Activity and enhancer binding factors for jc virus regulatory elements in differentiating embryonal carcinoma cells

VIROLOGY

(1990)

177,784-789

Activity and Enhancer Binding Factors for JC Virus Regulatory in Differentiating Embryonal Carcinoma Cells

Elements

HARIKRISHNA NAKSHATRI, ALAN PATER, AND MARY M. PATER’ Division

of Basic

Medical

Sciences,

faculty

of Medicine, Received

Memorial

December

University

of Newfoundland,

27, 1989; accepted

St. John’s,

Newfoundland,

Canada

A 15 3V6

April 25, 1990

We have studied cell-type-specific expression by JC virus (JCV) DNA regulatory sequences using embryonal carcinoma (EC) cells as a model system. In transient transfection assays, JCV enhancer demonstrated activity in retinoic acid-differentiated neuronal type cells but not in undifferentiated or DMSO-differentiated muscle type cells. To correlate in vivo activity with the binding of transcription factors, we performed DNasel footprinting experiments. Retinoic acidtreated EC cell extracts provided three completely protected regions, each containing sequences with homology to nuclear factor 1 (NFl) binding motifs and the partially protected TATA box. Oligonucleotide competition studies suggest that all three NFl binding motifs are bound by the same factors but with different affinities and that there are cooperative interactions between NFl proteins binding to adjacent regions. No protected region other than the partially protected TATA box was detected in undifferentiated and DMSO-differentiated EC cells in which JC regulatory sequences were not expressed. 0 1990 Academic Press, Inc.

The results of analysis of papovaviral regulatory elements, using plasmid constructs presented diagrammatically in Fig. 1, in three EC cell types are presented in Table 1. In these experiments the enhancer-promoterless pSVOcat serves as a negative control. Plasmids pSV2cat and pBKcat, both with wider tissue specificities, were included as positive controls. In undifferentiated P19 cells none of the regulatory elements expressed significant CAT activity. However, differentiation by retinoic acid allowed the expression of the CAT gene from all three viral constructs. Comparable transfection efficiencies in various cell types were confirmed by measuring the levels of the low-molecular-weight plasmid DNAs from the transfected cells by dot-blot hybridization (data not shown). The levels of CAT activity from pSV2cat and pBKcat were significantly higher than that from pJCcat. Both pSV2cat and pBKcat, but not pJCcat, were also expressed in DMSO-differentiated cells (Table 1). These results confirm a restricted cell specificity of JCV regulatory elements to cells of neuronal origin. Deletion mutation analysis of JCV regulatory element was undertaken to identify the minimal sequences required for JCV enhancer function. A derivative of pJCcat, pH6 (Fig. l), for which Sstl and ligase were used to delete sequences from nucleotides 57 to 155 (7) to generate a plasmid with one 98-bp repeat and the enhancer upstream region, produced 229/oof wildtype CATactivity(Table 1). Deletion mutant pH8, which contains only a portion of the first repeat (nucleotides 5112-57) encompassing the putative TATA box and CCAAT box, produced activity equivalent to mutant

The human papovavirus, JC virus (JCV), has been isolated from patients with the demyelinating brain disease, progressive multifocal leukoencephalopathy (I, 2). Although it has significant sequence homology to SV40 and BK virus, JCV displays strict host cell specificity (3, 4). For example, in transgenic mice, JCV DNA induces dysmylenation in the central nervous system but not the peripheral nervous system (3). In tissue culture, unlike SV40 and BK, JCV regulatory elements’ activity is restricted to human fetal glial cells (4). Structurally, the JCV regulatory element consists of a 98-bp tandem repeat within which is a 15-nucleotide AT-rich sequence believed to function as a TATA box. For functional analysis of JCV regulatory elements we have utilized the embryonal carcinoma (EC) cell system, P19, which can differentiate into a mixture of glial cells, neurons, and astrocytes upon retinoic acid (RA) treatment and into cardiac and skeletal muscle cells upon dimethyl sulfoxide (DMSO) treatment (5). In this study, we describe an analysis of the function of the JCV regulatory region in undifferentiated, retinoic aciddifferentiated, and DMSO-differentiated EC cells by transient transfection assays. Plasmids pSVOcat without enhancer-promoter and pSV2cat and pBKcat containing SV40 and BK virus enhancer-promoter, respectively, have been described previously (6). Plasmid pJCcat contains the HindIll-Pvull fragment (JCV Mad I strain, nucleotides 5112-270 (7)) of JC DNA inserted into the /-/indIll site of plasmid pSVOcat such that the CAT gene is expressed from the early JC promoter. ’ To whom

requests

for reprints

0042-.6822/90 $3.00 CopyrIght 0 1990 by Academic Press. Inc. All rights of reproduction in any form reserved.

should

be addressed.

