 and treated with 0.5 lM MG132 for 72 h (MG132) or left untreated (w/o) before measuring proteasome activity. (B and C) Influence of calpain and proteasome inhibitors on HCMV replication. HELF were infected at MOI 0.001 and analyzed for viral replication in the presence of (B) MG132 or lactacystin or (C) calpain inhibitor I or II using plaque assays. (D) Time-of-addition analysis of MG132. HCMV-infected HELF (MOI 0.001) were treated with 0.5 lM MG132 beginning from different time points p.i. (0–96 h). At day 7 p.i. cells were analyzed for viral replication using plaque assays. Error bars represent ± S.D. of three independent experiments.

Table 1 Antiviral activities of proteasome inhibitors Compound MG132 Lactacystin

Mean IC50 (lM)

Mean CC50 (lM)

Plaque reduction

Cell proliferation

0.165 13.530

1.05 57.00

1) treated with 0.5 lM MG132 or left untreated was carried out. All types of capsids (A-, B- and C-capsids; Table 2, Supplementary data) were found in the nuclei of untreated cells. Additionally, infectious particles and dense bodies were observed in the cytoplasm or in the extracellular space (Supplementary data). In contrast, MG132 treated cells did not show any viral particles (Table 2, Supplementary data), indicating that viral replication did not occur. These experiments suggested that MG132 prevents formation of viral particles. 3.5. Influence of MG132 on HCMV protein expression Since MG132 completely suppressed formation of infectious progeny virus we wanted to characterize the impact of protea-

Table 2 Nuclear capsids of infected cells in the absence or presence of 0.5 lM MG132 Virus

Number of nuclei counted

Number of capsid form per nucleus (%) A

B

C

AD169 AD169 + MG132

15 17

15.5 0.0

36.9 0.0

47.6 0.0

some inhibition on IE, E and L protein expression. Cover slip cultures were infected at MOI 0.1 and 1.0, respectively, in the presence or absence of 0.5 lM MG132 prior to immunofluorescence staining at 72 h p.i. Expression and distribution of viral proteins changed in the presence of MG132. At low MOI the number of IE1-positive cells was strongly reduced. However, the level of IE1 expression in these cells was comparable to the level achieved in the absence of the inhibitor. Additionally, distribution of pUL44 changed from replication compartments [16] into a

M. Kaspari et al. / FEBS Letters 582 (2008) 666–672

punctuate pattern representing pre-replication compartments. Moreover, a complete loss of MCP expression was observed (Fig. 2). Consistent with this MCP was no longer detectable in MG132 treated cells infected at high MOI (Fig. 2). Again, pUL44 accumulated in pre-replication compartments (Fig. 2). However, the expression of IE1 remained widely unchanged at high MOI. To further characterize the influence of MG132 on viral protein expression at different MOIÕs, extracts of mock- or HCMV infected HELF (MOI 0.1–2) treated with 0.5 lM MG132 or left untreated were subjected to Western blot analysis. At MOI 0.1 the expression of IE, E and L proteins was clearly inhibited by MG132 (Fig. 3). However, at higher MOI (MOI 0.5–2) expression of IE1 was no longer affected by proteasome inhibition, while IE2 expression levels were slightly reduced up to MOI 1 (Fig. 3). In contrast, expression of gB and MCP was completely inhibited at all MOIÕs examined (Fig. 3). Additionally, expression of pUL44 was strongly suppressed by MG132 at both low and high MOI (Fig. 3). Levels of pp65 were clearly reduced up to MOI 0.5 and gradually increased at higher MOIÕs but did not reach the level expressed in the untreated control (Fig. 3). These experiments revealed that the observed inhibition of infectious particle formation by MG132 is due to complete suppression of L protein expression. Furthermore, these data demonstrated that expression of IE and E proteins is reduced at low MOI. However, the effect of MG132 on IE protein expression was abolished at high MOI. 3.6. Influence of MG132 on cleavage of concatenated DNA To analyze the structure of viral DNA accumulated in the presence of increasing concentrations of MG132 (0.05–

MOI 0.1 IE1

UL44

669

mock

-

+

MOI 0.1

MOI 0.25

-

-

+

+

MG132 IE2 IE1

UL44

pp65

gB MCP Actin

MOI 0.5

MOI 1

-

-

+

MOI 2

+

-

+

MG132 IE2 IE1

UL44

pp65

gB MCP

MCP Actin

w/o Fig. 3. HCMV protein expression in MG132 treated cells is MOIdependent. HELF were mock- or HCMV infected (MOI 0.1–2) and cultured with or without 0.5 lM MG132. At 72 h p.i. cell extracts were subjected to Western blot analysis using antibodies against IE1/IE2, pUL44, pp65, gB and MCP. Membranes were re-probed for b-actin to verify equal loading. Arrows indicate the position of proteins.

