- Email: [email protected]
Dynamic localization of tripartite motif-containing 22 in nuclear and nucleolar bodies
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / y e x c r
Research Article
Dynamic localization of tripartite motif-containing 22 in nuclear and nucleolar bodies Gayathri Sivaramakrishnan, Yang Sun, Si Kee Tan, Valerie C.L. Lin⁎ School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore
A R T I C L E I N F O R M AT I O N
AB ST R AC T
Article Chronology:
Tripartite motif-containing 22 (TRIM22) exhibits antiviral and growth inhibitory properties, but
Received 30 October 2008
there has been no study on the localization and dynamics of the endogenous TRIM22 protein. We
Revised version received
report here that TRIM22 is dramatically induced by progesterone in MDA-MB-231-derived ABC28
21 January 2009
cells and T47D cells. This induction was associated with an increase in TRIM22 nuclear bodies (NB),
Accepted 29 January 2009
and an even more prominent increase in nucleolar TRIM22 bodies. Distinct endogenous TRIM22 NB
Available online 10 February 2009
were also demonstrated in several other cell lines including MCF7 and HeLa cells. These TRIM22 NB resemble Cajal bodies, co-localized with these structures and co-immunoprecipitated with p80-
Keywords:
coilin. However, IFNγ-induced TRIM22 in HeLa and MCF7 cells did not form NB, implying the forms
TRIM22
and distribution of TRIM22 are regulated by specific cellular signals. This notion is also supported by
Nuclear bodies
the observation that TRIM22 NB undergoes dynamic cell-cycle dependent changes in distribution
Progesterone
such that TRIM22 NB started to form in early G0/G1 but became dispersed in the S-phase. In light of
Cajal bodies
its potential antiviral and antitumor properties, the findings here provide an interesting gateway to
Breast cancer
study the relationship between the different forms and functions of TRIM22. © 2009 Elsevier Inc. All rights reserved.
Introduction Tripartite motif-containing 22 (TRIM22/Staf50) is a member of the TRIM family of proteins, which have a conserved domain structure consisting of, at their N-terminal ends, the Really interesting new gene (RING), B-box, and Coiled-coil (CC) domains [1], and a variable C-terminal region. In TRIM22, this C-terminal region has the SplA and ryanodine receptor (SPRY) domain. TRIM22 is an interferon (IFN)- and p53-inducible gene that inhibits clonogenic growth and HIV-1 infection when over-expressed [2–7]. There is also evidence that TRIM22 interferes with viral gene transcription and Gag protein localization [2,6]. The E3 ligase activity of the RING domain appears to be critical for the antiviral activity of TRIM22 [2].
Nuclear bodies (NB) are nuclear subdomains that contain an assortment of proteins and are involved in a number of vital nuclear functions. Cajal bodies (CB) are the better characterized NB marked by the presence of p80-coilin. CBs disassemble during mitosis, reassemble during the mid-G1 phase, and are found next to nucleoli and within the nucleoplasm [8,9]. CBs are also enriched in at least 30 proteins belonging to several different classes, including ribonuclear proteins, nucleolar proteins, transcription factors and proteins involved in the regulation of the cell cycle (reviewed in [10]). CBs are involved in a number of biological processes such as the biogenesis of small nuclear and nucleolar ribonuclear proteins, gene replication and transcription and biogenesis of human telomerase [11–16].
⁎ Corresponding author. Fax: +65 6791 3856. E-mail address: [email protected] (V.C.L. Lin). Abbreviations: CB, Cajal body; CC, Coiled-coil; Ct, Cycle threshold; DAPI, 4′,6-diamidino-2-phenylindole; EGFP, Enhanced green fluorescent protein; EYFP, Enhanced yellow fluorescent protein; FCS, Fetal calf serum; IFN, Interferon; NB, Nuclear bodies; PR, Progesterone receptor; RING, Really interesting new gene; SPRY, SplA and ryanodine receptor; TRIM, Tripartite motif-containing; TRIM22, Tripartite motif-containing 22 0014-4827/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2009.01.028
1522
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
There have been conflicting reports regarding the localization of ectopically expressed TRIM22 [1,17,18]. The shorter form of TRIM22 (442 aa) translated from a 1329 nt coding sequence appears to localize as cytoplasmic aggregates with some nuclear localization [1]. Recently, a c-myc-tagged protein product of the 1497 nt TRIM22 coding sequence amplified from human peripheral blood mononuclear cells localized exclusively as a nuclear protein [17]. However, the localization of endogenous TRIM22, which is essential for understanding its cellular function, is unknown. Using a specific antibody and siRNAs against TRIM22, we present evidence that TRIM22 localizes predominantly to the nucleus but is also found in the cytoplasm to some extent. TRIM22 forms distinct NB that are regulated by signaling molecules including progesterone and IFNγ and during the cell cycle. TRIM22 NB are adjacent to CB and interact with p80-coilin. The results provide an important gateway to understand the mechanism of action of TRIM22.
Materials and methods Chemicals and reagents Progesterone and 17β-estradiol were obtained from Sigma Chemical Co. (St. Louis, USA). IFNγ were purchased from ProSpec-Tany TechnoGene Ltd. (Rehovot, Israel). PCR reagents were from Fermentas, Inc. (Hanover, MD, USA). Restriction enzymes and buffers used in cloning were obtained from New England Biolabs, Inc. (Beverly, MA, USA). Fugene 6 transfection reagent was obtained from Roche Diagnostics (Indianapolis, IN, USA). Anti-FLAG M2 affinity beads and antibodies for FLAG (mouse-M2), p80-coilin (mouse, clone pδ), and EGFP (rabbit) were obtained from Sigma-Aldrich, Inc. (St. Louis, USA). Rabbit anti-fibrillarin antibody was obtained from Santa Cruz (Santa Cruz, CA, USA). Mouse anti-GAPDH antibody was obtained from Ambion Inc. (Austin, TX, USA). Rabbit anti-coilin antibody (204/ 10) was obtained from Dundee Cell Products (Dundee, Scotland, UK). Anti-mouse-IgG-HRP, anti-rabbit-IgG-HRP, and ECLplus were obtained from Amersham Biosciences (Piscataway, NJ, USA). Alexa Fluor 633 nm anti-mouse IgG, and anti-fade fluorescence mounting media were from Invitrogen Molecular Probes (Carlsbad, CA, USA). 488 nm anti-rabbit IgG and 4′,6diamidino-2-phenylindole (DAPI) were obtained from SigmaAldrich, Inc. All tissue culture plastics and most molecular biology reagents were obtained from Invitrogen Life Technologies (Carlsbad, CA, USA).
Cell lines Isolation and characterization of progesterone receptor (PR)transfected MDA-MB-231 cell clone ABC28 and vector-transfected
counterpart CTC15 have been described previously [19]. T47D cells were obtained from Dr. Suet Feung Ching (Hutchison/MRC Research Centre, University of Cambridge, UK). HeLa cells were obtained from Dr. Koh Cheng Gee (Nanyang Technological University, Singapore). MCF7 cells were obtained from A.T.C.C. (Manassas, VA, USA).
Cell culture and treatments All cells were routinely maintained in phenol-red containing Dulbecco's Modified Eagle Medium supplemented with 7.5% fetal calf serum (FCS), 2 mM glutamine and 40 μg/ml gentamycin (normal growth media). HeLa and MCF7 cells were treated with 250 or 500 IU/ml of IFNγ for 24 h in normal growth media. All hormone treatment experiments were conducted using phenol red-free Dulbecco's Modified Eagle Medium supplemented with 2 mM L-glutamine, 40 μg/ml gentamycin and 5% dextran-coated charcoal-treated FCS (Test Medium). FCS was treated with dextran-coated charcoal to remove the endogenous steroid hormones that might interfere with the effects of progesterone and 17β-estradiol. For hormone treatment, cells were grown in Test Medium for 48 h before they were treated with the hormone from a 1000-fold stock in ethanol. This gave a final concentration of ethanol of 0.1%. Treatment controls received 0.1% ethanol. To determine the effect of methotrexate on the distribution of endogenous TRIM22, MCF7 cells were grown on coverslips for 48 h in growth media containing 10% serum before treatment with 10 μM methotrexate in 10% serum containing media for 48 h.
Reverse-transcription and real-time PCR Total RNA was extracted using TRIzol reagent (Life Technologies, Inc., Gaithersburg, MD). cDNA was synthesized from 5 μg total RNA using random hexamers or oligo(dT) primers and SuperScript II reverse transcriptase (Invitrogen). Real-time PCR was performed using SYBR Green PCR reagent on an ABI Prism 7700 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) according to the manufacture's protocol. The TRIM22 and 36B4 (control to normalize for cDNA quantity) transcripts were amplified using specific primers (Table 1). Changes in fluorescence of the SYBR Green I dye in each cycle were monitored, and the cycle threshold (Ct), which is defined as the cycle number at which the amount of amplified target reaches a fixed threshold, was obtained for each gene. The relative amount of PCR products generated from each primer set was determined on the basis of the Ct value. The expression difference for each gene between ethanol (control) and progesterone-treated samples was calculated by normalizing Ct values with 36B4 gene expression according to the formula: Relative expression = 2[[Ct(control)TRIM22 − Ct (progesterone)TRIM22] − [Ct(control)36B4 − Ct(progesterone)36B4]] .
Table 1 – Forward and reverse primers used in the amplification of TRIM22 and 36B4 mRNA Gene (purpose) 36B4 TRIM22 (real-time PCR) TRIM22 (3′ end detection) TRIM22 probe for Northern blotting
Forward primer (5′ → 3′)
Reverse primer (5′ → 3′)
GATTGGCTACCCAACTGTTGCA CAATGTCACAAACCACGGAGCA GAAGGTGAGGTGAATGTGCTG TCACGAAGAGAAATATTGGCAGCC
CAGGGGCAGCAGCCACAAAGGC GGAAGAGATGGTCTCAGTAGAGAT CGGGGTACCTCAGGAGCTCGGTGGGCA ATGGATTTCTCAGTAAAGGTAG
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
1523
Northern blotting analysis
Whole cell lysate preparation and Western blotting
15 μg of total RNA from ABC28 cells treated for 16 or 24 h with 0.1% ethanol or 100 nM progesterone was resolved on a 1.5% formaldehyde-agarose gel. Integrity of gel-embedded RNA was confirmed prior to RNA denaturation by incubations in 1.5 M NaCl/ 0.5 M NaOH buffer and neutralization in 1.5 M NaCl/0.5 M Tris–Cl (pH 7.0) buffer. RNA was transferred overnight onto optimized nylon membranes (Amersham, Uppsala, Sweden) and covalently linked to the membrane by UV. A TRIM22 fragment corresponding to 1–1056 nt of the coding sequence was labeled with dCTP-32P by random priming reaction using the Redivue labeling kit (Amersham), and allowed to hybridize overnight to RNA at 42 °C in prewarmed ULTRAhyb buffer (Ambion). The hybridized membrane were exposed to X-ray film and developed using a phosphoimager (Molecular Imager FX, Biorad).
Cells were lysed with cold lysis buffer (50 mM Hepes, 150 mM NaCl, 1% Triton X-100, 5 μg/ml pepstatin A, 5 μg/ml leupeptin, 2 μg/ml aprotinin, 1 mM PMSF, 100 mM sodium fluoride and 1 mM sodium vanadate, pH 7.5) and passed through a 1/2 cc syringe 15 times. Lysate was cleared by centrifugation (12,000 g for 12 min) and protein quantitated. Western blotting analysis was carried out as previously described [21].
Cloning of TRIM22 The TRIM22 coding sequence was amplified by PCR from RZPD clone TRIM22-pCMV-SPORT6 (IMAGE id: 5583800). The amplified sequence tallied exactly with the TRIM22 sequence under the GenBank entry BC035582. The TRIM22 amplicons were cloned into pET-24(+)b, pEGFP-N1, pEYFP-C1 and pXJ-FLAG [20] vectors. Entire inserts and flanking regions were sequenced to ensure tags were in frame.