784

SHORT ,SV24AT

pBK*CAT

72b~

1

72bm [2ll2ll2l

COMMUNICATIONS

v

llbp

rJC.CAT

PHB

PH8

PH9

PHlO

785

We next performed DNasel footprinting to correlate in viva activity of the JCV enhancer with in vitro DNAprotein interaction. Footprints are shown in Fig. 2A and the protected sequences are summarized in Fig. 2B. In undifferentiated cell extracts (PlSUD), there was a weak protection principally of the TATA box. In addition to the partial protection of the TATA box, three completely and one partially protected regions were produced by retinoic acid-differentiated cell extracts (P19RA). All three completely protected regions contained sequences homologous to motifs for CCAAT binding protein, CTF/NFl (10, 1 I). Similar protection has been observed by others with extracts from glial cells and purified NFl (13- 15). Note that one protected region, region I (Figs. 2A and 2B), also contains sequences which share homology to the E4TFl binding site (12). It has been reported that the degree of protection afforded to a motif corresponding to region I in JCV expressing glial cells and nonexpressing HeLa cells differs markedly, with complete protection only in glial cells (13). This complete protection could be due to the interaction of a second factor, present only in glial cells, along with NFl. This might involve a factor(s) binding to the E4TFl motifs of this region in differentiated EC cells. Affinity purification and uv crosslinking experiments are required to examine this possibility. Footprints in DMSO-differentiated cell extracts are similarto those in undifferentiated cell extracts (Fig. 2A, P19DM)

FIG. 1. Structures of recombinant plasmids. Repeated sequences of SV40. JC, and BK enhancers are given as boxes. Internally deleted sequences of plasmid H6 and H9 are indicated by broken lines.

pH6 (19% of wild-type activity). Mutant pH9, which contains the second half of the second repeat (nucleotides 157-217) and repeat upstream sequences (nucleotides 2 18-270) did not have any significant activity. Thus, the single copy of TATA plus CCAAT boxes accounts for a fifth of the activity of the JCV enhancerpromoter sequences and no activity is seen when these sequences are deleted. To study whether the 72-bp repeat sequences of SV40 enhancer could affect the expression from the JCV early promoter, we cloned the Fokl fragment of pSV2cat (nucleotides 150424 (8)) upstream of the putative JCV TATA and CCAAT boxes (as in plasmid pH8) to generate the plasmid pH 10. This modification did not have any effect on the level of CAT expression in any of the three different cell types (Table 1). Therefore it would appear that the JCV promoter, represented by the TATA and CCAAT boxes, shows enhanced, cell-specific activity only in response to a duplication of these sequences but not in response to the generally very effective SV40 enhancer sequences.

TABLE

1

ACTIVIT~OFSV~O, JC, AND BK REGULATORY ELEMENTSIN TIATED AND RETINOIC ACID- AND DMSO-DIFFERENTIATED CARCINOMA CELL LINE P19

UNDIFFERENEMBRYONAL

Cell phenotype

PSVOcat pSV2cat pJCcat PH6 PH6 pH9 pHl0 pBKcat

P19UD

P19RA

Pl SDMSO

0.1 0.4 0.1 ND ND ND 0.07 1.2

0.1 85.8 6.6 1.5 1.3 0.2 1.31 372

0.1 58.4 0.3 ND ND ND 0.31 69.6

Note. P19UD, P19RA, and Pl 9DMSOare undifferentiated and retinoic acid- and DMSO-differentiated P19 cells, respectively. The cells were transfected with 10 pg of the indicated plasmids and 10 Pg of pUC-19. Units are defined as the percentage of conversion of [‘4C]chloramphenicol to acetylated chloramphenicol. Figures greater than 100% represent numbers for which diluted extracts had been used. The average of two experiments are presented. pSVZcatand pBKcat-transfected differentiated cell extracts are diluted 1 in 30 times and the activity is quantitated after considering the dilution factor. ND: not determined.