MG132

MOI 1 IE1

UL44

MCP

w/o

MG132

Fig. 2. Influence of MG132 on HCMV protein expression. HCMV infected HELF (MOI 0.1 or 1) were cultured in the presence (MG132) or absence (w/o) of 0.5 lM MG132 for 72 h and subjected to immunofluorescence using antibodies against IE1, pUL44 and MCP.

0.5 lM), DNA of infected HELF (MOI 1) was separated by PFGE and stained with ethidium bromide. PFGE analysis showed that cleavage of concatemeric DNA into unit-length genomes occurs in the presence of low MG132 concentrations (60.15 lM) (Fig. 4A, lanes 4–6). However, increasing MG132 concentrations (P0.25 lM) resulted in loss of unit-length genomes (Fig. 4A, lanes 7–8). These data indicated that high concentrations of MG132 inhibit viral DNA cleavage. 3.7. MG132 prevents viral DNA replication To determine whether any DNA replication occurs in the presence of MG132 we performed analyses using the nucleoside BrdU. Due to the virus-host shut-off in HCMV infected cells BrdU is preferably incorporated into newly synthesized viral DNA. HCMV infected HELF (MOI 1) were cultured with or without MG132 (0.05–0.5 lM) for 45 h before adding

670

M. Kaspari et al. / FEBS Letters 582 (2008) 666–672

0.50 µM

0.10 µM

0.15 µM 0.25 µM

w/o w/o w/o

M (kb)

0.05 µM

AD169

mock

A

242.5

+ MG132

monomeric DNA

194.0

145.5 1

B

EtOH

0.15 μM

2 3 4 5 6

0.05 μM

0.25 μM

7

8

0.1 μM

0.50 μM

Fig. 4. Analysis of viral DNA cleavage and de novo DNA synthesis. (A) HELF were mock- or HCMV infected (MOI 1) and cultured in the absence (w/o) or presence of the indicated MG132 concentrations. At 72 h p.i. cells were harvested for PFGE. Lambda Ladder PFG Marker was used to determine the size of HCMV genome monomers. (B) HCMV infected HELF (MOI 1) were cultured with the indicated MG132 concentrations or with solvent alone (EtOH) for 45 h before adding 10 lM BrdU. At 72 h p.i. cells were subjected to immunofluorescence using mAb a-BrdU.

10 lM BrdU. At 72 h p.i. cells were stained with mAb a-BrdU. Untreated cells showed a BrdU staining pattern of bright nuclear patches (Fig. 4B). This localization is consistent with ongoing viral DNA replication. Conversely BrdU staining was clearly reduced in the presence of MG132 and restricted to small dot-like areas in the nucleus (Fig. 4B), indicating that MG132 strongly inhibits viral DNA synthesis, hence causing an early block in viral maturation. 3.8. Role of ND10 components in the anti-HCMV effect of MG132 The observation that IE expression was sensitive to proteasome inhibition at low but not at high MOI led us to the question whether Nuclear Domain 10 (ND10) components including PML or hDaxx could be involved in the anti-HCMV

effect of MG132. Moreover, Saffert and Kalejta reported that stabilization of hDaxx by proteasome inhibitors blocks IE protein expression at low but not at high MOI [17]. To address this question we determined the influence of shRNA targeting hDaxx and PML [18] on IE1 expression. shRNA expressing HFF (hDaxx-kd, PML-kd cells) were infected at MOI 0.003 and treated with 0.5 lM MG132 or left untreated. HFF with integrated vector sequence (vector) or non-functional shRNA (siC) [18] served as control and were treated in the same way. At 24 h p.i. IE1 expression was analyzed by immunofluorescence. IE1 expression was still sensitive to MG132 in the absence of PML and hDaxx (Fig. 5A). Interestingly, the number of IE1positive cells in MG132 treated PML-kd and hDaxx-kd HFF was slightly increased compared to control HFF (Fig. 5A),

