You are on page 1of 13

Oncogene (2013) 32, 1626–1637

& 2013 Macmillan Publishers Limited All rights reserved 0950-9232/13

Cross-talk between phosphorylation and SUMOylation regulates
transforming activities of an adenoviral oncoprotein
P Wimmer1, P Blanchette2, S Schreiner1, W Ching1, P Groitl1, J Berscheminski1, PE Branton2,3,4, H Will1 and T Dobner1

Since the discovery of post-translational modification (PTM) by the small ubiquitin-related modifiers (SUMOs), a multitude of
proteins have been described to be reversibly modified, resulting in the alteration of several cellular pathways. Interestingly, various
pathogens gain access to this modification system, although the molecular mechanisms and functional consequences are barely
understood. We show here that the adenoviral oncoprotein E1B-55K is a substrate of the SUMO conjugation system, which is
directly linked to its C-terminal phosphorylation. This regulative connection is indispensable for modulation of the tumor
suppressor p53/chromatin-remodeling factor Daxx by E1B-55K and, consequently, its oncogenic potential in primary mammalian
cells. In virus infection, E1B-55K PTMs are necessary for localization to viral transcription/replication sites. Furthermore, we identify
the E2 enzyme Ubc9 as an interaction partner of E1B-55K, providing a possible molecular explanation for SUMO-dependent
modulation of cellular target proteins. In conclusion, these results for the first time provide evidence how E1B-55K PTMs are
regulated and subsequently facilitate exploitation of the host cell SUMOylation machinery.

Oncogene (2013) 32, 1626–1637; doi:10.1038/onc.2012.187; published online 21 May 2012
Keywords: transformation; oncoprotein; SUMO; phosphorylation

INTRODUCTION Intriguingly, several pathogens take advantage of the cellular
Ubiquitin-like proteins compose one of the most prominent SUMOylation machinery, either by modulating essential compo-
families of small modifiers.1 One of its members, termed small nents or being themselves targets of this PTM (reviewed in Boggio
ubiquitin-related modifier (SUMO), has been discovered in the et al.19 Wilson et al.20 and Wimmer et al.21). As described
1990s and extensively characterized by a lot of different previously, E1B-55K undergoes two principle PTMs (Figure 1b),
laboratories. Currently, modification by SUMO-1, SUMO-2 and namely SUMO-1 conjugation at a SUMO-1 conjugation motif (SCM;
SUMO-3 has been shown to regulate a diverse number of cellular c-K104-x-E/D)22–26 and phosphorylation by casein kinase 2 (CK2)
processes, including protein degradation, chromatin structure, within its C-terminal phosphorylation region (CPR).27–29
subnuclear localization, transcription factor activity, protein– Interestingly, mutational inactivation of either the E1B-55K SCM
protein interaction, nuclear/cytoplasmic shuttling and the integrity or CPR leads to remarkably similar alterations in the molecular
of promyelocytic leukemia (PML)-nuclear bodies (PML-NBs).2–12 functions of the viral protein. Both the SUMOylation-deficient
Although detailed data about the enzymatic machinery of mutant E1B-55K-SCM30 and the phosphorylation-deficient mutant
SUMO conjugation are available (Figure 1a), many questions E1B-55K-S490/1/T495A27,28 (Figure 1c), alternatively termed E1B-
remain to be answered about the regulation and corresponding 55K-3XPA,31 fail to efficiently repress p53-mediated transacti-
downstream effects of SUMOylation. Moreover, the observation vation, nucleo-cytoplasmic relocalization of p53 and thus cellular
that multiple cellular pathways are extensively regulated by this transformation of primary mammalian cells. Although phos-
post-translational modification (PTM), while only a low percentage phorylation of substrates has been shown consistently to be a
of effector proteins are modified, represents one of the most prerequisite for efficient SUMO modification,32–34 a comparable
puzzling questions, aptly termed as SUMO enigma.6 mechanism linking both PTMs of E1B-55K has not been described
The human Adenovirus type 5 early region 1B 55KDa (HAdV5 yet. Moreover, the viral protein harbors a leucine-rich nuclear
E1B-55K) is a multifunctional phosphoprotein, mediating several export sequence (NES) of the HIV-1 rev type (Figure 1b).22,25,26
critical steps during the early and late phase of infection Interestingly, functional inactivation of the E1B-55K-NES was
(reviewed in Berk 200713 and Berk 200514). Initially, E1B-55K shown to induce enhanced modification by SUMO-1,
counteracts antiproliferative processes induced by the host cell, modulation of p53 activity and augmented transformation of
including activation of p53-dependent and independent primary mammalian cells.22,23,35,36 Further results link E1B-55K
apoptosis, induction of cell cycle arrest and stimulation of SUMOylation to interact/modulate certain isoforms of the
cellular DNA damage response (reviewed in Weitzman 2005 tumor suppressor protein PML37 and to the degradation of the
et al.15 and White 200116). In the late phase, the adenoviral chromatin-remodeling factor Daxx.38
protein stimulates efficient cytoplasmic accumulation and In summary, these results suggest that SUMO modification
translation of viral late mRNAs (reviewed in Dobner 2001 directly impacts on several important functions of the adenoviral
et al.17and Flint 2003 et al.18). protein, although the respective regulatory mechanisms remain

Department of Molecular Virology, Heinrich–Pette–Institute—Leibniz-Institute for Experimental Virology, Hamburg, Germany; 2Department of Biochemistry, McGill University,
Montreal, Canada; 3Department of Oncology, McGill University, Montreal, Canada and 4The Goodman Cancer Centre, McGill University, Montreal, Quebec, Canada.
Correspondence: Dr T Dobner, Department of Molecular Virology, Heinrich–Pette–Institute—Leibniz-Institute for Experimental Virology, Martinistr. 52, 20251 Hamburg, Germany.
Received 6 December 2011; revised 29 March 2012; accepted 6 April 2012; published online 21 May 2012

Taken together. Species G: HAdV52). nuclear export sequence. Notably. The characteristic enzymatic machinery facilitating modification by SUMO is shown schematically. we analyzed differences in SUMO attachment to the nucleus and the cytoplasm. unknown. possibly including poly- of E1B-55K to be modified by SUMO paralogues other than SUMO-1. SCM. as well as poly-SUMOylation of motif (Figure 1a). however. we performed our experiments in living cells to evaluate whether SUMO-2/3 attachment occurs in vivo. representative types (Species A: HAdV12. Previous results highlighted the importance E1B-55K by SUMO-2 and SUMO-3 (Supplementary Figure 1B).4.L K G V K R E R G A C E A T E . conjugation (E2) and ligation (E3). involving protein maturation.27.107 E1B-55K-SCM (K104R) 488 . Previously published and newly established E1B-55K mutants.E A Y P E A R R I A T . and E1B-55K-3XPD. E1B-55K might be SUMO-3 attachment.V R R E . as well as the detailed amino acid substitutions are shown. SUMO-1 conjugation motif. SUMO-2 or that under the experimental conditions tested. E1B-55K-NES exhibited the highest levels of SUMO & 2013 Macmillan Publishers Limited Oncogene (2013) 1626 – 1637 . support the notion that E1B-55K might get modified by endogenous SUMO-1.31 respectively. Species E: HAdV4.40 we decided to investigate the potential modified by all three SUMO paralogues. Species D: HAdV46/49.39. Interestingly. Moreover. This of SUMOylation for E1B-55K function..12 (b) Conservation analysis of HAdV E1B-55K. Detailed information concerning the biochemistry have been reviewed before. E1B-55K phosphorylation effects subsequent SUMO modification To gain initial evidence whether E1B-55K is associated with As phosphorylation of substrates has been shown repeatedly to different SUMO paralogues. either with (Figures 2C and D) or without (Figures 2A and B) another mutant that constitutively mimicks phosphorylation in the inhibition of E1B-55K nuclear export.23. Species B: HAdV3/7/11/14/16/21/34/35/50. C-terminal phosphorylation region.22–24. -2 or -3 additionally suggest E1B-55K is modified by all SUMO paralogues. ATP-dependent activation (E1). Currently. be a prerequisite for efficient SUMO modification. Species C: HAdV1/2/5.25. The absolute conservation of E1B-55K is shown. these functions possibility is quite conceivable as both modifiers have an internal have so far been exclusively linked to SUMO-1 modifica. These results for the first time consequences such as viral-mediated cell transformation.stop E1B-55K-3XPA (S490/1T495A) 102 . As all paralogues can be conjugated to the SCM in vitro.22.1.35–38 As published data suggest overlapping. -2 or -3 in vivo.41 localization was analyzed E1B-55K species. (c) HAdV5 E1B-55K organization and mutations.stop E1B-55K-3XPD (S490/1T495D) Figure 1.E L Y P E L R R I L T . enabling multiple conjugation and formation of SUMO chains tion. the results indicate considerably different consequences of SUMO-1. E1B-55K-wt/ CPR by replacement of serine/threonine residues by aspartic acids E1B-55K-NES showed an association and accumulation in the (Figure 1c). The organization and mutations of E1B-55K are illustrated. including E1B-55K-3XPA. SUMOylation with SUMO-2 and SUMO-3. the present study aims to analyze closer the nature/effects colocalization with either SUMO-1 (Figures 2Ag–Ai/Cg–Ci) or of SUMO modification of E1B-55K and functional downstream SUMO-2/3 (Figure 2Bg–Bi/Dg–Di).32–34 we tion of E1B-55K with either endogenous SUMO-1 or SUMO-2/3 by wanted to evaluate a putative connection of both E1B-55K confocal microscopy. Simultaneously performed immunoprecipitation experiments with RESULTS exogenously overexpressed SUMO-1.11.. nucleus with both endogenous SUMO-1 (Figures 2Cd–Cf/Cj–Cl) Consistent with the data shown before (Supplementary and SUMO-2/3 (Figures 2Dd–Df/Dj–Dl) upon inhibition of nuclear Figure 1).26.G S S D E D T D .stop 81 . based on the analysis of 19 different. PTM-dependent transformation by a viral oncogene P Wimmer et al 1627 maturation activation: E1 conjugation: E2 ligation: E3 SUMO-GG SUMO-GG S SUMO-GG S E3 ligase Uba2 Ψ-K-x-E/D SUMO SUMO-GG Ubc9 Aos1 target consensus ψ-K-x-E/D n= 19 serotypes NES SCM CPR absolute complexity E1B-55K sequence 1 90 180 270 360 450 aa 0 80 160 240 320 400 480 aa NES SCM cystein/histidine-rich region CPR E1B-55K (496 aa) NES SCM 81 . NES. SCM. As HAdV5 E1B-55K represents a well-established model export. enabling covalent attachment of alternative SUMO paralogues to the SUMO-chain formation adenoviral protein at K104R (Supplementary Figure 1A).G D D D E D D D . (a) Simplified mechanism of SUMO modification. Therefore. which may be due to mono-. As the viral protein actively shuttles between PTMs. showing the NES.G A A D E D A D . CPR. but also similar to classical ubiquitination.115 488 .93 E1B-55K-NES (L83/87/91A) 488 . SCM and CPR. E1B-55K-SCM completely failed to show any system. we analyzed the subcellular associa. Species F: HAdV40/41. three different SUMO paralogues have been identified multiple E1B-55K cross-reactive bands are apparent in the immuno- that are conjugated to the respective target proteins via the c-K-x-E/D blots.28.

