| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Animal: DNA Viruses |
Queensland Institute of Medical Research and University of Queensland Joint Oncology Program, 300 Herston Road, Herston 4029, Brisbane, Australia1
Author for correspondence: Kenia Krauer. Fax +61 7 3362 0106. e-mail keniaK{at}qimr.edu.au
 |
Abstract |
|---|
 TOP Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
and RBP-2N. Fractionation studies and green fluorescent protein (GFP)-tagged expression studies demonstrated that both RBP isoforms were located predominantly in the cell nucleus. Interestingly, GFP-tagged RBP-J showed diffuse, uniform nuclear staining, whereas GFP-tagged RBP-2N showed a discrete nuclear pattern, demonstrating differences between the two isoforms. Within the nuclear fraction of EBV-negative BL cells, RBP existed both in a free form and bound to chromatin, whereas in LCLs the intranuclear RBP was predominantly chromatin-bound. Expression of the EBV latent proteins was found to lead to the sequestering of RBP from the cytoplasm into the cell nucleus and to an increase in the chromatin-bound forms of RBP. |
Introduction |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
) has been characterized as a transcriptional repressor in mammalian cells and a transcriptional activator in Drosophila melanogaster (Brou et al., 1994
). Previously, RBP was incorrectly identified as a recombinase that was potentially involved in V(D)J recombination in vertebrates (Hamaguchi et al., 1989 ). RBP is a 60 kDa DNA-binding protein that is expressed ubiquitously and binds to the DNA consensus sequence 5' GTGGGAA 3' (Dou et al., 1994 ; Tun et al., 1994 ). The protein is expressed from 11 exons and contains a nuclear-localization domain and a 40 amino acid region that is similar to the catalytic domain of integrases (Argos et al., 1986 ). In total, four cDNAs have been identified (Amakawa et al., 1993 ; Kawaichi et al., 1992 ), each varying only in the first exon, suggesting the possible existence of four different protein isoforms, RBP-J, RBP-2N, aPCR-2 and aPCR-3.
Studies of viral gene expression initially established RBP as a transcriptional repressor of adenovirus polypeptide IX gene expression (Dou et al., 1994 ). So far, only a few mammalian cellular genes have been shown to be targetted by RBP; IL-6 (Kannabiran et al., 1997 ; Plaisance et al., 1997 ), CD23, CD21 and IL-1
(Krauer et al., 1998 ). Kannabiran et al. (1997) demonstrated that RBP repressed the activation of the IL-6 promoter by NF-B and C-EBP and demonstrated that the location of the RBP site was critical in mediating repression. Plaisance et al. (1997) showed that RBP partially blocked access of NF-B to the promoter region and postulated that RBP was responsible for the low basal levels of IL-6. Our previous study identified an RBP-binding site, which overlapped a previously characterized NF-B-binding site, in the -300 region of the IL-1 promoter (Krauer et al., 1998 ). This study also demonstrated that RBP bound preferentially to the region over NF-B, suggesting that RBP may be responsible for constitutive repression of the IL-1 gene.
The Drosophila homologue of RBP is suppressor of hairless, Su(H), which plays a functional role in the development of the peripheral nervous system in Drosophila (Furukawa et al., 1994 ). Studies have gone on to demonstrate that Su(H) participates specifically in the developmental Notch receptor signalling pathway (Artavanis-Tsakonas et al., 1995 ). When co-expressed with Notch, Su(H) is sequestered in the cytoplasm. However, following Notch ligand binding, Su(H) is translocated to the nucleus where it can modulate gene expression. A number of proteins can interact with Su(H), including the different forms of Notch, which lead to transcriptional activation, and Hairless, which inhibits DNA binding (reviewed in Artavanis-Tsakonas et al., 1995 ). The developmental role of Su(H) therefore appears to be determined by interaction with other proteins, which can then regulate gene expression. Given the crucial role that this protein plays in the Notch signalling pathway and that RBP-J knock-out mice die at around 9 days of gestation (Oka et al., 1995 ), Su(H)/RBP is likely to play an important role in cell development and differentiation.
The herpesvirus Epstein–Barr virus (EBV) utilizes RBP to mediate transcriptional regulation of both viral and cellular genes. The nuclear proteins EBNA-2 and the EBNA-3 gene family of proteins (EBNA-3, -4 and -6) have all been shown to interact with RBP (Krauer et al., 1996 ; Robertson et al., 1995 ; Grossman et al., 1994 ). The EBNA-2 protein is responsible for transcriptional activation of viral proteins, including TP-1, TP-2 and LMP-1, and a number of cellular genes including CD23 and CD21. Following EBNA-2 binding to RBP, which leads to masking of the RBP repression domain, the complex then targets promoters containing the RBP DNA consensus sequence. The EBNA-3 family of proteins, however, appear to regulate RBP-mediated gene expression by binding to RBP and preventing RBP from binding to DNA (Robertson et al., 1995 ). Modulation of cellular gene expression has been shown following expression of the EBNA-3 gene family (Kienzle et al., 1996 ; Silins & Sculley, 1994 ). In addition, the EBNA-3 family proteins can repress EBNA-2-mediated transcriptional activation of the TP promoter (Le Roux et al., 1994 ). Given the crucial roles that EBNA-2, -3, -4 and -6 play in EBV transformation of B cells, and given that they all bind RBP, it is likely that RBP is a key player in the transcriptional regulation of genes following EBV infection and hence immortalization of B cells.