786

SHORT

COMMUNICATIONS

P19DM

Pl9RA P 19UD

A

!FB#W#!#IF

A

SV JC BKPU 8FBll 2t34/55)7 dF

% FB’/!F

TAl lATA i TATA

analysis of JC regulatory elements. The Hindlll-Pvull fragment (nucleotides 51 12-270 (7)) of JC was first cloned FIG. 2. (A) DNasel footprinting intoXba1 site of pUC-19. Inserted enhancer fragment was then liberated by digestion with restriction enzymes SalI and Smal and electroeluted from agarose gels. End labeling was by reverse transcriptase in the presence of #P]dCTP. End-labeled DNA was then subjected to DNasel footprinting with 25 Pg of protein of nuclear extracts from undifferentiated (Pl 9UD), retinoic acid-differentiated (P19RA). and DMSO-differentiated (P19DM) cells. Nuclear extract preparation and DNasel footprints were according to the methods of Henninghausen and Luban (9). Samples applied to the lanes are as follows. Chemical cleavage of purines is in lane A + G (24). DNA treated with DNasel in the absence and presence of nuclear extract is in lanes F and B, respectively. Lanes 1 to 8 represent competition experiments where the same amount of protein and probe were incubated with cold competitor DNA before DNasel digestion. Competitors were SV40 in lanes 1 and 2, JC in lanes 3 and 4, BK in lanes 5 and 6. and pUC-19 in lanes 7 and 8. In odd- and even-numbered lanes, 4 ng (1 O-fold excess) and 20 ng, respectively, of competitor DNAs were used. Competitor DNA fragments were the HindIll-Pvull fragment of pSV2cat for SV40 (nucleotides l-340 (8)), the HindIll-Pvull fragment of JCV (nucleotides 51 12-270 (7)) the Haelll fragment of BK (nucleotides 3186-3479 (25)) and the Pvull fragment of pUC-19 (nucleotides 306-638). Three completely protected regions (I, II, Ill) and one partially protected region (bracket with broken line) were detected in retinoic acid-differentiated cell nuclear extracts. (B) DNasel protected nucleotide sequences. Protected sequences are overlined and the designations, TATA, I, II, and Ill, above the lines correspond to those in (A). Consensus sequences for known transcription factors are underlined and the names of corresponding transcription factors are indicated below.

SHORT

COMMUNICATIONS

787

III P19-RA PlP-DH 5111

5120

5130/o

10

I I I II TAAGCTTGGAGGCGGAGGCGGCCTCGGCCTC~CTGTATATAT

20 I

30 40 50 60 I I I I I MAAMMGGGAAGGGATGGCTGCCAGCCAAGCATGAGCTCATACCTA NFl TATA II

P19-UD PlP-RA PlP-DM 70

80 90 100 I I I i GGGAGCCAACCAGCTAACAGCCAGTAAACAAAGCACAAGG

110 I

120

130

140

150

I

I

I

I

I

CTGTATATATAAAAAAAAGGGAAGGGATGGCTGCCAGCCAAGCATGAGC I NFI

P19-UD PIP-RA PIP-DM

I 160

170 I

I

180 I

190 I

200 I

I

TCATACCTAGGGAGCCAACCAGCTAACAGCCAGTAAACAAAGCACAAGG

PIP-UD PIP-RA PIP-DM

250

260

270

I

I

I

GAGCTGTTTTGGCTTGTCACCAG

210

220

230

240

I

I

I

GGAAGTGGAMGCAGCCAAGGGAACATGTTTTGCGAGCCA 'E4TFl NFl

FIG. P-Continued

and correlate with lack of activity of JCV regulatory elements in these cells. Specificity of DNA-protein interaction was tested using different competitor DNAs. Unlabeled JCV regulatory element DNA competed efficiently with all pro-

tected regions with the possible exception of the TATA sequence (Fig. 2A). Competition was also observed with NFl motif-containing BKV regulatory elements. Neither the regulatory elements of SV40 DNA, which has no NFl binding motif, nor the 323-bp Pvull frag-

TABLE2 TRANSREGULATION

OF SV40, JC, AND BK REGUMTORY

ELEMENTS BY JC T-ANTIGEN Test plasmids

Cotransfecting CMV-JCV-TA CMV-JCV-TA

plasmids”

pSV2cat

pJCEcatb

160 68.7

2.2 1.5

pJC,catC 3.0 5.6

pBKEcatd

pBK,caY

233 100

Note. Cotransfection was with CMV-JCV-TA in retinoic acid-treated P19 cells. Units are as in Table 1. Averages presented. a Concentrations of test and CMV-JCV-TA or CMV-JCV-TA plasmids were 1 and 10 Mg, respectively. Total quantity adjusted to 1 1 pg with pUC-19 DNA. b pJC,cat is the same as pJCcat. ’ pJC,cat has the same JC sequences as in pJC,cat, but in a reverse orientation, d pBK,cat is the same as pBKcat. e pBK,cat has the same BK sequences as pBK,cat, but in a reverse orientation

4.7 12.3 of two experiments of transfected

are

DNA was

788

SHORT

COMMUNICATIONS

ment of pUC-19 DNA competed for JCV DNA-protein interactions (Fig. 2A). These results suggest a closer relationship between JCV and BKV transcriptional control mechanisms than between those of JCV and sv40.