d. D) or dimethyl sulfoxide (A. PTM-dependent transformation by a viral oncogene P Wimmer et al 1628 A Mock Aa Ab Ac B Mock Ba Bb Bc α-E1B α-SUMO-1 merge α-E1B α-SUMO-2/3 merge E1B-55K-wt Ad Ae Af E1B-55K-wt Bd Be Bf α-E1B α-SUMO-1 merge α-E1B α-SUMO-2/3 merge E1B-55K-SCM Ag Ah Ai E1B-55K-SCM Bg Bh Bi α-E1B α-SUMO-1 merge α-E1B α-SUMO-2/3 merge E1B-55K-NES Aj Ak Al E1B-55K-NES Bj Bk Bl α-E1B α-SUMO-1 merge α-E1B α-SUMO-2/3 merge C Mock Ca Cb Cc D Mock Da Db Dc + LMB α-E1B α-SUMO-1 merge + LMB α-E1B α-SUMO-2/3 merge E1B-55K-wt Cd Ce Cf E1B-55K-wt Dd De Df + LMB α-E1B α-SUMO-1 merge + LMB α-E1B α-SUMO-2/3 merge E1B-55K-SCM Cg Ch Ci E1B-55K-SCM Dg Dh Di + LMB α-E1B α-SUMO-1 merge + LMB α-E1B α-SUMO-2/3 merge E1B-55K-NES Cj Ck Cl E1B-55K-NES Dj Dk Dl + LMB α-E1B α-SUMO-1 merge + LMB α-E1B α-SUMO-2/3 merge Figure 2. e. E1B-55K is modified by all SUMO paralogues. subpanels a. h. B) for 6 h and fixed with paraformaldehyde after 24 h. Representative a-E1B-55K (red. g. subpanels b. D). i and l. j) and a-SUMO (green. B. Oncogene (2013) 1626 – 1637 & 2013 Macmillan Publishers Limited .and Alexa488-conjugated secondary antibodies. (A–D) H1299 cells were transfected with vector/pE1B-55K-wt/-SCM/-NES. A. k) staining patterns of at least 50 analyzed cells are shown. C) or mAb 8A2 (a-SUMO-2/3. including visualization of the nuclei by 4’. respectively. The overlays of the single images (merge) are shown in subpanels c. Samples were double-labeled in situ with mAb 4E8 (a-E1B-55K) and mAb 21C7 (a-SUMO-1. f.6-diamidino-2-phenylindole staining. Primary antibodies were detected with Alexa549. alternatively treated with 20 nM leptomycin B (LMB) (C.

Coprecipitation of the viral protein by p53 HA-reactive bands with E1B-55K-wt/-NES/-3XPD (Figure 3B.29 or with to induce SUMO modification of p53 (Figure 4Ca. lanes 5 and 6). As anticipated. 4 and 6) induced CK2. column 7). we mutation of the E1B-55K CPR. whereas different degrees with Ubc9 (Figure 5). but additionally its ability C-terminal phosphorylation of E1B-55K (Figures 3E–G). lanes 2 and 6). Intriguingly.31 reduced focus-forming activity. column 9) activity was (Figure 5Aa) and/or endogenous Ubc9 (Figure 5Ba). p53-dependent on the E1B-55K SUMOylation status. In total. which in turn lane 5).49 As expected. which correlates with remodeling factor Daxx via a Cul5-dependent proteasomal the degree of E1B-55K SUMO modification (Figures 3Eb/Ea).22. as the inhibitory effects of DMAT are limited and not all SUMO modification of E1B-55K. different E1B-55K mutants. a SUMO-modified form precipitation with E1B-55K-specific antibodies (Abs) (Figure 3D). the functional correlation of both SUMO chains attached to the viral protein in a phosphorylation. 5 and 7). results by Penella et al. the results Finally. in particular for the ability of the adenoviral transcriptional repression. we tested the correlation of these mutations with the strongly indicated a cross-talk between C-terminal phosphorylation oncogenic potential of E1B-55K in primary baby rat kidney (pBRK) and SUMOylation of E1B-55K. which is frequently deregulated and/or utilized by As PTMs have been shown to be essential for E1B-55K pathogens. E1B-55K-wt assumed to either interact with Ubc9 and/or SUMO itself.29 some degree of phosphorylation and SUMO modification is reduction of endogenous Daxx. As expected. mechanism. Interaction of Ubc9 and E1B-55K is independent of the PTMs independent mechanisms of E1B-55K induced cellular SUMO modification involves a well-defined enzymatic cascade transformation. To determine further. To determine the p53-repression potential of the enzymes involved in SUMO attachment. Figure 3). Moreover. lanes 3 and 5). The pro-tumorigenic functions of E1B-55K had primarily been In total. Consistent pattern of endogenous SUMO-2/3 can be visualized after with previous findings (Figure 2. Interestingly. of both PTMs bars 5/7) than E1B-55K-wt (Figure 4F. showed that the viral E1B-55K protein induces SUMOylation of whereas PTM was barely detected on E1B-55K-3XPA (Figure 3A.36. lanes 4 and 7).23. E1B-55K-NES and E1B-55K- inactivation of the E1B-55K SCM or CPR significantly impaired 3XPD showed a slightly higher affinity for Ubc9 compared with p53-repression (Figures 4A and B. either abrogating (E1B-55K-3XPA) or analyzed SUMO modification of E1B-55K either after application of constitutively mimicking phosphorylation (E1B-55K-3XPD).28 we decided to analyze whether the SUMOylation machinery (Figures 2 and 3)22. proteasomal selectively combined mutations within the E1B-55K protein show degradation of p53 and Mre11 by the viral Cul5/Rbx1/ElonginB/C that inactivation of the CPR completely abolished SUMOylation of E3-ubiquitin ligase complex 50 occured independently of E1B-55K E1B-55K-NES (Figure 3H. lanes 2. columns 3–5). Interestingly.37. elimination of E1B-55K PTMs either previously described. indicating that PTM of E1B-55K is (Figure 3B. (Supplementary Figure 1.19.22. lanes 6 and 7). 3XPD) was evident (Figure 4Ca. bodies. which in turn induces p53 SUMOylation and facilitates repression of p53 by spatial restriction to PML-nuclear bodies. we tested the ability of DMAT leads to a reduction of E1B-55K phosphorylation (Figure 3Fb) E1B-55K to decrease endogenous levels of the chromatin- relative to the untreated controls (Figure 3Fa). these results provided strong evidence for a direct linked to the silencing of p53 tumor suppressor functions. Figure 3). failed to do so (Figure 4D. lane 4). lanes 3–7). of E1B-55K-wt and mutants (NES. columns 6 and 8). It has functional link between E1B-55K transformation potential and its been generally believed that E1B-55K induces p53 PTM and PTMs (Figure 4F). whereas E1B-55K-wt (Figure 4F. 3XPD) showed alterations in modifications directly impacts on repression of p53-dependent SUMO attachment (Figure 3A.22.36 however. Upon cotransfection transcription.48 (Figure 1a). a comparable band dispensible for this interaction (Figure 4Ca. Intriguingly. we performed reporter gene assays To address this hypothesis. whereas E1B-55K-SCM or E1B-55K-3XPA showed modified by SUMO-2/3 in vivo. which in turn leads to functional protein to modulate p53 tumor suppressor functions (Figures 4A–C) inactivation of the cellular tumor suppressor.47.29 application of the CK2 inhibitor apart from p53 regulation (Figures 4A–C). these & 2013 Macmillan Publishers Limited Oncogene (2013) 1626 – 1637 . as evident from multiple proteins (Figure 4C). all three proteins C-terminal residues of E1B-55K might be exclusively targeted by E1B-55K-wt/-NES/-3XPD (Figure 4D. lanes 3 and 5). bar 3) induced efficient transformation of pBRK cells in combina- tion with HAdV5 E1A. bar 4) different E1B-55K functions. no slower-migrating forms (Figure 4Ca. did not the kinase inhibitor DMAT. we analyzed the modulation of p53 by the viral dependent manner (Figure 3B). of any one of the three SUMO paralogues. we investigated whether E1B-55K using a luciferase reporter with an artificial (Figure 4A) or a cellular interacts with the E2 enzyme Ubc9 as many target proteins are p53-dependent promoter (Figure 4B). to prove a direct correlation between both PTMs. As shown previously.23. which partly inhibits CK2-mediated only alter E1B-55K PTM by SUMO itself. SUMOylated (Figure 3A. the less SUMOylated E1B-55K species both for exogenous E1B-55K-3XPD (Figures 4A and B.23 and at least linkage of E1B-55K SUMOylation and phosphorylation might partly depends on PTMs for certain functions (Figure 4). lanes 4 and 6). These data for the first time link cells (Figure 4F). E1B-55K-3XPD was Moreover. SUMO-ligase. lanes clearly showed that all adenoviral mutants were capable of 2. both E1B-55K indicating that interlinking of phosphorylation and SUMO proteins mutated in the CPR (3XPA.38. whereas E1B-55K-SCM was not comparable to that of E1B-55K-NES (Figures 4A and B. selectively combined mutations within the E1B-55K coding sequence To validate the functional connection of both E1B-55K PTMs (Figure 3H). induced a dose-dependent reduction in reporter activity in both Intriguingly.35–38 provide a unifying framework concerning the modulation of we hypothesized that E1B-55K itself may be able to modulate cellular proteins. again suggesting that the adenoviral protein might get multiply lanes 3.27. bar 6) or the SCM (Figure 4F. Moreover. Moreover. Intriguingly. further data suggested additional p53. PTM-dependent transformation by a viral oncogene P Wimmer et al 1629 modification (Figure 3A. but thus far unrelated observations concerning by mutating the CPR (Figure 4F.35 and Müller and Dobner36 comparably SUMOylated than E1B-55K-wt (Figure 3A.and poly. Taken together.23. whereas the latter has PTMs (Figure 4E). all tested adenoviral E1B-55K proteins interacted to the experiments (Figures 4A and B. whereas E1B-55K-SCM/-3XPA evident in the respective experiments (Figures 3E and F). lane 3). visualization of HA-tagged SUMO molecules augments p53 inactivation by spatial restriction to PML-nuclear after immunoprecipitation of E1B-55K revealed mono.38.42–46 Recently and the proteasome-mediated decrease of the chromatin- published data indicated that E1B-55K itself is a functional E3 remodeling factor Daxx (Figure 4D). thus indicating substrate-specific roles for PTMs repeatedly been shown to exhibit strongest PTM by SUMO in E1B-55K-dependent degradation processes. which has been shown to directly correlate with the However. Finally.35.28.20 As E1B-55K itself is a target of the cellular oncogenic potential. both E1B-55K-NES and E1B-55- E1B-55K oncogenic properties are dependent on the cross-talk 3XPD exhibited a comparable focus-forming activity (Figure 4F. E1B-55K PTMs. 4 and 6) and no detectable modification for E1B-55K-SCM/-3XPA binding the tumor suppressor. bar 3).