Little is known about the RBP protein in human B cells. This study has examined the cellular localization of RBP in B-cell lines and showed that the intranuclear forms of RBP are altered following the expression of EBV genes, such that the majority of RBP becomes bound to chromatin.
|
Methods |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Cell lines and maintenance. ), dG75 (Ben-Bassat et al., 1977 ), dG75 E346 (Kienzle et al., 1996 ) and the lymphoblastoid cell lines (LCLs) QIMR-CS-B95.8 (Krauer et al., 1996 ) and QIMR-LL-B95.8 were maintained in RPMI 1640 supplemented with 10% foetal calf serum (FCS), benzylpenicillin (0·7 mg/ml) and streptomycin (1 mg/ml) at 37 °C in a 5% CO2 atmosphere.
Transient transfection of dG75 and dG75 E346 cells.
dG75 and dG75 E346 cells were transfected by electroporation under conditions that have been described previously (Kienzle et al., 1996 ). Briefly, 3x106 cells (exponential-phase growth) were resuspended in 0·4 ml RPMI supplemented with 10% FCS and placed in a Gene pulser cuvette (Bio-Rad). Plasmid DNA (10 µg) was introduced into the cells by electroporation in a Bio-Rad Gene Pulser with 0·23 kV at 960 mF (time constant ~30 ms). The cells were incubated for 5 min at room temperature before transfer into 10 ml RPMI 1640 with 10% FCS and then incubated at 37 °C. Transfected cells were analysed for gene expression by Western blotting and immunofluorescence after 48 h incubation.
Construction of green fluorescent protein (GFP) expression plasmids and fluorescent microscopy
pEGFP-RBP-J.
The RBP-J cDNA was excised from pSG5-J (a gift from Clare Sample, St Jude Children’s Research Hospital, Memphis, TN, USA) by restriction digestion with BamHI and BglII and subcloned into the BglII site of pEGFP-C1. This plasmid contains the enhanced GFP (EGFP) (Clontech) under the control of a CMV promoter. The resulting plasmid was sequenced to ensure correct reading frame and orientation of the RBP-J cDNA.
pEGFP-RBP-2N.
The RBP-2N cDNA was excised from pACT-2N clone G4(3) (Young et al., 1997 ) by restriction digestion with XhoI and subcloned into the XhoI site of pEGFP-C1. The resulting plasmid was sequenced to ensure correct reading frame and orientation of the RBP-2N cDNA.
Fluorescent microscopy.
Expression of the GFP–RBP-J and GFP–RBP-2N fusions was determined by visualization under a fluorescent microscope. Transfected cells were harvested 48 h after transfection and placed on a slide in a minimal amount of medium, covered with a coverslip and then visualized. The percentage of fluorescent cells was analysed by FACScan analysis and the results are shown as means±SD.
Immunoblotting.
Protein extracts from 1x106 cells were electrophoresed on 7·5% SDS–polyacrylamide gels (Sambrook et al., 1989 ) and electrotransferred onto nitrocellulose filters (Hybond-ECL nitrocellulose; Amersham). The membrane was processed as described previously (Krauer et al., 1996 ). Expression of the EBNA-1, -2, -3, -4 and -6 proteins was detected by incubation with human serum (MCr serum; Sculley et al., 1984 ) diluted 1:200 in 5% Blotto in PBS, while expression of RBP was detected by addition of anti-RBP antibody (supplied by T. Honjo; Sakai et al., 1995 ) diluted 1:1000. The proteins were visualized by using the ECL Western blotting detection system (Amersham).
Cell extracts and gel retardation assays.
Extracts were prepared by using a previously described method (Sambrook et al., 1989 ). Briefly, 2x107 cells were washed in PBS and resuspended in 1 ml cold buffer A [10 mM HEPES–NaOH, pH 8, 50 mM NaCl, 500 mM sucrose, 1 mM EDTA, 0·25 mM EGTA, 0·6 mM spermidine hydrochloride, 0·5% (v/v) Triton X-100, 1 mM PMSF and 7 mM -mercaptoethanol]. The cells were then homogenized in a Dounce homogenizer by using 15 strokes and transferred to a fresh tube. After centrifugation (650 g, 10 min, 4 °C), the supernatant (cytoplasmic extract) was collected. The pellet, containing the nuclei, was then resuspended in 300 µl buffer B [10 mM HEPES–NaOH, pH 8, 400 mM NaCl, 25% (v/v) glycerol, 0·1 mM EDTA, 0·1 mM EGTA, 0·6 mM spermidine hydrochloride, 1 mM PMSF and 7 mM -mercaptoethanol] and incubated on a rocking platform at 4 °C for 40 min. After centrifugation (1100 g, 10 min, 4 °C), the supernatant (nuclear extract) was collected. Gel retardation assays were performed using the Bandshift kit (Pharmacia) according to the manufacturer’s instructions. Briefly, the binding reactions were performed in 10 mM Tris–HCl, pH 7·5, 50 mM NaCl, 0·5 mM DTT, 1·5 mM EDTA, 1 mM MgCl2, 4% glycerol and 0·5 µg poly(dI–dC).poly(dI–dC) in a volume of 15 µl. Competitors or antibodies were incubated with the nuclear protein extracts on ice prior to the addition of the radiolabelled probe. 32P-end-labelled, double-stranded oligonucleotide 5' TCTTCTAACGTGGGAAAATCCAGT 3' was used as the probe.