M. Kaspari et al. / FEBS Letters 582 (2008) 666–672

100

671

mutant should be resistant to proteasome inhibition. However, replication of HB5DNFjB was suppressed by MG132 (Fig. 5B), suggesting that the block of viral replication is achieved via a different mechanism than inhibition of NFjB activation.

w/o

IE1 positive cells/sample

MG132 80

4. Discussion 60

40

20

0 vector

siC

siPML2

siDaxx1

plaque number (% control)

140 120 100

HCMV-BACHB5 HCMV-BAC-HB5ΔNF-κB

80 60 40 20 0 0

0.1

0.2

0.3

0.4

0.5

MG132 (µM)

Fig. 5. Target analysis of MG132. (A) PML-kd HFF (siPML2), hDaxx-kd HFF (siDaxx1) and control HFF (vector, siC) were infected with HCMV AD169 at MOI 0.003 and treated with 0.5 lM MG132 (MG132) or left untreated (w/o). At 24 h p.i. cells were analyzed for IE1 expression using immunofluorescence. (B) HELF were infected with HCMV AD169 BACmids HB5 or HB5DNFjB at MOI 0.001 and analyzed for viral replication in the presence of the indicated MG132 concentrations using plaque assays. Error bars represent ± S.D. of three independent experiments.

supporting the hypothesis that PML and hDaxx exert antiviral functions [19,20]. However, knockdown of these proteins did not abrogate the anti-HCMV effect of MG132 suggesting the existence of additional viral targets for proteasome inhibitors. 3.9. Effect of MG132 on replication of an HCMV NFjB null mutant NFjB is known to stimulate transcription of IE genes due to four NFjB binding sites within the MIE enhancer/promoter [21]. Pro¨sch et al. previously established the hypothesis that suppression of HCMV replication by MG132 results from blocking NFjB activation [6]. In order to test this hypothesis we performed plaque assays using the HCMV AD169 BACmid HB5 and the mutant HB5DNFjB in which all four NFjB binding sites are abolished by site directed mutagenesis. If the antiviral effect of MG132 is caused by blocking NFjB activation, this

It was recently demonstrated that proteasome inhibitors suppress HCMV replication [6]. On the basis of this observation, we further explored the role of the proteasome in HCMV replication. Since most proteasome inhibitors also block other cellular proteases [22], we wanted to confirm that specifically 20 S proteasome activity is required for virus replication. Lactacystin, the most specific proteasome inhibitor known was shown to inhibit viral replication. Conversely, calpain inhibitors did not block viral replication, suggesting that it is inhibition of the proteasome that causes suppression of HCMV replication by MG132. Further evidence for the importance of the proteasome during HCMV replication was obtained by analyzing 20 S proteasome activity in infected HELF. Proteasome activity increased in infected HELF, which might be beneficial for HCMV as it could lead to a rapid turnover of viral or cellular proteins required for viral replication. To determine the stage of viral replication that is sensitive to proteasome inhibition we performed a time-of-addition study with MG132. Interestingly, all phases of viral replication turned out to be sensitive to MG132 treatment. The most noticeable effect was observed when the inhibitor was added in the IE or E phase. However, even if MG132 was added at late times post infection, viral replication was reduced by approximately 50%. Thus, proteasome activity seems to be required at multiple steps in HCMV replication. We also assessed whether viral replication is still accomplished in the presence of proteasome inhibitors by electron microscopy analyses. However, neither infectious nor noninfectious viral particles were detected in MG132 treated sections, indicating a complete breakdown in virus assembly. Subsequently we investigated whether this block was caused by suppression of viral protein expression using immunofluorescence and Western blot analyses. Expression of late structural proteins including MCP and gB was completely abolished at both high and low MOI, thus explaining the absence of progeny virus. In contrast to the results observed with L proteins, IE protein expression was suppressed at low MOI; however, this repression was abolished at MOI 1. Our observations are consistent with a report from Saffert et al. showing that stabilization of hDaxx by proteasome inhibitors blocks IE1 expression at low but not at high MOI [17]. However, we showed that IE1 expression remained sensitive to proteasome inhibition in HFF with a stable knockdown of hDaxx and PML, respectively, suggesting the existence of other targets which have an additional impact on viral replication. Expression of early proteins was severely affected by MG132 at both low and high MOI. Using immunofluorescence staining against pUL44, we observed only pre-replication compartments in the presence of MG132, indicating a block in viral replication. To verify this result, incorporation studies of BrdU into newly synthesized viral DNA were carried out. These studies showed that de novo DNA synthesis was

