- Email: [email protected]
Effects of repetitive and non-repetitive rat rDNA enhancer elements on in vivo transcription by RNA polymerases I and II
Gene, 141 (1994) 271-275 0 1994 Elsevier Science B.V. All rights reserved.
271
0378-l 119/94/$07.00
GENE 07769
Effects of repetitive and non-repetitive rat rDNA enhancer elements on in vivo transcription by RNA polymerases I and II (Ribosomal RNA; transfection; primer extension; CAT assay)
Asish K. Ghosh, Milko Kermekchiev
and Samson T. Jacob
Department of Pharmacology and Molecular Biology, The Chicago Medical School, North Chicago, IL 60064, USA Received by R. Padmanabhan:
14 July 1993; Revised/Accepted:
21 October/3
November
1993; Received at publishers:
14 December
1993
SUMMARY
Previous study has demonstrated that a far upstream 174-bp spacer sequence of the rat rRNA-encoding (rDNA) gene can function as an enhancer in vitro in an orientation- and distance-independent manner [Dixit et al., J. Biol. Chem. 262 (1987) 11616-116221. To demonstrate that this element can also function in vivo, two rat rDNA-cat plasmids, one with the 174-bp element and the other without this sequence, were constructed and transfected into CHO cells. Primer extension analysis of the transcripts produced after transfection showed that transcription initiation occurred at the + 1 site of the rDNA. The 174-bp sequence stimulated the rat poll promoter activity in cis 44S-fold over the control (with the promoter alone). This RNA polymerase (~011) enhancer also stimulated the mouse metallothionein-I (MT-Z) and SV40 promoter activities in vivo, irrespective of its distance and orientation. Further dissection of the 174-bp element revealed that the stimulatory activity on the RNA polymerase II (~0111) promoter resides within the 37-bp and 43-bp domains at the 3’ end of the 174-bp element. Unlike this spacer enhancer, the 130-bp repeat element (RE) proximal to the rat promoter [Ghosh et al., Gene 125 (1993) 217-2221 was unable to modulate the ~0111promoter activity in vivo. These data show that while the non-repetitive enhancer sequence of rat rDNA is interchangeable for the ~011 and ~0111 promoters, the RE is polI-specific.
INTRODUCTION
The basal rRNA-encoding gene (rDNA) transcription requires a species-specific factor (TFID/SLl ) and RNA polymerase I (see Paule, 1993). Maximal transcription of rDNA, however, is dependent upon other c&acting upstream spacer sequences and trans-acting factors. Tjian and co-workers identified in human cells an upstream control element (UCE) and characterized a protein factor Correspondence to: Dr. S.T. Jacob,
Department
of Pharmacology
Molecular Biology. The Chicago Medical School, 3333 Green Road, North Chicago, IL 60064, USA. Tel. (l-708) 578-3271; (l-708) 578-3268.
and Bay Fax
Abbreviations: AMV, avian myeloblastosis virus; bp, base pair(s); CAT, chloramphenicol acetyltransferase; cat, gene encoding CAT, CHO, Chinese hamster ovary; d, deoxyribo; E,BF, enhancer-l-binding SSDI 0378-1119(93)E0765-6
(UBF, meaning upstream binding factor) that interacts with both UCE and the downstream core promoter sequences (Learned et al., 1986). In the human system, UBF is essential for SLI interaction with the promoter (Bell et al., 1990). In addition to UCE, other spacer elements such as enhancers have been identified in the spacer regions of Xenopus (Moss, 1983; Reeder, 1984), rat (Dixit et al., 1987; Ghosh et al., 1993), yeast (Johnson and Warner, 1989) and more recently, in mouse (Pikaard factor; kb, kilobase gene; nt, nucleotide(s);
or 1000 bp; MT-I, metallothionein-I encoding oligo, oligodeoxyribonucleotide; pol, RNA poly-
merase; PolIk, Klenow (large) fragment of E. coli DNA polymerase I; r, ribosomal; rDNA, DNA (gene) coding for rRNA; RE, repeat element(s); SDS, sodium dodecyl sulfate; SLl, species-specificity factor for polI transcription; SV40, simian virus 40; tsp, transcription start point(s); UBF, upstream binding factor: UCE, upstream control element(s).
272 et al., 1990) rDNA. We have identified three spacer elements within the 4%kb region upstream from the rat rDNA transcription start point (tsp), which can stimulate rDNA gene transcription in vitro (Dixit et al., 1987). One element that is located between the -2183-kb and - 2357-kb positions has been fully characterized in vitro (Dixit et al., 1987). This enhancer sequence could function independent of orientation and distance and has, therefore, met all the criteria established for a poll1 enhancer. Subsequent studies have demonstrated that the 174-bp enhancer element consists of multiple motifs (Garg et al., 1989; Jacob et al., 1991). We have characterized a protein (E,BF) that interacts with the promoter sequence, 174-bp enhancer (Zhang and Jacob, 1990) and 130-bp repeat enhancer sequence (Ghosh et al., 1993) of rat rDNA and stimulates transcription from the rRNA promoter (Zhang and Jacob, 1990 and Ghosh et al., 1993). This protein is related to the human Ku autoantigen with respect to immunological and certain structural characteristics (Hoff and Jacob, 1993). Recent study has demonstrated that E,BF is required for the basal ~011 transcription and that it acts primarily in the preinitiation complex formation (Hoff et al., 1994). Previous study in our laboratory demonstrated that the 174-bp rat rDNA spacer element can stimulate polII promoter (MTI) activity in vitro (Dixit et al., 1989). Similar results were also obtained in yeast where a ~011enhancer element was found to stimulate ~0111promoter activity (Larch et al., 1990). The purpose of present study was to determine whether (a) the 174”bp rat rDNA spacer element can function in vivo; (b) it can also modulate transcription from poll1 promoters; and (c) unlike this &-acting sequence, the 130-bp RE recently characterized in our laboratory (Ghosh et al., 1993) can function only with ~011 promoter.