Generation of TRIM22 antibody in mice 498 aa His-tagged TRIM22 protein was expressed from the pET-24 (+)b vector in BL21(DE3) cells. The protein was purified using NiNTA agarose (Qiagen, GmBH, Germany) under denaturing conditions and refolded in cold refolding buffer (0.5 M L-arginine, 50 mM Tris–Cl, 20 mM CaCl2, 5 mM EDTA, 0.5 mM Cysteine, pH 7.4) before being used as an immunogen in Balb/c mice. Mice were handled following the guidelines on the Care and Use of Animals for Scientific Purposes (2004) set by National Advisory Committee for Laboratory Animal Research in Singapore.
siRNA transfection Three pre-designed TRIM22 siRNA sequences (IDs #107935, 107936, and 107937) and control siRNA sequences (ID #AM4611) were purchased from Ambion. 10 or 80 nM siRNA were transfected into ABC28 and HeLa cells using Lipofectamine 2000 (Invitrogen) for 8 h before treatment with progesterone (10 or 100 nM, ABC28) for 48 h or IFNγ (250 IU/ml, HeLa) for 24 h.
Immunofluorescence Cells were fixed with 3.7% formaldehyde and permeabilized with 0.2% Triton X-100. After washing with PBS, cells were blocked with 2% FCS/PBS, and incubated with primary antibodies in 2% FCS/PBS for 2 h at 37 °C. Cells were incubated with secondary antibodies and DAPI (0.1 μg/ml) in 2% FCS/PBS for 2 h at 37 °C before mounting on glass slides using anti-fade fluorescence mounting media. All images were acquired as black and white TIFF files with an Olympus Manual Reflected Fluorescence System attached to an Olympus DP30BW Digital Camera (Olympus Singapore Pte Ltd., Singapore) using a 60 × UPlanSApo oil objective. Color was added using DP Manager (Olympus).
Coimmunoprecipitation MCF7 cells were transfected with p80-coilin-EGFP (provided by Dr. Joseph Gall) and FLAG or FLAG-TRIM22 using Fugene 6 for 40 h. 400 μg of whole cell lysates, prepared as described earlier, were incubated with anti-FLAG M2 affinity beads overnight with rotation. Beads were pelleted by centrifugation, washed three times with cold TBS buffer (150 mM NaCl, 50 mM Tris–HCl, pH 7.4) and resuspended in an equal volume of 2 × sample buffer before analysis by Western blotting.
Cell synchronization and flow cytometry HeLa-TRIM22-EGFP cells were synchronized using the double thymidine block method modified from Pederson and Robbins and Li et al. [22,23]. Cells were exposed to 2.5 mM thymidine for 16 h, with thymidine removed for 12 h, after which time thymidine was added for an additional 16 h. Cells were harvested at 0, 3, 5, 8, 13 or 14 h after the second thymidine block. To determine the percentage of HeLa-TRIM22-EGFP cells in each stage of the cell cycle, cells were stained with propidium iodide (10 mM Tris–Cl, pH 8.8, 10 mM NaCl, 50 μg/ml propidium iodide, 10 μg/ml RNase A, 0.1% NP-40). Cells were analyzed by flow cytometry (FACSCalibur using CellQuest, BD Biosciences) after gating to remove cell debris and aggregates.
Plasmid transfection Cells were transiently transfected with constructs for 24–48 h using Fugene 6. For the generation of MCF7 and HeLa cells stably expressing TRIM22-EGFP (MCF7-TRIM22-EGFP and HeLaTRIM22-EGFP), cells were transfected with the TRIM22-pEGFPN1 construct using Lipofectamine reagent (Invitrogen) and selected in normal growth media containing 400–500 μg/ml Geneticin (Invitrogen). After clone selection and expansion, selected clones were sorted by FACSorter (BD Biosciences, San Jose, CA) to obtain cell populations enriched with TRIM22-EGFP expressing cells.
Results and discussion TRIM22 is a progesterone-inducible protein It is well-established that TRIM22 is a target of both Type I (α, β, ω) and Type II IFNs (γ). Here, we found that the TRIM22 transcript was also up-regulated by progesterone by up to 25-fold in a timedependent manner in MDA-MB-231 cells ectopically expressing the PR (ABC28) [19] (Fig. 1A) and in T47D cells expressing endogenous PR (Supplementary Fig. 1A). This up-regulation was
1524
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
Fig. 1 – The TRIM22 protein is greatly enhanced by progesterone and interferon. (A) TRIM22 mRNA expression levels were up-regulated in a time-dependent manner by progesterone in ABC28 cells, but not in CTC15 vector-transfected controls. Results are presented as fold changes after hormone treatment relative to ethanol-treated controls which is set at 1. (B) Northern blotting analysis on 15 μg of total RNA from ABC28 cells treated with ethanol (E) or progesterone (P) confirmed that progesterone induces the TRIM22 mRNA. (C) Progesterone, unlike 17β-estradiol, increases TRIM22 protein levels. ABC28 cells were treated for 48 h with ethanol and different concentrations of progesterone (1–100 nM). T47D cells were treated with ethanol, progesterone (100 nM), 17β-estradiol (1 nM) or both hormones for 24 h or 48 h. Whole cell lysates (25 μg) were analyzed by Western blotting using anti-TRIM22 and anti-GAPDH antibodies. GAPDH was analyzed as a loading control. The identity of the progesterone-induced band (−) (absent in ethanol-treated sample, E) was confirmed to be TRIM22 from its reduction in the presence of three different TRIM22 siRNA (S5, S6, S7) but not control siRNA (SC). (D) Interferon enhances TRIM22 protein levels. HeLa cells and MCF7 cells were treated for 24 h with 250 or 500 IU/ml interferon (IFN)γ. The identity of the IFNγ-induced band was confirmed by its dramatic reduction in the presence of TRIM22 siRNA 7 (S7) but not control siRNA (SC). Lysates were analyzed as above.
further confirmed in ABC28 cells by Northern blotting analysis which revealed a single up-regulated band at both the 16 h and 24 h time points (Fig. 1B). The TRIM22 transcript was initially reported to contain a 1329 nt coding region and this cDNA was used in the earlier studies [1,6]. However, this 1329 nt TRIM22 entry has since been replaced in the NCBI database by a transcript containing a 1497 nt coding region (a 168 nt 3′ extension of the first) (NM_006074.3) which codes for a 498 aa protein. To investigate if progesterone-regulated TRIM22 transcript contained the 1497 nt coding sequence, we performed RT-PCR with a pair of primers that would amplify a 879 nt sequence from the 3′ end of the 1497 nt coding sequence but would not yield any product if the coding sequence was 1329 nt. Supplementary Fig. 1B shows a distinct band of approximately 879 nt that was distinctively upregulated by progesterone and sequencing analysis verified that the 879 nt product corresponded to the 3′ sequence of the 1497 nt transcript. Northern blotting analysis indicated that progesterone up-regulated the expression of only one TRIM22 transcript. We can conclude therefore that progesterone induced the expression of a TRIM22 transcript that codes for a 498 aa protein.
To investigate the regulation and localization of endogenous TRIM22 protein, we generated mouse polyclonal antibodies against recombinant 498 aa His-tagged TRIM22 protein. This antibody detected a band at around 50–60 kDa that was drastically upregulated by progesterone in both ABC28 and T47D cells in a dosedependent manner (Fig. 1C). The identity of the up-regulated protein was validated with three pre-designed TRIM22 siRNA that reduced levels of the induced protein to near-basal levels (Fig. 1C). These TRIM22 siRNA were able to knockdown exogenously expressed FLAG-TRIM22 (Supplementary Fig. 2), thereby confirming their specificity. Whereas the effect of progesterone on TRIM22 protein levels in ABC28 cells was observed at both the 24 h (data not shown) and 48 h time points, in T47D cells, the effect was only prominent at the 24 h time point. We suspect that the protein was unstable in T47D cells. 17β-estradiol, an established mitogen in breast cancer cells including T47D cells, had no effect on TRIM22 protein levels either alone or in combination with progesterone. As expected, IFNγ increased TRIM22 protein levels in both HeLa cells and MCF7 cells; the up-regulated TRIM22 protein was reduced by TRIM22 siRNA (Fig 1D). IFNβ likewise increased TRIM22 protein levels (data not shown).
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
Progesterone has previously been shown to have strong growth-inhibitory effects in ABC28 cells and T47D cells both in vitro and in vivo [24–26]. The finding that progesterone greatly upregulates TRIM22 mRNA and protein levels supports a growth inhibitory role for TRIM22. A possible growth inhibitory role for TRIM22 is also supported by general observations that IFN β and γ are growth-inhibitory in several cell types including MCF7 cells and HeLa cells [27–29]. The effect of TRIM22 on cell growth and cell cycle progression are being investigated.
TRIM22 localizes in the nucleus and forms nuclear and nucleolar bodies The high specificity of the mouse polyclonal antibody against TRIM22 allowed us to use the antibody to investigate the cellular
1525
localization of the endogenous TRIM22 protein, which has not been reported before. In ABC28 cells, endogenous TRIM22 localized as a nucleoplasmic protein and as distinct NB, besides some cytoplasmic localization (ABC28 – E – TRIM22 panel, Fig. 2A). Both the nuclear forms of the protein were evidently increased by progesterone treatment (ABC28 – P – TRIM22 panel, Fig. 2A). More interestingly, about half of all progesterone-treated cells had TRIM22 NB within nucleoli, which were identified by the location of endogenous fibrillarin (Fig. 2A). In contrast, TRIM22 NB were not obvious in T47D cells and treatment of T47D cells with progesterone only increased nucleoplasmic TRIM22 (Supplementary Fig. 3A). The inability of TRIM22 to form NB in T47D cells may be related to the instability of the TRIM22 protein which could not be detected after 48 h progesterone treatment (Fig. 1C).
Fig. 2 – The TRIM22 protein localizes as a nucleoplasmic protein and forms distinct nuclear and nucleolar bodies. (A) ABC28 cells treated with ethanol (E) or progesterone (P; 10 nM) for 48 h were immunostained to detect TRIM22 (red) and fibrillarin (green) as a nucleolar marker. Merged images are presented showing co-localizations (yellow). (B) MCF7 and HeLa cells were immunostained to detect TRIM22 (red). (C) Localization of 498 aa TRIM22 was analyzed in MCF7 cells stably expressing TRIM22-EGFP or transiently (24–48 h) expressing EYFP-TRIM22 (yellow) or FLAG-TRIM22 (red). (A–C) Cells were counterstained with DAPI (blue) and visualized by fluorescence microscopy at a 60× or 96× magnification. Bars, 10 or 20 μm. Insets show representative cells (arrows) enlarged 2×.
1526
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
TRIM22 also localized as a nucleoplasmic protein and formed NB in HeLa and MCF7 cells (Fig. 2B). Like that observed in control ABC28 cells (ABC28 – E – TRIM22, Fig. 2A), TRIM22 NB was not found in all cells and, while the bodies were similarly sized, they were far less prevalent than in progesterone-treated ABC28 cells (ABC28 – P – TRIM22, Fig. 2A). Treatment with IFNγ did not noticeably enhance TRIM22 body formation in HeLa and MCF7 cells but enhanced nucleoplasmic staining (Supplementary Fig. 3B). In both HeLa and MCF7 cells, we also noted an increased cytoplasmic staining upon IFNγ treatment. The dispersed form of TRIM22 in HeLa cells and MCF7 cells predominated despite the greater induction of TRIM22 by IFNγ than by progesterone in ABC28 cells where TRIM22 is capable of forming NB. This suggests
that the ability of the TRIM22 protein to localize as NB and/or nucleoplasmic protein depends on certain cellular signal(s) or protein modifications. It is worth mentioning here that sumoylation of TRIM19/PML is essential for the formation of PML NB through non-covalent binding of PML to sumoylated PML [30]. There are also several highly probable sumoylation sites in TRIM22 and it remains to be seen if TRIM22 is sumoylated and if this modification affects TRIM22 NB formation. Enhanced green fluorescent protein (EGFP) tagged TRIM22 also localized as distinct NB and as a nucleoplasmic protein in MCF7 cells (Fig. 2C), confirming our observations of TRIM22's nuclear localization and NB formation. To exclude the possibility that the large EGFP tag on the C-terminus of TRIM22 modified the
Fig. 3 – Nucleolar bodies are drastically reduced upon silencing. (A) ABC28 cells, left untransfected or transfected with control siRNA or two different TRIM22 siRNA sequences (10 nM) before treatment with progesterone (10 nM) for 48 h, were immunostained with anti-TRIM22 antibody and visualized by fluorescence microscopy at 60× magnification. Images were taken with the same exposure. Bars, 20 μm. Insets show representative cells (arrows) enlarged 2×. (B) Numbers of ABC28 cells expressing nuclear and nucleolar TRIM22 bodies were quantitated in two independent experiments (untransfected, n = 144; control siRNA, n = 259; TRIM22 siRNA 6, n = 211; TRIM22 siRNA 7, n = 205). Percentage of cells having nuclear and nucleolar TRIM22 bodies are presented. Error bars reflect standard error of the mean.