. t . also modulation of the host cell 55 72 55 72 55 55 72 95 55 72 95 55 72 55 kDa kDa kDa kDa kDa 1 1 1 + + + + + 1 1 em em em em em pt pt pt pt pt + + y y 2 2 y y y 2 + + + 2 2 pE ve pE ve pE ve pE ve pE ve 1B cto 1 ct o 1B cto 1 ct o 1 ct o + . . possibly. 1 -S 1 - 5 5 5 5 IP α-E1B IP α-E1B + B-5 SC + B-5 SC C + B-5 SC 5 IP α-E1B IP α-E1B 5 + B-5 SC + B-5 SC + B-5 + pHA-SUMO-1 pE 5K M pE 5K M pE 5K M pE 5K M pE 5K M pE 5K M 1 . 1 . allowing concluded that this interaction is mostly independent of E1B-55K correlation between E1B-55K PTMs and the affinity for Ubc9. which strates and preferential late viral mRNA export. . 1 + pHA-SUMO-3 + pHA-SUMO-1 -N 6 6 6 6 + B-5 NE + B-5 NE E + B-5 NE 6 6 pE 5K S pE 5K S + B-5 NE + B-5 NE + B-5 pE 5K S pE 5K S pE 5K S pE 5K S 1B -3 1B -3 1B -3 1B -3 1B -3 1B -3 -5 XP -5 XP -5 XP -5 XP -5 XP -5 XP 5K A 5K A 5K A 5K A 5K A 5K A IgH -3 -3 -3 IgH IgH IgH IgH -3 -3 IgH XP XP XP XP -3 XP XP D D D β-actin β-actin D D D E1B-55K E1B-55K E1B-55K E1B-55K HA-SUMO HA-SUMO conjugates conjugates E1B-55K-SUMO E1B-55K-SUMO E1B-55K-SUMO E1B-55K-SUMO P Wimmer et al PTM-dependent transformation by a viral oncogene SUMOylation machinery.52 These & 2013 Macmillan Publishers Limited . r + . 1 . 1B -w -5 t E1B-55K E1B-55K E1B-55K E1B-55K + 4 -5 pE 5K pE 5K 1 -S 1 C 5 -S + B-5 C 5 + B-5 pE 5K M E1B-55K-SUMO 1 E1B-55K-SUMO E1B-55K-SUMO pE 5K M -N 6 1 E input -N IP α-E1B + B-5 6 + B-5 E pE 5K S pE 5K S 1B -3 7 1B -3 + -5 XP 7 + -5 XP A + pHA-SUMO-1/2/3-GG pE 5K A + pHA-SUMO-1/2/3-GG pE 5K 1B -3X 1B -3X -5 PD -5 PD 5K IgH 5K -N -N ES ES /3 β-actin E1B-55K /3 XP E1B-55K XP A A E1B-55K PTMs regulate subnuclear localization during virus E1B-55K-SUMO E1B-55K-SUMO virus infection involving selective degradation of cellular sub- mediates essential functions during the early and late phases of The HAdV5 E1B-55K protein is a multifunctional regulator. r 3 3 3 3 3 + B-5 r + B-5 r + B-5 r infection pE 55K pE 5K pE 55K pE 5K pE 5K 1B -w 1B -w 1B -w 1B -w 1B -w input + t + t + t + t + t 4 4 4 4 4 -5 -5 -5 -5 -5 + DMSO + DMSO + DMSO IP α-pST IP α-E1B pE 5K pE 5K pE 5K pE 5K pE 5K 1 1 input 1B -S 1B -S 1B -S -S -S C C C C C 5 5 IP α-E1B -5 -5 -5 + B-5 + B-5 5K M 5K M 5K M pE 5K M pE 5K M -N -N -N 1 -N 1 -N 6 6 ES ES ES + B-5 ES + B-5 ES pE 5 pE 5 1B K-3 1B K-3 -5 XP -5 XP 5K A 5K A -3 -3 IgH IgH 55 72 XP XP 55 72 55 72 55 D D β-actin long expo. 1 1 1 1 1 + + + + + + 1 em em em em em em 2 pt pt pt pt 2 2 2 y y pt pt y y 2 + + + y + y + + 2 pE ve pE ve pE ve pE ve pE ve pE ve 1 ct o 1 ct o 1 ct 1 ct 1 ct o 1 ct o o o 3 3 3 3 3 + B-5 r + B-5 r + B-5 r 3 pE 5K pE 5K + B-5 r + B-5 r + B-5 r pE 5K pE 5K pE 5K pE 5K 1B -w 1B -w 1B -w 1B -w 1B -w 1B -w 4 4 + + 4 t t + 4 . based on the lack of a strict data indicated that Ubc9 interacts with the viral protein. 4 + short expo. t 4 + t + t + - 4 -5 -5 input pE 55K pE 55K pE 5K pE 5K pE 55K pE 55K 1 input . 1 - + pHA-SUMO-3 + pHA-SUMO-3 + pHA-SUMO-1 1 . we direct access and. short expo. - pE 55K pE 55K pE 55K pE 55K pE 55K pE 55K 1 -w 1 -w 1 -w 1 -w 1 -w 1 -w 4 4 4 t t t t t t 4 4 4 + B-5 + B-5 + B-5 + B-5 + B-5 + B-5 pE 5K pE 5K pE 5K pE 5K pE 5K pE 5K 1 1 1 1 1 1 input input -S -S -S -S -S -S 5 5 C C 5 C C C C 5 5 5 IP α-E1B + B-5 + B-5 + B-5 + B-5 IP α-E1B + B-5 + B-5 IP α-E1B IP α-E1B pE 5K M pE 5K M pE 5K M pE 5K M pE 5K M pE 5K M 1 1 1 1 1 1 + empty vector + empty vector + empty vector + pHA-SUMO-2 + pHA-SUMO-2 -N -N + pHA-SUMO-2 -N -N -N -N 6 6 E E 6 6 E E E E 6 6 + B-5 + B-5 + B-5 + B-5 + B-5 + B-5 pE 5K S pE 5K S pE 5K S pE 5K S pE 5K S pE 5K S 1B -3 1B -3 1B -3 1B -3 1B -3 1B -3 Oncogene (2013) 1626 – 1637 -5 XP -5 XP -5 XP -5 XP -5 XP -5 XP 5K A 5K A 5K A 5K A 5K A 5K A -3 -3 -3 -3 -3 -3 XP XP XP XP XP XP D D D D D D 55 72 55 72 55 72 55 72 72 95 72 95 kDa kDa kDa kDa kDa kDa SUMOylation and/or phosphorylation. 1 . 1 . kDa kDa kDa SUMO-2/3 SUMO-2/3 conjugates conjugates E1B-55K 1 + + + 1 1 em em em pt pt pt + y y + y 2 + 2 2 55 72 pE ve pE ve 55 72 pE ve kDa 1B cto 1 ct o 1B cto kDa + . 1 . r 3 3 3 + B-5 r 1 pE 55K pE 5K pE 55K + 1 1B -w 1B -w 1B -w + input em + t + t + t 4 4 4 -5 -5 -5 em IP α-pST IP α-E1B pt + y 2 pE 5K pE 5K pE 5K pt + 5μM DMAT + 5μM DMAT + 5μM DMAT + y pE ve 2 1B -S 1B -S 1B -S -5 C -5 C -5 C pE ve 1 ct 1 ct 3 5K M 5K M 5K M + B-5 or IgH IgH IgH 3 -N -N -N + B-5 or pE 5K ES ES ES pE 5K 1B -w β-actin t long expo. 1630 55 72 72 95 55 72 55 72 55 72 72 95 kDa kDa kDa kDa kDa kDa 1 1 1 1 + + + + + + 1 1 em em em em em em pt pt pt pt pt pt 2 2 y y y y 2 y y 2 2 + + + + + + 2 pE ve pE ve pE ve pE ve pE ve pE ve 1B cto 1B cto 1B cto 1B cto 1B cto 1B cto r r 3 r r 3 r r 3 3 + + + + + + 3 . - 3 .51. r + .17. However.