To further examine the nature of the DNA-protein interactions, we used oligonucleotide competition of DNasel protection. Region I oligonucleotide (sequences BAAGGGGAAGTGGAAAGCAGCCAA3’; Fig. 3) competed effectively for all three regions (lane 4). Competition by region II oligonucleotide (sequences 5’ATGGCTGCCAGCCAAG3’) was weak and observed for only regions I and II (Fig. 3, lane 3). Stronger competition of region II by region I than by region II oligonucleotide may be due to the cooperative nature of NFl interactions: binding to region I may be important for subsequent, stronger binding to region II. These results suggest that while there is interaction of the same factor(s) to all three regions and with different affinities (as also supported by gel mobility shift experiments; data not shown), cooperative interaction with the factors on neighboring sites is also a major determinant of binding. Control oligonucleotide, sequence 5’AAGGGAAGGGATGG3’, a sequence present between the TATA box and the CCAAT box of JCV DNA, did not affect any protected regions (Fig. 3, lane J). Some proteins, such as NFl , are involved in cell-specific transregulation of promoter-enhancers through direct interactions with the target DNA sequences (11, 16-18). However, many other proteins have transregulating functions that are independent of such interactions (for a review, see (19)). To initiate studies of mechanisms operative for expression of JC regulatory sequences in retinoic acid-differentiated P19 cells, we examined the influence of JCVT-antigen on papovaviral transcription by performing transient cotransfection experiments using JCV T-antigen. Plasmid CMV-JCV-TA, containing the JCV T-antigen coding sequences under the control of cytomegalovirus (CMV) regulatory elements, was used as a T-antigen expression plasmid while CMV-JCV-TA, which has the deletion of nucleotides 4958 to 3013 in the T-antigen coding region, served as a negative control. The plasmid CMV-JCVTA was constructed by inserting the Eael fragment, nucleotides 64 to 810 (20) of CMV into the partial HindIll plus Ball digested fragment of JCV DNA from nucleotides 5 1 17 to 17 16 (7) in the sense orientation. CMV-JCV-TA was constructed by digestion of CMVJCV-TA with Stul and subsequent recircularization of the largest of the two fragments. Cotransfection with CMV-JCV-TA plasmid resulted in the.suppression of the expression from the early promoters of JC, SV40, and BK viruses (Table 2). To determine the effect of the T-antigen on the late control regions, cotransfection

FIG. 3. Oligonucleotide competition experiments Competition was carried out with 100 ng of the oligonucleotides corresponding to region II (lane 3) and region I (lane 4). DNasel digestion in the absence and presence of cell extracts without any competitor are in lanes F and B, respectively. In lane J, a control oligonucleotide of sequence 5’AAGGGAAGGGATGGJ was used. Competition was performed in retinoic acid-differentiated cell extracts. Competition with oligonucleotide of region I was strong compared to competition with region II oligonucleotide.

experiments were done with JC and BK late promoterCAT plasmids. Approximately two- and three-fold increases in the activities of JCV and BKV late regions, respectively, were observed (Table 2). These data, along with the previously published results (27-23), suggest that tumor antigens of all three viruses are ca-

SHORT

COMMUNICATIONS

pable of transactivating the expression of homologous and heterologous papovaviral late control regions. It remains unclear, however, how JCV T-antigen affects transcription from all three viral regulatory elements, especially SV40, since these elements contain very limited sequence homologies. An interesting possibility is that the JCV T-antigen may act by stimulating the expression of or modifying a preexisting cellular transcription factor(s) rather than by interacting directly with specific DNA sequences. In support of this hypothesis, Mitchell et a/. (22) have observed that SV40 T-antigen interacts with transcription factor AP2 and affects the activity of SV40 regulatory elements. The use of embryonal carcinoma cells in this study is a powerful system for identification of tissue-specific transcription factors induced upon differentiation. We are currently seeking the identity of specific factors and the mechanism(s) of their induction during the differentiation process. If the mechanism of induction of transcription factors is by de nova synthesis, differential cDNA screening using viral regulatory elements as probes may serve as a convenient method for cloning genes encoding transcription factors restricted to the RA-differentiated EC cells. ACKNOWLEDGMENTS We thank H. Hamada for P19 cell line and plasmid pCMS 207 and K. Ross for typing the manuscript. Harikrishna Nakshatri has been a predoctoral fellow of the Cancer Research Society, Inc., Montreal, Quebec, Canada, and more recently of the National Cancer Institute of Canada. This work was supported in part by the Medical Research Council and the National Cancer Institute of Canada.