Extraction of chromatin- and non-chromatin-bound nuclear extracts.
Nuclear proteins that are not associated with the chromatin can be extracted with 150 mM NaCl, while non-histone DNA-binding proteins (chromatin-bound) can be extracted with 500 mM NaCl (Busch et al., 1967 ). Nuclei were prepared as described above and were then fractionated by extraction in either nuclear extraction buffer B containing 150 mM NaCl or a buffer containing 500 mM NaCl.
DNase treatment of cell nuclei to release DNA-binding proteins.
Nuclei were prepared as described above and were then extracted with buffer B containing 150 mM NaCl, which led to the extraction of nuclear proteins that were not tightly associated with nuclear components. The nuclei were then treated for 15 min at room temperature with 150 U RNase-free DNase (Boehringer Mannheim) in buffer B containing 150 mM NaCl and supplemented with 5 mM MgCl2. The resulting extract contained DNA-binding proteins that were released following DNase treatment.
Co-immunoprecipitation.
Immunoprecipitations were performed according to a method modified from that of Harlow & Lane (1988) . Co-immunoprecipitations were prepared from nuclear extracts prepared with buffer B containing 500 mM NaCl (see above). The nuclear extracts were pre-cleared with protein-G–Sepharose (Pharmacia) for 1 h at 4 °C with constant rocking. An appropriate antibody (5 µg) was added to the supernatant and the mixture was incubated for 2 h at 4 °C with rocking. Protein-G–Sepharose (30 µl) was added and then incubated for 1 h at 4 °C with rocking. The samples were then centrifuged at 10000 g for 1 min, the supernatant was discarded and the pellet was washed twice with 1·5 ml nuclear lysis buffer. The pellet was resuspended in 40 µl SDS–PAGE loading buffer, heated at 85 °C for 10 min and centrifuged at 10000 g for 5 min and 15 µl of the supernatant was electrophoresed on a 7·5 or 10% SDS–polyacrylamide gel.
Two-dimensional (2D) gel electrophoresis.
2D gel electrophoresis was performed with the Immobline Drystrip (pH 3–10·5) and ExcelGel SDS gradient (8–18%) according to the manufacturer’s instructions (Pharmacia). LCL cells (1x107 dG75 and CS-B95.8) were lysed in a buffer containing 9 M urea, 2% Triton X-100, 2% -mercaptoethanol, 1·4 µg/ml PMSF and 2% pharmalyte 3–10 (Bio-Rad). Sample solution (8 M urea, 2% -mercaptoethanol, 0·5% Triton X-100 and 2% pharmalyte 3–10) was added to each lysate and then loaded onto the rehydrated Immobline Drystrip for isoelectric focussing. After the first dimension, the drystrip was placed on the ExcelGel SDS gradient (8–18%). After electrophoresis through the second dimension, the proteins were transferred onto nitrocellulose filters and subjected to Western blot analysis with the anti-RBP antibody as described above.
|
Results |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
). It has not yet been determined whether protein is expressed from each of these isoforms or if there are differences in the level of protein expression between the isoforms. Therefore, 2D gel electrophoresis and immunoblotting have been used to identify the different isoforms and their expression levels in human B-cell lines. Our observations suggest that the anti-RBP antibody is capable of binding to regions shared between the proteins and, as the different isoforms only vary in the first exon, the antibody would bind to all RBP isoforms. Protein extracts derived from the LCL CS-B95.8 were separated by 2D gel electrophoresis and the RBP isoforms were identified by Western blotting (Fig. 1). Two bands were identified by Western blotting at the expected size and isoelectric points for the isoforms RBP-2N and RBP-J. The strongest band, based on the predicted isoelectric point of 7·47, corresponded to RBP-2N, while the second, weaker band corresponded to the isoelectric point (predicted 6·85) of RBP-J. This result was consistent with our previous study (Krauer et al., 1996 ), which demonstrated that RBP-2N mRNA was expressed at a higher level than RBP-J mRNA in B-cell lines. Proteins were not detected in the positions that would have corresponded to the largest isoform, PCR-2 (predicted isoelectric point 8·22), or the smallest isoform, PCR-3 (predicted isoelectric point 7·8). The results showed that RBP-2N and RBP-J were the only isoforms of RBP expressed in B lymphocytes and that RBP-2N is smaller and slightly more basic in charge and is the dominant form of RBP in B-cell lines.