672

strongly inhibited by MG132. Using PFGE we analyzed whether MG132 also has an influence on DNA packaging. Interestingly, MG132 inhibited cleavage of concatemeric viral DNA in a concentration-dependent manner. In order to identify the underlying mechanism we investigated the role of NFjB in the anti-HCMV effect of MG132. Plaque assays were performed using the HCMV mutant HB5DNFjB carrying a deletion of the four NFjB binding sites in the MIE enhancer/promoter. However, replication of this mutant was still sensitive to proteasome inhibition. Our results indicate that inhibition of NFjB activation is not the mode of action for proteasome inhibitors in HCMV replication. In summary, these results indicate that proteasome activity is required for multiple steps in viral replication, thus representing a new target for antiviral therapy. Further analyses would aim to identify both the key protein whose stabilization by proteasome inhibitors blocks viral replication and the corresponding E3 ubiquitin ligase. By inhibiting the particular E3 enzyme, interference with HCMV replication would be more specific and unwanted side effects could be avoided. Acknowledgements: This study was supported by Deutsche Forschungsgemeinschaft (Bo 1214/5-4) and Sonnenfeld-Stiftung. We thank N. Bannert (Robert Koch-Institut, Berlin) for unlimited access to his EM-laboratory and are grateful to G. Holland for performing the sectioning. We acknowledge W. Britt (University of Alabama, Birmingham) for providing mAbs p63-27, 28-4 and BS510. We are grateful to T. Stamminger (Friedrich-Alexander Universita¨t, Erlangen-Nu¨rnberg) for providing knockdown HFF. E.B. thanks D. Kru¨ger for kind support.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet. 2008.01.040. References [1] Khor, R., McElroy, L.J. and Whittaker, G.R. (2003) The ubiquitin-vacuolar protein sorting system is selectively required during entry of influenza virus into host cells. Traffic 4, 857– 868. [2] La Frazia, S., Mici, C. and Santoro, M.G. (2006) Antiviral activity of proteasome inhibitors in herpes simplex virus-1 infection: role of nuclear factor-kB. Antiviral Ther. 11, 995–1004. [3] Watanabe, H., Tanaka, Y., Shimazu, Y., Sugahara, F., Kuwayama, M., Hiramatsu, A., Kiyotani, K., Yoshida, T. and Sakaguchi, T. (2005) Cell-specific inhibition of paramyxovirus maturation by proteasome inhibitors. Microbiol. Immunol. 49, 835–844. [4] Harty, R.N., Brown, M.E., McGettigan, J.P., Wang, G., Jayakar, H.R., Huibregtse, J.M., Whitt, M.A. and Schnell, M.J. (2001) Rhabdoviruses and the cellular ubiquitin-proteasome system: a budding interaction. J. Virol. 75, 10623–10629. [5] Schubert, U., Ott, D.E., Chertova, E.N., Welker, R., Tessmer, U., Princiotta, M.F., Bennink, J.R., Kra¨usslich, H.G. and Yewdell, J.W. (2000) Proteasome inhibition interferes with Gag polyprotein processing, release, and maturation of HIV-1 and HIV-2. Proc. Natl. Acad. Sci. USA 97, 13057–13062.