REFERENCES 7. PADGET, B. L., ROGERS, C. M.. and WALKER, D. L., infect Immun. 15,656-662 (1977). 2. PADGET, B. L., and WALKER, D. L., 1. infect. Dis. 127, 467-470 (1973). 3. SMALL, J. A., SEANGOS, G. A., CORK, L., JAY, G., and KHOURY, G., Ce//46, 13-18(1986).

789

4. KENNEY, S., NATARAIAN, V., STRIKE, D., KHOURY, G., and SALZMAN, N. P., Science 1337-l 339 (1984). 5. RUDNICKI, M. A., and MCBURNEY, M. W. In “Teratocarcinoma and Embryonic Stem Cells: A Practical Approach” (E. J. Robertson, Ed.), pp. 19-49. IRL Press, Oxford 1987. 6. PATER, A., and PATER, M. M., Virology 163,625-628 (1988). 7. FRISQUE, R. J., BREAM, G. L., and CANNELLA. M. T., /. viral. 51, 458-469 (1984). 8. GORMAN, C. M., MOFFAT, L. F., and HOWARD, B. H., Mol. Ce//. Biol. 2, 1044-1051 (1982). 9. HENNINGHAUSEN, L.. and LUBAN. H. In “Methods in Enzymology” (S. L. Berger and A. R. Kimmel, Eds.), Vol. 152, pp. 721-735. Academic Press, San Diego, 1987. 70. JONES, K. A., YAMAMOTO, K. R., ~~~TJIAN, R., Cell42, 559-572 (1985). 11. JONES, K. A., KADONAGA, J. T., ROSENFELD, P. J., KELLY, T. J., and TJIAN. R., Cell48, 79-89 (1987). 72. JONES, N. C., RIGBY, P. W. J., and ZIFF, E. B., Genes Dev. 2, 267281 (1988). 13. AMEMIYA, K., TRAUB, R., DURHAM, L., and MAJOR, E. O., J. Biol. Chem. 264,7025-7032 (1989). 14. TADA, T.. LASHGARI, M., RAPPAPORT. J., and KHALILI, K., /. Viol. 63,463-466 (1989). 75. TAMURA, T., INOUE, T.. NAGATA, K., and MIKOSHIBA, K., Biochem. Biophys. Res. Commun. 157,419-425 (1988). 76. RAYMONDJEAN, M., CEREGHINI, S., and YANIV. M., Proc. Nat/. Acad. SC;. USA 85,757-76 1 (1988). 77. CHRIST/, R. J., YANG, V. W., NTAMBI, J. M.. GEIMAN, D. E., LANDSCHULZ, W. H., FRIEDMAN,A. D., NAKABEPU,~., KELLY,T. J., and LANE, D. M., Genes Dev. 3,1323-l 335 (1989). 18. BIRKENMEIER, E. H., GWYNN, B., HOWARD, S.. JERRY, J., GORDON, J. I., LANDSCHULZ, W. H., and MCKNIGHT, S. L., Genes Dev. 3, 1146&1156(1989). 79. JOHNSON, P. F., and MCKNIGHT, S. L., Annu. Rev. Biochem. 58, 799-839 (1989). 20. BOSHART, M., WEBER, F., JOHN, G., DORSCH-HASTER. K.. FLECKENSTEIN, B., and SCHAFFNER, W.. Ce1/41,521-530 (1985). 21. CAPUTO, A., BARBANTI-BRODANO, G., WANG, E.. and RICCIARDI, R. P., Virology 152, 459-465 (1986). 22. MITCHELL, P. J., WANG, C., and TJIAN, R., Cell 50, 847-861 (1987). 23. LASHGARI, M. S., TADA. H., AMINI, S., and KHALILI, K., Virology 170,292-295 (1989). 24. MAXAM, A. M., and GILBERT, W. In “Methods in Enzymology” (L. Grossman and K. Moldave, Eds.), Vol. 65, pp. 499-560. Academic Press, San Diego, 1980. 25. YANG, R. C. A., and Wu, R., Science 206, 456-46 1 (1979).