IP of E1B-55K was performed using mAb 2A6 (a-E1B-55K) and resolved on 10% SDS–PAGE.36 Intriguingly. presumably.19–21 3XPD (Figure 6Aa. tumor suppressor p53 has been suggested to represent the Localization of E1B-55K during infection is dependent on predominant mechanism by which E1B-55K facilitates multiple factors.24 To determine the functional consequences of E1B-55K PTMs for productive viral infection.61 localize to E2A-positive viral replication centers. E1B-55K-wt showed a laboratories.66 targeting (Figure 6Bo). (D) H1299 cells were transfected with vector/pE1B-55K-wt/-SCM/-NES/-3XPA/-3XPD. functional inactivation of E1B-55K SCM abolished the dent of E1B-55K SUMOylation and/or phosphorylation in the formation of nuclear aggregates (Figure 6Bg) and colocalization presence of E4orf6 (Figure 4E).27. it remains subject for accumulation of the viral protein in the replication centers further studies to evaluate whether E1B-55K PTM-dependent (Figure 6Br). harvested after 48 h and total cell extracts were prepared. clearly occur independently of E4orf638 and dependent on E1B-55K PTMs colocalizing with the centers of viral replication (Figure 6Bl). whereas input levels (G) were detected using mAb 2A6 (a-E1B-55K) and mAb AC-15 (a–b-actin).67 subnuclear localization and. alternatively treated with 5 mM DMAT or DMSO directly after transfection. E. we analyzed the viruses H5pg4100 (wild DISCUSSION type). E1B-55K-3XPD which might accordingly result in alterations of the degradation showed higher levels of SUMO modification compared with E1B.14 although and PTMs. enhanced SUMOylation of target proteins such as p53. viral mRNA transport and efficient virus replication. however. explaining the so far unrelated data of a number of (Figure 6b). well-developed E2A-positive viral the adenoviral E4orf6 protein. isms utilized by viruses. IP of E1B-55K was performed using mAb 2A6 (a-E1B-55K) and resolved on 10% SDS–PAGE. showed higher levels of SUMO. H5pm4149 (E1B null). Directly precipitated proteins (a) and input levels (b) were detected by using mAb 2A6 (a-E1B-55K) and mAb AC-15 (a–b-actin). However. very little data is with E2A-positive replication centers (Figure 6Bi). Immunoprecipitation of E1B-55K was performed using mAb 2A6 (a-E1B-55K) and resolved on 10% SDS–PAGE. capabilities of E1B-55K and/or the E1B-55K/E4orf6 complex. mAb 2A6 (a-E1B-55K) and mAb AC-15 (a–b-actin). including the presence of other viral proteins deregulation of cell cycle control and apoptosis. E1B-55K phosphorylation effects subsequent SUMO modification. (E–G) H1299 cells were transfected with vector/pE1B-55K-wt/-SCM/-NES. Immunoprecipitation of E1B-55K or pST was performed using mAb 2A6 (a-E1B-55K. whereas degradation of p53 and Mre11 is indepen- Furthermore.54 Importantly. regulation system. lanes 4 and 6). Precipitated proteins were detected using mAb 2A6 (a-E1B-55K.28. lane 3). 55K-3XPA (Figure 6Aa. lane 6) and subnuclear been suggested to have a role in nuclear protein degradation.27.24 (Figure 4D). H5pm4174 (E1B-55K-3XPA. either the SCM or the CPR interferes with the ability of E1B-55K to Future detailed studies are being planned to characterize the modulate the tumor suppressor protein p53 and chromatin- differential contributions of E1B-55K PTMs to basic protein remodeling factor Daxx. Bp-Br)). Intriguingly. PTM-dependent transformation by a viral oncogene P Wimmer et al 1631 functions partly correlate with proper localization of E1B-55K functions. harvested after 48 h and total cell extracts were prepared.22.65.53 whereas efficient infection was indicated by F-box (SCF)-like E3-ubiquitin ligase complex in combination with formation of several. certain functions of E1B-55K more efficient mechanisms might be present during viral have been suggested to depend on its potential to transiently infection. Precipitated proteins were detected using either mAb 2A6 (a-E1B-55K) (A) or mAb 3F10 (a-HA) (B).22–24.55–59 As SUMO PTM of E1B-55K is essential for the modulation of p53. H5pm4101 The last 15 years of research have generated a tremendous (E1B-NES) and newly established mutants H5pm4174 (E1B-3XPA). we have shown that mutational inactivation of during infection. tion machinery. These results showed that the complicated has a role in Daxx degradation. harvested after 24 h and total cell extracts were prepared. (H) H1299 cells were transfected with vector/pE1B-55K-wt/-SCM/-NES/- 3XPA/-3XPD/-NES-3XPD and pHA-SUMO-1/2/3-GG. (A–C) H1299 cells were transfected with vector/pE1B-55K-wt/- SCM/-NES/-3XPA/-3XPD and vector/pHA-SUMO-1/-2/-3. inducing proteasomal degradation replication centers (Figure 6Bb). PTM of E1B-55K might also be complex intracellular distribution pattern. leading to an enhanced Although clearly speculative at this point. modulation of the (Figure 6Ab. whereas the opposite phenotype is Figure 3. Molecular weight in kDa is indicated on the left. both viruses expressing E1B-55K-NES or E1B-55K. To confirm the correlation of both E1B-55K lighted its importance in basic cell metabolism. whereas input levels (C) were detected using mAb 2A6 (a-E1B-55K) and mAb AC-15 (a–b-actin). which is dependent on the PTMs of the viral proteins.64 however. F). E1B-55K forms a Skp1/Cullin/ (Figure 6Ba). all E1B-55K mutants mediates a large variety of different molecular activities during coprecipitated with the E2 enzyme Ubc9 during infection early/late phases of infection. because it has been shown that E1B-55K (E1B-55K-3XPD.62 DNA-ligase IV63 and of E1B-55K by mutating the NES induced a relocalization to integrin a3. SUMO modification of the viral protein and surprising that intracellular pathogens take advantage of this interaction with Ubc9 was analyzed during infection (Figure 6A).52 In both contexts. As described previously. ranging from small a key determinant in selective degradation of the chromatin- accumulations to a diffuse/slightly granular staining in the nucleus remodeling factor Daxx. E1B-55K) have been linked in some detail to the SUMOyla- E1B-55K-SCM or E1B-55K-3XPA (Figure 6Aa. modification has been implicated in subnuclear targeting. This would be CPR (Figure 6.53. as well as in transformation of (Figure 6Ab). lanes 7 and 6). we our present data on the functional connection between E1B-55K analyzed steady-state localization of E1B-55K to validate the PTMs provide the first evidence for a unifying framework (Figures regulative connection of both E1B-55K PTMs during infection 3 and 4). Unfortunately. whereas the E1B-55K-3XPA mutant was both subnuclear targeting to PML-nuclear bodies. matched the predicted possibility of a SUMO modification is gradually lost upon expression of E4orf6. little is known about the molecular mechan- As expected.22–24. except in the case of the E1B-55K null virus primary mammalian cells. while proteins are labeled on the right.51. lanes 5 and 7).14. structures that have defective in SUMOylation (Figure 6Aa. harvested after 48 h and total cell extracts were prepared. H5pm4227 of particular interest. & 2013 Macmillan Publishers Limited Oncogene (2013) 1626 – 1637 . D) or mAb a-pST (E) and resolved on 10% SDS–PAGE.60 E1B-55K is a multifunctional regulator that Consistent with previous results (Figure 5).23.36. Although it is not PTMs shown above.13. degradation of Daxx has been shown to brightly stained nuclear accumulations (Figure 6Bj). available comparing the detailed kinetics of substrate degrada- both E1B-55K proteins.28. Usually. H5pm4102 (E1B-K104R). two proteins of the E1 region modified E1B-55K. Precipitated SUMO-conjugated proteins (a) and input levels (b) were detected by using mAb 8A2 (a-SUMO-2/3).54 functional relationship between the E1B-55K PTMs. amount of data concerning PTM by SUMO proteins and high- H5pm4227 (E1B-3XPD). whereas degradation of p53 and interplay of a viral protein with the host cell PTM system alters Mre11 mainly occurs in cytoplasmic aggresomes. harboring mutations within the tion/inactivation during productive viral infection. In the context of HAdV5 infection. Bm-Bo). whereas no equivalent species was evident for (E1A. concerning selective degradation of cellular target during infection. protein.50 Mre11. viral protein function In summary.

8 + + + + + + + + + + + + short expo. 0. (A/B) H1299 cells were transfected with 1 mg pRenilla-Luc. harvested after 36 h and total cell extracts were prepared. total cell extracts were prepared and resolved on 10% SDS–PAGE. pAb 1807 (a-E4orf6) and mAb AC-15 (a–b-actin).2 1.6 E1 ve pE 5K 0.cto + B-5 t -w 1 K 1 K K 5K 0.0 pE1B-55K-3XPA 1. p53-dependent promotor 1. 0. + + + + + + E1A (genomic) 72 E1B-55K-SUMO + E1B-wt 72 E1B-55K-SUMO + E1B-SCM 55 E1B-55K + E1B-NES 55 E1B-55K + E1B-3XPA long expo.5 1.4 0. Molecular weight in kDa is indicated on the left. 0.0 pE1B-55K-3XPD + pE1B-55K-NES input input + pE1B-55K-3XPA + p53SN3 + p53SN3 kDa + pE1B-55K-3XPD 75 *) ) D D A A pE -55 t (** 1 K. Input levels were detected by using mAb 2A6 (a-E1B-55K).0 0.3 mg artifical (A)/endogenous (B) p53-dependent reporter construct (pRE-Luc/pcyclinG-Luc).2 p53-SUMO p53-SUMO 72 72 p53 p53 0.5 mg*.0 pE1B-55K-wt + + + + + + + + p53SN3 1.2 E1B-55K 55 β-actin 0. harvested after 72 h.4 IgH long expo.6 55 p53 55 p53 0. (E) H1299 cells were transfected with different combinations of 0.0 mg**).0 pE1B-55K-NES 1. The mean and s.d. 72 p53-SUMO 72 p53-SUMO 0. quantified and normalized to pXC15-E1B-wt.0 pE1B-55K-SCM 1. Input levels were detected by using pAb a-Mre11. harvested after 24 h. 1. and grown for 21 days.25 mg p53SN53. 0.0 pE1B-55K-3XPD 1. short expo.0 55 55 1 2 3 4 5 6 7 IgH + no DNA short expo.5 wt 1B K-N 1B K-N 1B K-3 -3 1B K-3 -3 1 K- - 1 K- 1 K- K 5K 5K pE -55 pE 5 pE 5 pE 5 pE -55 pE 5 pE 5 l1 l2 l3 l1 l2 l3 55 1B 1B p53 ro ro ro ro ro ro 1 pE pE nt nt nt nt nt nt co co co co co co kDa kDa + + + + + + + 55 E1B-55K E1B-55K E1B-55K 55 55 26 β-actin E4orf6 β-actin 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 β-actin 1 2 3 4 5 6 7 8 9 IP α-p53 (DO-1) input 1. (F) pBRK cells were transfected with pXC15. Forty-eight hours after transfection total cell extracts were prepared and luciferase activity was determined. E1B-55K oncogenic properties are dependent on the cross-talk of both PTMs.0 pE1B-55K-SCM + + + + + + pE4orf6 1.2 0.2 n=3 artificial. Oncogene (2013) 1626 – 1637 & 2013 Macmillan Publishers Limited . total cell extracts were prepared and resolved on 10% SDS–PAGE. mAb DO-I (a-p53).0 pE1B-55K-NES + + pE1B-55K-wt 1. Representative results of three independent experiments are shown. Asterisks reflect the different amounts of transfected pE1B-55K-wt (0. pAb a-Daxx and mAb AC-15 (a–b-actin).) 1B K-w ) Mre11 M M pE -55 ES pE -55 ES (* (* (* -5 XP XP -5 XP XP + B-5 SC + B-5 SC + B-5 wt + B-5 wt + B . IgH + E1B-3XPD 72 E1B-55K-SUMO 1 2 3 4 5 6 7 55 E1B-55K β-actin 1 2 3 4 5 6 7 Figure 4.5 1. Luciferase samples were immunoblotted and probed with mAb 2A6 (a-E1B-55K) and mAb AC-15 (a–b-actin). Immunoprecipitation of p53 was performed using mAB DO-I (a-p53) and resolved on 10% SDS–PAGE. mAb 2A6 (a-E1B-55K).0 em y ve ve ve pE 5 pE 5 pE 55 + B-5 -5 -5 -5 - - - y y y 1B 1B 1B pt pt pt 1 em em em pE pE pE pE kDa kDa 0. (C) H1299 cells were transfected with vector/pE1B-55K-wt/-SCM/-NES/-3XPA/-3XPD and p53SN3. Finally.75 mg pE4orf6 and pE1B-55K-wt/-SCM/-NES/-3XPA/-3XPD.0 1.6 pE 5 pE 5 pE 5 5 y pt 1 em kDa + + + 0. PTM-dependent transformation by a viral oncogene P Wimmer et al 1632 1.0 relative Firefly-luciferase activity relative Firefly-luciferase activity input 0. long expo. mAb DO-I (a-p53) and mAb AC-15 (a–b-actin). p53-dependent promotor n=3 endogenous.0 1 2 3 4 5 6 7 8 9 column 1 2 3 4 5 6 7 8 9 column 1 2 3 4 5 6 + + + + + + + + pRenilla-Luc + + + + + + + + + + pRenilla-Luc + + + + + + + pRE-Luc + + + + + + + + + pCyclinG-Luc input + + + + + + p53SN3 + + + + + + + p53SN3 + + + + + empty vector 0. harboring the alternative mutations within the E1B ORF as indicated. (D) H1299 cells were transfected with vector/pE1B-55K-wt/-SCM/-NES/-3XPA/-3XPD.0 pE1B-55K-3XPA + pE1B-55K-SCM 1.8 D A M S -5 XP XP + B-5 -SC + B-5 -NE r 1B -3 -3 B.’s are presented for three independent experiments.4 hDaxx 95 0.0 pE1B-55K-wt 0. focus formation was visualized by crystal violet staining.4 n=3 + p53SN3 + p53SN3 1.2 D D A A M M pE 55K S ES relative focus-forming activity -5 XP XP -5 XP XP + B-5 SC E C t t pE ve r r or or -w 1B K-N -w 1B -N -S 1B -3 -3 1B -3 -3 pt cto 1B cto 1 K- ct ct pE 55K 5K 5K 5K 5K K 5K 1.02 mg p53SN3 and alternatively vector/ pE1B-55K-wt/-SCM/-NES/-3XPA/-3XPD as indicated. Precipitated proteins (a) and input levels (b) were detected using mAb 2A6 (a-E1B-55K). while proteins are labeled on the right.8 0.