|
and RBP-2N ). However, once in the nucleus, RBP was not detected by the anti-RBP antibody, possibly due to the antibody epitope being masked once RBP bound to nuclear proteins or to DNA. Therefore, to determine the cellular distribution of RBP-J and RBP-2N in human B-cell lines and to avoid problems associated with use of anti-RBP antibody, these proteins were expressed in fusions with EGFP. Transient expression of the pEGFP-C1 plasmid in dG75 cells showed an even fluorescence throughout the cell cytoplasm and nucleus (Fig. 2a). After transient expression, EGFP–RBP-J was predominantly observed in the cell nucleus, with only minimal staining evident in the cytoplasm (Fig. 2b). In addition, there was no obvious staining pattern within the nucleus to suggest binding to defined structures. Therefore, RBP-J appeared to be distributed evenly throughout the cell nucleus. In contrast, transient expression of EGFP–RBP-2N in dG75 cells (Fig. 2c, d) typically showed a different nuclear pattern to that of EGFP–RBP-J. Fluorescent microscopy was performed on days 1 and 2 following transfection and showed that the nuclear distribution pattern of EGFP–RBP-2N changed with time. On day 1, EGFP–RBP-2N appeared to be localized to a number of regions throughout the cell nucleus, while on day 2, the fluorescence become focussed to one or two discrete regions of the nucleus (Fig. 2d). Western blot analysis of total cell lysates prepared from the transiently transfected dG75 cells confirmed that full-length fusion proteins of EGFP–RBP-J and EGFP–RBP-2N were expressed in the cells, in addition to the cellular RBP (Fig. 2e, f). Expression of EGFP–RBP-J can be seen clearly by Western blotting on days 1 and 2 after transient transfection of dG75 cells (Fig. 2e). pEGFP-RBP-2N transfectants analysed by Western blotting on days 1 and 2 after transfection (Fig. 2f, lanes 6 and 7) did, however, show a reduced amount of EGFP–RBP-2N fusion protein in comparison to EGFP–RBP-J (Fig. 2f, lanes 4 and 5). In order to be able to visualize EGFP fusion proteins by Western blotting, the gel had to be over-loaded: the lanes therefore appear as smears rather than as discrete bands. These results indicate that the two isoforms RBP-J and RBP-2N localize to different regions of the nucleus.
|
fusion protein is capable of binding its DNA consensus sequence ). pEGFP-C1 and pEGFP-RBP-J were transiently transfected into dG75 cells and nuclear extracts were prepared 2 days later as described in Methods. The 32P-labelled probe was incubated with the nuclear extracts and binding complexes were analysed on polyacrylamide gels (Fig. 3). These results show that one major complex was observed following the addition of the extract from dG75 cells transfected with the pEGFP-C1 plasmid (lane 2): the binding complex was due to RBP binding (described previously in Krauer et al., 1998 ). The subsequent addition of anti-RBP antibody led to this RBP complex being supershifted, demonstrating the presence of RBP in the complex (lane 3). The specificity of the binding complex was also confirmed, as competition with 100-fold excess of a probe lacking the RBP consensus sequence had no effect (lane 4) and neither did the addition of a control irrelevant antibody (lane 5). Incubation of the labelled probe with extracts prepared from dG75 cells containing EGFP–RBP-J fusion protein led to the formation of two binding complexes, the lower complex due to cellular RBP and the larger, upper complex due to EGFP–RBP-J (lane 6). Addition of anti-RBP antibody led to supershifting of both binding complexes (lane 7). These results demonstrate that the EGFP–RBP-J fusion protein is capable of binding to its DNA consensus sequence in gel-shift analysis, suggesting that the protein is functionally active.
|
, it became evident that over-expression led to a reduction in viable fluorescent cells. To analyse this effect, transient transfection studies were performed in dG75 cells with different amounts (5 and 10 µg) of pEGFP-RBP-2N, pEGFP-RBP-J and pEGFP-C1. Transfected cells were analysed for fluorescence on the FACScan on days 1 and 2 post-transfection (Table 1). In two independent experiments, the results showed that expression of EGFP–RBP-J led to 7·10±0·56% and 7·72±0·63% fluorescent cells on days 1 and 2, respectively. In contrast, expression of EGFP–RBP-2N resulted in only 3·58±0·18% of cells showing fluorescence on day 1, which decreased to only 2·05±0·64% fluorescent cells on day 2. The amount of plasmid transfected did not alter the percentage of fluorescent cells observed. The reduced numbers of EGFP–RBP-2N/-J fluorescent cells relative to EGFP alone suggested either that expression of EGFP–RBP-2N/-J was toxic to dG75 cells or that EGFP–RBP-2N/-J had a shorter half-life than EGFP alone and that EGFP–RBP-2N had a shorter half-life than EGFP–RBP-J. This result was also supported by the reduced levels of EGFP–RBP-2N fusion protein observed in the transfected cells by Western blotting compared with EGFP–RBP-J (Fig. 2e
, f).