EXPERIMENTAL AND DISCUSSION
(a) The 174-hp rat rDNA enhancer element stimulates pol1 promoter activity in vivo To investigate whether the 174-bp spacer enhancer element (located between bp positions -2357 and -2183 with respect to tsp) can also enhance the rat rDNA promoter activity in vivo, we constructed two rDNA-cat plasmids, one containing cat gene and rat ~011 promoter region, (pRrPcat) and another containing the 174-bp sequence in addition to the promoter and cat gene (pRrEPcat) (see Fig. lA, a and b) and transfected them into CHO cells. Analysis of the RNA from the transfected cells by primer extension revealed a 192-nt product that corresponds to a correctly initiated (+ I tsp) transcript (Fig. 2, lanes 5 and 6). When the transfected construct
contained the enhancer element (pRrEPcu~), transcription was significantly stimulated (Fig. 2, compare lanes 5 and 6). Densitometric scanning of the transcripts showed a 4-5-fold increase in the amount of RNA initiated at the correct site in the presence of the enhancer. This result suggested that the 174”bp region of rat rDNA spacer can augment ~011 promoter activity in vivo. (b) The 174-bp rDNA enhancer can stimulate poll1 promoter activity in vivo Next, we investigated whether the 174-bp spacer sequence of rat rDNA can also stimulate ~0111 promoter activity in vivo. Previous study in our laboratory has shown that this upstream element can modulate mouse MT-Z gene promoter activity in vitro (Dixit et al., 1989). To test whether the poll enhancer functions with ~0111 promoters also in vivo, CHO cells were transfected with two plasmids, one that contains mouse MT-I gene promoter (- 148 to +67) ligated to the cut gene and the other consisting of the 174-bp rDNA fragment placed in front of the mouse MT-f promoter and cat gene (see Fig. lA, c and d). Analysis of the CAT activity in CHO cells transfected with these constructs showed that the 174-bp sequence can also augment the ~0111 promoter TABLE I Relative CAT activity in CHO cells transfected with different plasmid constructs Plasmids used in transfectiot?
c. pMTlcat d. pRrEnMTlcu~ e. peat promoter f. pRrEn,SV4OPcat g. pRrEn,SV40Pcat h. pRrEn, +,SV40Pcat i. pRrEn,SV40Pcat j. pRrEn(wt./dir.)SV40Pcat k. pRrEn(wt./rev.)SVllOPcnt I. pRrEn130SV40Pcar m. pSV2cat
Relative CAT activity” (% of CAT conversion relative to that of pSV2cat) 24 51 21 20 30 10 3.5 33 10 l1100)
aDesignations c-m correspond to the schematic representation of constructs c-m in Fig. 1A. CsCl-gradient-purifi~ plasmid DNA (10 ug) was used for transfection as described by Gorman et al. (1982). Different plasmids (for details, see Fig. lA, c-m) were transfected into CHO cells and after 48 h of transfection, cells were harvested and lysate was made by three successive freezing and thawing. Protein concentrations were determined by BCA protein reagent assay kit (Pierce). bCAT activities from different constructs were normalized to the value of CAT activity obtained with control pSV2cat plasmid and presented as relative CAT activity. Reaction mixture contained 0.2 pCi [*“C]chloramphenicol/5 mM acetyl CoA/0.25 M Tris.Cl (pH 7S)/cell extract (100-200 ug total protein). Samples were incubated at 37°C for 30 to 60 min and then analyzed by thin-layer chromatography as described (Gorman et al., 1982).
273
a.
A.
-167
+1
cat
+124
i.
b.
Enhancer
k. Enhancer
Enhancer
rDNA promoter
cat
MTI Promoter
I.
C.
Enhancer sv40
cat
SV40 Promoter
m. Enhancer
d. Enhancer SV40 Promoter
e.
cat
B. f.
HpaI
AACCGATTCGAGAGTGCATGTCCTTTCACTTAGAGGTGGCCTGT :2351 AP-2 site C CTlGA S T;GG
TTTCACCGCTCCGTCCTACTTTCCTTTTCTGGGT,G
SV40 Promoter
II.
Ava I TTGGCC&GAGTCGCCATCCCGCGTTCACTGTGTTCC
cat
,,%&
&iI Poly. Consensus sequence ACAACCGGGACCtXYWX6CCATTCGGGAGAAGTGGTGGGTA~
i.
KP+ -2183
Fig. 1. Plasmids, enhancer elements and their sequences. (A) Schematic representation of the recombinant plasmids used: (a) pRrPcat that contains Sall-Hind111 rat rDNA fragment (nt - 167 to + 124) in front of cat gene of pSVOcat but lacks SV40 promoter and enhancer. (b) pRrEPcat that contains
Hpal-KpnI
(-2357
to -2183)
fragment
of rat rDNA
ligated
to WI-Hind111
(- 167 to + 124) fragment
of the promoter
and cat gene of
pSVOcat. (c) pMTlcat that contains mouse MT-I gene promoter (- 148 to +67) in front of cat gene. (d) pRrEnMTlPcat that contains rat rDNA 174-bp enhancer, mouse MT-l gene promoter (- 148 to + 67) and cat gene. The 174-bp element is separated from the - 148-bp position of MT-l promoter
by 383-bp
rat rDNA
spacer
sequence.
