- Email: [email protected]
Co-localization of the retinoblastoma protein and the epstein-barr virus-encoded nuclear antigen EBNA-5
EXPERIMENTAL
CELL RESEARCH
197,314-318
(1991)
Co-localization of the Retinoblastoma Protein and the Epstein-Barr Virus-Encoded Nuclear Antigen EBNA-5 W.-Q. JIANG,* Departments
L. SZEKELY,~ of *Medical
V. WENDEL-HANSEN,*
Cell Genetics and tTumor
Biology, Karolinska
Znstitutet,
G. KLEIN,?
AND A. ROS$N*
S-104 01 Stockholm,
Sweden
known, except for the fact that EBNA-1 is essential for the maintenance of the viral episomes and EBNA-2 is needed for the viral activation of the resting B-lymphocytes and subsequent immortalization. EBNA-5, also designated as EBNA-LP, is encoded by the same polycistronic message as EBNA-2, but is probably not required for immortalization. Earlier work in our [18] and Elliott Kieff s [ 171 laboratory has shown that EBNA-5 is expressed as relatively large foci within the dispersed chromatin of EBV-LCL. This distribution differs markedly from the evenly
A monoclonal antibody (aRBlC1) raised against an Rb fusion protein detects a limited number (4-10) of relatively large intranuclear foci in an EBV-immortalized cord blood cell line (IB4). These domains also bind an anti-EBNA-5 monoclonal antibody. The Rb antibody reactive sites also co-localize with the SV40 T antigen in transformed monkey cells (COS). The nuclear structures stained by aRBlC1 and EBNA-5 antibodies are distinct from the structures detected with antibodies against centromeric proteins and certain snRNP epitopes. EBNA-SlRb-positive domains do not selectively react with antibodies against the La antigen known to associate with the small EBV-encoded nuclear RNA species designated as the EBERs. o 1991 Academic press, IIIC.
dispersed, granular fluorescence of EBNA-1, 2, and 3 [16, 181. In view of the interactions between Rb and transforming proteins of the papova and adenoviruses, it is of great interest to explore whether any spatial or interactive relationship exists between Rb and the growth transformation associated EBV-encoded proteins. As a first step toward such an analysis, we have compared the intranuclear distribution of Rb and EBNA-5. For comparison, we have used SV40 T antigen, known to complex with Rb. We have also examined the distribution of centromeric proteins, a snRNP antigen (Sm) and the La antigen, known to associate with the EBV-encoded small nuclear RNA species (EBERs) [27]. We found co-localization of EBNA-5 and Rb and of SV40-T and Rb.
INTRODUCTION
The retinoblastoma susceptibility gene (Rb) codes for a nuclear protein (Rb) that appears to play a key role in the control of the cell cycle [ 1, 21. Deletion or mutation of the Rb gene is an essential event for the development of both familial and sporadic retinoblastomas [3] and may also contribute to the development of other forms of neoplasia [4, 51. Rb is a DNA-binding nuclear llO-kDa protein [6] that is nonphosphorylated during GO and Gl but becomes phosphorylated at the GUS boundary of the cell cycle [7]. The distribution of the Rb protein within the cell nucleus is largely unknown, although our recent immunofluorescence studies [8] suggest that part of the protein is concentrated to granules in the dispersed, transcriptionally active euchromatin compartment of the interphase nucleus. Rb forms complexes with the transforming proteins of several small DNA tumor viruses including the SV40 [9], adenovirus ElA [lo], and papilloma virus E7 [ll]. Specific amino acid sequences in Rb recognize a conserved sequence motif present in several of the viral transforming proteins [12]. It has been suggested that this complexing may interfere with the normal growth regulatory function of the Rb protein [13]. EBV-immortalized lymphoblastoid cell lines (LCL) express six virally encoded nuclear proteins, the EBNAs (references see [14, 151). Their function is largely un0014~4S27/91 $3.00 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
N. RINGERTZ,*
MATERIALS
AND
METHODS
Cell lines. The EBV-immortalized cord blood cell line, IB4 [19], was used. Like other EBV-LCLs, this cell line expresses all six EBVencoded nuclear antigens. It was chosen for its relatively prominent EBNA-5 expression. There is considerable variation in the expression of this antigen among cell lines, with regard to both the presence of the epitopes that can react with the mouse anti-EBNA-5 mAb’ JF186 [20] and the level of expression. The transformed monkey kidney epithelial cell line (COS) was also used to examine the intranuclear distribution of Rb and SV40 T antigens. This cell line contains the SV40 T gene but lacks the viral origin of replication [21].