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
conformation and/or the localization of the protein, the TRIM22 coding sequence was also cloned into the pEYFP-C1 and the pXJFLAG vectors such that the protein would have the enhanced yellow fluorescent protein (EYFP) or FLAG tags, respectively, on its Nterminus. TRIM22 localized as distinct NB regardless of tag size or whether the tag is fused to the C-terminus or N-terminus of TRIM22 (Fig. 2C). It should be noted that ectopically expressed TRIM22 localized exclusively as a nuclear protein in cells expressing low to
1527
moderate levels of the protein. The cytoplasmic signal observed with the anti-TRIM22 antibody could therefore either reflect nonspecific binding of the antibody or the recognition of continuously expressed TRIM22 before its translocation into the nucleus. The effect of two TRIM22 siRNA (siRNA 6 and 7) on the existence of TRIM22 as a nucleoplasmic protein and as nuclear and nucleolar bodies was investigated in ABC28 cells treated with progesterone. Interestingly, while TRIM22 siRNA did not decrease the numbers of
Fig. 4 – TRIM22 co-localizes partially with p80-coilin. (A) MCF7 cells stably expressing TRIM22-EGFP were immunostained with anti-p80-coilin antibody. Insets show representative cells (arrows) enlarged 2.5×. (B) Untreated and progesterone-treated ABC28 cells and (C) untreated MCF7 cells were immunostained with anti-TRIM22 and anti-p80-coilin antibodies. Endogenous TRIM22 (red) and p80-coilin (green) co-localize in nuclear bodies (arrows in ABC28 and MCF7 panels) and in strand-like structures in the cytoplasm (arrows in ABC28–48P panel). (A–C) All cells were counterstained with DAPI (blue) and visualized by fluorescence microscopy at 60× or 96× magnifications. Bars, 10 or 20 μm. Insets show representative cells (arrows) enlarged 2×. (D) Whole cell lysates from MCF7 cells co-expressing p80-coilin-EGFP and pXJ-FLAG (FLAG) or FLAG-TRIM22 were subjected to immunoprecipitation using anti-FLAG M2 affinity beads (400 μg IP). Immunoprecipitated proteins were analyzed alongside whole cell lysates (20 μg) by Western blotting using anti-EGFP and anti-FLAG antibodies.
1528
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
Fig. 5 – Dynamics of TRIM22 bodies during the cell cycle. (A) HeLa cells stably expressing TRIM22-EGFP were synchronized by the double thymidine block and cell cycle distribution at different time points after synchronization was analyzed by flow cytometry. Percentages of cells in each phase of the cycle are indicated. (B) Localization of TRIM22-EGFP at specific time-points corresponding to synchrony at the G1/S, early S, mid-late S, G2/M, and G0/G1 phases was analyzed by fluorescence microscopy at a 60× magnification. Bars, 20 μm. Insets show representative cells (arrows) enlarged 2.5×. (C) TRIM22-EGFP expression levels through the cell cycle were analyzed by Western blotting of whole cell lysates (20 μg) extracted from cells not subjected to the double thymidine block (asynchronous, AS) and cells collected at different time points (0, 3, 5, 8, 13 h) after synchronization. GAPDH was analyzed as a loading control.
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
cells expressing TRIM22 NB, the intensities of these bodies and nucleoplasmic staining were reduced (Fig. 3A). Most evidently, expression of TRIM22 siRNA led to disappearance of regular bodies within nucleoli of the majority of progesterone-treated cells (Fig. 3A). Quantitation analysis of random fields revealed that at least 50% of untransfected and control siRNA transfected cells contained nucleolar TRIM22 bodies. In the presence of TRIM22 siRNA 6 or 7, however, only 7–12% of cells had nucleolar TRIM22 bodies (Fig. 3B). Since much of the TRIM22 nucleolar bodies in ABC28 cells were reduced by TRIM22 siRNA, one possibility is that they were formed from nascent, progesterone-induced TRIM22 mRNA whereas the TRIM22 NB in the nucleus were mostly TRIM22 protein existing before progesterone treatment. It is therefore plausible that the formation of TRIM22 nucleolar bodies precedes that of NB. TRIM22 nucleolar bodies may be translocated out of the nucleoli to form NB. The findings also raise an interesting possibility that TRIM22 is involved in the specific function of the nucleolus, notably ribosome synthesis, cell-cycle regulation, and biogenesis of most cellular ribonuclear particles [31].
TRIM22 bodies partially overlap with the Cajal body signature protein p80-coilin As TRIM22 protein formed regular bodies similar in appearance to CBs which are also frequently observed next to and within nucleoli [32], the co-localization of TRIM22 with endogenous p80-coilin was investigated. TRIM22-EGFP bodies in stably-transfected MCF7 cells overlapped partially with endogenous CBs (Fig. 4A). We found that 144 (35%) of 415 endogenous CBs counted overlapped with TRIM22 NB in MCF7-TRIM22-EGFP cells. If cells with numerous TRIM22 speckles were excluded, all CBs overlapped with TRIM22 NB in 23 (18%) of 130 cells. The TRIM22 speckles may be the dispersed form of TRIM22 NB and numerous TRIM22 speckles in the cells would mean a greater chance of random encounter and may therefore erroneously increase the percentage of co-localization with CB. There was no obvious relationship between the numbers of CBs within the cell and the numbers of these CBs co-localizing with TRIM22 bodies. Analysis of confocal Z-stacks confirmed the co-localization of TRIM22-EGFP with endogenous CBs (Supplementary Fig. 4). While there were no detectable overlapping between TRIM22 NB and CB in untreated ABC28 cells and HeLa cells (results not shown), endogenous TRIM22 NB overlapped with CBs in a fraction of progesterone-treated ABC28 cells (Fig. 4B) and in MCF7 cells (Fig. 4C). In contrast, none of the TRIM22 NB co-localized with PML bodies (data not shown). Another interesting observation is that in almost all progesterone-treated ABC28 cells observed, TRIM22 colocalized with p80-coilin in the strand-like structures within cytoplasm (Fig. 4B). As the nuclear protein p80-coilin is a critical component of multifunctional CBs, the association between p80coilin and TRIM22 outside nuclei in progesterone-treated ABC28 cells is likely to have important implications for their function. It is to be noted that MCF7 cells with intense cytoplasmic staining of TRIM22 antibody in Fig. 4C had probably just undergone mitosis. As will be seen later in the cell synchronization study, HeLa cells stably expressing TRIM22-EGFP (Fig. 5B, 8 h — G2/M image) also had intense diffuse fluorescent signal in the cytoplasm during and after mitosis and this probably reflects dynamic changes of TRIM22 localization during cell-cycle progression. Through a co-immunoprecipitation experiment in MCF7 cells co-expressing both proteins, we were able to verify that FLAG-
1529
TRIM22 interacted with p80-coilin-EGFP (Fig. 4D). The finding that not all TRIM22 NB were associated with CBs suggests that these two types of NB may associate during specific phases of NB formation or during certain functional states of the NB.
TRIM22 nuclear bodies undergo dynamic changes during progression through the cell cycle Since TRIM22 co-localizes with CBs which undergo dynamic assembly and disassembly during cell cycle progression [8], we investigated if the formation of TRIM22 NB and nucleoplasmic TRIM22 would vary with cell-cycle progression. HeLa cells are easily synchronized by the double thymidine block method. HeLa cells stably expressing TRIM22-EGFP were sorted by FACS into three different populations depending on the level of fluorescence (high, medium and low). The cells chosen for the cell cycle analysis expressed low to medium levels of TRIM22-EGFP. We were able to achieve a relatively high percentage of synchronized cells in the G0/G1 (72.2% at 14 h post-release), S (68.9–70.8% at 3–5 h postrelease), and G2/M (66.4% at 8 h post-release) phases after releasing these cells from the double thymidine block (Fig. 5A). It appeared that G0/G1-phase cells, before entering S-phase, displayed regular TRIM22-EGFP bodies (Fig. 5B, 0 h). When cells were allowed to progress to the S-phase, the vast majority of these cells expressed TRIM22-EGFP in a speckled fashion (Fig. 5B, 3 h and 5 h). The speckled distribution was also observed in cells that had distinct nuclear boundaries in the G2/M population. However, during and immediately after mitosis, when DNA condenses and segregates, TRIM22-EGFP became completely diffused in the nucleus and cytoplasm (Fig. 5B, 8 h). The corresponding DAPI images of synchronized cells at the different time points have been presented in Supplementary Fig. 5. The diffused localization of TRIM22-EGFP was also observed in cells in telophase (panel ⁎ in 8 h). Multiple regular TRIM22 bodies re-appeared in cells undergoing cytokinesis (panel ⁎⁎ in 14 h) and were present after the cells entered G0/G1-phase (Fig. 5B, 14 h). These changes in the distribution of TRIM22-EGFP are unlikely to be due to differing TRIM22-EGFP expression levels as these were relatively constant during the different phases (Fig. 5C). Interestingly, TRIM22-EGFP bodies were found to be localized frequently within and next to nucleoli in cells that had entered the G0/G1-phase (Fig. 5B, 14 h). This is consistent with the observation that 50% of progesteronetreated ABC28 cells, which are mostly arrested in the G0/G1-phase [19], contained nucleolar TRIM22 bodies. The nuclear dynamic changes of TRIM22 NB are remarkably similar to CBs which disassemble during mitosis and reassemble during the mid-G1 phase [8]. The cycle-dependent dynamic changes of CB are associated with its function in cell cycle regulation. For instance, the CB-associated protein p220NPAT is essential for G0/G1 progression to S-phase in HCT116 cells [33]. Cells lacking p220NPAT also displayed defects in the focal localization of p80-coilin as CB. The dynamic changes of TRIM22 NB during the cell cycle may likewise reflect its role in cell cycle regulation. Since TRIM22-EGFP localizes in a speckled fashion in S-phase synchronized HeLa cells, we tested if endogenous TRIM22 dispersed in MCF7 cells arrested in the S-phase by methotrexate [34]. Indeed, in contrast to the bright regular TRIM22 NB seen in control cells, methotrexate-treated MCF7 cells contained TRIM22 distributed in a speckled manner (Fig. 6). However, methotrexate did not cause any obvious change of TRIM22 NB in ABC28 cells
1530
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
Fig. 6 – Methotrexate induces a dispersion of TRIM22 bodies. MCF7 cells were plated in growth media containing 10% serum and treated with or without 10 μM methotrexate in 10% serum for 48 h. Cells were immunostained with the anti-TRIM22 antibody and counterstained with DAPI before fluorescence microscopy at a 60× magnification. Bars, 20 μm. Insets show representative cells (arrows) enlarged 2.5×.
compared to untreated controls (data not shown). Therefore, the dynamic changes in TRIM22 NB and cellular localization also vary depending on cell types.