aptly termed CK2. into ring-like E1B-55K positive nuclear areas during productive This motif is characterized by the consensus sequence c-K-x-E-x(2-5) virus infection. allowing mammalian cells (Figure 4F). thus facilitating covalent modification of the viral protein. despite their These findings are consistent and integrate both p53-dependent continuing ability to bind the E2 enzyme Ubc9. Penella et al. harvested after 48 h and total cell extracts were prepared. induced by mutations that enhance the SUMOylation of E1B-55K. Immunoprecipitation of endogenous Ubc9 was performed using mAb a-Ubc9 and resolved on 10% SDS–PAGE.SC 1B -3 -3 pt cto 1B cto - 1B 5K pE 55K 5K em y ve E1 5K- 1 1 1 pE 5K k pE oc 5 - y kDa pt M + + em kDa 72 E1B-55K-SUMO + + + + + + E1B-55K-SUMO 72 55 E1B-55K E1B-55K 20 55 Ubc9 20 HA-Ubc9 β-actin β-actin 1 2 3 4 5 6 1 2 3 4 5 6 7 Figure 5. However. Figure 7). F & 2013 Macmillan Publishers Limited Oncogene (2013) 1626 – 1637 . Immunoprecipitation of HA- tagged Ubc9 was performed using mAb 3F10 (a-HA) and resolved on 10% SDS–PAGE. Following the assumption that phosphorylation of -D/E-x(2)-D/E and is present on all E1B-55K proteins of species C E1B-55K preferentially occurs at these nuclear accumulations.72. thus strongly suggesting that phosphor- the interplay of the viral protein with the host cell SUMO ylation is of general importance for E1B-55K function as these sites modification system. these studies do not studies are dedicated to establish a suitable in vitro system to exclude modification by other SUMO paralogues under specific obtain more resilient results on the functional interplay between conditions or of additional. PTM-dependent transformation by a viral oncogene P Wimmer et al 1633 IP α-HA IP α-Ubc9 + pHA-Ubc9 D A pE 5K M pE 5K S -5 XP XP D A M C E S -5 XP XP t r + B-5 r -w -N 1B -S + B-5 SC + B-5 NE 1B -3 -3 pt cto o + B-5 wt ct 1B K-3 -3 pE 5K pE 5K 5K 1 K- 1 K- 1 K- ve pE ve 5K + B-5 -5 + B-5 pE 55 pE 5 pE 5 pE 5 y y pt - 1 1 1 1B em em k pE oc kDa + + + + kDa M + + 72 55 E1B-55K E1B-55K 55 1 2 3 4 5 6 7 1 2 3 4 5 6 input input + pHA-Ubc9 D A pE 5K M S -5 XP XP C + B-5 NE t 1B -w -S 1B -3 -3 D A - pE 5K pE 55K K 5K -5 XP XP S M pE 5 -5 -wt pE ve r +E -5 r 1B -NE + B-5 + B-5 B. (B) Alternatively. although the interaction of conserved among HAdVs.69 as in particular the catalytic subunit CK2a is redistributed negatively charged amino-acid dependent SUMOylation motif. modification of the tumor suppressor p53. (A) H1299 cells were transfected with vector/pE1B-55K- wt/-SCM/-NES/-3XPA/-3XPD and vector/pHA-Ubc9. so far unknown.75 we assume that phosphorylation of E1B-55K is at least in part regulated by E1B-55K harbors a specific kind of SCM motif. our ongoing The molecular 3-dimensional protein structure of Ubc9 is known studies may soon reveal whether this applies also to modification in detail74 and led to the characterization of determinants of E1B-55K by different SUMO paralogues. H1299 cells were transfected with vector/pE1B-55K-wt/-SCM/-NES/-3XPA/-3XPD. mAb 3F10 (a-HA).73 Ongoing modified by SUMO-1 during viral infection. a Although. thus enabling simultaneous our results of sequential SUMOylation would explain the interaction with the catalytic groove and the nearby basic patch of previously described phenomenon of SUMO-dependent nuclear Ubc9. Further studies need to be carried SCM or E1B-55K-3XPA to be SUMOylated and/or to induce SUMO out to determine if SUMOylation of E1B-55K occurs in species A.47. adenoviruses (HAdV) species. the observation that E1B-55K itself interacts with that putatively phosphorylatable residues are present at the the SUMO E2 enzyme Ubc9 offers a new opportunity to analyze C-terminus in all species.29.54.76 Figure 7 shows In addition. implicated in interactions with target proteins. In the transcription factors MEF2 and HSF1 showed that phosphorylation case of SUMOylation. Coprecipitated proteins (Aa/Ba) and input levels (Ab/Bb) were detected by using mAb 2A6 (a-E1B-55K).36 showed recently that p53 is only mechanism similar to the one described for Daxx. but the in vivo relevance assists in non-covalent interaction with SUMO and/or SUMO-Ubc9 and functional consequences remain to be demonstrated. Thus E1B-55K SUMOylation might not occur with all human observation is somehow reminiscent of the inability of E1B-55K. newest data convincingly show that C-terminal recent descriptions of Ubc9 interaction partners.71 This G. Interestingly. We currently and -independent mechanisms in their importance for the ability favor the model that phosphorylation of E1B-55K might be of E1B-55K to induce malignant transformation of primary necessary to stimulate the enzymatic activity of Ubc9. F and both partners is independent of phosphorylation. with the exception of species A. both the structure and functions of target proteins. harvested after 48 h and total cell extracts were prepared. E1B-55K contains a LK dipeptide directly matrix association of E1B-55K during the very early phase of N-terminal of the SCM (Figures 1b and 7B). The potential target for E1B-55K modification by all three SUMO it additionally remains possible that phosphorylation of E1B-55K paralogues is an interesting observation. Based on more Interestingly. the consensus c-K-x-E motif is also highly facilitates SUMO modification by Ubc9.68 SUMO transfer onto the viral protein itself and/or p53.70 whereas delayed expression of E4orf6 initiates a to its binding to the E2 enzyme in a fashion similar to gradual loss of E1B-55K PTM and nuclear matrix association. previous reports concerning have been maintained throughout adenovirus evolution.29 adenoviruses (Figure 1b. cellular proteins. while proteins are labeled on the right.38. Molecular weight in kDa is indicated on the left. mAb a-Ubc9 and mAb AC-15 (a–b-actin).37. which may contribute infection. As E1B-55K and components of the cellular SUMO modification attachment of alternative SUMO paralogues can drastically modify system. Additionally. Interaction of Ubc9 and the viral protein is independent of E1B-55K’s PTMs.54 that observed with the HPV E1 protein.

g. e. The overlays of the single images (merge) are shown in subpanels c. E1B-55K-3XPA (H5pm4174) and E1B-55K-3XPD (H5pm4227) virus (moi 10). m w N K- K- K- K- K- K 55 55 55 55 55 55 k B- B- B- B- B- B- oc α-E2A E1 E1 E1 E1 E1 E1 kDa α-E1B >70% merge M E1B-55K-NES 55 E1B-55K H5pm4101 1 2 3 4 5 6 7 input moi 10 us PD PA M ES 36h p. SC 3X 3X t m w N K- K- K- K- K- H5pm4149 K 55 55 55 55 55 55 k B- B- B- B- B- B- oc E1 E1 E1 E1 E1 E1 kDa M short expo. f. H5pm4102. H5pm4174. Primary antibodies were detected with Cy3.i. h. harvested after 36 h p. respectively. including visualization of the nuclei by 4’. j.6-diamidino-2-phenylindole staining. E1B-55K’s PTMs regulate subnuclear localization during productive infection. in SC 3X 3X t m w N K- K- K- K- K- K 55 55 55 55 55 55 k α-E1B α-E2A B- B- B- B- B- B- >70% merge oc E1 E1 E1 E1 E1 E1 kDa M E1B-55K-3XPA 20 Ubc9 H5pm4174 55 E1B-55K 72 E2A β-actin α-E1B α-E2A >70% merge E1B-55K-3XPD 1 2 3 4 5 6 7 H5pm4227 α-E1B α-E2A >70% merge Figure 6. representative single cell images are provided. Samples were double-labeled in situ with mAb 4E8 (a-E1B-55K) and mAb B6-8 (a-E2A). mAb B6-8 (a-E2A).i. Oncogene (2013) 1626 – 1637 & 2013 Macmillan Publishers Limited . 72 E1B-55K 55 72 long expo. q) staining patterns are shown. l. (B) H1299 cells were infected with wt (H5pg4100) and mutant viruses (H5pm4149. Coprecipitated proteins (Aa/Ab) and input levels (A(c)) were detected by using mAb 2A6 (a-E1B-55K). (A) H1299 cells were infected with mock. in E1B-55K neg. α-E1B α-E2A 100% merge 55 E1B-55K E1B-55K-SCM 1 2 3 4 5 6 7 H5pm4102 IP α-Ubc9 moi 10 us PD PA M ES in SC 3X 3X t 36h p. E1B-55K-SCM (H5pm4102). d. Immunoprecipitation of E1B-55K or Ubc9 was performed using mAb 2A6 (a-E1B-55K) or mAb a-Ubc9 and resolved on 10% SDS–PAGE. i. H5pm4227) (moi 10) and fixed with methanol after 48 h p. whereas the frequency of observed subcellular localization patterns are documented as percentages of at least 50 analyzed cells. E1B-55K-minus (H5pm4149). To illustrate the subnuclear localization of E1B-55K and E2A in sufficient detail.i.i. H5pm4101. E1B-55K-NES (H5pm4101).i. m. while proteins are labeled on the right. and total cell extracts were prepared.and FITC-conjugated secondary antibodies. PTM-dependent transformation by a viral oncogene P Wimmer et al 1634 E1B-55K-wt H5pg4100 IP α-E1B moi 10 α-E1B α-E2A >90% merge us PD PA M ES 36h p. subpanels b. n. mAb a-Ubc9 and mAb AC-15 (a–b-actin). p) and a-E2A (green. Molecular weight in kDa is indicated on the left. subpanels a. E1B-55K-wt (H5pg4100). Representative a-E1B-55K (red. k. o and r.