(e) peat contains
SV40 promoter
and cur gene. (f) pRrEn,SV40Pcat
was constructed
one copy of 37-bp domain of the 174-bp enhancer at the Bgrll site of peat. (g) pRrEn,SV40Pcat plasmid was made by inserting 174-bp enhancer at the Bgrll site of peat. (b) pRrEn, + z SV4OP cat was constructed by inserting a 37-bp and 43-bp containing
by inserting
43-bp domain of the fragment in front of
SV40 promoter of pcut promoter plasmid in direct orientation and separated from SV40 promoter by a 383-bp rat rDNA spacer fragment (SallKpnl). (i) pRrEn,SV40P cat was constructed by inserting 96-bp domain of 174-bp enhancer at the Bglll site of peat. (j) pRrEn (wt/ dir.) SV40Pcat that contains the 174-bp orientation and separated (repetitive
element)
enhancer inserted at the Bgrll site of pcut. (k) pRrEn (wt/rev.) SV4OP cut that contains the 174-bp element in reverse from SV40 promoter by a 1.4-kb pBR322 vector fragment. (I) pRrEn130SV40Pcat that contains rat rDNA 130-bp RE
in front of cat gene. (m) pSV2cut
that contains
SV40 promoter
and enhancer
in front of cut gene. (B) Nucleotide
174-bp rat rDNA enhancer (upper strand) and its different functional domains. The ~0111 &-elements are shown in shaded sequences and the poll, 11, Ill consensus sequences are in the opposite orientation. The AP-2 site differs from the consensus of a G in the 6th postion
sequence
of
letters. Note that the Spl sequence by replacement
with A
activity in vivo (Table I), which is consistent with our in vitro data (Dixit et al., 1989). The level of stimulation observed in vivo (compare Table I, c and d) is not as high as observed in vitro. It is conceivable that this difference in the magnitude of stimulation observed in vivo vs. in vitro is due to the relatively high level of MT-I promoter activity in CHO cells. To test this possibility, we used another ~0111 promoter, SV40 promoter, under which cat expression is relatively low. The 174-bp ~011 enhancer was inserted in front of SV40 promoter in both orientation and at a different distance from the promoter and cat gene (see Fig lA, j and k) and transfected into CHO cells. As shown in Table I, (compare j with e) the CAT activity was approx. fivefold higher in CHO cells
transfected with the plasmid containing the spacer sequence than that with the control plasmid. Insertion of the 174-bp spacer sequence in reverse orientation (Table I, compare k with j) or at a distance of 1.4 kb from the promoter (Table I, compare k with j) did not significantly affect the enhancer efficiency of the 174-bp element. This enhancer element has, therefore, met all the criteria established for a typical ~0111 enhancer although the degree of stimulation of the ~0111 promoter by the ~011 enhancer element was not as high as that usually found for typical ~0111 enhancers (Serfling et al., 1985), but it is comparable with that observed with some ~0111 &-acting/enhancer elements (Ehrlich et al., 1988; Zhang et al., 1991). Since the spacer sequence can enhance both
274
4-
2k1234 oUU
3*c3 u ii&h
5 +l tsp
-167
6 +124
cat
vT_ 192 nt
32
P-labeled cat oligo
Fig. 2. Effect of 174-bp rat rDNA spacer element on poll promoter activity in vivo. Equimolar amounts of pRrPcat (lane 5) and pRrEPcut (lane 6) plasmids were transfected into CHO cells and transient expression of cat specific RNA was measured by primer extension. Transcripts derived from + 1 site of rat rDNA promoter is 192-nt long. Lanes l-4 show G, A, T and C dideoxy sequencing reactions of pRrEPcat plasmid using the 20-mer cat primer. Hatched and lined boxes indicate the rat rDNA sequence from - 167 to + 124 and cat gene, respectively. Solid thin and thick lines indicate the reverse transcriptase-directed elongation product and 20-mer cnt oligo as primer, respectively. The arrow indicates the primer extended product. Methods: CHO cells were used for transfection studies. Cells were split into 1 x lo5 per lGO-mm plate 24 h before transfection and grown in Dulbecco‘s modified Eagle media supplemented with 1% streptomycin/lO~ units penicillin per ml/lo% fetal bovine serum. Old medium was removed 3 h prior to transfection and cells were fed with fresh media. CsCl-purified plasmid DNA (10 pg) was used for transfection. After 48 h, cells were harvested and the total RNA was isolated from transfected CHO cells following a protocol developed by Stallcup and Washington (see Sambrook et al., 1989). To characterize the 5’ end of the rDNA-cat hybrid RNA, primer extension analysis was used. An oligo corresponding to one region of cat structural gene (Y-CAACGGTGGTATATCCAGTG) was used as the primer. RNA samples were denatured for 5 min at 80°C in a solution containing 10OmM NaCI/O.l mM EDTA/20mM TrisCl (pH 8.0)/ approx. 10 ng of 5’ end-labelled 20-mer cat oligo. The samples were then allowed to hybridize for 4 h at 45°C. Following hybridization, a mixture containing 10 mM Mgt&/4 mM DTT/SO mM TrisCl (pH 8.0)/0.55 mM of each dTTP, dATP, dCTP and dGTP/tO units of AMV reverse transcriptase was added to each tube. The samples were incubated at 37°C for 1 h and then incubated with RNase A (50 p&/ml) for additional 15 min, dried, resuspended in denaturing dye and electrophoresed on 7 M urea-6% polyacrylamide gel with sequencing markers.