Concluding remarks NB are subnuclear domains that contain an assortment of components involved in diverse nuclear functions including ribosome biogenesis, pre-mRNA processing, splicing, apoptosis and transport [35]. By generating effective TRIM22 antibodies and using specific siRNA, the present study provides strong evidence to add TRIM22 NB to the growing list of NB. Like other NB, TRIM22 NB are dynamic structures that can form and disintegrate in response to cellular signals. This notion is supported by studies with progesterone and IFNγ. While progesterone increased both nucleoplasmic TRIM22 protein and TRIM22 nuclear and nucleolar bodies in ABC28 cells, IFNγ treatment only enhanced nucleoplasmic TRIM22 with no evident increase in TRIM22 NB in HeLa cells and MCF7 cells. This is despite the fact that the induction of TRIM22 by IFNγ was of much greater magnitude than by progesterone and that endogenous TRIM22 does form NB in these cells. We suggest that these different forms are related to the functionality of TRIM22 in response to these agents in different cell types. One of the most prominent aspects of NB dynamics is their changes in forms during the cell cycle. In synchronized HeLa cells, TRIM22 typically formed NB during the G0/G1-phase. The TRIM22 NB was transformed into speckle-like structures during S-phase and probably early G2/M-phase. Consistent with this finding in HeLa cells, we also observed that TRIM22 formed speckle-like structures in MCF7 cells arrested in the S-phase by methotrexate. During the M-phase, TRIM22 became diffused but the NB was assembled again by the end of cytokinesis. This cell cycle-dependent dynamism of TRIM22 NB and an association of TRIM22 induction by progesterone and IFN with their growth-
inhibitory effects in breast cancer cells [19,25,28] support a role for TRIM22 in the inhibition of cell cycle progression. It is also important to highlight the prominent presence of TRIM22 NB in the nucleolus following progesterone-treatment. Approximately 50% of progesterone-treated ABC28 cells contained nucleolar TRIM22 NB that can be readily eliminated by specific TRIM22 siRNA. Nucleolar localization of TRIM22 NB is by no means a meaningless sequestration of abundant protein because the protein levels of transfected TRIM22-EGFP or FLAGTRIM22 were much higher than progesterone-induced TRIM22, yet much less TRIM22-EGFP or FLAG-TRIM22 NB appeared in the nucleolus. It is likely therefore that TRIM22 is involved in one or more nucleolar functions, which include rDNA transcription, prerRNA processing, ribosome assembly and formation of the telomerase ribonuclear protein complex. It is interesting to mention in this context that TRIM22 is well-known for its antiviral properties and the nucleolus has been suggested as a gateway for viral replication [36]. Viral proteins are known to interact with nucleolar proteins such as B23 and fibrillarin and use the nucleolus as a site of replication [37–39]. Although we did not observe an increase of TRIM22 NB within the nucleolus in IFNγ-treated cells, we cannot exclude the possibility that IFNγinduced TRIM22 has a diffuse localization or transient aggregation in the nucleolus, which is difficult to ascertain. It is also possible that progesterone- and IFNγ-induced TRIM22 may elicit antiviral effects via different mechanisms depending on the presence of the co-induced partner proteins. Future work in this area should include an investigation into the association between TRIM22's nucleolar localization and its antiviral properties.
Acknowledgments The work is supported by grants from the Ministry of Education, the Republic of Singapore, to Valerie CL Lin.
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
We wish to thank Dr. E. Manser et al. for the pXJ-FLAG plasmid and Dr. Joseph Gall for the p80-coilin-EGFP plasmid. We also wish to thank Dr. Stephen Ogg and Dr. Li Hoi Yeung for helpful discussions.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.yexcr.2009.01.028.
REFERENCES
[1] A. Reymond, G. Meroni, A. Fantozzi, G. Merla, S. Cairo, L. Luzi, D. Riganelli, E. Zanaria, S. Messali, S. Cainarca, A. Guffanti, S. Minucci, P.G. Pelicci, A. Ballabio, The tripartite motif family identifies cell compartments, EMBO J. 20 (2001) 2140–2151. [2] S.D. Barr, J.R. Smiley, F.D. Bushman, The interferon response inhibits HIV particle production by induction of TRIM22, PLoS Pathog. 4 (2008) e1000007. [3] A. Bouazzaoui, M. Kreutz, V. Eisert, N. Dinauer, A. Heinzelmann, S. Hallenberger, J. Strayle, R. Walker, H. Rubsamen-Waigmann, R. Andreesen, H. von Briesen, Stimulated trans-acting factor of 50 kDa (Staf50) inhibits HIV-1 replication in human monocyte-derived macrophages, Virology 356 (2006) 79–94. [4] E. Izmailova, F.M. Bertley, Q. Huang, N. Makori, C.J. Miller, R.A. Young, A. Aldovini, HIV-1 Tat reprograms immature dendritic cells to express chemoattractants for activated T cells and macrophages, Nat. Med. 9 (2003) 191–197. [5] S. Obad, H. Brunnstrom, J. Vallon-Christersson, A. Borg, K. Drott, U. Gullberg, Staf50 is a novel p53 target gene conferring reduced clonogenic growth of leukemic U-937 cells, Oncogene 23 (2004) 4050–4059. [6] C. Tissot, N. Mechti, Molecular cloning of a new interferon-induced factor that represses human immunodeficiency virus type 1 long terminal repeat expression, J. Biol. Chem. 270 (1995) 14891–14898. [7] J. Zhang, S.C. Das, C. Kotalik, A.K. Pattnaik, L. Zhang, The latent membrane protein 1 of Epstein–Barr virus establishes an antiviral state via induction of interferon-stimulated genes, J. Biol. Chem. 279 (2004) 46335–46342. [8] L.E. Andrade, E.M. Tan, E.K. Chan, Immunocytochemical analysis of the coiled body in the cell cycle and during cell proliferation, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 1947–1951. [9] M. Platani, I. Goldberg, J.R. Swedlow, A.I. Lamond, In vivo analysis of Cajal body movement, separation, and joining in live human cells, J. Cell Biol. 151 (2000) 1561–1574. [10] M. Cioce, A.I. Lamond, Cajal bodies: a long history of discovery, Annu. Rev. Cell. Dev. Biol. 21 (2005) 105–131. [11] B.E. Jady, E. Bertrand, T. Kiss, Human telomerase RNA and box H/ACA scaRNAs share a common Cajal body-specific localization signal, J. Cell Biol. 164 (2004) 647–652. [12] S. Marcand, V. Brevet, C. Mann, E. Gilson, Cell cycle restriction of telomere elongation, Curr. Biol. 10 (2000) 487–490. [13] D. Stanek, K.M. Neugebauer, The Cajal body: a meeting place for spliceosomal snRNPs in the nuclear maze, Chromosoma 115 (2006) 343–354. [14] A. Narayanan, W. Speckmann, R. Terns, M.P. Terns, Role of the box C/D motif in localization of small nucleolar RNAs to coiled bodies and nucleoli, Mol. Biol. Cell. 10 (1999) 2131–2147. [15] L.S. Shopland, M. Byron, J.L. Stein, J.B. Lian, G.S. Stein, J.B. Lawrence, Replication-dependent histone gene expression is related to Cajal body (CB) association but does not require sustained CB contact, Mol. Biol. Cell. 12 (2001) 565–576. [16] M. Cioce, A.I. Lamond, Cajal bodies: a long history of discovery, Annu. Rev. Cell. Dev. Biol. 21 (2005) 105–131.
1531
[17] Z. Duan, B. Gao, W. Xu, S. Xiong, Identification of TRIM22 as a RING finger E3 ubiquitin ligase, Biochem. Biophys. Res. Commun. 374 (2008) 502–506. [18] X. Li, B. Gold, C. O'HUigin, F. Diaz-Griffero, B. Song, Z. Si, Y. Li, W. Yuan, M. Stremlau, C. Mische, H. Javanbakht, M. Scally, C. Winkler, M. Dean, J. Sodroski, Unique features of TRIM5alpha among closely related human TRIM family members, Virology 360 (2007) 419–433. [19] V.C. Lin, E.H. Ng, S.E. Aw, M.G. Tan, E.H. Ng, V.S. Chan, G.H. Ho, Progestins inhibit the growth of MDA-MB-231 cells transfected with progesterone receptor complementary DNA, Clin. Cancer Res. 5 (1999) 395–403. [20] E. Manser, H.Y. Huang, T.H. Loo, X.Q. Chen, J.M. Dong, T. Leung, L. Lim, Expression of constitutively active alpha-PAK reveals effects of the kinase on actin and focal complexes, Mol. Cell. Biol. 17 (1997) 1129–1143. [21] Z.Y. Zheng, B.H. Bay, S.E. Aw, V.C. Lin, A novel antiestrogenic mechanism in progesterone receptor-transfected breast cancer cells, J. Biol. Chem. 280 (2005) 17480–17487. [22] L. Li, K. Roy, S. Katyal, X. Sun, S. Bleoo, R. Godbout, Dynamic nature of cleavage bodies and their spatial relationship to DDX1 bodies, Cajal bodies, and gems, Mol. Biol. Cell. 17 (2006) 1126–1140. [23] T. Pederson, E. Robbins, A method for improving synchrony in the G2 phase of the cell cycle, J. Cell. Biol. 49 (1971) 942–945. [24] S.D. Groshong, G.I. Owen, B. Grimison, I.E. Schauer, M.C. Todd, T.A. Langan, R.A. Sclafani, C.A. Lange, K.B. Horwitz, Biphasic regulation of breast cancer cell growth by progesterone: role of the cyclin-dependent kinase inhibitors, p21 and p27(Kip1), Mol. Endocrinol. 11 (1997) 1593–1607. [25] V.C. Lin, A.S. Eng, N.E. Hen, E.H. Ng, S.H. Chowdhury, Effect of progesterone on the invasive properties and tumor growth of progesterone receptor-transfected breast cancer cells MDA-MB-231, Clin. Cancer Res. 7 (2001) 2880–2886. [26] V.C. Lin, E.H. Ng, S.E. Aw, M.G. Tan, B.H. Bay, Progesterone induces focal adhesion in breast cancer cells MDA-MB-231 transfected with progesterone receptor complementary DNA, Mol. Endocrinol. 14 (2000) 348–358. [27] L. Fransen, J. Van der Heyden, R. Ruysschaert, W. Fiers, Recombinant tumor necrosis factor: its effect and its synergism with interferon-gamma on a variety of normal and transformed human cell lines, Eur. J. Cancer Clin. Oncol. 22 (1986) 419–426. [28] J.L. Gooch, R.E. Herrera, D. Yee, The role of p21 in interferon gamma-mediated growth inhibition of human breast cancer cells, Cell Growth Differ. 11 (2000) 335–342. [29] A.C. Hobeika, W. Etienne, P.E. Cruz, P.S. Subramaniam, H.M. Johnson, IFNgamma induction of p21WAF1 in prostate cancer cells: role in cell cycle, alteration of phenotype and invasive potential, Int. J. Cancer 77 (1998) 138–145. [30] T.H. Shen, H.K. Lin, P.P. Scaglioni, T.M. Yung, P.P. Pandolfi, The mechanisms of PML-nuclear body formation, Mol. Cell. 24 (2006) 331–339. [31] I. Raska, P.J. Shaw, D. Cmarko, Structure and function of the nucleolus in the spotlight, Curr. Opin. Cell Biol. 18 (2006) 325–334. [32] R.L. Ochs, T.W. Stein Jr., E.M. Tan, Coiled bodies in the nucleolus of breast cancer cells, J. Cell. Sci. 107 (Pt 2) (1994) 385–399. [33] X. Ye, Y. Wei, G. Nalepa, J.W. Harper, The cyclin E/Cdk2 substrate p220(NPAT) is required for S-phase entry, histone gene expression, and Cajal body maintenance in human somatic cells, Mol. Cell Biol. 23 (2003) 8586–8600. [34] J.H. van Dierendonck, J.H. Wijsman, R. Keijzer, C.J. van de Velde, C.J. Cornelisse, Cell-cycle-related staining patterns of anti-proliferating cell nuclear antigen monoclonal antibodies. Comparison with BrdUrd labeling and Ki-67 staining, Am. J. Pathol. 138 (1991) 1165–1172. [35] A.G. Matera, Nuclear bodies: multifaceted subdomains of the interchromatin space, Trends Cell Biol. 9 (1999) 302–309. [36] J.A. Hiscox, The nucleolus—a gateway to viral infection? Arch. Virol. 147 (2002) 1077–1089.
1532
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
[37] S.H. Kim, E.V. Ryabov, N.O. Kalinina, D.V. Rakitina, T. Gillespie, S. MacFarlane, S. Haupt, J.W. Brown, M. Taliansky, Cajal bodies and the nucleolus are required for a plant virus systemic infection, EMBO J. 26 (2007) 2169–2179. [38] S.H. Kim, S. Macfarlane, N.O. Kalinina, D.V. Rakitina, E.V. Ryabov, T. Gillespie, S. Haupt, J.W. Brown, M. Taliansky, Interaction of a plant
virus-encoded protein with the major nucleolar protein fibrillarin is required for systemic virus infection, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 11115–11120. [39] A. Michienzi, S. Li, J.A. Zaia, J.J. Rossi, A nucleolar TAR decoy inhibitor of HIV replication, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 14047–14052.