’ BSS, balanced salt solution; E5, Epstein-Barr virus-encoded nuclear antigen-5; FITC, fluorescein isothiocyanate; Ig, immunoglobulin; mAb, monoclonal antibody; and RHOD, rhodamine. 314
Rb AND
EBNA-5
TABLE Two-Color
315
CO-LOCALIZATION
1
Immunofluorescence
Procedures’
Reagents (dilution)
Step
Procedure I for IB4 cells Mouse anti-Rb mAb aRBlC1 (1:30) or pMG3 245 (1:200) FITC-conjugated rabbit anti-mouse Ig (1:40)
Procedure III for IB4 cells
Procedure II* for IB4 cells Mouse anti-Sm Y12 (1:l)
mAb
FITC-conjugated rabbit anti-mouse Ig (1:20) -
FITC-conjugated swine anti-rabbit Ig (1:120) Normal mouse serum (1:lO) Biotinylated mouse anti-E5 mAb (1:5)
Normal mouse serum (1:lO) Biotinylated mouse anti-E5 mAb (15)
RHOD-conjugated streptavidin (1:50)
RHOD-conjugated streptavidin (1:50)
Human antisera LU (1:500) or La (1:l) FITC-conjugated rabbit antihuman IgG ychains (1:20) -
Normal mouse serum (1:lO) Biotinylated mouse anti-E5 mAb (1:5) RHOD-conjugated streptavidin (1:50)
Procedure IV for COS cells Mouse anti-Rb mAb RblCl (1:30) or pMG3 245 (1:200) FITC-conjugated rabbit anti-mouse Ig (1:40) FITC-conjugated swine anti-rabbit Ig (1:120) Normal mouse serum (1:lO) Biotinylated mouse antiSV40 T mAb(l:25) RHOD-conjugated streptavidin (1:50)
“All incubation steps were carried out in room temperature for 45 min. ‘Before step 1, the cells were incubated with RNase A (2.00 mg/ml, Sigma Co.) for 15 min.
All cells were cultured at 37’C in a 5% CO, atmosphere, in Iscove’s. medium containing 5% fetal calf serum, 1 n&f L-glutamine, 100 IUI ml penicillin, and 100 gglml streptomycin. Antibodies. The anti-Rb mouse mAb was clone aRBlCl[8] purified from ascites using a Sepharose-protein G column. The mouse
anti-Rb mAb, pMG3-245 [28] was the generous gift of Dr. Wen-Hwa Lee. The anti-EBNA-5 mouse mAb was clone JF186 [22]. It was biotinyiated with biotinamidocaproate N-hydroxy-succinimide ester (Sigma Chemicals, St. Louis) following the manufacturers protocol. The specificity of the anti-Rb antibodies has been controlled on reti-
FIG. 1. Immunofluorescence patterns of IB4 cells double stained with mouse anti-EBNA-5 mAb and anti-Rb mAbs (aRBlC1 or pMG3245). (A) Anti-Rb (aRBlCl), (B) anti-EBNA-5, same cell as in (A). Figure (C) anti-Rb (aRBlCl), (D) anti-EBNA-5 same cell as in (C). Rb (aRBlC1) and EBNA-5 antigens are present in a limited number of distinct foci. (E) Anti-Rb (pMG3-245) and (F) anti-EBNA-5, same cell as in (E). Staining with anti-Rb (pMG3-245) gives a diffuse nucleoplasmic fluorescence in contrast to anti-Rb (aRBlC1). Bar = 10 Cm.
316
JIANG
ET AL.
Rb AND
EBNA-5
noblastoma cells reconstituted with human Rb cDNA expressed from a retrovirus vector compared to nonreconstituted cells. Lymphoid cell8 expressing or lacking EBNA-5 were uaed to test the specificity of anti-EBNA-5 mAb. The mou8e anti-Sm mAb was clone Y12 [22]. Serum LU, derived from a Stockholm patient suffering from the CREST form of scleroderma, contained anti-centromere antibodies [23]. La antibodies were obtained from a human autoantiserum [24] and affinity purified by absorption to a recombinant La antigen [25] expressed in Escherichia coli from a full length construct inserted into an pEX2 vector. The Pab 416 mou8e anti-SV 40 large T mAb and it8 biotinylated derivative were from Oncogene Science (New York). Fluorescein isothiocyanate (FITC)-conjugated rabbit anti-mouse Ig, FITC-conjugated swine anti-rabbit Ig, FITC-conjugated rabbit antihuman IgG (y chains) were obtained from Dakopatts (Glostrup, Denmark). The rhodamine (RHOD)-conjugated streptavidin was obtained from Immunotech (Marseille, France). Double immunofhorescence staining. The cell8 were either growing on cover glasses or prepared as dry smears from BSS solution. Different fixation protocol8 have been tested (methanokacetone 1:2, l:l, or 4% paraformaldehyde applied either on dry or Cytospin-prepared wet cells). All fixation protocol8 gave essentially the Same staining pattern and differed only in the intensity of the staining. The double immunofluorescence staining procedure8 for different combination8 are described in Table 1. Each incubation step was followed by three washes in BSS. The stained slides were mounted in an antifading mounting solution (1,4-diazabicyclo-octane, Sigma Chemicals, St Louis, 25 mg/ml in glycerol/BSS, l:l, pH 8.6). To exclude nonspecific staining, two control8 for each combination were performed by: (1) using irrelevant reagent8 anti-SV40 T mAb for IB4 cell8 or BSS for COS cell8 instead of the antibodies in step 1 (Table 1); (2) using irrelevant reagents, biotinylated anti-SV40 T mAb (IB4 cells) or BSS (COS cells) instead of the antibodies in step 5 (Table 1). Nonspecific staining was at a very low level and did not present a problem in any of the staining routine8 described here. Digital imagingfluorescence microscopy. Two-color-stained preparation8 were examined in a Zeiss Universal fluorescence microscope (Carl Zeiss, Inc., Thornwood, NY) together with a silicon-intensified target (SIT) video camera (J8rgen Andersen Ingeniijrfirma AS, Glostrup, Denmark). A Zeiss 100X Planapo oil immersion objective (NA 1.3, iris) and a Zeiss 100X Neufluor oil immersion objective (NA 1.3) were used. The image8 transferred from the SIT camera were digitized, stored, and prOCe88ed in a Kontron/Zeiss IBAS computer. Two image8 were recorded in the same optical plane using different filters. One represented a RHOD-conjugated antibody while the other corresponded to a FITC-conjugated antibody. Lateral displacement between the two image8 due to the switch of filter8 on the microscope wa8 corrected for. Light emitted from one Auorochrome did not leak into the image recorded for the other. Using a dynamic discrimination followed by a small-sized median filter, the haze and very small speckle8 in the image8 were removed and two binary pattern8 were generated. Subsequently, the two binary pattern8 were assigned different gray level8 and added together. Thus the overlapping region8 have a third gray level. The visualization of the difference8 in gray level8 was facilitated by using the pseudocolor function in the IBAS computer. Using pseudocolor, we also co-display one contrast-enhanced image and one binary image at the Same time by setting one
317
CO-LOCALIZATION color for each image. The detailed scribed [18,24].