~011 and ~0111 promoter activity, we analyzed this sequence for any potential ~0111 regulatory elements. Indeed, this rDNA region contains several poll1 tran-
scription-factor-binding domains, such as AP-Zlike sequence (a G replaced by A), Spl binding sequence and polymerase I, II and III common consensus sequence in opposite orientation (see Fig. iB). We investigated the nature of the domain(s) within the spacer sequence that is responsible for the enhancer activity in vivo. To address this issue, the enhancer element was dissected into three domains: 37 bp (NciI-KpnI), 43 bp (AvaI-AM) and 96 bp (HpaI-AvaI) (see also Dixit et al., 1987; 1989; Garg et al., 1989; Jacob et al., 1991). Individual fragments were inserted in front of SV40 promoter of pear plasmid and transfected into CHO cells (see Fig. 1A, f>g and if. Both 37-bp and 43-bp domains of the 174-bp element were capable of activating the SV40 promoter in cis, whereas the 96bp element remained silent (see Table I, compare i with f and g). It, therefore, appeared that the enhancer activity of 174-bp element in ~0111 resides within 37-bp and 43-bp domains. To test the combined effect of 37-bp and 43-bp domains on SV40 promoter activity, we placed 37-bp and 43-bp sequence in front of SV40 promoter (see Fig. lA, h) and transfected these constructs into CHO cells. The effect of combined 37-bp and 43-bp elements on SV40 promoter was almost the same as that of the 174-bp element (see Table I, compare j with h). To rule out the possibility that differential transfection efficiency of different constructs is responsible for different CAT activities, we repeated the transfection experiment four times along with a control plasmid (pSV2cat) for transfection efficiency. In each experiment, CAT activities from different constructs were normalized to the value of CAT activity obtained with control pSV2cnt plasmid and presented as relative CAT activity. Consistent results were obtained from experiment to experiment with only a slight vacation in the stimulatory effect which is typical for transient transfection assay. Further, the lack of stimulatory activity of the 96bp domain at the 5’ end of 174-bp element on SV40 promoter activity indicates that the enhancer activity of 37-bp, 43-bp and complete 174-bp elements on SV40 promoter is not due to differential transfection efficiency. These results indicate that certain &-sequences within the intergenic spacer of rat rDNA may act as potential regulatory elements of protein coding genes when they are placed in front of polI1 promoter. These elements may interact with a transcription factor(s) that is shared by both poff- and -II-directed genes. A recent study has shown that a 192-bp non-repeat yeast RNA ~011 enhancer also contains several ~0111enhancer motifs namely, GRF2/REBl/RBPl_binding site, ABFl-binding site and a thymidine-rich element. The GRF2/REBl/RBPl is known to modulate both poll and ~0111transcription whereas the T-rich sequence can only augment activity of a poll1 (CYC1) promoter (Larch et.al., 1990). Similarly, a PO1111upstream element consisting of
275
sequences from - 217 to - 315 of mouse U6-encoding gene can replace the upstream control element of a human U2-encoding gene (Bark et al., 1987), suggesting that specific upstream sequences can be functionally utilized by both RNA polymerases II and III. Moon and Krause (1991) have demonstrated the existence of a common upstream element for ~011,II and III genes that contributes to their transcriptional efficiency. (c) The 130-hp repetitive enhancer element (RE) of rat rDNA cannot modulate polII-mediated
transcription in
vivo
Finally, a comparison of the enhancer element studied here with the enhancer activity of the repeat sequences in rDNA is in order. Since the discovery that frog rDNA repeat sequence (60/81-bp element) can enhance ~011 transcription in vivo, attempts have been made to identify similar elements in other species. We identified such an element in the rat rDNA and showed that the intergenic spacer region that contains 113 repeats of the 130-bp sequence can stimulate ~011 transcription in vitro (Dixit et al., 1987). Recently, we demonstrated the enhancer activity of this repeat sequence in vivo as well (Ghosh et al., 1993). It was of interest to investigate whether this RE element can also stimulate ~0111 transcription. To test this possibility experimentally, three constructs, peat, pRrEn130SV40Pcat and pSV2cat (for details, see Fig. 1A, e, 1 and m) were used for transfection into CHO cells. CAT activity measurements after transfection showed that the ~011 RE had no effect on the ~0111 promoter activity (Table I, compare 1 with e). Based on these observations, it is reasonable to propose the existence of two types of enhancers at least for the rat rRNA gene, a polIspecific element comprising a repetitive sequence proximal to the core promoter and the other consisting of a single copy element far upstream from the core promoter, which appears to have a general enhancer function.
ACKNOWLEDGEMENTS
This work was supported by USPHS grant CA 31894 from the National Cancer Institute. We thank Barbara J. Worrell and Joanne Schwerman for assistance in the preparation of this manuscript.
REFERENCES Bark, C., Weller, P., Zabielski, J., Janson, L. and Pettersson, U.: A distant enhancer element is required for polymerase III transcription of a U6-RNA gene. Nature 328 (1987) 3566359. Bell, S.P., Jantzen, H.M. and Tjian, R.: Assembly of alternative multiprotein complexes directs rRNA promoter selectively. Genes Dev. 4(1990)943-954.