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / y e x c r
Research Article
Dynamic localization of tripartite motif-containing 22 in nuclear and nucleolar bodies Gayathri Sivaramakrishnan, Yang Sun, Si Kee Tan, Valerie C.L. Lin⁎ School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore
A R T I C L E I N F O R M AT I O N
AB ST R AC T
Article Chronology:
Tripartite motif-containing 22 (TRIM22) exhibits antiviral and growth inhibitory properties, but
Received 30 October 2008
there has been no study on the localization and dynamics of the endogenous TRIM22 protein. We
Revised version received
report here that TRIM22 is dramatically induced by progesterone in MDA-MB-231-derived ABC28
21 January 2009
cells and T47D cells. This induction was associated with an increase in TRIM22 nuclear bodies (NB),
Accepted 29 January 2009
and an even more prominent increase in nucleolar TRIM22 bodies. Distinct endogenous TRIM22 NB
Available online 10 February 2009
were also demonstrated in several other cell lines including MCF7 and HeLa cells. These TRIM22 NB resemble Cajal bodies, co-localized with these structures and co-immunoprecipitated with p80-
Keywords:
coilin. However, IFNγ-induced TRIM22 in HeLa and MCF7 cells did not form NB, implying the forms
TRIM22
and distribution of TRIM22 are regulated by specific cellular signals. This notion is also supported by
Nuclear bodies
the observation that TRIM22 NB undergoes dynamic cell-cycle dependent changes in distribution
Progesterone
such that TRIM22 NB started to form in early G0/G1 but became dispersed in the S-phase. In light of
Cajal bodies
its potential antiviral and antitumor properties, the findings here provide an interesting gateway to
Breast cancer
study the relationship between the different forms and functions of TRIM22. © 2009 Elsevier Inc. All rights reserved.
Introduction Tripartite motif-containing 22 (TRIM22/Staf50) is a member of the TRIM family of proteins, which have a conserved domain structure consisting of, at their N-terminal ends, the Really interesting new gene (RING), B-box, and Coiled-coil (CC) domains [1], and a variable C-terminal region. In TRIM22, this C-terminal region has the SplA and ryanodine receptor (SPRY) domain. TRIM22 is an interferon (IFN)- and p53-inducible gene that inhibits clonogenic growth and HIV-1 infection when over-expressed [2–7]. There is also evidence that TRIM22 interferes with viral gene transcription and Gag protein localization [2,6]. The E3 ligase activity of the RING domain appears to be critical for the antiviral activity of TRIM22 [2].
Nuclear bodies (NB) are nuclear subdomains that contain an assortment of proteins and are involved in a number of vital nuclear functions. Cajal bodies (CB) are the better characterized NB marked by the presence of p80-coilin. CBs disassemble during mitosis, reassemble during the mid-G1 phase, and are found next to nucleoli and within the nucleoplasm [8,9]. CBs are also enriched in at least 30 proteins belonging to several different classes, including ribonuclear proteins, nucleolar proteins, transcription factors and proteins involved in the regulation of the cell cycle (reviewed in [10]). CBs are involved in a number of biological processes such as the biogenesis of small nuclear and nucleolar ribonuclear proteins, gene replication and transcription and biogenesis of human telomerase [11–16].
⁎ Corresponding author. Fax: +65 6791 3856. E-mail address: [email protected] (V.C.L. Lin). Abbreviations: CB, Cajal body; CC, Coiled-coil; Ct, Cycle threshold; DAPI, 4′,6-diamidino-2-phenylindole; EGFP, Enhanced green fluorescent protein; EYFP, Enhanced yellow fluorescent protein; FCS, Fetal calf serum; IFN, Interferon; NB, Nuclear bodies; PR, Progesterone receptor; RING, Really interesting new gene; SPRY, SplA and ryanodine receptor; TRIM, Tripartite motif-containing; TRIM22, Tripartite motif-containing 22 0014-4827/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2009.01.028
1522
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
There have been conflicting reports regarding the localization of ectopically expressed TRIM22 [1,17,18]. The shorter form of TRIM22 (442 aa) translated from a 1329 nt coding sequence appears to localize as cytoplasmic aggregates with some nuclear localization [1]. Recently, a c-myc-tagged protein product of the 1497 nt TRIM22 coding sequence amplified from human peripheral blood mononuclear cells localized exclusively as a nuclear protein [17]. However, the localization of endogenous TRIM22, which is essential for understanding its cellular function, is unknown. Using a specific antibody and siRNAs against TRIM22, we present evidence that TRIM22 localizes predominantly to the nucleus but is also found in the cytoplasm to some extent. TRIM22 forms distinct NB that are regulated by signaling molecules including progesterone and IFNγ and during the cell cycle. TRIM22 NB are adjacent to CB and interact with p80-coilin. The results provide an important gateway to understand the mechanism of action of TRIM22.
Materials and methods Chemicals and reagents Progesterone and 17β-estradiol were obtained from Sigma Chemical Co. (St. Louis, USA). IFNγ were purchased from ProSpec-Tany TechnoGene Ltd. (Rehovot, Israel). PCR reagents were from Fermentas, Inc. (Hanover, MD, USA). Restriction enzymes and buffers used in cloning were obtained from New England Biolabs, Inc. (Beverly, MA, USA). Fugene 6 transfection reagent was obtained from Roche Diagnostics (Indianapolis, IN, USA). Anti-FLAG M2 affinity beads and antibodies for FLAG (mouse-M2), p80-coilin (mouse, clone pδ), and EGFP (rabbit) were obtained from Sigma-Aldrich, Inc. (St. Louis, USA). Rabbit anti-fibrillarin antibody was obtained from Santa Cruz (Santa Cruz, CA, USA). Mouse anti-GAPDH antibody was obtained from Ambion Inc. (Austin, TX, USA). Rabbit anti-coilin antibody (204/ 10) was obtained from Dundee Cell Products (Dundee, Scotland, UK). Anti-mouse-IgG-HRP, anti-rabbit-IgG-HRP, and ECLplus were obtained from Amersham Biosciences (Piscataway, NJ, USA). Alexa Fluor 633 nm anti-mouse IgG, and anti-fade fluorescence mounting media were from Invitrogen Molecular Probes (Carlsbad, CA, USA). 488 nm anti-rabbit IgG and 4′,6diamidino-2-phenylindole (DAPI) were obtained from SigmaAldrich, Inc. All tissue culture plastics and most molecular biology reagents were obtained from Invitrogen Life Technologies (Carlsbad, CA, USA).
Cell lines Isolation and characterization of progesterone receptor (PR)transfected MDA-MB-231 cell clone ABC28 and vector-transfected
counterpart CTC15 have been described previously [19]. T47D cells were obtained from Dr. Suet Feung Ching (Hutchison/MRC Research Centre, University of Cambridge, UK). HeLa cells were obtained from Dr. Koh Cheng Gee (Nanyang Technological University, Singapore). MCF7 cells were obtained from A.T.C.C. (Manassas, VA, USA).
Cell culture and treatments All cells were routinely maintained in phenol-red containing Dulbecco's Modified Eagle Medium supplemented with 7.5% fetal calf serum (FCS), 2 mM glutamine and 40 μg/ml gentamycin (normal growth media). HeLa and MCF7 cells were treated with 250 or 500 IU/ml of IFNγ for 24 h in normal growth media. All hormone treatment experiments were conducted using phenol red-free Dulbecco's Modified Eagle Medium supplemented with 2 mM L-glutamine, 40 μg/ml gentamycin and 5% dextran-coated charcoal-treated FCS (Test Medium). FCS was treated with dextran-coated charcoal to remove the endogenous steroid hormones that might interfere with the effects of progesterone and 17β-estradiol. For hormone treatment, cells were grown in Test Medium for 48 h before they were treated with the hormone from a 1000-fold stock in ethanol. This gave a final concentration of ethanol of 0.1%. Treatment controls received 0.1% ethanol. To determine the effect of methotrexate on the distribution of endogenous TRIM22, MCF7 cells were grown on coverslips for 48 h in growth media containing 10% serum before treatment with 10 μM methotrexate in 10% serum containing media for 48 h.
Reverse-transcription and real-time PCR Total RNA was extracted using TRIzol reagent (Life Technologies, Inc., Gaithersburg, MD). cDNA was synthesized from 5 μg total RNA using random hexamers or oligo(dT) primers and SuperScript II reverse transcriptase (Invitrogen). Real-time PCR was performed using SYBR Green PCR reagent on an ABI Prism 7700 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) according to the manufacture's protocol. The TRIM22 and 36B4 (control to normalize for cDNA quantity) transcripts were amplified using specific primers (Table 1). Changes in fluorescence of the SYBR Green I dye in each cycle were monitored, and the cycle threshold (Ct), which is defined as the cycle number at which the amount of amplified target reaches a fixed threshold, was obtained for each gene. The relative amount of PCR products generated from each primer set was determined on the basis of the Ct value. The expression difference for each gene between ethanol (control) and progesterone-treated samples was calculated by normalizing Ct values with 36B4 gene expression according to the formula: Relative expression = 2[[Ct(control)TRIM22 − Ct (progesterone)TRIM22] − [Ct(control)36B4 − Ct(progesterone)36B4]] .
Table 1 – Forward and reverse primers used in the amplification of TRIM22 and 36B4 mRNA Gene (purpose) 36B4 TRIM22 (real-time PCR) TRIM22 (3′ end detection) TRIM22 probe for Northern blotting
Forward primer (5′ → 3′)
Reverse primer (5′ → 3′)
GATTGGCTACCCAACTGTTGCA CAATGTCACAAACCACGGAGCA GAAGGTGAGGTGAATGTGCTG TCACGAAGAGAAATATTGGCAGCC
CAGGGGCAGCAGCCACAAAGGC GGAAGAGATGGTCTCAGTAGAGAT CGGGGTACCTCAGGAGCTCGGTGGGCA ATGGATTTCTCAGTAAAGGTAG
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
1523
Northern blotting analysis
Whole cell lysate preparation and Western blotting
15 μg of total RNA from ABC28 cells treated for 16 or 24 h with 0.1% ethanol or 100 nM progesterone was resolved on a 1.5% formaldehyde-agarose gel. Integrity of gel-embedded RNA was confirmed prior to RNA denaturation by incubations in 1.5 M NaCl/ 0.5 M NaOH buffer and neutralization in 1.5 M NaCl/0.5 M Tris–Cl (pH 7.0) buffer. RNA was transferred overnight onto optimized nylon membranes (Amersham, Uppsala, Sweden) and covalently linked to the membrane by UV. A TRIM22 fragment corresponding to 1–1056 nt of the coding sequence was labeled with dCTP-32P by random priming reaction using the Redivue labeling kit (Amersham), and allowed to hybridize overnight to RNA at 42 °C in prewarmed ULTRAhyb buffer (Ambion). The hybridized membrane were exposed to X-ray film and developed using a phosphoimager (Molecular Imager FX, Biorad).
Cells were lysed with cold lysis buffer (50 mM Hepes, 150 mM NaCl, 1% Triton X-100, 5 μg/ml pepstatin A, 5 μg/ml leupeptin, 2 μg/ml aprotinin, 1 mM PMSF, 100 mM sodium fluoride and 1 mM sodium vanadate, pH 7.5) and passed through a 1/2 cc syringe 15 times. Lysate was cleared by centrifugation (12,000 g for 12 min) and protein quantitated. Western blotting analysis was carried out as previously described [21].
Cloning of TRIM22 The TRIM22 coding sequence was amplified by PCR from RZPD clone TRIM22-pCMV-SPORT6 (IMAGE id: 5583800). The amplified sequence tallied exactly with the TRIM22 sequence under the GenBank entry BC035582. The TRIM22 amplicons were cloned into pET-24(+)b, pEGFP-N1, pEYFP-C1 and pXJ-FLAG [20] vectors. Entire inserts and flanking regions were sequenced to ensure tags were in frame.