procedure8
were previously
de-
RESULTS The EBV-transformed lymphoblastoid cell line, IB4, known to express EBNA-5 and the nonpermissive SV4O-transformed monkey cell line, COS, served as our targets to compare the intranuclear distribution of Rb and EBNA-5 antigens (IB4 cells) and Rb and SV40 T antigens (COS-cells). A monoclonal mouse antibody directed against the Rb protein (aRBlC1) identified well-defined intranuclear foci (Figs. 1A and 1C) in IB4 cells. Similar structures were also seen after staining with EBNA-5 antibodies (Figs. 1B and 1D). Foci were present in cells dried on glass slides and wet cells that had attached to slides by cytocentrifugation in a salt solution. Similar findings were made on cells fixed in methanol:acetone (1:2 and 1:l) and on cells fixed in 4% paraformaldehyde. Comparison of the localization of Rb and EBNA-5 nuclear domains in a number of IB4 cells as well as digital image analysis of double-stained preparations (Fig. 2) showed that Rb and EBNA-5 co-localized to the same nuclear structures. Rb antigens detected by mAb aRBlC1 also co-localized with SV40 T-positive structures in COS cells (Fig. 3). The pattern was somewhat different from the Rb and EBNA-5 patterns in IB4 cells (Fig. 1 and 2). Rb and SV40 T antibodies identified a larger number of smaller foci in COS cells. In IB4 cells the Rb and EBNA-5 antibodies detected a more limited number of larger dot-like domains. Another monoclonal Rb antibody (pMG3-245) stained the nuclei of IB4 cells diffusely, without distinct nuclear foci. The nucleoli of IB4 or COS cells remained unstained by both types of Rb antibodies (Fig. 1). The EBNA-5 foci in IB4 ceils were not selectively stained by anti-centromere (Fig. 4), anti-Sm (Fig. 5), and anti-La antibodies (data not shown). The anti-centromere antibodies (Fig. 4) stained distinct nuclear domains, corresponding to the interphase position of centromeres. These regions were different from and had no apparent topological relationship to the EBNA-5-positive bodies (Fig. 4). The anti-Sm (Fig. 5) and anti-La antibodies give a relatively even immunofluorescence throughout the nucleus except for the nucleoli. The
FIG. 2. Co-localization of Rb (A and C) and EBNA-5 (B and D) antigen8 in IB4 cells. (A and B) Unedited pictures; (C and D) enhanced pictures. (E) Superimposed enhanced pictures. White pseudocolor represents overlapping between Rb (green) and EBNA-5 (orange). FIG. 3. Co-localization of Rb (A and C) and SV40 T (B and D) in COS cells. Picture8 are organized a8 in Fig. 2. White in (E) represents overlapping between Rb and SV40 T. FIG. 4. Different intranuclear distributions of centromeric antigen8 (A and C) and EBNA-5 (B and D). Picture8 are organized a8 in Fig. 2. Practically no white in (E). FIG. 5. Different intranuclear distributions of an snRNP antigen (Sm) (A and C) and EBNA-5 (B and D). In (E) the EBNA-5 (orange) is superimposed on the Sm (green) pattern. The Sm pattern contains a few diffuse region8 of stronger fluorescence (arrows) which do not coincide with the EBNA-5 foci.
318
JIANG
anti-Sm also detected a small number of more intensely fluorescing regions (arrows in Fig. 5). These structures did not coincide with any of the EBNA-g-positive foci. There was no apparent relationship between the distribution of the La-antigen and the EBNA-5 positive foci (data not shown). DISCUSSION The present results show co-localization of Rb (mAb aRBlC1) and EBNA-5 in EBV-transformed IB4 cells and of SV40 T and Rb in COS cells. More detailed and biochemically oriented studies are needed to examine the question whether the co-localization is associated with a direct complexing between the two proteins. Recent finding of a limited sequence homology between EBNA-5 and one of the Rb-binding sites in adenoviral ElA [lo] may be relevant in this connection. For further studies directed toward an understanding of the co-localization of Rb and EBNA-5 in IB4 cells, our earlier finding that EBNA-5 foci are mainly localized in the dispersed, transcriptionally active chromatin rather than in the tightly condensed, inactive heterochromatin may be relevant [18]. Rb is expressed by a wide spectrum of normal cells. Its involvement in normal growth control is suggested by cell cycle-related changes in its state of phosphorylation [13] and by the causative or contributory role of its inactivation or loss in several neoplastic diseases. The distinct Rb-EBNA-5 containing foci resemble, at least at the light microscopic level, the inclusion bodies found in cells that overexpress certain nuclear proteins. Further ultrastructural and immunocytochemical studies are required to show whether this resemblance is true or spurious. The intranuclear topology of the RBI EBNA-5 foci needs to be mapped in relation to specific chromosomal regions. Distinct areas involved in the assembly of molecular complexes associated with DNA replication, transcription, or RNA processing are of particular interest in this context. It is also noteworthy that our studies showed no relationship to snRNP containing complexes (references see accompanying paper by Nyman et al. [26]). We did not find a direct relationship to the La protein, known to be associated with the EBVencoded small nuclear RNA species EBER 1 and 2 [27]. This research was supported by grants from the Swedish Cancer Society, the Swedish Medical Research Council, the Karolinska Institutet, and AB ASTRA and by NIH Grant 5 ROlCA5222502. We thank Gunilla Kjellstrom and Evi Mellqvist for excellent technical assistance and Marie Wahren for immunopurification of the La antiserum.