Hay) were expressed from pcDNA3. Billerica. a-SUMO-1 mouse mAb 21C7.77 HEK29378 and pBRK cells79 were grown in Dulbecco’s modified mutations disrupting the SCM (K104R.79 Subsequently. Wilmington. the E1B mutant proteins E1B-55K-NES (L83/87/91A).23.91 a-SUMO-2/3 mouse mAb 8A2. USA). Conserved putative SCM and CPR consensus sequences are highlighted in gray. pcyclinG-Luc) for a-Ubc9 mouse mAb (BD Transduction. Germany). titered in HEK293. p53. BD Biosciences. 100 U penicillin and H5pm4101) of E1B.85 H5pm4102 and H5pm4101 carry H1299. aligned according to sequence similarity and grouped according to ICTV defined adenovirus species (A–G). and H5pm4227 (S490/1T495D) were produced as described.83 Total cell Germany). Heidelberg. MATERIALS AND METHODS Viruses Cell lines and culture conditions H5pg4100 served as wild type HAdV5. host cell SUMOylation machinery. Taken together. Germany). expression of Firefly luciferase have been described.24 Newly established viruses H5pm4174 (S490/1T495A) 100 mg streptomycin per ml in 5% CO2 atmosphere at 37 1C.0 mg of pRL-TK (Promega. the MA. plasmids (see Figure 4) and 1. (a) Representative sections of 32 HAdV serotype- specific E1B-55K sequences are shown. H5pm4102) or the NES (L83/87/91A. Eppelheim.87 a-E4orf6 rabbit pAb 1807 89 and Europe GmbH. and G viruses. our data provide the first evidence for a Transformation assays functional relationship between two so far unlinked PTMs of the To test the oncogenic potential of E1B-55K plus E1A. USA). Cells were expanded onto 150-mm culture plates after 48 h.27. cell factors (for example. PTM-dependent transformation by a viral oncogene P Wimmer et al 1635 pylogenetic relationship HAdV12 (based on ClustalW alignment of E1B-55K) A HAdV31 HAdV18 HAdV40 SCM CPR F HAdV12 SCADDRDKQEKKESLKEAA----DHLTLSCLRTDYESSDEDDN HAdV41 A HADV18 SCAAVNIKRERETVLSRLA----DHLTLSCLRTDYESSDEDGN HAdV52 G HAdV31 SCAVVKKRERKESLKETVL----DHLTLSCLRTDYESSDEDN HAdV3 SGQNRGIKRERNPSGNNSR----DHLVLACTGAEFGSSGEETD HAdV1 HAdV7 SGQNRGIKRERNPSGNNSR----DHLVLACTGAEFGSSGEETD HAdV2 HAdV16 SGQDRGIKRERNPSGNNSR----DHLVLACTGAEFGSSGEETD C HAdV6 HAdV21 SGQDRGIKRERNPSGNNSR----DHLVLACTGAEFGSSGEETD B HAdV11 TGRDRGVKRERASSGTDAR----DHLVIARTGAEFGSSGEETD HAdV5 HAdV14 TGRDRGVKRERASSGTDAR----DHLVIARTGAEFGSSGEETD HAdV3 HAdV34 TGRDRGVKRERASSGTDAR----DHLVIARTGAEFGSSGEETD HAdV35 TGRDRGVKRERASSGTDAR----DHLVLACTGAEFGSSGEETD HAdV7 HAdV50 SGQDRGIKRERNPSGNNSR----DHLVLACTGAEFGSSGEETD HAdV16 HAdV1 GQGLKGVKRERGAFEATEE----DHLVLACTRAEFGSSDEDTD HAdV21 C HAdV2 HAdV5 GQGLKGVKRERGASEATEE----DHLVLACTRAEFGSSDEDTD GQGLKGVKRERGACEATEE----DHLVLACTRAEFGSSDEDTD HAdV50 B HAdV6 GQGLKGVKRERGASEATEE----DHLVLACTRAEFGSSDEDTD HAdV11 HAdV8 TSMARGVKRERSDGGNTGL----DHLVMACTGTEFSSSGEDTD HAdV35 HAdV9 TSMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV14 HAdV15 TTMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV19 TSMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV34 HAdV22 TSMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV4 E HAdV28 TSMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD D HAdV29 TSMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV15 HAdV36 TSMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV8 HAdV37 TSMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV54 HAdV46 TTMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV53 TSMARGVKRERSDGGNTGM----DHLVMACTGTEFSSSGEDTD HAdV19 HAdV54 TSMARGVKRERSDGGNTGL----DHLVMACTGTEFSSSGEDTD HAdV37 E HAdV4 SGRERGIKRERHDETNHRI----DHLVLSCTGTEFGSSGEESD HAdV53 D HAdV40 SGQRGGINGQRGTKRKMEN----DHQMMSCLRTDYESSDED HAdV22 F HAdV41 SGQRGEKRKLENDGADFLK----DHQMLSCLRTDYESSDED HAdV36 G HAdV52 SGGARKKQKTEPEPRNFLN----DHHMLSCLRTDYESSDEE HAdV46 HAdV28 HAdV9 HAdV29 Figure 7. WI. putatively modified lysine and serine/threonine residues are shown in bold.85 E1B-55K-SCM (K104R). cultivated for 2–3 weeks changing the medium every 3–4 days and multilayered cell accumulations (foci) were visualized by crystal violet staining. Madison.28 and Schwartz et al.31 E4orf6. Eagle’s medium with 10% FCS (fetal calf serum). MA. (b) The phylogenetic relationship of serotype-specific E1B-55K sequences based on ClustalW analysis is shown.90 Primary Abs for the detection of cellular proteins cells were transfected with indicated amounts of reporter and effector included a-p53 mouse mAb DO-I (Sigma-Aldrich. freshly prepared pBRK viral E1B-55K oncoprotein. H1299 a-E2A mouse mAb B6-8. Bad Wildbad. Taufkirchen.KG. Germany).81 N-terminally HA-tagged Ubc982 and SUMO-1/2/3 lacking amino acid 98–101/94-95/93-103 from the Abs and protein analysis C-terminus (provided by R. Primary Abs for HAdV5 proteins included a-E1B-55K mouse monoclonal Ab Subconfluent cells were transfected using 25 kDa PEI (Polysciences (mAb) 2A6.37 For dual luciferase assays. Moreover. This could be a characteristic that has been gained Co. 3  106 pBRK cells were transfected with pXC15 identified interaction of E1B-55K with Ubc9 offers a new insight plasmid84 or versions respective mutations in the E1B open reading frame into how HAdVs in general may actively interfere/modulate the (ORF). in different tissues. Berthold Technologies GmbH & (Figure 7B). a-Daxx rabbit pAb (Merck Millipore. All samples were normalized for or lost with the evolution of adenoviruses that replicate efficiently transfection efficiency by measuring Renilla activity.31 in the E1B- CDS (coding sequence). USA). Comparative analysis and phylogeny of serotype-specific E1B-55K sequences.88 a-E1B-55K rat mAb 4E8. The E1B-55K protein sequences without classic SCM motif (c-K-x-E) are highlighted in gray.87 All viruses were propagated and HAdV5 E1B-55K-wt. Ubc9).23 E1B-55K-3XPA (S490/1T495A)27.80 p53. enabling modulation of various host cells were generated from 3–5 days old CD rats (Charles River.28 and E1B-55K- 3XPD (S490/1T495D). transient transfections and reporter gene assays and does not express the viral protein. H5pm4149 carries stop codons in the E1B-55K ORF Plasmids. GmbH. which are phylogenetically grouped separately and extracts were prepared 36 h after transfection and Firefly activity was exhibit different tissue specificity during infection in vivo assayed in a luminometer (Lumat LB9510. Daxx.86 harboring mutations described by Teodoro et al. a-GFP & 2013 Macmillan Publishers Limited Oncogene (2013) 1626 – 1637 .92 The p53-dependent reporter constructs (pRE-Luc.