|
fusion proteins, transient transfection studies were performed in dG75 cells stably expressing the EBNA-3, -4 and -6 proteins (Fig. 4). Expression levels from pEGFP-RBP-J (relative to pEGFP-C1 expression) were not statistically different in dG75 cells (25·1±13·3%) and dG75 cells expressing EBNA-3, -4 and -6 (39·2±15·0%), whereas expression of EGFP–RBP-2N was 4-fold higher in dG75 cells expressing the EBNA-3 gene family proteins (22·0±4·4%) compared with dG75 cells (5·5±3·3%). These results highlight an important difference between RBP-2N and RBP-J and suggest that EBNA-3, -4 and -6 may interact preferentially with or affect RBP-2N.
|
). To examine the status of intranuclear RBP in B cells and, secondly, to determine whether these intranuclear forms of RBP were altered following EBV infection, nuclei were purified from EBV-positive and EBV-negative cell lines and were extracted with 150 mM and 500 mM NaCl and the protein extracts were analysed by Western blotting (Fig. 5). As expected, EBNA-1 was present in the cytoplasmic fraction, 150 mM NaCl nuclear extract and 500 mM NaCl nuclear extract of EBV-infected cells (LL B95.8, CS B95.8, MUTU I and MUTU III). In contrast, EBNA-3, -4 and -6 were restricted to the 500 mM NaCl nuclear extract of EBV-positive cell lines (MUTU III, LL B95.8 LCL and CS B95.8 LCL), showing the purity of these extracts. Western blotting for the presence of RBP, which was performed on equivalent amounts of extracted protein, showed that the majority of RBP was solubilized with 500 mM NaCl, indicating that it was chromatin-bound (Fig. 5b). Free intranuclear forms of RBP (150 mM extract) were also present in all the cell lines: however, cell lines expressing the full set of EBV latent proteins (LL B95.8, CS B95.8 and MUTU III) had reduced amounts of cytoplasmic RBP compared with the EBV-negative dG75 cell line and MUTU I cells (which only express EBNA-1). These results could be seen more clearly when electrophoretic mobility shift assays (EMSA) were performed on the same extracts (Fig. 5c). These results demonstrate that, in B-cell lines expressing the full set of EBV latent proteins, RBP does not exist as the free intranuclear form (150 mM salt extract), in comparison with B-cell lines not expressing these EBV latent proteins, where free intranuclear forms are present. In addition, EMSA revealed reduced levels of RBP in the cytoplasm of cells expressing the full set of EBV latent proteins compared with EBV-negative cell lines or cell lines expressing only EBNA-1 (MUTU I). These results suggest that the sequestering of RBP from the cytoplasm and its association with chromatin can be mediated by the EBV latent proteins, EBNA-2, -3, -4 and -6.
|
and EBNA-2, -3 and -6 with the chromatin fraction is DNA-dependent were extracted from nuclei in buffer containing 500 mM NaCl, suggesting that these proteins were associated with chromatin. It is, however, possible that the proteins were associated with subnuclear components other than chromatin. Therefore, to distinguish between these alternatives, nuclei were purified and extracted with buffer containing 150 mM NaCl and were then incubated with DNase in the same buffer containing MgCl2. After DNase treatment, the nuclei were collected and subjected to a further extraction with buffer containing 500 mM NaCl. Proteins present in each extract were analysed by Western blotting using antibodies against RBP and EBNA-2 and -6 (Fig. 6). As shown previously, fractionation of nuclei in the absence of DNase treatment (Fig. 6, lanes 1–4) resulted in the majority of RBP being present in the 500 mM NaCl extract (lane 3), with some residual RBP found in the matrix fraction (lane 4). In contrast, after DNase treatment (lanes 5–8), significant amounts of RBP were found in the buffer containing 150 mM NaCl (lane 6), with the remaining RBP extracted with 500 mM NaCl (lane 7). This result suggested that a large proportion of RBP (80%) was bound to chromatin in a DNA-dependent manner. A similar effect was also observed for EBNA-2 and -6 (Fig. 6). Approximately half of the EBNA-6 protein and a quarter of the EBNA-2 protein were extracted by buffer containing 150 mM NaCl after DNase treatment, indicating that a proportion of these proteins was also associated with chromatin in a DNA-dependent manner. Interestingly, after DNase treatment, a large percentage of EBNA-2 and -6 was found within the cell-matrix fraction. These results confirm the finding that RBP and EBNA-2 and -6 are associated with chromatin, as digestion with DNase led to the release of these proteins, and demonstrate that binding of these proteins is DNA-dependent.