REFERENCES 1.
DeCaprio, J. A., Ludlow, J. W., Lynch, D., Furukawa, Y., Griffin, J., Piwnica-Worms, H., Huang, C.-M., and Livingston, D. M. (1989) Cell 58,1085-1095.
Received June 28,199l
ET AL.
2. Howe, J. A., Mymryk, S. T. (1990) hoc.
J. S., Egan, C., Branton, P. E., and Bayley, N&l. Acad. Sci. USA 87, 5883-5887.
3. Benedict,
W. F., Morphree, A. L., Banerjee, A., Spina, C. A., Sparkes, M. C., and Sparkes, R. S. (1985) Science 219,973-975.
4. Reissman, P. T., Simon, M. A., Lee, W-H., and Slamon, D. (1989)Oncogene4,839-843. 5. Harbour, J. W., Lai, S. L., Whang-Peng, J. Gazdar, A. F., Minna, I. D., and Kaye, F. J. (1988) Science 241,353-357.
6. Lee, W.-H., Shew, J. Y., Hong, F. D., Serv, T. W., Donoso, L. A., Young, L. J., Bookstein,
R., and Lee, E. (1987) Nature
329,
642-645. 7. Buchkovich,
K., Duffy,
L. A., and Harlow,
E. (1989) Cell 58,
1097-1105.
8. Szekely,
L., Uzvolgyi, E., Jiang, W.-Q., Durko, M., Wiman, K. G., Klein, G., and Stimegi, J. (1991) J. Cell Growth Difier. 2, 287-295.
9. DeCaprio,
J. A., Ludlow, J. W., Figge, J., Shew, J. Y., Huang, C. M., Lee, W. H., Marsilio, E., Paucha, E., and Livingston, D. M. (1988) Cell 54,275-283.
10.
Whyte,
P., Williamson,
N. M., and Harlow,
E. (1989) Nature
334,124-129. 11. 12. 13.
Munger, K., Werness, B. A., Dyson, N., Phelps, W. C., Harlow, E., and Howley, P. M. (1989) EMBO J. 8,4099-4105. Huen, D. S., Grand, R. J., and Young, L. S. (1988) Oncogene 3, 729-730. Mihara, K., Cao, X-R., Murphree, A. L., T’Ang,
Yen, A., Chandler, S., Driscoll, B., A., and Fung, Y-K. T. (1989) Science
246,1300-1303. 14.
Klein, G. (1989) Cell 58, 5-8.
15.
Dillner, J., and Kallin, B. (1988) in Advances in Cancer Research (G. Klein and S. Weinhouse, Ed.), Vol. 50, pp. 95-158, Academic Press, San Diego.
16. Dillner,
J., Kallin, B., Ehlin-Henriksson, B., Rymo, L., Henle, W., Henle, G., and Klein, G. (1986) Int. J. Cancer 37, 195-200. 17. Petti, L., Sample, C., and Kieff, E. (1990) Virology 176, 563-
574. 18.
Jiang, W-Q., Wendel-Hansen, V., Lundkvist, A., Ringertz, Klein, G., and Rosen, A. (1991) J Cell Sci. 99,497-502.
19.
King, W., Thomas-Powell, A. L., Raab-Traub, and Kieff, E. (1980) J. Viral. 36, 506-518.
N. R.,
N., Hawke,
M.,
20. Finke, J., Rowe, M., Kallin, B., Ernberg, I., Rosen, A., Dillner, J., and Klein, G. (1987) J. Virol. 61, 3870-3878. 21. Gluzman, Y. (1981) Cell 23,175-182. 22. Lerner, E. A., Lerner, M. R., Janeway, C. A., and Steitz, J. A. (1981) Proc. Natl. Acad. Sci. USA 78, 2737-2741.
23. Hadlaczky, G., Went, M., and Ringertz, N. R. (1986) Exp. Cell Res. 167,1-15. 24. Nyman, U., Hallman, H., Hadlaczky, G., Pettersson, I., Sharp, G., and Ringertz, N. R. (1986) J. Cell Biol. 102,137-144. 25. Nyman, U., Ringertz, N. R., and Pettersson, I. (1989) Immunol. Lett. 22, 65-72.
26. Nyman,
U., Mellqvist, E., Pettersson, (1991) Exp. Cell Res. 197,307-313.
I., and Ringertz,
N. R.
27. Rosa, M. D., Gottlieb, E., Lerner, M. R., and Steitz, J. A. (1981) Mol. Cell. Biol. 1, 785-796. 28. Shew, J-Y., Chen, P-L., Bookstein, R., Lee, E., and Lee, W-H. (1990) Cell Growth Differ. 1, 17-25.