Dixit, A., Garg, L.C., Chao, W. and Jacob, S.T.: An enhancer element in the far upstream spacer region of rat ribosomal RNA gene. J. Biol. Chem. 262 (1987) 11616-11622. Dixit, A., Garg, L.C. and Jacob, ST.: A c&acting sequence within the rat ribosomal DNA enhancer region can modulate RNA polymerase II directed transcription of the metallothionein I gene in vitro. DNA 8 (1989) 311-320. Ehrlich, R., Maguire, J.E. and Singer, D.S.: Identification of negative and positive regulatory elements associated with a class I major histocompatibility complex gene. Mol. Cell. Biol. 8 (1988) 695-703. Garg, L.C., Dixit, A. and Jacob, S.T.: A 37-bp pair element in the far upstream spacer region can enhance transcription of rat rDNA in vitro and can bind to the core promoter binding factor(s). J. Biol. Chem. 264 (1989) 220-224. Gorman, C.M., Moffat, L.F. and Howard, B.H.: Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol. Cell. Biol. 2 (1982) 10441051. Ghosh, A.K., Hoff, C.M. and Jacob, S.T.: Characterization of the 130-bp repeat enhancer element of the rat ribosomal gene: Functional interaction with transcription factor E,BF. Gene 125 (1993) 217-222. Hoff, C.M. and Jacob, S.T.: Characterization of the factor EiBF from a rat hepatoma that modulates ribosomal RNA gene transcription and its relationship to the human Ku autoantigen. Biochem. Biophys. Res. Commun. 190 (1993) 747-753. Hoff, C.M., Ghosh, A.K., Prabhakar, B.S. and Jacob, S.T.: E,BF, Ku-related protein, is a positive regulator of RNA polymerase I transcription initiation. Proc. Natl. Acad. Sci. USA 91 (1994) 762-766. Jacob, S.T., Zhang, J., Garg, L.C. and Book, C.B.: Multiple functional enhancer motifs of rat ribosomal gene. Mol. Cell. Biochem. 104 (1991) 155-162. Johnson, S.P. and Warner, J.R.: Unusual enhancer function in yeast rRNA transcription. Mol. Cell. Biol. 9 (1989) 498664993. Learned, R.M., Learned, T.K., Haltiner, M.M. and Tjian, R.T.: Human rRNA transcription is modulated by the coordinate binding of two factors to an upstream control element. Cell 45 (1986) 847-857. Larch, Y., Lue, N.F. and Kornberg, R.D.: Interchangeable RNA polymerase I and II enhancers. Proc. Natl. Acad. Sci. USA 87 (1990) 8202-8206. Moon, I.L. and Krause, M.O.: Common RNA polymerase I, II, III upstream elements in mouse 7SK gene locus revealed by the inverse polymerase chain reaction. DNA Cell Biol. 10 (1991) 23-32. Moss, T.: A transcriptional function for the repetitive ribosomal spacer in Xenopus heuis. Nature 302 (1983) 223-228. Paule, M.R.: Polymerase I transcription termination and processing: a meeting review. Gene Expr. 3 (1993) l-9. Pikaard, C.S., Pape, L., Henderson, S., Ryan, K., Pwalman, M., Lopata, M., Reeder, R.H. and SollnerWebb, B.: Enhancers for polymerase I in mouse ribosomal DNA. Mol. Cell. Biol. 10 (1990) 4816-4825. Reeder, R.H.: Enhancers and ribosomal gene spacers. Cell 38 (1984) 3499351. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual, Vol. 1. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989, pp. 7.10-7.11. Serfling E., Jasin M. and Schaffner, W.: Enhancer and eukaryotic gene transcription. Trends Genet. 1 (1985) 224-230. Zhang, J. and Jacob, S.T.: Purification and characterization of a novel factor which stimulates rat ribosomal gene transcription in vitro by interacting with enhancer and core promoter elements. Mol. Cell. Biol. 10 (1990) 5177-5286. Zhang, X.-K., Dong, J.-M. and Chiu, J.-F.: Regulation of a-fetoprotein gene expression by antagonism between AP-1 and the glucocorticoid receptor at their overlapping binding site. J. Biol. Chem. 266 (1991) 8248-8254.
271
0378-l 119/94/$07.00
GENE 07769
Effects of repetitive and non-repetitive rat rDNA enhancer elements on in vivo transcription by RNA polymerases I and II (Ribosomal RNA; transfection; primer extension; CAT assay)
Asish K. Ghosh, Milko Kermekchiev
and Samson T. Jacob
Department of Pharmacology and Molecular Biology, The Chicago Medical School, North Chicago, IL 60064, USA Received by R. Padmanabhan:
14 July 1993; Revised/Accepted:
21 October/3
November
1993; Received at publishers:
14 December
1993
SUMMARY
Previous study has demonstrated that a far upstream 174-bp spacer sequence of the rat rRNA-encoding (rDNA) gene can function as an enhancer in vitro in an orientation- and distance-independent manner [Dixit et al., J. Biol. Chem. 262 (1987) 11616-116221. To demonstrate that this element can also function in vivo, two rat rDNA-cat plasmids, one with the 174-bp element and the other without this sequence, were constructed and transfected into CHO cells. Primer extension analysis of the transcripts produced after transfection showed that transcription initiation occurred at the + 1 site of the rDNA. The 174-bp sequence stimulated the rat poll promoter activity in cis 44S-fold over the control (with the promoter alone). This RNA polymerase (~011) enhancer also stimulated the mouse metallothionein-I (MT-Z) and SV40 promoter activities in vivo, irrespective of its distance and orientation. Further dissection of the 174-bp element revealed that the stimulatory activity on the RNA polymerase II (~0111) promoter resides within the 37-bp and 43-bp domains at the 3’ end of the 174-bp element. Unlike this spacer enhancer, the 130-bp repeat element (RE) proximal to the rat promoter [Ghosh et al., Gene 125 (1993) 217-2221 was unable to modulate the ~0111promoter activity in vivo. These data show that while the non-repetitive enhancer sequence of rat rDNA is interchangeable for the ~011 and ~0111 promoters, the RE is polI-specific.