Generation of TRIM22 antibody in mice 498 aa His-tagged TRIM22 protein was expressed from the pET-24 (+)b vector in BL21(DE3) cells. The protein was purified using NiNTA agarose (Qiagen, GmBH, Germany) under denaturing conditions and refolded in cold refolding buffer (0.5 M L-arginine, 50 mM Tris–Cl, 20 mM CaCl2, 5 mM EDTA, 0.5 mM Cysteine, pH 7.4) before being used as an immunogen in Balb/c mice. Mice were handled following the guidelines on the Care and Use of Animals for Scientific Purposes (2004) set by National Advisory Committee for Laboratory Animal Research in Singapore.
siRNA transfection Three pre-designed TRIM22 siRNA sequences (IDs #107935, 107936, and 107937) and control siRNA sequences (ID #AM4611) were purchased from Ambion. 10 or 80 nM siRNA were transfected into ABC28 and HeLa cells using Lipofectamine 2000 (Invitrogen) for 8 h before treatment with progesterone (10 or 100 nM, ABC28) for 48 h or IFNγ (250 IU/ml, HeLa) for 24 h.
Immunofluorescence Cells were fixed with 3.7% formaldehyde and permeabilized with 0.2% Triton X-100. After washing with PBS, cells were blocked with 2% FCS/PBS, and incubated with primary antibodies in 2% FCS/PBS for 2 h at 37 °C. Cells were incubated with secondary antibodies and DAPI (0.1 μg/ml) in 2% FCS/PBS for 2 h at 37 °C before mounting on glass slides using anti-fade fluorescence mounting media. All images were acquired as black and white TIFF files with an Olympus Manual Reflected Fluorescence System attached to an Olympus DP30BW Digital Camera (Olympus Singapore Pte Ltd., Singapore) using a 60 × UPlanSApo oil objective. Color was added using DP Manager (Olympus).
Coimmunoprecipitation MCF7 cells were transfected with p80-coilin-EGFP (provided by Dr. Joseph Gall) and FLAG or FLAG-TRIM22 using Fugene 6 for 40 h. 400 μg of whole cell lysates, prepared as described earlier, were incubated with anti-FLAG M2 affinity beads overnight with rotation. Beads were pelleted by centrifugation, washed three times with cold TBS buffer (150 mM NaCl, 50 mM Tris–HCl, pH 7.4) and resuspended in an equal volume of 2 × sample buffer before analysis by Western blotting.
Cell synchronization and flow cytometry HeLa-TRIM22-EGFP cells were synchronized using the double thymidine block method modified from Pederson and Robbins and Li et al. [22,23]. Cells were exposed to 2.5 mM thymidine for 16 h, with thymidine removed for 12 h, after which time thymidine was added for an additional 16 h. Cells were harvested at 0, 3, 5, 8, 13 or 14 h after the second thymidine block. To determine the percentage of HeLa-TRIM22-EGFP cells in each stage of the cell cycle, cells were stained with propidium iodide (10 mM Tris–Cl, pH 8.8, 10 mM NaCl, 50 μg/ml propidium iodide, 10 μg/ml RNase A, 0.1% NP-40). Cells were analyzed by flow cytometry (FACSCalibur using CellQuest, BD Biosciences) after gating to remove cell debris and aggregates.
Plasmid transfection Cells were transiently transfected with constructs for 24–48 h using Fugene 6. For the generation of MCF7 and HeLa cells stably expressing TRIM22-EGFP (MCF7-TRIM22-EGFP and HeLaTRIM22-EGFP), cells were transfected with the TRIM22-pEGFPN1 construct using Lipofectamine reagent (Invitrogen) and selected in normal growth media containing 400–500 μg/ml Geneticin (Invitrogen). After clone selection and expansion, selected clones were sorted by FACSorter (BD Biosciences, San Jose, CA) to obtain cell populations enriched with TRIM22-EGFP expressing cells.
Results and discussion TRIM22 is a progesterone-inducible protein It is well-established that TRIM22 is a target of both Type I (α, β, ω) and Type II IFNs (γ). Here, we found that the TRIM22 transcript was also up-regulated by progesterone by up to 25-fold in a timedependent manner in MDA-MB-231 cells ectopically expressing the PR (ABC28) [19] (Fig. 1A) and in T47D cells expressing endogenous PR (Supplementary Fig. 1A). This up-regulation was
1524
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
Fig. 1 – The TRIM22 protein is greatly enhanced by progesterone and interferon. (A) TRIM22 mRNA expression levels were up-regulated in a time-dependent manner by progesterone in ABC28 cells, but not in CTC15 vector-transfected controls. Results are presented as fold changes after hormone treatment relative to ethanol-treated controls which is set at 1. (B) Northern blotting analysis on 15 μg of total RNA from ABC28 cells treated with ethanol (E) or progesterone (P) confirmed that progesterone induces the TRIM22 mRNA. (C) Progesterone, unlike 17β-estradiol, increases TRIM22 protein levels. ABC28 cells were treated for 48 h with ethanol and different concentrations of progesterone (1–100 nM). T47D cells were treated with ethanol, progesterone (100 nM), 17β-estradiol (1 nM) or both hormones for 24 h or 48 h. Whole cell lysates (25 μg) were analyzed by Western blotting using anti-TRIM22 and anti-GAPDH antibodies. GAPDH was analyzed as a loading control. The identity of the progesterone-induced band (−) (absent in ethanol-treated sample, E) was confirmed to be TRIM22 from its reduction in the presence of three different TRIM22 siRNA (S5, S6, S7) but not control siRNA (SC). (D) Interferon enhances TRIM22 protein levels. HeLa cells and MCF7 cells were treated for 24 h with 250 or 500 IU/ml interferon (IFN)γ. The identity of the IFNγ-induced band was confirmed by its dramatic reduction in the presence of TRIM22 siRNA 7 (S7) but not control siRNA (SC). Lysates were analyzed as above.
further confirmed in ABC28 cells by Northern blotting analysis which revealed a single up-regulated band at both the 16 h and 24 h time points (Fig. 1B). The TRIM22 transcript was initially reported to contain a 1329 nt coding region and this cDNA was used in the earlier studies [1,6]. However, this 1329 nt TRIM22 entry has since been replaced in the NCBI database by a transcript containing a 1497 nt coding region (a 168 nt 3′ extension of the first) (NM_006074.3) which codes for a 498 aa protein. To investigate if progesterone-regulated TRIM22 transcript contained the 1497 nt coding sequence, we performed RT-PCR with a pair of primers that would amplify a 879 nt sequence from the 3′ end of the 1497 nt coding sequence but would not yield any product if the coding sequence was 1329 nt. Supplementary Fig. 1B shows a distinct band of approximately 879 nt that was distinctively upregulated by progesterone and sequencing analysis verified that the 879 nt product corresponded to the 3′ sequence of the 1497 nt transcript. Northern blotting analysis indicated that progesterone up-regulated the expression of only one TRIM22 transcript. We can conclude therefore that progesterone induced the expression of a TRIM22 transcript that codes for a 498 aa protein.
To investigate the regulation and localization of endogenous TRIM22 protein, we generated mouse polyclonal antibodies against recombinant 498 aa His-tagged TRIM22 protein. This antibody detected a band at around 50–60 kDa that was drastically upregulated by progesterone in both ABC28 and T47D cells in a dosedependent manner (Fig. 1C). The identity of the up-regulated protein was validated with three pre-designed TRIM22 siRNA that reduced levels of the induced protein to near-basal levels (Fig. 1C). These TRIM22 siRNA were able to knockdown exogenously expressed FLAG-TRIM22 (Supplementary Fig. 2), thereby confirming their specificity. Whereas the effect of progesterone on TRIM22 protein levels in ABC28 cells was observed at both the 24 h (data not shown) and 48 h time points, in T47D cells, the effect was only prominent at the 24 h time point. We suspect that the protein was unstable in T47D cells. 17β-estradiol, an established mitogen in breast cancer cells including T47D cells, had no effect on TRIM22 protein levels either alone or in combination with progesterone. As expected, IFNγ increased TRIM22 protein levels in both HeLa cells and MCF7 cells; the up-regulated TRIM22 protein was reduced by TRIM22 siRNA (Fig 1D). IFNβ likewise increased TRIM22 protein levels (data not shown).
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
Progesterone has previously been shown to have strong growth-inhibitory effects in ABC28 cells and T47D cells both in vitro and in vivo [24–26]. The finding that progesterone greatly upregulates TRIM22 mRNA and protein levels supports a growth inhibitory role for TRIM22. A possible growth inhibitory role for TRIM22 is also supported by general observations that IFN β and γ are growth-inhibitory in several cell types including MCF7 cells and HeLa cells [27–29]. The effect of TRIM22 on cell growth and cell cycle progression are being investigated.
TRIM22 localizes in the nucleus and forms nuclear and nucleolar bodies The high specificity of the mouse polyclonal antibody against TRIM22 allowed us to use the antibody to investigate the cellular
1525
localization of the endogenous TRIM22 protein, which has not been reported before. In ABC28 cells, endogenous TRIM22 localized as a nucleoplasmic protein and as distinct NB, besides some cytoplasmic localization (ABC28 – E – TRIM22 panel, Fig. 2A). Both the nuclear forms of the protein were evidently increased by progesterone treatment (ABC28 – P – TRIM22 panel, Fig. 2A). More interestingly, about half of all progesterone-treated cells had TRIM22 NB within nucleoli, which were identified by the location of endogenous fibrillarin (Fig. 2A). In contrast, TRIM22 NB were not obvious in T47D cells and treatment of T47D cells with progesterone only increased nucleoplasmic TRIM22 (Supplementary Fig. 3A). The inability of TRIM22 to form NB in T47D cells may be related to the instability of the TRIM22 protein which could not be detected after 48 h progesterone treatment (Fig. 1C).
Fig. 2 – The TRIM22 protein localizes as a nucleoplasmic protein and forms distinct nuclear and nucleolar bodies. (A) ABC28 cells treated with ethanol (E) or progesterone (P; 10 nM) for 48 h were immunostained to detect TRIM22 (red) and fibrillarin (green) as a nucleolar marker. Merged images are presented showing co-localizations (yellow). (B) MCF7 and HeLa cells were immunostained to detect TRIM22 (red). (C) Localization of 498 aa TRIM22 was analyzed in MCF7 cells stably expressing TRIM22-EGFP or transiently (24–48 h) expressing EYFP-TRIM22 (yellow) or FLAG-TRIM22 (red). (A–C) Cells were counterstained with DAPI (blue) and visualized by fluorescence microscopy at a 60× or 96× magnification. Bars, 10 or 20 μm. Insets show representative cells (arrows) enlarged 2×.
1526
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
TRIM22 also localized as a nucleoplasmic protein and formed NB in HeLa and MCF7 cells (Fig. 2B). Like that observed in control ABC28 cells (ABC28 – E – TRIM22, Fig. 2A), TRIM22 NB was not found in all cells and, while the bodies were similarly sized, they were far less prevalent than in progesterone-treated ABC28 cells (ABC28 – P – TRIM22, Fig. 2A). Treatment with IFNγ did not noticeably enhance TRIM22 body formation in HeLa and MCF7 cells but enhanced nucleoplasmic staining (Supplementary Fig. 3B). In both HeLa and MCF7 cells, we also noted an increased cytoplasmic staining upon IFNγ treatment. The dispersed form of TRIM22 in HeLa cells and MCF7 cells predominated despite the greater induction of TRIM22 by IFNγ than by progesterone in ABC28 cells where TRIM22 is capable of forming NB. This suggests
that the ability of the TRIM22 protein to localize as NB and/or nucleoplasmic protein depends on certain cellular signal(s) or protein modifications. It is worth mentioning here that sumoylation of TRIM19/PML is essential for the formation of PML NB through non-covalent binding of PML to sumoylated PML [30]. There are also several highly probable sumoylation sites in TRIM22 and it remains to be seen if TRIM22 is sumoylated and if this modification affects TRIM22 NB formation. Enhanced green fluorescent protein (EGFP) tagged TRIM22 also localized as distinct NB and as a nucleoplasmic protein in MCF7 cells (Fig. 2C), confirming our observations of TRIM22's nuclear localization and NB formation. To exclude the possibility that the large EGFP tag on the C-terminus of TRIM22 modified the
Fig. 3 – Nucleolar bodies are drastically reduced upon silencing. (A) ABC28 cells, left untransfected or transfected with control siRNA or two different TRIM22 siRNA sequences (10 nM) before treatment with progesterone (10 nM) for 48 h, were immunostained with anti-TRIM22 antibody and visualized by fluorescence microscopy at 60× magnification. Images were taken with the same exposure. Bars, 20 μm. Insets show representative cells (arrows) enlarged 2×. (B) Numbers of ABC28 cells expressing nuclear and nucleolar TRIM22 bodies were quantitated in two independent experiments (untransfected, n = 144; control siRNA, n = 259; TRIM22 siRNA 6, n = 211; TRIM22 siRNA 7, n = 205). Percentage of cells having nuclear and nucleolar TRIM22 bodies are presented. Error bars reflect standard error of the mean.