CELL RESEARCH
197,314-318
(1991)
Co-localization of the Retinoblastoma Protein and the Epstein-Barr Virus-Encoded Nuclear Antigen EBNA-5 W.-Q. JIANG,* Departments
L. SZEKELY,~ of *Medical
V. WENDEL-HANSEN,*
Cell Genetics and tTumor
Biology, Karolinska
Znstitutet,
G. KLEIN,?
AND A. ROS$N*
S-104 01 Stockholm,
Sweden
known, except for the fact that EBNA-1 is essential for the maintenance of the viral episomes and EBNA-2 is needed for the viral activation of the resting B-lymphocytes and subsequent immortalization. EBNA-5, also designated as EBNA-LP, is encoded by the same polycistronic message as EBNA-2, but is probably not required for immortalization. Earlier work in our [18] and Elliott Kieff s [ 171 laboratory has shown that EBNA-5 is expressed as relatively large foci within the dispersed chromatin of EBV-LCL. This distribution differs markedly from the evenly
A monoclonal antibody (aRBlC1) raised against an Rb fusion protein detects a limited number (4-10) of relatively large intranuclear foci in an EBV-immortalized cord blood cell line (IB4). These domains also bind an anti-EBNA-5 monoclonal antibody. The Rb antibody reactive sites also co-localize with the SV40 T antigen in transformed monkey cells (COS). The nuclear structures stained by aRBlC1 and EBNA-5 antibodies are distinct from the structures detected with antibodies against centromeric proteins and certain snRNP epitopes. EBNA-SlRb-positive domains do not selectively react with antibodies against the La antigen known to associate with the small EBV-encoded nuclear RNA species designated as the EBERs. o 1991 Academic press, IIIC.
dispersed, granular fluorescence of EBNA-1, 2, and 3 [16, 181. In view of the interactions between Rb and transforming proteins of the papova and adenoviruses, it is of great interest to explore whether any spatial or interactive relationship exists between Rb and the growth transformation associated EBV-encoded proteins. As a first step toward such an analysis, we have compared the intranuclear distribution of Rb and EBNA-5. For comparison, we have used SV40 T antigen, known to complex with Rb. We have also examined the distribution of centromeric proteins, a snRNP antigen (Sm) and the La antigen, known to associate with the EBV-encoded small nuclear RNA species (EBERs) [27]. We found co-localization of EBNA-5 and Rb and of SV40-T and Rb.
INTRODUCTION
The retinoblastoma susceptibility gene (Rb) codes for a nuclear protein (Rb) that appears to play a key role in the control of the cell cycle [ 1, 21. Deletion or mutation of the Rb gene is an essential event for the development of both familial and sporadic retinoblastomas [3] and may also contribute to the development of other forms of neoplasia [4, 51. Rb is a DNA-binding nuclear llO-kDa protein [6] that is nonphosphorylated during GO and Gl but becomes phosphorylated at the GUS boundary of the cell cycle [7]. The distribution of the Rb protein within the cell nucleus is largely unknown, although our recent immunofluorescence studies [8] suggest that part of the protein is concentrated to granules in the dispersed, transcriptionally active euchromatin compartment of the interphase nucleus. Rb forms complexes with the transforming proteins of several small DNA tumor viruses including the SV40 [9], adenovirus ElA [lo], and papilloma virus E7 [ll]. Specific amino acid sequences in Rb recognize a conserved sequence motif present in several of the viral transforming proteins [12]. It has been suggested that this complexing may interfere with the normal growth regulatory function of the Rb protein [13]. EBV-immortalized lymphoblastoid cell lines (LCL) express six virally encoded nuclear proteins, the EBNAs (references see [14, 151). Their function is largely un0014~4S27/91 $3.00 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
N. RINGERTZ,*
MATERIALS
AND
METHODS
Cell lines. The EBV-immortalized cord blood cell line, IB4 [19], was used. Like other EBV-LCLs, this cell line expresses all six EBVencoded nuclear antigens. It was chosen for its relatively prominent EBNA-5 expression. There is considerable variation in the expression of this antigen among cell lines, with regard to both the presence of the epitopes that can react with the mouse anti-EBNA-5 mAb’ JF186 [20] and the level of expression. The transformed monkey kidney epithelial cell line (COS) was also used to examine the intranuclear distribution of Rb and SV40 T antigens. This cell line contains the SV40 T gene but lacks the viral origin of replication [21].