|
/RBP-2N and EBNA-2, -3 and -6 ), which is consistent with our finding that expression of the EBNA-3 gene family increased the expression levels of EGFP–RBP-2N fusion protein in comparison with EGFP–RBP-J. To determine whether, like EBNA-3, -4 and -6, EBNA-2 also associated preferentially with RBP-2N, co-immunoprecipitation experiments were performed. Nuclei were prepared from CS-B95.8 LCL and the chromatin-bound nuclear proteins were extracted with buffer containing 500 mM NaCl. EBNA-2, -3 and -6 proteins were immunoprecipitated from this fraction and the presence of co-immunoprecipitated RBP isoforms was analysed by Western blot analysis (Fig. 7). Both RBP-J and RBP-2N were co-precipitated with EBNA-2, -3 and -6 (lanes 2–4, respectively). However, RBP-2N appeared to be preferentially precipitated by EBNA-3 and -6, in comparison to EBNA-2 (lane 4). The majority of EBNA-2, -3 and -6 was precipitated, whereas each co-immunoprecipitation brought down only a fraction of RBP. Total cell lysate from CS-B95.8 LCLs (lane 5) and a control immunoprecipitation (no antibody added; lane 1) are also shown. These results suggest that the EBNA-3 family proteins may associate preferentially with the RBP-2N isoform.
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
, aPCR-2 and aPCR-3. Our previous study, using RT–PCR, showed that RBP-2N is the predominantly expressed mRNA in human B-cell lines (Krauer et al., 1996 ). The present study shows that only two RBP isoforms are detected by 2D gel electrophoresis, which match closely the expected size and isoelectric points of RBP-2N and RBP-J (Fig. 1). At the protein level, RBP-2N was expressed at considerably higher levels than RBP-J, which is consistent with our previous RT–PCR results. These results are also in agreement with an earlier report showing the presence of two RBP isoforms in EBV-transformed B-cell lines (Johannsen et al., 1996 ). While the results indicate that the aPCR-2 and aPCR-3 isoforms are not expressed in B cells, they could be expressed in other tissues.
The cellular location of RBP has been linked to the differentiation status of cells (Sakai et al., 1995 ). In the human B-cell lines dG75 and MUTU I, fractionation studies showed that RBP was predominantly located within the cell nucleus, with small amounts being observed in the cell cytoplasm (Figs 2 and 5). RBP was absent from the membrane fractions, however (data not shown). Analysis of the intranuclear fractions demonstrated that RBP was distributed between free, nuclear forms and chromatin-bound forms, with a predominance in the chromatin-bound fraction (Fig. 5). The B-cell lines used in this study show a germinal B-cell phenotype and are considered to be at the intermediate stage of B-cell differentiation (Cushley & Harnett, 1993 ). RBP might therefore be expected to be distributed between the cytoplasmic and nuclear fractions.
Differentiation of embryonic carcinoma cells leads to nuclear translocation of RBP (Sakai et al., 1995 ) and an increase in the chromatin-bound form of the protein. Given that RBP acts as a transcriptional repressor, these results suggest that nuclear translocation of RBP leads to decreased expression of cellular genes that control differentiation. An increase in nuclear RBP and a concomitant reduction in cytoplasmic RBP were observed in EBV-transformed B-cell lines (LCLs) in comparison to dG75 EBV-negative BL cells (Fig. 5). In addition, a reduction in free forms of RBP and an increase in chromatin-bound RBP were also evident following expression of the EBV latent proteins (Fig. 5). Expression of the EBV latent proteins led to a sequestering or recruitment of chromatin-bound forms of RBP in the nucleus. This resulted in the majority of cellular RBP existing either bound to chromatin (Fig. 5, upper panel) or complexed to the EBV proteins, suggesting that RBP-mediated cellular gene expression may be partially under the control of EBV. This is not unexpected, as previous studies have suggested that approximately half of the cellular RBP is associated with the EBNAs in LCLs (Johannsen et al., 1996 ). As RBP controls/regulates cell differentiation, the sequestering of free RBP by EBV latent proteins may lead to a blockage in further cellular differentiation. This would agree with an earlier finding that, following EBV infection, B cells do not differentiate further to become antibody-secreting plasma cells (Rickinson & Kieff, 1996 ).