INTRODUCTION
The basal rRNA-encoding gene (rDNA) transcription requires a species-specific factor (TFID/SLl ) and RNA polymerase I (see Paule, 1993). Maximal transcription of rDNA, however, is dependent upon other c&acting upstream spacer sequences and trans-acting factors. Tjian and co-workers identified in human cells an upstream control element (UCE) and characterized a protein factor Correspondence to: Dr. S.T. Jacob,
Department
of Pharmacology
Molecular Biology. The Chicago Medical School, 3333 Green Road, North Chicago, IL 60064, USA. Tel. (l-708) 578-3271; (l-708) 578-3268.
and Bay Fax
Abbreviations: AMV, avian myeloblastosis virus; bp, base pair(s); CAT, chloramphenicol acetyltransferase; cat, gene encoding CAT, CHO, Chinese hamster ovary; d, deoxyribo; E,BF, enhancer-l-binding SSDI 0378-1119(93)E0765-6
(UBF, meaning upstream binding factor) that interacts with both UCE and the downstream core promoter sequences (Learned et al., 1986). In the human system, UBF is essential for SLI interaction with the promoter (Bell et al., 1990). In addition to UCE, other spacer elements such as enhancers have been identified in the spacer regions of Xenopus (Moss, 1983; Reeder, 1984), rat (Dixit et al., 1987; Ghosh et al., 1993), yeast (Johnson and Warner, 1989) and more recently, in mouse (Pikaard factor; kb, kilobase gene; nt, nucleotide(s);
or 1000 bp; MT-I, metallothionein-I encoding oligo, oligodeoxyribonucleotide; pol, RNA poly-
merase; PolIk, Klenow (large) fragment of E. coli DNA polymerase I; r, ribosomal; rDNA, DNA (gene) coding for rRNA; RE, repeat element(s); SDS, sodium dodecyl sulfate; SLl, species-specificity factor for polI transcription; SV40, simian virus 40; tsp, transcription start point(s); UBF, upstream binding factor: UCE, upstream control element(s).
272 et al., 1990) rDNA. We have identified three spacer elements within the 4%kb region upstream from the rat rDNA transcription start point (tsp), which can stimulate rDNA gene transcription in vitro (Dixit et al., 1987). One element that is located between the -2183-kb and - 2357-kb positions has been fully characterized in vitro (Dixit et al., 1987). This enhancer sequence could function independent of orientation and distance and has, therefore, met all the criteria established for a poll1 enhancer. Subsequent studies have demonstrated that the 174-bp enhancer element consists of multiple motifs (Garg et al., 1989; Jacob et al., 1991). We have characterized a protein (E,BF) that interacts with the promoter sequence, 174-bp enhancer (Zhang and Jacob, 1990) and 130-bp repeat enhancer sequence (Ghosh et al., 1993) of rat rDNA and stimulates transcription from the rRNA promoter (Zhang and Jacob, 1990 and Ghosh et al., 1993). This protein is related to the human Ku autoantigen with respect to immunological and certain structural characteristics (Hoff and Jacob, 1993). Recent study has demonstrated that E,BF is required for the basal ~011 transcription and that it acts primarily in the preinitiation complex formation (Hoff et al., 1994). Previous study in our laboratory demonstrated that the 174-bp rat rDNA spacer element can stimulate polII promoter (MTI) activity in vitro (Dixit et al., 1989). Similar results were also obtained in yeast where a ~011enhancer element was found to stimulate ~0111promoter activity (Larch et al., 1990). The purpose of present study was to determine whether (a) the 174”bp rat rDNA spacer element can function in vivo; (b) it can also modulate transcription from poll1 promoters; and (c) unlike this &-acting sequence, the 130-bp RE recently characterized in our laboratory (Ghosh et al., 1993) can function only with ~011 promoter.
EXPERIMENTAL AND DISCUSSION
(a) The 174-hp rat rDNA enhancer element stimulates pol1 promoter activity in vivo To investigate whether the 174-bp spacer enhancer element (located between bp positions -2357 and -2183 with respect to tsp) can also enhance the rat rDNA promoter activity in vivo, we constructed two rDNA-cat plasmids, one containing cat gene and rat ~011 promoter region, (pRrPcat) and another containing the 174-bp sequence in addition to the promoter and cat gene (pRrEPcat) (see Fig. lA, a and b) and transfected them into CHO cells. Analysis of the RNA from the transfected cells by primer extension revealed a 192-nt product that corresponds to a correctly initiated (+ I tsp) transcript (Fig. 2, lanes 5 and 6). When the transfected construct
contained the enhancer element (pRrEPcu~), transcription was significantly stimulated (Fig. 2, compare lanes 5 and 6). Densitometric scanning of the transcripts showed a 4-5-fold increase in the amount of RNA initiated at the correct site in the presence of the enhancer. This result suggested that the 174”bp region of rat rDNA spacer can augment ~011 promoter activity in vivo. (b) The 174-bp rDNA enhancer can stimulate poll1 promoter activity in vivo Next, we investigated whether the 174-bp spacer sequence of rat rDNA can also stimulate ~0111 promoter activity in vivo. Previous study in our laboratory has shown that this upstream element can modulate mouse MT-Z gene promoter activity in vitro (Dixit et al., 1989). To test whether the poll enhancer functions with ~0111 promoters also in vivo, CHO cells were transfected with two plasmids, one that contains mouse MT-I gene promoter (- 148 to +67) ligated to the cut gene and the other consisting of the 174-bp rDNA fragment placed in front of the mouse MT-f promoter and cat gene (see Fig. lA, c and d). Analysis of the CAT activity in CHO cells transfected with these constructs showed that the 174-bp sequence can also augment the ~0111 promoter TABLE I Relative CAT activity in CHO cells transfected with different plasmid constructs Plasmids used in transfectiot?