’ BSS, balanced salt solution; E5, Epstein-Barr virus-encoded nuclear antigen-5; FITC, fluorescein isothiocyanate; Ig, immunoglobulin; mAb, monoclonal antibody; and RHOD, rhodamine. 314
Rb AND
EBNA-5
TABLE Two-Color
315
CO-LOCALIZATION
1
Immunofluorescence
Procedures’
Reagents (dilution)
Step
Procedure I for IB4 cells Mouse anti-Rb mAb aRBlC1 (1:30) or pMG3 245 (1:200) FITC-conjugated rabbit anti-mouse Ig (1:40)
Procedure III for IB4 cells
Procedure II* for IB4 cells Mouse anti-Sm Y12 (1:l)
mAb
FITC-conjugated rabbit anti-mouse Ig (1:20) -
FITC-conjugated swine anti-rabbit Ig (1:120) Normal mouse serum (1:lO) Biotinylated mouse anti-E5 mAb (1:5)
Normal mouse serum (1:lO) Biotinylated mouse anti-E5 mAb (15)
RHOD-conjugated streptavidin (1:50)
RHOD-conjugated streptavidin (1:50)
Human antisera LU (1:500) or La (1:l) FITC-conjugated rabbit antihuman IgG ychains (1:20) -
Normal mouse serum (1:lO) Biotinylated mouse anti-E5 mAb (1:5) RHOD-conjugated streptavidin (1:50)
Procedure IV for COS cells Mouse anti-Rb mAb RblCl (1:30) or pMG3 245 (1:200) FITC-conjugated rabbit anti-mouse Ig (1:40) FITC-conjugated swine anti-rabbit Ig (1:120) Normal mouse serum (1:lO) Biotinylated mouse antiSV40 T mAb(l:25) RHOD-conjugated streptavidin (1:50)
“All incubation steps were carried out in room temperature for 45 min. ‘Before step 1, the cells were incubated with RNase A (2.00 mg/ml, Sigma Co.) for 15 min.
All cells were cultured at 37’C in a 5% CO, atmosphere, in Iscove’s. medium containing 5% fetal calf serum, 1 n&f L-glutamine, 100 IUI ml penicillin, and 100 gglml streptomycin. Antibodies. The anti-Rb mouse mAb was clone aRBlCl[8] purified from ascites using a Sepharose-protein G column. The mouse
anti-Rb mAb, pMG3-245 [28] was the generous gift of Dr. Wen-Hwa Lee. The anti-EBNA-5 mouse mAb was clone JF186 [22]. It was biotinyiated with biotinamidocaproate N-hydroxy-succinimide ester (Sigma Chemicals, St. Louis) following the manufacturers protocol. The specificity of the anti-Rb antibodies has been controlled on reti-
FIG. 1. Immunofluorescence patterns of IB4 cells double stained with mouse anti-EBNA-5 mAb and anti-Rb mAbs (aRBlC1 or pMG3245). (A) Anti-Rb (aRBlCl), (B) anti-EBNA-5, same cell as in (A). Figure (C) anti-Rb (aRBlCl), (D) anti-EBNA-5 same cell as in (C). Rb (aRBlC1) and EBNA-5 antigens are present in a limited number of distinct foci. (E) Anti-Rb (pMG3-245) and (F) anti-EBNA-5, same cell as in (E). Staining with anti-Rb (pMG3-245) gives a diffuse nucleoplasmic fluorescence in contrast to anti-Rb (aRBlC1). Bar = 10 Cm.
316
JIANG
ET AL.
Rb AND
EBNA-5
noblastoma cells reconstituted with human Rb cDNA expressed from a retrovirus vector compared to nonreconstituted cells. Lymphoid cell8 expressing or lacking EBNA-5 were uaed to test the specificity of anti-EBNA-5 mAb. The mou8e anti-Sm mAb was clone Y12 [22]. Serum LU, derived from a Stockholm patient suffering from the CREST form of scleroderma, contained anti-centromere antibodies [23]. La antibodies were obtained from a human autoantiserum [24] and affinity purified by absorption to a recombinant La antigen [25] expressed in Escherichia coli from a full length construct inserted into an pEX2 vector. The Pab 416 mou8e anti-SV 40 large T mAb and it8 biotinylated derivative were from Oncogene Science (New York). Fluorescein isothiocyanate (FITC)-conjugated rabbit anti-mouse Ig, FITC-conjugated swine anti-rabbit Ig, FITC-conjugated rabbit antihuman IgG (y chains) were obtained from Dakopatts (Glostrup, Denmark). The rhodamine (RHOD)-conjugated streptavidin was obtained from Immunotech (Marseille, France). Double immunofhorescence staining. The cell8 were either growing on cover glasses or prepared as dry smears from BSS solution. Different fixation protocol8 have been tested (methanokacetone 1:2, l:l, or 4% paraformaldehyde applied either on dry or Cytospin-prepared wet cells). All fixation protocol8 gave essentially the Same staining pattern and differed only in the intensity of the staining. The double immunofluorescence staining procedure8 for different combination8 are described in Table 1. Each incubation step was followed by three washes in BSS. The stained slides were mounted in an antifading mounting solution (1,4-diazabicyclo-octane, Sigma Chemicals, St Louis, 25 mg/ml in glycerol/BSS, l:l, pH 8.6). To exclude nonspecific staining, two control8 for each combination were performed by: (1) using irrelevant reagent8 anti-SV40 T mAb for IB4 cell8 or BSS for COS cell8 instead of the antibodies in step 1 (Table 1); (2) using irrelevant reagents, biotinylated anti-SV40 T mAb (IB4 cells) or BSS (COS cells) instead of the antibodies in step 5 (Table 1). Nonspecific staining was at a very low level and did not present a problem in any of the staining routine8 described here. Digital imagingfluorescence microscopy. Two-color-stained preparation8 were examined in a Zeiss Universal fluorescence microscope (Carl Zeiss, Inc., Thornwood, NY) together with a silicon-intensified target (SIT) video camera (J8rgen Andersen Ingeniijrfirma AS, Glostrup, Denmark). A Zeiss 100X Planapo oil immersion objective (NA 1.3, iris) and a Zeiss 100X Neufluor oil immersion objective (NA 1.3) were used. The image8 transferred from the SIT camera were digitized, stored, and prOCe88ed in a Kontron/Zeiss IBAS computer. Two image8 were recorded in the same optical plane using different filters. One represented a RHOD-conjugated antibody while the other corresponded to a FITC-conjugated antibody. Lateral displacement between the two image8 due to the switch of filter8 on the microscope wa8 corrected for. Light emitted from one Auorochrome did not leak into the image recorded for the other. Using a dynamic discrimination followed by a small-sized median filter, the haze and very small speckle8 in the image8 were removed and two binary pattern8 were generated. Subsequently, the two binary pattern8 were assigned different gray level8 and added together. Thus the overlapping region8 have a third gray level. The visualization of the difference8 in gray level8 was facilitated by using the pseudocolor function in the IBAS computer. Using pseudocolor, we also co-display one contrast-enhanced image and one binary image at the Same time by setting one
317
CO-LOCALIZATION color for each image. The detailed scribed [18,24].