Do the two RBP isoforms (RBP-2N and RBP-J) expressed in B cells have different functions? This study makes a number of important observations with regard to functional differences between RBP-2N and RBP-J. Firstly, RBP-J and RBP-2N show different nuclear localization patterns (Fig. 2). Secondly, the over-expression studies with EGFP–RBP-2N and EGFP–RBP-J suggested either that B-cell lines show different sensitivities to RBP-2N and RBP-J (Table 1) or that RBP-2N and RBP-J are processed differently within the cell. This suggests that RBP-2N and RBP-J have different functions since, if their functions were interchangeable, cells would not display differential sensitivity to the two isoforms. That the EBNA-3 gene family of proteins was capable of increasing greatly the number of cells expressing EGFP–RBP-2N, but not EGFP–RBP-J, supports the existence of functional differences between the two isoforms and suggests that RBP-2N interacts preferentially with the EBNA-3, -4 and -6 proteins (Fig. 4). As our studies were performed on EGFP fusion proteins, in order to ensure that they were functionally active, gel-shift analysis was performed on nuclear extracts of DG75 cells transfected with EGFP–RBP-J (Fig. 3). These results showed that the fusion protein was able to bind its DNA consensus sequence and that the complex could be supershifted with anti-RBP antibodies. As RBP-J and RBP-2N have identical DNA-binding domains, it is logical to predict that both isoforms can bind their DNA binding sites. The finding of preferential association of the EBNA-3 family proteins with RBP-2N would also agree with the co-immunoprecipitation results, where the smaller, more predominant isoform, RBP-2N, was shown to associate more readily with EBNA-3 and -6 in comparison to EBNA-2. The first exon of RBP-J encodes 19 amino acids, whereas the first exon of RBP-2N encodes only six amino acids. It is possible that the differences between the first exons could influence either the half-life of the proteins, post-translational modification or their three-dimensional structure, which could therefore affect interaction of these proteins with other proteins or DNA.
It remains to be determined whether the cellular genes regulated by EBV are the same as those that play a role in determining cellular differentiation status. However, from the results presented in this study, it is apparent that each of the RBP isoforms performs different functions and that EBV (or at least the EBNA-3, -4 and -6 gene products of EBV) target the RBP-2N isoform. This observation is supported by a previous study, which showed that the EBNA-3 gene family associated preferentially with the smaller RBP isoform in B-cell lines (Johannsen et al., 1996 ). In addition, this earlier study showed that the majority of EBNA-2 was associated with RBP, whereas approximately 20% of EBNA-6 was associated with RBP. As RBP-2N and RBP-J show different nuclear localization patterns and differ in their ability to associate with the EBNAs, it is possible that one isoform may control cellular differentiation whereas the other may have a more generalized function in gene regulation.
|
Acknowledgments |
|---|
|
References |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Argos, P., Landy, A., Abremski, K., Egan, J. B., Haggard-Ljungquist, E., Hoess, R. H., Kahn, M. L., Kalionis, B., Narayana, S. V. & Pierson, L. S.III (1986). The integrase family of site-specific recombinases: regional similarities and global diversity. EMBO Journal 5, 433-440.[Medline]
Artavanis-Tsakonas, S., Matsuno, K. & Fortini, M. E. (1995). Notch signaling. Science 268, 225-232.
Ben-Bassat, H., Goldblum, N., Mitrani, S., Goldblum, T., Yoffey, J. M., Cohen, M. M., Bentwich, Z., Ramot, B., Klein, E. & Klein, G. (1977). Establishment in continuous culture of a new type of lymphocyte from a ‘Burkitt like’ malignant lymphoma (line D.G.-75). International Journal of Cancer 19, 27-33.
Brou, C., Logeat, F., Lecourtois, M., Vandekerckhove, J., Kourilsky, P., Schweisguth, F. & Israel, A. (1994). Inhibition of the DNA-binding activity of Drosophila suppressor of hairless and of its human homolog, KBF2/RBP-J, by direct protein–protein interaction with Drosophila hairless. Genes & Development 8, 2491-2503.
Busch, H., Narayan, K. S. & Hamilton, J. (1967). Isolation of nucleoli in a medium containing spermine and magnesium acetate. Experimental Cell Research 47, 329-336.[Medline]
Cushley, W. & Harnett, M. M. (1993). Cellular signalling mechanisms in B lymphocytes. Biochemical Journal 292, 313-332.
Dou, S., Zeng, X., Cortes, P., Erdjument-Bromage, H., Tempst, P., Honjo, T. & Vales, L. D. (1994). The recombination signal sequence-binding protein RBP-2N functions as a transcriptional repressor. Molecular and Cellular Biology 14, 3310-3319.
Furukawa, T., Kimura, K., Kobayakawa, Y., Tamura, K., Kawaichi, M., Tanimura, T. & Honjo, T. (1994). Genetic characterization of Drosophila RBP-J (suppressor of hairless) as a neurogenic gene in adult PNS development. Japanese Journal of Genetics 69, 701-711.[Medline]
Gregory, C. D., Rowe, M. & Rickinson, A. B. (1990). Different Epstein–Barr virus-B cell interactions in phenotypically distinct clones of a Burkitt’s lymphoma cell line. Journal of General Virology 71, 1481-1495.
Grossman, S. R., Johannsen, E., Tong, X., Yalamanchili, R. & Kieff, E. (1994). The Epstein–Barr virus nuclear antigen 2 transactivator is directed to response elements by the J recombination signal binding protein. Proceedings of the National Academy of Sciences, USA 91, 7568-7572.
Hamaguchi, Y., Matsunami, N., Yamamoto, Y. & Honjo, T. (1989). Purification and characterization of a protein that binds to the recombination signal sequence of the immunoglobulin J segment. Nucleic Acids Research 17, 9015-9026.