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
conformation and/or the localization of the protein, the TRIM22 coding sequence was also cloned into the pEYFP-C1 and the pXJFLAG vectors such that the protein would have the enhanced yellow fluorescent protein (EYFP) or FLAG tags, respectively, on its Nterminus. TRIM22 localized as distinct NB regardless of tag size or whether the tag is fused to the C-terminus or N-terminus of TRIM22 (Fig. 2C). It should be noted that ectopically expressed TRIM22 localized exclusively as a nuclear protein in cells expressing low to
1527
moderate levels of the protein. The cytoplasmic signal observed with the anti-TRIM22 antibody could therefore either reflect nonspecific binding of the antibody or the recognition of continuously expressed TRIM22 before its translocation into the nucleus. The effect of two TRIM22 siRNA (siRNA 6 and 7) on the existence of TRIM22 as a nucleoplasmic protein and as nuclear and nucleolar bodies was investigated in ABC28 cells treated with progesterone. Interestingly, while TRIM22 siRNA did not decrease the numbers of
Fig. 4 – TRIM22 co-localizes partially with p80-coilin. (A) MCF7 cells stably expressing TRIM22-EGFP were immunostained with anti-p80-coilin antibody. Insets show representative cells (arrows) enlarged 2.5×. (B) Untreated and progesterone-treated ABC28 cells and (C) untreated MCF7 cells were immunostained with anti-TRIM22 and anti-p80-coilin antibodies. Endogenous TRIM22 (red) and p80-coilin (green) co-localize in nuclear bodies (arrows in ABC28 and MCF7 panels) and in strand-like structures in the cytoplasm (arrows in ABC28–48P panel). (A–C) All cells were counterstained with DAPI (blue) and visualized by fluorescence microscopy at 60× or 96× magnifications. Bars, 10 or 20 μm. Insets show representative cells (arrows) enlarged 2×. (D) Whole cell lysates from MCF7 cells co-expressing p80-coilin-EGFP and pXJ-FLAG (FLAG) or FLAG-TRIM22 were subjected to immunoprecipitation using anti-FLAG M2 affinity beads (400 μg IP). Immunoprecipitated proteins were analyzed alongside whole cell lysates (20 μg) by Western blotting using anti-EGFP and anti-FLAG antibodies.
1528
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
Fig. 5 – Dynamics of TRIM22 bodies during the cell cycle. (A) HeLa cells stably expressing TRIM22-EGFP were synchronized by the double thymidine block and cell cycle distribution at different time points after synchronization was analyzed by flow cytometry. Percentages of cells in each phase of the cycle are indicated. (B) Localization of TRIM22-EGFP at specific time-points corresponding to synchrony at the G1/S, early S, mid-late S, G2/M, and G0/G1 phases was analyzed by fluorescence microscopy at a 60× magnification. Bars, 20 μm. Insets show representative cells (arrows) enlarged 2.5×. (C) TRIM22-EGFP expression levels through the cell cycle were analyzed by Western blotting of whole cell lysates (20 μg) extracted from cells not subjected to the double thymidine block (asynchronous, AS) and cells collected at different time points (0, 3, 5, 8, 13 h) after synchronization. GAPDH was analyzed as a loading control.
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
cells expressing TRIM22 NB, the intensities of these bodies and nucleoplasmic staining were reduced (Fig. 3A). Most evidently, expression of TRIM22 siRNA led to disappearance of regular bodies within nucleoli of the majority of progesterone-treated cells (Fig. 3A). Quantitation analysis of random fields revealed that at least 50% of untransfected and control siRNA transfected cells contained nucleolar TRIM22 bodies. In the presence of TRIM22 siRNA 6 or 7, however, only 7–12% of cells had nucleolar TRIM22 bodies (Fig. 3B). Since much of the TRIM22 nucleolar bodies in ABC28 cells were reduced by TRIM22 siRNA, one possibility is that they were formed from nascent, progesterone-induced TRIM22 mRNA whereas the TRIM22 NB in the nucleus were mostly TRIM22 protein existing before progesterone treatment. It is therefore plausible that the formation of TRIM22 nucleolar bodies precedes that of NB. TRIM22 nucleolar bodies may be translocated out of the nucleoli to form NB. The findings also raise an interesting possibility that TRIM22 is involved in the specific function of the nucleolus, notably ribosome synthesis, cell-cycle regulation, and biogenesis of most cellular ribonuclear particles [31].
TRIM22 bodies partially overlap with the Cajal body signature protein p80-coilin As TRIM22 protein formed regular bodies similar in appearance to CBs which are also frequently observed next to and within nucleoli [32], the co-localization of TRIM22 with endogenous p80-coilin was investigated. TRIM22-EGFP bodies in stably-transfected MCF7 cells overlapped partially with endogenous CBs (Fig. 4A). We found that 144 (35%) of 415 endogenous CBs counted overlapped with TRIM22 NB in MCF7-TRIM22-EGFP cells. If cells with numerous TRIM22 speckles were excluded, all CBs overlapped with TRIM22 NB in 23 (18%) of 130 cells. The TRIM22 speckles may be the dispersed form of TRIM22 NB and numerous TRIM22 speckles in the cells would mean a greater chance of random encounter and may therefore erroneously increase the percentage of co-localization with CB. There was no obvious relationship between the numbers of CBs within the cell and the numbers of these CBs co-localizing with TRIM22 bodies. Analysis of confocal Z-stacks confirmed the co-localization of TRIM22-EGFP with endogenous CBs (Supplementary Fig. 4). While there were no detectable overlapping between TRIM22 NB and CB in untreated ABC28 cells and HeLa cells (results not shown), endogenous TRIM22 NB overlapped with CBs in a fraction of progesterone-treated ABC28 cells (Fig. 4B) and in MCF7 cells (Fig. 4C). In contrast, none of the TRIM22 NB co-localized with PML bodies (data not shown). Another interesting observation is that in almost all progesterone-treated ABC28 cells observed, TRIM22 colocalized with p80-coilin in the strand-like structures within cytoplasm (Fig. 4B). As the nuclear protein p80-coilin is a critical component of multifunctional CBs, the association between p80coilin and TRIM22 outside nuclei in progesterone-treated ABC28 cells is likely to have important implications for their function. It is to be noted that MCF7 cells with intense cytoplasmic staining of TRIM22 antibody in Fig. 4C had probably just undergone mitosis. As will be seen later in the cell synchronization study, HeLa cells stably expressing TRIM22-EGFP (Fig. 5B, 8 h — G2/M image) also had intense diffuse fluorescent signal in the cytoplasm during and after mitosis and this probably reflects dynamic changes of TRIM22 localization during cell-cycle progression. Through a co-immunoprecipitation experiment in MCF7 cells co-expressing both proteins, we were able to verify that FLAG-
1529
TRIM22 interacted with p80-coilin-EGFP (Fig. 4D). The finding that not all TRIM22 NB were associated with CBs suggests that these two types of NB may associate during specific phases of NB formation or during certain functional states of the NB.
TRIM22 nuclear bodies undergo dynamic changes during progression through the cell cycle Since TRIM22 co-localizes with CBs which undergo dynamic assembly and disassembly during cell cycle progression [8], we investigated if the formation of TRIM22 NB and nucleoplasmic TRIM22 would vary with cell-cycle progression. HeLa cells are easily synchronized by the double thymidine block method. HeLa cells stably expressing TRIM22-EGFP were sorted by FACS into three different populations depending on the level of fluorescence (high, medium and low). The cells chosen for the cell cycle analysis expressed low to medium levels of TRIM22-EGFP. We were able to achieve a relatively high percentage of synchronized cells in the G0/G1 (72.2% at 14 h post-release), S (68.9–70.8% at 3–5 h postrelease), and G2/M (66.4% at 8 h post-release) phases after releasing these cells from the double thymidine block (Fig. 5A). It appeared that G0/G1-phase cells, before entering S-phase, displayed regular TRIM22-EGFP bodies (Fig. 5B, 0 h). When cells were allowed to progress to the S-phase, the vast majority of these cells expressed TRIM22-EGFP in a speckled fashion (Fig. 5B, 3 h and 5 h). The speckled distribution was also observed in cells that had distinct nuclear boundaries in the G2/M population. However, during and immediately after mitosis, when DNA condenses and segregates, TRIM22-EGFP became completely diffused in the nucleus and cytoplasm (Fig. 5B, 8 h). The corresponding DAPI images of synchronized cells at the different time points have been presented in Supplementary Fig. 5. The diffused localization of TRIM22-EGFP was also observed in cells in telophase (panel ⁎ in 8 h). Multiple regular TRIM22 bodies re-appeared in cells undergoing cytokinesis (panel ⁎⁎ in 14 h) and were present after the cells entered G0/G1-phase (Fig. 5B, 14 h). These changes in the distribution of TRIM22-EGFP are unlikely to be due to differing TRIM22-EGFP expression levels as these were relatively constant during the different phases (Fig. 5C). Interestingly, TRIM22-EGFP bodies were found to be localized frequently within and next to nucleoli in cells that had entered the G0/G1-phase (Fig. 5B, 14 h). This is consistent with the observation that 50% of progesteronetreated ABC28 cells, which are mostly arrested in the G0/G1-phase [19], contained nucleolar TRIM22 bodies. The nuclear dynamic changes of TRIM22 NB are remarkably similar to CBs which disassemble during mitosis and reassemble during the mid-G1 phase [8]. The cycle-dependent dynamic changes of CB are associated with its function in cell cycle regulation. For instance, the CB-associated protein p220NPAT is essential for G0/G1 progression to S-phase in HCT116 cells [33]. Cells lacking p220NPAT also displayed defects in the focal localization of p80-coilin as CB. The dynamic changes of TRIM22 NB during the cell cycle may likewise reflect its role in cell cycle regulation. Since TRIM22-EGFP localizes in a speckled fashion in S-phase synchronized HeLa cells, we tested if endogenous TRIM22 dispersed in MCF7 cells arrested in the S-phase by methotrexate [34]. Indeed, in contrast to the bright regular TRIM22 NB seen in control cells, methotrexate-treated MCF7 cells contained TRIM22 distributed in a speckled manner (Fig. 6). However, methotrexate did not cause any obvious change of TRIM22 NB in ABC28 cells
1530
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
Fig. 6 – Methotrexate induces a dispersion of TRIM22 bodies. MCF7 cells were plated in growth media containing 10% serum and treated with or without 10 μM methotrexate in 10% serum for 48 h. Cells were immunostained with the anti-TRIM22 antibody and counterstained with DAPI before fluorescence microscopy at a 60× magnification. Bars, 20 μm. Insets show representative cells (arrows) enlarged 2.5×.
compared to untreated controls (data not shown). Therefore, the dynamic changes in TRIM22 NB and cellular localization also vary depending on cell types.