procedure8
were previously
de-
RESULTS The EBV-transformed lymphoblastoid cell line, IB4, known to express EBNA-5 and the nonpermissive SV4O-transformed monkey cell line, COS, served as our targets to compare the intranuclear distribution of Rb and EBNA-5 antigens (IB4 cells) and Rb and SV40 T antigens (COS-cells). A monoclonal mouse antibody directed against the Rb protein (aRBlC1) identified well-defined intranuclear foci (Figs. 1A and 1C) in IB4 cells. Similar structures were also seen after staining with EBNA-5 antibodies (Figs. 1B and 1D). Foci were present in cells dried on glass slides and wet cells that had attached to slides by cytocentrifugation in a salt solution. Similar findings were made on cells fixed in methanol:acetone (1:2 and 1:l) and on cells fixed in 4% paraformaldehyde. Comparison of the localization of Rb and EBNA-5 nuclear domains in a number of IB4 cells as well as digital image analysis of double-stained preparations (Fig. 2) showed that Rb and EBNA-5 co-localized to the same nuclear structures. Rb antigens detected by mAb aRBlC1 also co-localized with SV40 T-positive structures in COS cells (Fig. 3). The pattern was somewhat different from the Rb and EBNA-5 patterns in IB4 cells (Fig. 1 and 2). Rb and SV40 T antibodies identified a larger number of smaller foci in COS cells. In IB4 cells the Rb and EBNA-5 antibodies detected a more limited number of larger dot-like domains. Another monoclonal Rb antibody (pMG3-245) stained the nuclei of IB4 cells diffusely, without distinct nuclear foci. The nucleoli of IB4 or COS cells remained unstained by both types of Rb antibodies (Fig. 1). The EBNA-5 foci in IB4 ceils were not selectively stained by anti-centromere (Fig. 4), anti-Sm (Fig. 5), and anti-La antibodies (data not shown). The anti-centromere antibodies (Fig. 4) stained distinct nuclear domains, corresponding to the interphase position of centromeres. These regions were different from and had no apparent topological relationship to the EBNA-5-positive bodies (Fig. 4). The anti-Sm (Fig. 5) and anti-La antibodies give a relatively even immunofluorescence throughout the nucleus except for the nucleoli. The
FIG. 2. Co-localization of Rb (A and C) and EBNA-5 (B and D) antigen8 in IB4 cells. (A and B) Unedited pictures; (C and D) enhanced pictures. (E) Superimposed enhanced pictures. White pseudocolor represents overlapping between Rb (green) and EBNA-5 (orange). FIG. 3. Co-localization of Rb (A and C) and SV40 T (B and D) in COS cells. Picture8 are organized a8 in Fig. 2. White in (E) represents overlapping between Rb and SV40 T. FIG. 4. Different intranuclear distributions of centromeric antigen8 (A and C) and EBNA-5 (B and D). Picture8 are organized a8 in Fig. 2. Practically no white in (E). FIG. 5. Different intranuclear distributions of an snRNP antigen (Sm) (A and C) and EBNA-5 (B and D). In (E) the EBNA-5 (orange) is superimposed on the Sm (green) pattern. The Sm pattern contains a few diffuse region8 of stronger fluorescence (arrows) which do not coincide with the EBNA-5 foci.
318
JIANG
anti-Sm also detected a small number of more intensely fluorescing regions (arrows in Fig. 5). These structures did not coincide with any of the EBNA-g-positive foci. There was no apparent relationship between the distribution of the La-antigen and the EBNA-5 positive foci (data not shown). DISCUSSION The present results show co-localization of Rb (mAb aRBlC1) and EBNA-5 in EBV-transformed IB4 cells and of SV40 T and Rb in COS cells. More detailed and biochemically oriented studies are needed to examine the question whether the co-localization is associated with a direct complexing between the two proteins. Recent finding of a limited sequence homology between EBNA-5 and one of the Rb-binding sites in adenoviral ElA [lo] may be relevant in this connection. For further studies directed toward an understanding of the co-localization of Rb and EBNA-5 in IB4 cells, our earlier finding that EBNA-5 foci are mainly localized in the dispersed, transcriptionally active chromatin rather than in the tightly condensed, inactive heterochromatin may be relevant [18]. Rb is expressed by a wide spectrum of normal cells. Its involvement in normal growth control is suggested by cell cycle-related changes in its state of phosphorylation [13] and by the causative or contributory role of its inactivation or loss in several neoplastic diseases. The distinct Rb-EBNA-5 containing foci resemble, at least at the light microscopic level, the inclusion bodies found in cells that overexpress certain nuclear proteins. Further ultrastructural and immunocytochemical studies are required to show whether this resemblance is true or spurious. The intranuclear topology of the RBI EBNA-5 foci needs to be mapped in relation to specific chromosomal regions. Distinct areas involved in the assembly of molecular complexes associated with DNA replication, transcription, or RNA processing are of particular interest in this context. It is also noteworthy that our studies showed no relationship to snRNP containing complexes (references see accompanying paper by Nyman et al. [26]). We did not find a direct relationship to the La protein, known to be associated with the EBVencoded small nuclear RNA species EBER 1 and 2 [27]. This research was supported by grants from the Swedish Cancer Society, the Swedish Medical Research Council, the Karolinska Institutet, and AB ASTRA and by NIH Grant 5 ROlCA5222502. We thank Gunilla Kjellstrom and Evi Mellqvist for excellent technical assistance and Marie Wahren for immunopurification of the La antiserum.