Hamburg. Proc Natl Acad Sci USA 2001. Dobner T. PEB and Signal 2010. Mol Cell. Nat Rev Mol Cell Biol 11: 861–871. Liu Y.. ubiquitin’s mysterious factor Daxx is required for efficient transformation of primary rodent cells. Shih HM. Heger P. 7 Johnson ES. 8: 947–956. 104: 6684–6689. Ann DK. pressor protein PML. J Virol cousins. 150 mM NaCl. 33 Tan JA. For immunoprecipitation (IP) proteinA-Sepharose/ 18 Flint J. Gonzalez R. Coverslips were washed three times in phosphate-buffered saline (0. REFERENCES 34 Stehmeier P. adenovirus type 5 E1B-55k oncoprotein is a highly active shuttle protein and conjugated secondary Abs (Dianova GmbH. 98: 11312–11317. 1 mM DTT. Hochstrasser M. Blanchette P. washed 75: 5677–5683. and ubiquitin-like proteins.1% aprotinin. J Virol 2001. Proteins bound to 19 Boggio R. Phospho-regulated SUMO interaction modules connect the 1 Kerscher O. Graham FL. 12210–12225. Hauber J. Rockford. Curr Top Microbiol 1 mg/ml pepstatin. 86: 642–654. 17 Dobner T. 16 White E. 33: 400–409. Melchior F. Curr Top Microbiol Immunol 2003. Dobner T. The human Adenovirus type 5 E1B-55K protein is phosphorylated by protein kinase CK2.or Alex549/Alesxa488. Dobner T. Müller S. Dobner T. Hartl B. coverslips were treated for 1 h with primary Abs. Ade- 8 Melchior F. Nat Rev Mol Cell Biol 39 Zhu J. Mol Cell Biol 2010. Ornelles DA. Germany).nonclassical ubiquitin. protein input levels/IP solutions equal amounts were separated by Virus Res 2001. cite/refer to due to space limitations. Dobner T. Schneider G. Härtl B. Germany. Intranuclear York. J Virol 1994.1% 27 Teodoro JG. Cell Cycle 2008. IL. CONFLICT OF INTEREST 30 Endter C. J Virol 2008. Durrin LK. 68: 776–786. conjugation and recognition. Methods Mol Biol 2009. Cell Div 2006. Jentsch S. Woo JL. SUMOylation of the transcriptional ‘Freie und Hansestadt Hamburg’ and the ‘Bundesministerium für Gesundheit’. nuclear export of the adenovirus type 5 early region 1B 55-kDa protein augments Autoradiograms were scanned. Blanchette P. a-mre11 rabbit pAb (Novus 14 Berk AJ. Rosorius O. SUMO modification ification by SUMO. Aldrich) and a-HA rat mAb 3F10 (provided by E Kremmer). Germany). is a p53-SUMO1 E3 ligase that represses p53 and stimulates its nuclear export 4 Geiss-Friedlander R. 12 Wilkinson KA. 23 Endter C. a–b-actin mouse mAb AC-15 (Sigma. Chiocca S. The SUMO system: an overview. Oncogene 2005. Weitzman MD. 32: 286–295. 0. Kzhyshkowska J. 3 Gareau JR. SUMO . After 1 h blocking in Tris-buffered saline-BSA/glycine transport of E4orf6 is mediated by a CRM1 independent-mechanism. Hartl B. The adenovirus E1B-55K oncoprotein induces SUMO 2 Bossis G. Dasso M. modifier (SUMO) binding determines substrate recognition and paralog-selective 11 Ullrich HD. Toxicology 278: 288–293. Chen SY. Sirma H. Hay RT. Krätzer F. J Virol 2010. 98: 11312–11317. 29: 5511–5522. 2: 202–210. 35 Müller S. Concepts in sumoylation: a decade on. Eshleman HD. Stauber R. 9: 430–436.5% Triton type 5 E1B-55K oncoprotein actively shuttles in virus-infected cells.1% sodium dodecyl sulfate. Nat Rev through interactions with promyelocytic leukemia nuclear bodies. For immunoblotting of 20 Wilson VG. 1 mM dithiothreitol (DTT). 1: 13. Stauber R. (TBS-BG) buffer. J Virol 1997. three times in phosphate-buffered saline (0. PhotoshopCS4/IllustratorCS4 (Adobe Systems GmbH. Intranuclear For IF. 24: 7686–7696. Blanchette P. repression. Chen Y. Melchior F. Regulation of the cell cycle and apoptosis by the oncogenes of ade- 1% NP-40. Regulation of mRNA production by the adenoviral E1B 55-kDa proteinG-agarose was coupled with primary Ab for 1 h at 4 1C. 38 Schreiner S. 5 mM EDTA. TD was co-repressor KAP1 is regulated by the serine and threonine phosphatase PP1. J Virol 2012. SUMO modification. 71: 3620–3627. 2007. Oncogene 2010. Branton PE et al. Jaffray E. added and E4 Orf6 proteins. 3: ra32. Lakdawala SS. UK). UK). Protein modification by SUMO. Curr Opin beads were precipitated by centrifugation. Guzzo CM. cropped and prepared using Adobe oncogenic transformation of primary rat cells. Branton PE. shuttling is independent of E4orf6. 29 Ching W. Oncogene 2005. Kzhyshkowska J. Small ubiquitin-related 2003. Thermo 22 Endter C. Xu X. Sung KS. 0. SUMO. phorylation at the carboxy terminus of the 55-kilodalton adenovirus type 5 E1B protein regulates transforming activity. Dejean A. Munich. Proc Natl Acad Sci USA 2007. cell cycle control. Groitl P. All protein 15 Weitzman MD. SUMO: a history of modification. Viruses and sumoylation: recent highlights. 24: 7673–7685. 2011. Dobner T. Kim CA. Hauber J et al. Hauber J. 85: 8752–8765. 497: 3–16. Mönchen. Choi CY et al. 18: 1–12. Dobner T. Branton PE. USA) on X-ray films (CEA RP X-ray film). Felberbaum R. Mol cell 2009. Hay DC. 84: Mol Cell Biol 2007. Dobner T. Kzhyshkowska J. Takayesu D. Koyuncu E. microscope. NY. Regulation of p53-dependent apoptosis. Inactivating intracellular antiviral responses during extracts were prepared in RIPA buffer (50 mM Tris-HCl/pH 8. Henley JM. 20: 7836–7846. whereas X-100) for 30 min. Song J. 13 Berk A. Biochem J 2010. Cambridge.22 Cells were fixed targeting and nuclear export of the adenovirus E1B-55K protein are regulated by in ice-cold methanol at 20 1C for 15 min for regular or alternatively at SUMO1 conjugation. Hoege C. Groitl P. Spruss T. Adenoviridae: The Viruses and Their Replication. Horn F. The Heinrich–Pette–Institute is supported by the 32 Li X. ACKNOWLEDGEMENTS 31 Schwartz RA. Recent lessons in gene expression. Sci supported by a grant from the Wilhelm Sander-Stiftung. 0. Lin HH. 283: 29405–29415. Dobner T. for 3 min at 95 1C in 2xLaemmli buffer. room temperature for 15 min with 4% paraformaldehyde for confocal 25 Dosch T. adenovirus infection. Whalen SG. Ellis NA. Pyrowolakis G. Viral interaction with the host cell sumoylation system. Schreiner S. Blockage of CRM1-dependent Scientific. 81: 17–27. Nuclear export of adenovirus RNA. transferred to nitrocellulose 21 Wimmer P. Annu Rev Cell Dev Biol 2006. Göttingen. Lima CD. Nature 2001. Biochem Sci 2007. Oncogene 2001. 1 mg/ml leupeptin. Distinct requirements of Adenovirus E1b55K protein for degradation of cellular We apologize for all the important publications and excellent reviews we could not substrates. N-ethylmaleimide. Images were cropped using Adobe PhotoshopCS4/ 28 Teodoro JG. 73: 355–382. Mechanisms. 82: 9043–9055. Annu Rev Cell Biol 2000. Pro- ceedings of the National Academy of Sciences of the United States of America 2001. 2006. Raven Press: New 41 Kindsmüller K. transcriptional Tween20). 7: 754–758. 272: 287–330. Dobner T. Modification in reverse: the SUMO proteases.1% Tween20) followed by 26 Krätzer F. to precleared extracts and rotated for 2 h at 4 1C. Indirect immunofluorescence 24 Kindsmüller K. Post-translational mod. regulation and consequences of protein 40 Mukhopadhyay D. were visualized by enhanced chemoluminescence (ECL) (Pierce. targeting and nuclear export of the adenovirus E1B-55K protein are regulated by Oncogene (2013) 1626 – 1637 & 2013 Macmillan Publishers Limited . and cell transformation by phosphorylation of the 55-kilodalton E1B Germany) and digital images were acquired on a DMRB or LSM3 confocal protein of human adenovirus type 5. Oncogene 2005. 5 Hannoun Z. mounted in Glow medium (Energene. SUMO-1 modification required for transformation by adenovirus type 5 early region 1B 55-kDa oncoprotein. 10 Seeler JS. The adenovirus microscopy and permeabilized in phosphate-buffered saline (0. Hauber J. Rangasamy D. J Biol Chem 2008. Annu Rev Biochem 2004. Trends SUMOylation. Groitl P. cells were grown on glass coverslips as described. Hirschmann N. 4: 690–699. Everett R. from adenovirus. Groitl P. Phosphorylation-dependent interaction of SATB1 and PIAS1 directs SUMO-regulated caspase cleavage of SATB1. novirus type 5 early region 1B 55K oncoprotein dependent degradation of cellular 9 Müller S. Schreiner S. SUMO-1 modification required for The authors declare no conflict of interest. 5th edn. Berk AJ. Modification of proteins by ubiquitin SUMO system to CK2 signaling. phenylmethylsulfonyl fluoride (PMSF). 428: 133–145. MoBiTec. Zhu S. 37 Wimmer P. The SUMO pathway: emerging mechanisms that shape 36 Pennella MA. 30: 2823–2835.1% Triton X-100. p53 and Mdm2. Adenovirus E1B 55-kilodalton protein specificity. Greenhough S. SDS–polyacrylamide gel electrophoresis. Russel MR. Wimmer P. modification of p53. 19: 850–857. The incubation with the corresponding FITC-/Cy3. 2005. 24: 55–64.5% sodiumdeoxycholate) containing 1% novirus. washed three times and boiled Microbiol. Cambridge. Carson CT. Phos- IllustratorCS4. 25 mM iodacetamide and 25 mM Immunol 2001. and cell biology Biologicals Ltd.0. transformation by adenovirus type 5 early region 1B 55-kDa oncoprotein. Dobner T. SUMO: regulating the regulator. Oncogene 2000. of E1B-55K oncoprotein regulates isoform-specific binding to the tumour sup- 6 Hay RT. 16: 591–626. Human pathogens and the host cell SUMOy- (Schleicher & Schüll/Whatman) and incubated as described. 22: 159–180. PB were supported through grants from the Canadian Institutes of Health Research. Nuclear and unclear functions of SUMO. PTM-dependent transformation by a viral oncogene P Wimmer et al 1636 rabbit pAb (Abcam.87 Bands lation system. 259: 25–54. Chen H. 0. Hauber J et al. J Virol 2011. Halliday T.