Concluding remarks NB are subnuclear domains that contain an assortment of components involved in diverse nuclear functions including ribosome biogenesis, pre-mRNA processing, splicing, apoptosis and transport [35]. By generating effective TRIM22 antibodies and using specific siRNA, the present study provides strong evidence to add TRIM22 NB to the growing list of NB. Like other NB, TRIM22 NB are dynamic structures that can form and disintegrate in response to cellular signals. This notion is supported by studies with progesterone and IFNγ. While progesterone increased both nucleoplasmic TRIM22 protein and TRIM22 nuclear and nucleolar bodies in ABC28 cells, IFNγ treatment only enhanced nucleoplasmic TRIM22 with no evident increase in TRIM22 NB in HeLa cells and MCF7 cells. This is despite the fact that the induction of TRIM22 by IFNγ was of much greater magnitude than by progesterone and that endogenous TRIM22 does form NB in these cells. We suggest that these different forms are related to the functionality of TRIM22 in response to these agents in different cell types. One of the most prominent aspects of NB dynamics is their changes in forms during the cell cycle. In synchronized HeLa cells, TRIM22 typically formed NB during the G0/G1-phase. The TRIM22 NB was transformed into speckle-like structures during S-phase and probably early G2/M-phase. Consistent with this finding in HeLa cells, we also observed that TRIM22 formed speckle-like structures in MCF7 cells arrested in the S-phase by methotrexate. During the M-phase, TRIM22 became diffused but the NB was assembled again by the end of cytokinesis. This cell cycle-dependent dynamism of TRIM22 NB and an association of TRIM22 induction by progesterone and IFN with their growth-
inhibitory effects in breast cancer cells [19,25,28] support a role for TRIM22 in the inhibition of cell cycle progression. It is also important to highlight the prominent presence of TRIM22 NB in the nucleolus following progesterone-treatment. Approximately 50% of progesterone-treated ABC28 cells contained nucleolar TRIM22 NB that can be readily eliminated by specific TRIM22 siRNA. Nucleolar localization of TRIM22 NB is by no means a meaningless sequestration of abundant protein because the protein levels of transfected TRIM22-EGFP or FLAGTRIM22 were much higher than progesterone-induced TRIM22, yet much less TRIM22-EGFP or FLAG-TRIM22 NB appeared in the nucleolus. It is likely therefore that TRIM22 is involved in one or more nucleolar functions, which include rDNA transcription, prerRNA processing, ribosome assembly and formation of the telomerase ribonuclear protein complex. It is interesting to mention in this context that TRIM22 is well-known for its antiviral properties and the nucleolus has been suggested as a gateway for viral replication [36]. Viral proteins are known to interact with nucleolar proteins such as B23 and fibrillarin and use the nucleolus as a site of replication [37–39]. Although we did not observe an increase of TRIM22 NB within the nucleolus in IFNγ-treated cells, we cannot exclude the possibility that IFNγinduced TRIM22 has a diffuse localization or transient aggregation in the nucleolus, which is difficult to ascertain. It is also possible that progesterone- and IFNγ-induced TRIM22 may elicit antiviral effects via different mechanisms depending on the presence of the co-induced partner proteins. Future work in this area should include an investigation into the association between TRIM22's nucleolar localization and its antiviral properties.
Acknowledgments The work is supported by grants from the Ministry of Education, the Republic of Singapore, to Valerie CL Lin.
EX P ER I ME NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21 –15 3 2
We wish to thank Dr. E. Manser et al. for the pXJ-FLAG plasmid and Dr. Joseph Gall for the p80-coilin-EGFP plasmid. We also wish to thank Dr. Stephen Ogg and Dr. Li Hoi Yeung for helpful discussions.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.yexcr.2009.01.028.
REFERENCES
[1] A. Reymond, G. Meroni, A. Fantozzi, G. Merla, S. Cairo, L. Luzi, D. Riganelli, E. Zanaria, S. Messali, S. Cainarca, A. Guffanti, S. Minucci, P.G. Pelicci, A. Ballabio, The tripartite motif family identifies cell compartments, EMBO J. 20 (2001) 2140–2151. [2] S.D. Barr, J.R. Smiley, F.D. Bushman, The interferon response inhibits HIV particle production by induction of TRIM22, PLoS Pathog. 4 (2008) e1000007. [3] A. Bouazzaoui, M. Kreutz, V. Eisert, N. Dinauer, A. Heinzelmann, S. Hallenberger, J. Strayle, R. Walker, H. Rubsamen-Waigmann, R. Andreesen, H. von Briesen, Stimulated trans-acting factor of 50 kDa (Staf50) inhibits HIV-1 replication in human monocyte-derived macrophages, Virology 356 (2006) 79–94. [4] E. Izmailova, F.M. Bertley, Q. Huang, N. Makori, C.J. Miller, R.A. Young, A. Aldovini, HIV-1 Tat reprograms immature dendritic cells to express chemoattractants for activated T cells and macrophages, Nat. Med. 9 (2003) 191–197. [5] S. Obad, H. Brunnstrom, J. Vallon-Christersson, A. Borg, K. Drott, U. Gullberg, Staf50 is a novel p53 target gene conferring reduced clonogenic growth of leukemic U-937 cells, Oncogene 23 (2004) 4050–4059. [6] C. Tissot, N. Mechti, Molecular cloning of a new interferon-induced factor that represses human immunodeficiency virus type 1 long terminal repeat expression, J. Biol. Chem. 270 (1995) 14891–14898. [7] J. Zhang, S.C. Das, C. Kotalik, A.K. Pattnaik, L. Zhang, The latent membrane protein 1 of Epstein–Barr virus establishes an antiviral state via induction of interferon-stimulated genes, J. Biol. Chem. 279 (2004) 46335–46342. [8] L.E. Andrade, E.M. Tan, E.K. Chan, Immunocytochemical analysis of the coiled body in the cell cycle and during cell proliferation, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 1947–1951. [9] M. Platani, I. Goldberg, J.R. Swedlow, A.I. Lamond, In vivo analysis of Cajal body movement, separation, and joining in live human cells, J. Cell Biol. 151 (2000) 1561–1574. [10] M. Cioce, A.I. Lamond, Cajal bodies: a long history of discovery, Annu. Rev. Cell. Dev. Biol. 21 (2005) 105–131. [11] B.E. Jady, E. Bertrand, T. Kiss, Human telomerase RNA and box H/ACA scaRNAs share a common Cajal body-specific localization signal, J. Cell Biol. 164 (2004) 647–652. [12] S. Marcand, V. Brevet, C. Mann, E. Gilson, Cell cycle restriction of telomere elongation, Curr. Biol. 10 (2000) 487–490. [13] D. Stanek, K.M. Neugebauer, The Cajal body: a meeting place for spliceosomal snRNPs in the nuclear maze, Chromosoma 115 (2006) 343–354. [14] A. Narayanan, W. Speckmann, R. Terns, M.P. Terns, Role of the box C/D motif in localization of small nucleolar RNAs to coiled bodies and nucleoli, Mol. Biol. Cell. 10 (1999) 2131–2147. [15] L.S. Shopland, M. Byron, J.L. Stein, J.B. Lian, G.S. Stein, J.B. Lawrence, Replication-dependent histone gene expression is related to Cajal body (CB) association but does not require sustained CB contact, Mol. Biol. Cell. 12 (2001) 565–576. [16] M. Cioce, A.I. Lamond, Cajal bodies: a long history of discovery, Annu. Rev. Cell. Dev. Biol. 21 (2005) 105–131.
1531
[17] Z. Duan, B. Gao, W. Xu, S. Xiong, Identification of TRIM22 as a RING finger E3 ubiquitin ligase, Biochem. Biophys. Res. Commun. 374 (2008) 502–506. [18] X. Li, B. Gold, C. O'HUigin, F. Diaz-Griffero, B. Song, Z. Si, Y. Li, W. Yuan, M. Stremlau, C. Mische, H. Javanbakht, M. Scally, C. Winkler, M. Dean, J. Sodroski, Unique features of TRIM5alpha among closely related human TRIM family members, Virology 360 (2007) 419–433. [19] V.C. Lin, E.H. Ng, S.E. Aw, M.G. Tan, E.H. Ng, V.S. Chan, G.H. Ho, Progestins inhibit the growth of MDA-MB-231 cells transfected with progesterone receptor complementary DNA, Clin. Cancer Res. 5 (1999) 395–403. [20] E. Manser, H.Y. Huang, T.H. Loo, X.Q. Chen, J.M. Dong, T. Leung, L. Lim, Expression of constitutively active alpha-PAK reveals effects of the kinase on actin and focal complexes, Mol. Cell. Biol. 17 (1997) 1129–1143. [21] Z.Y. Zheng, B.H. Bay, S.E. Aw, V.C. Lin, A novel antiestrogenic mechanism in progesterone receptor-transfected breast cancer cells, J. Biol. Chem. 280 (2005) 17480–17487. [22] L. Li, K. Roy, S. Katyal, X. Sun, S. Bleoo, R. Godbout, Dynamic nature of cleavage bodies and their spatial relationship to DDX1 bodies, Cajal bodies, and gems, Mol. Biol. Cell. 17 (2006) 1126–1140. [23] T. Pederson, E. Robbins, A method for improving synchrony in the G2 phase of the cell cycle, J. Cell. Biol. 49 (1971) 942–945. [24] S.D. Groshong, G.I. Owen, B. Grimison, I.E. Schauer, M.C. Todd, T.A. Langan, R.A. Sclafani, C.A. Lange, K.B. Horwitz, Biphasic regulation of breast cancer cell growth by progesterone: role of the cyclin-dependent kinase inhibitors, p21 and p27(Kip1), Mol. Endocrinol. 11 (1997) 1593–1607. [25] V.C. Lin, A.S. Eng, N.E. Hen, E.H. Ng, S.H. Chowdhury, Effect of progesterone on the invasive properties and tumor growth of progesterone receptor-transfected breast cancer cells MDA-MB-231, Clin. Cancer Res. 7 (2001) 2880–2886. [26] V.C. Lin, E.H. Ng, S.E. Aw, M.G. Tan, B.H. Bay, Progesterone induces focal adhesion in breast cancer cells MDA-MB-231 transfected with progesterone receptor complementary DNA, Mol. Endocrinol. 14 (2000) 348–358. [27] L. Fransen, J. Van der Heyden, R. Ruysschaert, W. Fiers, Recombinant tumor necrosis factor: its effect and its synergism with interferon-gamma on a variety of normal and transformed human cell lines, Eur. J. Cancer Clin. Oncol. 22 (1986) 419–426. [28] J.L. Gooch, R.E. Herrera, D. Yee, The role of p21 in interferon gamma-mediated growth inhibition of human breast cancer cells, Cell Growth Differ. 11 (2000) 335–342. [29] A.C. Hobeika, W. Etienne, P.E. Cruz, P.S. Subramaniam, H.M. Johnson, IFNgamma induction of p21WAF1 in prostate cancer cells: role in cell cycle, alteration of phenotype and invasive potential, Int. J. Cancer 77 (1998) 138–145. [30] T.H. Shen, H.K. Lin, P.P. Scaglioni, T.M. Yung, P.P. Pandolfi, The mechanisms of PML-nuclear body formation, Mol. Cell. 24 (2006) 331–339. [31] I. Raska, P.J. Shaw, D. Cmarko, Structure and function of the nucleolus in the spotlight, Curr. Opin. Cell Biol. 18 (2006) 325–334. [32] R.L. Ochs, T.W. Stein Jr., E.M. Tan, Coiled bodies in the nucleolus of breast cancer cells, J. Cell. Sci. 107 (Pt 2) (1994) 385–399. [33] X. Ye, Y. Wei, G. Nalepa, J.W. Harper, The cyclin E/Cdk2 substrate p220(NPAT) is required for S-phase entry, histone gene expression, and Cajal body maintenance in human somatic cells, Mol. Cell Biol. 23 (2003) 8586–8600. [34] J.H. van Dierendonck, J.H. Wijsman, R. Keijzer, C.J. van de Velde, C.J. Cornelisse, Cell-cycle-related staining patterns of anti-proliferating cell nuclear antigen monoclonal antibodies. Comparison with BrdUrd labeling and Ki-67 staining, Am. J. Pathol. 138 (1991) 1165–1172. [35] A.G. Matera, Nuclear bodies: multifaceted subdomains of the interchromatin space, Trends Cell Biol. 9 (1999) 302–309. [36] J.A. Hiscox, The nucleolus—a gateway to viral infection? Arch. Virol. 147 (2002) 1077–1089.
1532
EX PE R IM E NTAL CE LL RE S E ARCH 315 ( 20 0 9) 15 21– 15 3 2
[37] S.H. Kim, E.V. Ryabov, N.O. Kalinina, D.V. Rakitina, T. Gillespie, S. MacFarlane, S. Haupt, J.W. Brown, M. Taliansky, Cajal bodies and the nucleolus are required for a plant virus systemic infection, EMBO J. 26 (2007) 2169–2179. [38] S.H. Kim, S. Macfarlane, N.O. Kalinina, D.V. Rakitina, E.V. Ryabov, T. Gillespie, S. Haupt, J.W. Brown, M. Taliansky, Interaction of a plant
virus-encoded protein with the major nucleolar protein fibrillarin is required for systemic virus infection, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 11115–11120. [39] A. Michienzi, S. Li, J.A. Zaia, J.J. Rossi, A nucleolar TAR decoy inhibitor of HIV replication, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 14047–14052.