c. pMTlcat d. pRrEnMTlcu~ e. peat promoter f. pRrEn,SV4OPcat g. pRrEn,SV40Pcat h. pRrEn, +,SV40Pcat i. pRrEn,SV40Pcat j. pRrEn(wt./dir.)SV40Pcat k. pRrEn(wt./rev.)SVllOPcnt I. pRrEn130SV40Pcar m. pSV2cat
Relative CAT activity” (% of CAT conversion relative to that of pSV2cat) 24 51 21 20 30 10 3.5 33 10 l1100)
aDesignations c-m correspond to the schematic representation of constructs c-m in Fig. 1A. CsCl-gradient-purifi~ plasmid DNA (10 ug) was used for transfection as described by Gorman et al. (1982). Different plasmids (for details, see Fig. lA, c-m) were transfected into CHO cells and after 48 h of transfection, cells were harvested and lysate was made by three successive freezing and thawing. Protein concentrations were determined by BCA protein reagent assay kit (Pierce). bCAT activities from different constructs were normalized to the value of CAT activity obtained with control pSV2cat plasmid and presented as relative CAT activity. Reaction mixture contained 0.2 pCi [*“C]chloramphenicol/5 mM acetyl CoA/0.25 M Tris.Cl (pH 7S)/cell extract (100-200 ug total protein). Samples were incubated at 37°C for 30 to 60 min and then analyzed by thin-layer chromatography as described (Gorman et al., 1982).
273
a.
A.
-167
+1
cat
+124
i.
b.
Enhancer
k. Enhancer
Enhancer
rDNA promoter
cat
MTI Promoter
I.
C.
Enhancer sv40
cat
SV40 Promoter
m. Enhancer
d. Enhancer SV40 Promoter
e.
cat
B. f.
HpaI
AACCGATTCGAGAGTGCATGTCCTTTCACTTAGAGGTGGCCTGT :2351 AP-2 site C CTlGA S T;GG
TTTCACCGCTCCGTCCTACTTTCCTTTTCTGGGT,G
SV40 Promoter
II.
Ava I TTGGCC&GAGTCGCCATCCCGCGTTCACTGTGTTCC
cat
,,%&
&iI Poly. Consensus sequence ACAACCGGGACCtXYWX6CCATTCGGGAGAAGTGGTGGGTA~
i.
KP+ -2183
Fig. 1. Plasmids, enhancer elements and their sequences. (A) Schematic representation of the recombinant plasmids used: (a) pRrPcat that contains Sall-Hind111 rat rDNA fragment (nt - 167 to + 124) in front of cat gene of pSVOcat but lacks SV40 promoter and enhancer. (b) pRrEPcat that contains
Hpal-KpnI
(-2357
to -2183)
fragment
of rat rDNA
ligated
to WI-Hind111
(- 167 to + 124) fragment
of the promoter
and cat gene of
pSVOcat. (c) pMTlcat that contains mouse MT-I gene promoter (- 148 to +67) in front of cat gene. (d) pRrEnMTlPcat that contains rat rDNA 174-bp enhancer, mouse MT-l gene promoter (- 148 to + 67) and cat gene. The 174-bp element is separated from the - 148-bp position of MT-l promoter
by 383-bp
rat rDNA
spacer
sequence.
(e) peat contains
SV40 promoter
and cur gene. (f) pRrEn,SV40Pcat
was constructed
one copy of 37-bp domain of the 174-bp enhancer at the Bgrll site of peat. (g) pRrEn,SV40Pcat plasmid was made by inserting 174-bp enhancer at the Bgrll site of peat. (b) pRrEn, + z SV4OP cat was constructed by inserting a 37-bp and 43-bp containing
by inserting
43-bp domain of the fragment in front of
SV40 promoter of pcut promoter plasmid in direct orientation and separated from SV40 promoter by a 383-bp rat rDNA spacer fragment (SallKpnl). (i) pRrEn,SV40P cat was constructed by inserting 96-bp domain of 174-bp enhancer at the Bglll site of peat. (j) pRrEn (wt/ dir.) SV40Pcat that contains the 174-bp orientation and separated (repetitive
element)
enhancer inserted at the Bgrll site of pcut. (k) pRrEn (wt/rev.) SV4OP cut that contains the 174-bp element in reverse from SV40 promoter by a 1.4-kb pBR322 vector fragment. (I) pRrEn130SV40Pcat that contains rat rDNA 130-bp RE
in front of cat gene. (m) pSV2cut
that contains
SV40 promoter
and enhancer
in front of cut gene. (B) Nucleotide
174-bp rat rDNA enhancer (upper strand) and its different functional domains. The ~0111 &-elements are shown in shaded sequences and the poll, 11, Ill consensus sequences are in the opposite orientation. The AP-2 site differs from the consensus of a G in the 6th postion
sequence
of
letters. Note that the Spl sequence by replacement
with A
activity in vivo (Table I), which is consistent with our in vitro data (Dixit et al., 1989). The level of stimulation observed in vivo (compare Table I, c and d) is not as high as observed in vitro. It is conceivable that this difference in the magnitude of stimulation observed in vivo vs. in vitro is due to the relatively high level of MT-I promoter activity in CHO cells. To test this possibility, we used another ~0111 promoter, SV40 promoter, under which cat expression is relatively low. The 174-bp ~011 enhancer was inserted in front of SV40 promoter in both orientation and at a different distance from the promoter and cat gene (see Fig lA, j and k) and transfected into CHO cells. As shown in Table I, (compare j with e) the CAT activity was approx. fivefold higher in CHO cells
transfected with the plasmid containing the spacer sequence than that with the control plasmid. Insertion of the 174-bp spacer sequence in reverse orientation (Table I, compare k with j) or at a distance of 1.4 kb from the promoter (Table I, compare k with j) did not significantly affect the enhancer efficiency of the 174-bp element. This enhancer element has, therefore, met all the criteria established for a typical ~0111 enhancer although the degree of stimulation of the ~0111 promoter by the ~011 enhancer element was not as high as that usually found for typical ~0111 enhancers (Serfling et al., 1985), but it is comparable with that observed with some ~0111 &-acting/enhancer elements (Ehrlich et al., 1988; Zhang et al., 1991). Since the spacer sequence can enhance both
274
4-
2k1234 oUU
3*c3 u ii&h