49-kilodalton isoform of the adenovirus type 5 early region 1B 55-kilodalton 61 Soria C. weight protein repression of basal transcription. J Bioll Chem 2011. Hanakahi LA. Scheffner M et al. Scott GE. D’Amico D. 74: 575–582. Groitl P. 56: 465–478. J Virol 2008. The adenovirus E1B 55 kd protein influences mRNA Activation of p53 by conjugation to the ubiquitin-like protein SUMO-1.nature. J Virol 2007. conjugase UBC9 and likely interferes with polySUMOylation. J Virol 2009. Adenovirus E1B oncoprotein tethers a transcriptional RNA-activated protein kinase (PKR) into viral inclusions of the cell nucleus. Blanchette P. Adenovirus E1B 55K represses p53 activation in vitro. 81: 575–587. Branton PE. Beach D. Groitl P. Groitl P. Hossain J. Bonni A. Rosa-Calatrava M. Sandy P. 8: 76–87. Cell 1982. Genes Dev 2001. 2329–2336. Construction of adenovirus type 5 early region 1 and 4 virus 59 Woo JL. Dobner T. 74 Haas AL. Humana Press Inc. Liao D. Rohleder KJ. Cell Cycle 2009. A novel ubiquitin-like modification modulates 66 Latonen L. BioEssays: News and Reviews in Molecular. 63 Baker A. J Virol 2009. Levine AJ. Virology 65 Wojcik C. Humana Press Inc. Proteasome inhibitors induce nuclear ribonucleoprotein cytosol and the nuclear pore complex. Groitl P et al. pendent SUMO-interacting motifs in Daxx: evolutionary conservation from J Virol 2007. 15: 3104–3117. Sullivan CA. Domains required for in vitro association between the 79: 14004–14016. Grand RJ. 57 Leppard KN. transcriptional activation by the p53 tumor suppressor. Blockage by adenovirus E4orf6 of infection. Adenovirus type 5 early Struct Mol Biol 2009. Nature 1992. 272: 56 Leppard KN. host cell Ubc9 protein. Cancer Cells 1984. Adenovirus E1B 55-kilodalton protein: multiple roles in 78 Graham FL. A 4693–4704. EMBO J 1994. 51 Gonzalez RA. Horikoshi N. J Biol Chem 2000. Nucleolar aggresomes as counterparts of cytoplasmic aggresomes in the partitioning of the Ran-GTPase-activating protein RanGAP1 between the proteotoxic stress. Genes Dev 1994. p53 target genes by a small viral protein. Coutavas E. motifs of the scaffold protein DAXX. Shenk T. 83: 9045–9056. J Virol 2007. Adenovirus infection targets the cellular protein kinase CK2 and 44 Yew PR. Querido E. Nature 2010. 8: 301–307. Hengstermann A. An extended consensus motif enhances 50 Querido E. Oncogene 2010. 52 Blackford AN. Res Tech 2002. J Gen Virol 1999. EMBO J transport via an intranuclear effect on RNA metabolism. Monoclonal antibodies which recognize ligase complex. Selective effects on adenovirus late gene expression of deleting the 1470–1473. 88 Sarnow P. 1115–1123. Nuclear matrix localization and SUMO-1 80 Rubenwolf S. native and denatured forms of the adenovirus DNA-binding protein. phosphorylation. J Virol 1999. NJ. 357: 82–85. Virology 1990. 27: 174–175. Everett RD. Dobner T. 85 Groitl P. Nevels M. Lau DK. 418: 348–352. Dick FA et al. Espantman KC.: Totowa. Structural analysis of the modification of adenovirus type 5 E1b 55K protein are controlled by E4 Orf6 adenovirus type 5 E1B 55-kilodalton-E4orf6 protein complex. 128: 480–484. Zeller T. Corepressor required for adenovirus E1B 55. 68 Yew PR. Liu X. J Gen Virol 2003. 2006. region 1B 55-kDa oncoprotein can promote cell transformation by a mechanism 72 Escobar-Cabrera & 2013 Macmillan Publishers Limited Oncogene (2013) 1626 – 1637 . Inhibition of p53 transactivation required for transformation by 43 Sarnow P. Cyclin G is a transcriptional target of the p53 tumor 58 Blanchette P. J Virol 1997. Smiley J. Wolf H. Leinenkugel F. 54 Lethbridge KJ. McIntosh LP. DeMartino GN. 275: 30487–30495. 82 Gostissa M. PTM-dependent transformation by a viral oncogene P Wimmer et al 1637 SUMO1 conjugation. Dobner T. Dart CF. Virology 1982. 29: 87 Kindsmüller K. Berk AJ. cellular p53 and the adenovirus 2 E1B 55K proteins. J Virol 1991. J Virol 2009. 27: acterizing the N. Adenovirus oncoproteins inactivate the type 5-E1b-58Kd tumor antigen: characterization of the E1b-58Kd tumor antigen in Mre11 Rad50 NBS1 DNA repair complex. In: Wold WS. 83: 4000–4012. Godsey AC. Methods Mol Med 2007. 91 Matunis MJ. Branton PE et al. 286: 19816–19829. pp 187–195. inclusions that accumulate several key factors of neurodegenerative diseases and 92 Zhang XD. Yamada T. Int J Biochem 1983. of p53 by adenovirus E4orf6 and E1B55K proteins occurs via a novel mechanism 76 Rangasamy D. a molecular recognition feature in the E1A oncoprotein that binds the SUMO NJ. Supplementary Information accompanies the paper on the Oncogene website (http://www. Adenovirus type 5 early region 1B 156R protein promotes cell 73 Santiago A. Wimmer P. Cell Biol 2003. Characteristics of a human cell line viral infection and cell transformation. Galanis A. edn. AG et al. 36: 59–72. 4): 997–1008. Mol cell 2007. Regulated RNA processing and RNA transport during adenovirus 81 Dobner T. Construction of adenovirus mutants by direct cloning. Kremmer E. Schreiner S. Nat 47 Härtl B. 45 Martin ME. Weitzman MD. 80(Part 46 Martin ME. Okon M. 2nd early region 4 34-kilodalton protein. Shenk T. Levine AJ. Adenovirus ubiquitin-protein ligase stimulates viral late mRNA mutants. Rubenwolf S. suppressor protein. Morrison M. adenovirus-infected cells. Embo J 2006. Bovine papillomavirus E1 protein is sumoylated by the involving a Cullin-containing complex. Proceedings of the National Academy of Sciences of the United 67 Liu Y. Schütt H. Localization of the adenovirus early region 1B 55-kilodalton 79 Nevels M. Capili AD. Fonseca GJ. J Virol 2005. Shenk T. nuclear export. basis for phosphorylation-dependent SUMO conjugation by the E2 UBC9. 33: 386–395. 48 Sieber T. 65: 424–429. Yew PR. stability. 466: 1076–1081. SV40 large tumor antigen are physically associated with the same 54 kd cellular 69 Souquere-Besse S. 2005. adenovirus-infected and -transformed cells. Fogal V. Shenk T. Hovanessian protein in transformed cells. J Virol 2010. J Cell Biol 1996. EMBO J 1989. Sharrocks AD. alpha3 as a new substrate of the adenovirus E4orf6/E1B 55-kilodalton E3 ubiquitin 90 Reich NC. Identification of integrin novirus type 5. Oncogene 1992. Char- independent from blocking p53-activated transcription. Flint SJ.: Totowa. Berk AJ. Nature 2002. Williams J. Analysis of synthesis. E1b 55K protein. 135(6 Part 1): 1457–1470. 19: 71 Mohideen F. 86 Groitl P. Carbone D. Oncogene 2008. J Virol 2002. 120: 510–517. 75 Yang SH. Cellular and Developmental ification and binding regulate the association of CENP-E with kinetochores and Biology 2011. Adenovirus exploits the cellular States of America 2007. Berk AJ. 179: 806–814. 72: 3146–3154. 8: 1999. Branton PE. Blanchette P. 55 Leppard KN. 42 Kao CC. Sarnow P. Witty J. 60 Yousef AF. 28: 387–394. J Virol 70 Leppard KN. Estermann FE. 18: 6462–6471. Microsc repression domain to p53. Leppard KN. Identification of two inde- transformation independently of repression of p53-stimulated transcription. Matunis MJ. Control of mRNA export by adenovirus E4orf6 and E1B55K proteins during pro. associate in infected cells and affect ND10 components. 8: 190–202. Berk AJ. Bilimoria PM. Zeller T. 84 Logan J. SUMO-2/3 mod- cancer. 7: 171–180. Blanchette P. 83 Okamoto K. Semin Virol 1998. Mol cell 2008. 49 Schreiner S. aggresome response to accelerate inactivation of the MRN complex. Zhang H. 82: 1B. Speiseder T. Duprey E. Ho YS. 83: 5329–5338. Blobel G. Boivin D et al. Ketner G. Effects of mutations in the adenoviral E1B 55-kilodalton 77 Mitsudomi T. Yan Q. Yen TJ. Legrand V. p53 protein coding sequence on viral late mRNA metabolism. 76: gene mutations in non-small-lung cell cancer cell lines and their correlation with 4507–4519. A monoclonal antibody detecting the adenovirus 62 Stracker TH. 84: 7029–7038. 35: 579–589. Dallaire F. Porter AC. Dobner T. Bodner HK et al. 53 Ornelles DA. J Gen Virol 1977. Walsh C. Berk AJ. Berk AJ. Pilder S. Functional analysis of adenovirus type 5 early region ductive infection requires E4orf6 ubiquitin ligase activity. 130: 29–39. O’Shea CC. 2642–2651. the presence of ras mutations and clinical features. 16: 945–952. 73: 1245–1253. Drosophila to humans and their biochemical functions. Blanchette P. 13: 4816–4822. A molecular 3403–3414. Goeres J. 25: 5083–5093. 71: protein. 104: 6684–6689. Bonvin AM. Kremmer E. Marcellus RC. Filhol O. Tollefson AE (eds) Adenovirus Methods and Protocols. 84: 259–268. Determination of the transforming activities of adenovirus protein during lytic infection: association with nuclear viral inclusions requires the oncogenes. Schwarz SE. Protea.and C-terminal Small ubiquitin-like modifier (SUMO)-interacting 3673–3684. The adenovirus type 5 E1b 55K and E4 Orf3 proteins 1998. Kamura T. Degradation the specificity of substrate modification by SUMO. 2: 527–532. Mol Cell Biol 1999. Science 1996. Adenovirus E1b-58kd tumor antigen and adenovirus early 1B protein. Shevchenko A.000-molecular. oncoproteins target host DNA ligase IV for proteasomal degradation. Dobner T. Everett RD. Adenovirus E4 34k and E1b 55k 89 Boivin D. product of open reading frame 6 of the early region 4 protein of human ade- 64 Dallaire F. Nairn R. Carson CT. Nau MM. Shevchenko A. Steinberg SM. progression through mitosis. Wilson VG. Heterochromatin silencing of protein is sufficient to support virus replication. Pelka P. Kindsmuller K. Morrison MR. 81: 95–105. Levine AJ. Russel WC. Dobner T. Pichard E. Structural insights into early events in the conjugation of ubiquitin and some-dependent degradation of Daxx by the viral E1B-55K protein in human ubiquitin-like proteins. J Gen Virol 1993. Sirma H. Lima CD. and interacting polypeptides of the 34-kilodalton 81: 7034–7040. Ablack JN. Berk AJ. transformed by DNA from human adenovirus type 5. Dobner T. Dobner T. Identification of In: Wold WS (ed) Adenovirus Methods and Protocols. 29: 729–741. Zhao LY. Intracellular localization of proteasomes.

Copyright of Oncogene is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. . However. download. users may print. or email articles for individual use.