M. Kaspari et al. / FEBS Letters 582 (2008) 666–672 [6] Pro¨sch, S., Priemer, C., Ho¨flich, C., Liebenthal, C., Babel, N., Kru¨ger, D.H. and Volk, H.D. (2003) Proteasome inhibition: a novel tool to suppress human cytomegalovirus replication and virus-induced immune modulation. Antiviral Ther. 8, 555–567. [7] Dittmer, D. and Mocarski, E.S. (1997) Human cytomegalovirus infection inhibits G1/S transition. J. Virol. 71, 1629–1634. [8] Bresnahan, W.A., Boldogh, I., Thompson, A.E. and Albrecht, T. (1996) Human cytomegalovirus inhibits cellular DNA synthesis and arrests productively infected cells in late G1. Virology 224, 150–160. [9] Bogner, E., Reschke, M., Reis, B., Reis, E., Britt, W. and Radsak, K. (1992) Recognition of compartmentalized intracellular analogs of glycoprotein H of human cytomegalovirus. Arch. Virol. 126, 67–80. [10] Mocarski, E.S. and Courcelle, C.T. (2001) Cytomegalovirus and their replication, in: Fields Virology (Knipe, D. and Howley, T., Eds.), pp. 2629–2673, Lippincott, Williams and Wilkins, Philadelphia. [11] Borst, E.M., Hahn, G., Koszinowski, U.H. and Messerle, M. (1999) Cloning of the human cytomegalovirus (HCMV) genome as an infectious bacterial artificial chromosome in Escherichia coli: a new approach for construction of HCMV mutants. J. Virol. 73, 8320–8329. [12] Wagner, M., Gutermann, A., Podlech, J., Reddehase, M.J. and Koszinowski, U.H. (2002) Major histocompatibility complex class I allele-specific cooperative and competitive interactions between immune evasion proteins of cytomegalovirus. J. Exp. Med. 196, 805–816. [13] Messerle, M., Crnkovic, I., Hammerschmidt, W., Ziegler, H. and Koszinowski, U.H. (1997) Cloning and mutagenesis of a herpesvirus genome as an infectious bacterial artificial chromosome. Proc. Natl. Acad. Sci. USA 94, 14759–14763. [14] Smuda, C., Bogner, E. and Radsak, K. (1996) The human cytomegalovirus glycoprotein B gene (ORF UL55) is expressed early in the infectious cycle. J. Gen. Virol. 78, 1981–1992. [15] Dittmer, A., Drach, J.C., Townsend, L.B., Fischer, A. and Bogner, E. (2005) Interaction of the putative HCMV portal protein pUL104 with the large terminase subunit pUL56 and its inhibition by Benzimidazole-D -ribonucleosides. J. Virol. 79, 14660–14667. [16] Ahn, J.H., Jang, J.W. and Hayward, G.S. (1999) The human cytomegalovirus IE2 and UL112-113 proteins accumulate in viral DNA replication compartments that initiate from the periphery of promyelocytic leukemia protein-associated nuclear bodies (PODs or ND10). J. Virol. 73, 10458–10471. [17] Saffert, R.T. and Kalejta, R.F. (2006) Inactivating a cellular intrinsic immune defense mediated by Daxx is the mechanism through which the human cytomegalovirus pp71 protein stimulates viral immediate-early gene expression. J. Virol. 80, 3863–3871. [18] Tavalai, N., Papior, P., Rechter, S., Leis, M. and Stamminger, T. (2006) Evidence for a role of the cellular ND10 protein PML in mediating intrinsic immunity against human cytomegalovirus infections. J. Virol. 80, 8006–8018. [19] Croxton, R., Puto, L.A., de Belle, I., Thomas, M., Torii, S., Hanaii, F., Cuddy, M. and Reed, J.C. (2006) Daxx represses expression of a subset of antiapoptotic genes regulated by nuclear factor-kappaB. Cancer Res. 66, 9026–9035. [20] Everett, R.D., Rechter, S., Papior, P., Tavalai, N., Stamminger, T. and Orr, A. (2006) PML contributes to a cellular mechanism of repression of herpes simplex virus type 1 infection that is inactivated by ICP0. J. Virol. 80, 7995–8005. [21] Pro¨sch, S., Staak, K., Stein, J., Liebenthal, C., Stamminger, T., Volk, H.D. and Kru¨ger, D.H. (1995) Stimulation of human cytomegalovirus IE enhancer/promoter in HL60 cells by TNFa is mediated via induction of NF-kB. Virology 208, 197–206. [22] Tsubuki, S., Saito, Y., Tamioka, M., Ito, H. and Kawashima, S. (1996) Differential inhibition of calpain and proteasome activities by peptidyl aldehydes of dileucine and tri-leucine. J. Biochem. 119, 572–576.