REFERENCES 1.
DeCaprio, J. A., Ludlow, J. W., Lynch, D., Furukawa, Y., Griffin, J., Piwnica-Worms, H., Huang, C.-M., and Livingston, D. M. (1989) Cell 58,1085-1095.
Received June 28,199l
ET AL.
2. Howe, J. A., Mymryk, S. T. (1990) hoc.
J. S., Egan, C., Branton, P. E., and Bayley, N&l. Acad. Sci. USA 87, 5883-5887.
3. Benedict,
W. F., Morphree, A. L., Banerjee, A., Spina, C. A., Sparkes, M. C., and Sparkes, R. S. (1985) Science 219,973-975.
4. Reissman, P. T., Simon, M. A., Lee, W-H., and Slamon, D. (1989)Oncogene4,839-843. 5. Harbour, J. W., Lai, S. L., Whang-Peng, J. Gazdar, A. F., Minna, I. D., and Kaye, F. J. (1988) Science 241,353-357.
6. Lee, W.-H., Shew, J. Y., Hong, F. D., Serv, T. W., Donoso, L. A., Young, L. J., Bookstein,
R., and Lee, E. (1987) Nature
329,
642-645. 7. Buchkovich,
K., Duffy,
L. A., and Harlow,
E. (1989) Cell 58,
1097-1105.
8. Szekely,
L., Uzvolgyi, E., Jiang, W.-Q., Durko, M., Wiman, K. G., Klein, G., and Stimegi, J. (1991) J. Cell Growth Difier. 2, 287-295.
9. DeCaprio,
J. A., Ludlow, J. W., Figge, J., Shew, J. Y., Huang, C. M., Lee, W. H., Marsilio, E., Paucha, E., and Livingston, D. M. (1988) Cell 54,275-283.
10.
Whyte,
P., Williamson,
N. M., and Harlow,
E. (1989) Nature
334,124-129. 11. 12. 13.
Munger, K., Werness, B. A., Dyson, N., Phelps, W. C., Harlow, E., and Howley, P. M. (1989) EMBO J. 8,4099-4105. Huen, D. S., Grand, R. J., and Young, L. S. (1988) Oncogene 3, 729-730. Mihara, K., Cao, X-R., Murphree, A. L., T’Ang,
Yen, A., Chandler, S., Driscoll, B., A., and Fung, Y-K. T. (1989) Science
246,1300-1303. 14.
Klein, G. (1989) Cell 58, 5-8.
15.
Dillner, J., and Kallin, B. (1988) in Advances in Cancer Research (G. Klein and S. Weinhouse, Ed.), Vol. 50, pp. 95-158, Academic Press, San Diego.
16. Dillner,
J., Kallin, B., Ehlin-Henriksson, B., Rymo, L., Henle, W., Henle, G., and Klein, G. (1986) Int. J. Cancer 37, 195-200. 17. Petti, L., Sample, C., and Kieff, E. (1990) Virology 176, 563-
574. 18.
Jiang, W-Q., Wendel-Hansen, V., Lundkvist, A., Ringertz, Klein, G., and Rosen, A. (1991) J Cell Sci. 99,497-502.
19.
King, W., Thomas-Powell, A. L., Raab-Traub, and Kieff, E. (1980) J. Viral. 36, 506-518.
N. R.,
N., Hawke,
M.,
20. Finke, J., Rowe, M., Kallin, B., Ernberg, I., Rosen, A., Dillner, J., and Klein, G. (1987) J. Virol. 61, 3870-3878. 21. Gluzman, Y. (1981) Cell 23,175-182. 22. Lerner, E. A., Lerner, M. R., Janeway, C. A., and Steitz, J. A. (1981) Proc. Natl. Acad. Sci. USA 78, 2737-2741.
23. Hadlaczky, G., Went, M., and Ringertz, N. R. (1986) Exp. Cell Res. 167,1-15. 24. Nyman, U., Hallman, H., Hadlaczky, G., Pettersson, I., Sharp, G., and Ringertz, N. R. (1986) J. Cell Biol. 102,137-144. 25. Nyman, U., Ringertz, N. R., and Pettersson, I. (1989) Immunol. Lett. 22, 65-72.
26. Nyman,
U., Mellqvist, E., Pettersson, (1991) Exp. Cell Res. 197,307-313.
I., and Ringertz,
N. R.
27. Rosa, M. D., Gottlieb, E., Lerner, M. R., and Steitz, J. A. (1981) Mol. Cell. Biol. 1, 785-796. 28. Shew, J-Y., Chen, P-L., Bookstein, R., Lee, E., and Lee, W-H. (1990) Cell Growth Differ. 1, 17-25.