Harlow, E. & Lane, D. (1988). Antibodies: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Johannsen, E., Miller, C. L., Grossman, S. R. & Kieff, E. (1996). EBNA-2 and EBNA-3C extensively and mutually exclusively associate with RBPJ in Epstein–Barr virus-transformed B lymphocytes. Journal of Virology 70, 4179-4183.[Abstract]
Kannabiran, C., Zeng, X. & Vales, L. D. (1997). The mammalian transcriptional repressor RBP (CBF1) regulates interleukin-6 gene expression. Molecular and Cellular Biology 17, 1-9.[Abstract]
Kawaichi, M., Oka, C., Shibayama, S., Koromilas, A. E., Matsunami, N., Hamaguchi, Y. & Honjo, T. (1992). Genomic organization of mouse J recombination signal binding protein (RBP-J) gene. Journal of Biological Chemistry 267, 4016-4022.
Kienzle, N., Young, D., Silins, S. L. & Sculley, T. B. (1996). Induction of pleckstrin by the Epstein–Barr virus nuclear antigen 3 family. Virology 224, 167-174.[Medline]
Krauer, K. G., Kienzle, N., Young, D. B. & Sculley, T. B. (1996). Epstein–Barr nuclear antigen-3 and -4 interact with RBP-2N, a major isoform of RBP-J in B lymphocytes. Virology 226, 346-353.[Medline]
Krauer, K. G., Belzer, D. K., Liaskou, D., Buck, M., Cross, S., Honjo, T. & Sculley, T. (1998). Regulation of interleukin-1 transcription by Epstein–Barr virus involves a number of latent proteins via their interaction with RBP. Virology 252, 418-430.[Medline]
Le Roux, A., Kerdiles, B., Walls, D., Dedieu, J. F. & Perricaudet, M. (1994). The Epstein–Barr virus determined nuclear antigens EBNA-3A, -3B, and -3C repress EBNA-2-mediated transactivation of the viral terminal protein 1 gene promoter. Virology 205, 596-602.[Medline]
Oka, C., Nakano, T., Wakeham, A., de la Pompa, J. L., Mori, C., Sakai, T., Okazaki, S., Kawaichi, M., Shiota, K., Mak, T. W. & Honjo, T. (1995). Disruption of the mouse RBP-J gene results in early embryonic death. Development 121, 3291-3301.[Abstract]
Plaisance, S., Vanden Berghe, W., Boone, E., Fiers, W. & Haegeman, G. (1997). Recombination signal sequence binding protein J is constitutively bound to the NF-B site of the interleukin-6 promoter and acts as a negative regulatory factor. Molecular and Cellular Biology 17, 3733-3743.[Abstract]
Rickinson, A. B. & Kieff, E. (1996). Epstein–Barr virus. In Fields Virology, pp. 2397-2446. Edited by B. N. Fields, D. M. Knipe & P. M. Howley. Philadelphia: Lippincott–Raven.
Robertson, E. S., Grossman, S., Johannsen, E., Miller, C., Lin, J., Tomkinson, B. & Kieff, E. (1995). Epstein–Barr virus nuclear protein 3C modulates transcription through interaction with the sequence-specific DNA-binding protein J. Journal of Virology 69, 3108-3116.[Abstract]
Sakai, T., Furukawa, T., Iwanari, H., Oka, C., Nakano, T., Kawaichi, M. & Honjo, T. (1995). Loss of immunostaining of the RBP-J transcription factor upon F9 cell differentiation induced by retinoic acid. Journal of Biochemistry (Tokyo) 118, 621-628.
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Sculley, T. B., Walker, P. J., Moss, D. J. & Pope, J. H. (1984). Identification of multiple Epstein–Barr virus-induced nuclear antigens with sera from patients with rheumatoid arthritis. Journal of Virology 52, 88-93.
Silins, S. L. & Sculley, T. B. (1994). Modulation of vimentin, the CD40 activation antigen and Burkitt’s lymphoma antigen (CD77) by the Epstein–Barr virus nuclear antigen EBNA-4. Virology 202, 16-24.[Medline]
Tun, T., Hamaguchi, Y., Matsunami, N., Furukawa, T., Honjo, T. & Kawaichi, M. (1994). Recognition sequence of a highly conserved DNA binding protein RBP-J. Nucleic Acids Research 22, 965-971.
Young, D. B., Krauer, K., Kienzle, N. & Sculley, T. (1997). Both A type and B type Epstein–Barr virus nuclear antigen 6 interact with RBP-2N. Journal of General Virology 78, 1671-1674.[Abstract]
Received 4 May 1999;
accepted 10 August 1999.
This article has been cited by other articles:
|
|
 |
|
  P. P. Singh, B. Voleti, and A. Agrawal A Novel RBP-J{kappa}-Dependent Switch from C/EBPbeta to C/EBP{zeta} at the C/EBP Binding Site on the C-Reactive Protein Promoter J. Immunol., June 1, 2007; 178(11): 7302 - 7309. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