5 +l tsp
-167
6 +124
cat
vT_ 192 nt
32
P-labeled cat oligo
Fig. 2. Effect of 174-bp rat rDNA spacer element on poll promoter activity in vivo. Equimolar amounts of pRrPcat (lane 5) and pRrEPcut (lane 6) plasmids were transfected into CHO cells and transient expression of cat specific RNA was measured by primer extension. Transcripts derived from + 1 site of rat rDNA promoter is 192-nt long. Lanes l-4 show G, A, T and C dideoxy sequencing reactions of pRrEPcat plasmid using the 20-mer cat primer. Hatched and lined boxes indicate the rat rDNA sequence from - 167 to + 124 and cat gene, respectively. Solid thin and thick lines indicate the reverse transcriptase-directed elongation product and 20-mer cnt oligo as primer, respectively. The arrow indicates the primer extended product. Methods: CHO cells were used for transfection studies. Cells were split into 1 x lo5 per lGO-mm plate 24 h before transfection and grown in Dulbecco‘s modified Eagle media supplemented with 1% streptomycin/lO~ units penicillin per ml/lo% fetal bovine serum. Old medium was removed 3 h prior to transfection and cells were fed with fresh media. CsCl-purified plasmid DNA (10 pg) was used for transfection. After 48 h, cells were harvested and the total RNA was isolated from transfected CHO cells following a protocol developed by Stallcup and Washington (see Sambrook et al., 1989). To characterize the 5’ end of the rDNA-cat hybrid RNA, primer extension analysis was used. An oligo corresponding to one region of cat structural gene (Y-CAACGGTGGTATATCCAGTG) was used as the primer. RNA samples were denatured for 5 min at 80°C in a solution containing 10OmM NaCI/O.l mM EDTA/20mM TrisCl (pH 8.0)/ approx. 10 ng of 5’ end-labelled 20-mer cat oligo. The samples were then allowed to hybridize for 4 h at 45°C. Following hybridization, a mixture containing 10 mM Mgt&/4 mM DTT/SO mM TrisCl (pH 8.0)/0.55 mM of each dTTP, dATP, dCTP and dGTP/tO units of AMV reverse transcriptase was added to each tube. The samples were incubated at 37°C for 1 h and then incubated with RNase A (50 p&/ml) for additional 15 min, dried, resuspended in denaturing dye and electrophoresed on 7 M urea-6% polyacrylamide gel with sequencing markers.
~011 and ~0111 promoter activity, we analyzed this sequence for any potential ~0111 regulatory elements. Indeed, this rDNA region contains several poll1 tran-
scription-factor-binding domains, such as AP-Zlike sequence (a G replaced by A), Spl binding sequence and polymerase I, II and III common consensus sequence in opposite orientation (see Fig. iB). We investigated the nature of the domain(s) within the spacer sequence that is responsible for the enhancer activity in vivo. To address this issue, the enhancer element was dissected into three domains: 37 bp (NciI-KpnI), 43 bp (AvaI-AM) and 96 bp (HpaI-AvaI) (see also Dixit et al., 1987; 1989; Garg et al., 1989; Jacob et al., 1991). Individual fragments were inserted in front of SV40 promoter of pear plasmid and transfected into CHO cells (see Fig. 1A, f>g and if. Both 37-bp and 43-bp domains of the 174-bp element were capable of activating the SV40 promoter in cis, whereas the 96bp element remained silent (see Table I, compare i with f and g). It, therefore, appeared that the enhancer activity of 174-bp element in ~0111 resides within 37-bp and 43-bp domains. To test the combined effect of 37-bp and 43-bp domains on SV40 promoter activity, we placed 37-bp and 43-bp sequence in front of SV40 promoter (see Fig. lA, h) and transfected these constructs into CHO cells. The effect of combined 37-bp and 43-bp elements on SV40 promoter was almost the same as that of the 174-bp element (see Table I, compare j with h). To rule out the possibility that differential transfection efficiency of different constructs is responsible for different CAT activities, we repeated the transfection experiment four times along with a control plasmid (pSV2cat) for transfection efficiency. In each experiment, CAT activities from different constructs were normalized to the value of CAT activity obtained with control pSV2cnt plasmid and presented as relative CAT activity. Consistent results were obtained from experiment to experiment with only a slight vacation in the stimulatory effect which is typical for transient transfection assay. Further, the lack of stimulatory activity of the 96bp domain at the 5’ end of 174-bp element on SV40 promoter activity indicates that the enhancer activity of 37-bp, 43-bp and complete 174-bp elements on SV40 promoter is not due to differential transfection efficiency. These results indicate that certain &-sequences within the intergenic spacer of rat rDNA may act as potential regulatory elements of protein coding genes when they are placed in front of polI1 promoter. These elements may interact with a transcription factor(s) that is shared by both poff- and -II-directed genes. A recent study has shown that a 192-bp non-repeat yeast RNA ~011 enhancer also contains several ~0111enhancer motifs namely, GRF2/REBl/RBPl_binding site, ABFl-binding site and a thymidine-rich element. The GRF2/REBl/RBPl is known to modulate both poll and ~0111transcription whereas the T-rich sequence can only augment activity of a poll1 (CYC1) promoter (Larch et.al., 1990). Similarly, a PO1111upstream element consisting of
275
sequences from - 217 to - 315 of mouse U6-encoding gene can replace the upstream control element of a human U2-encoding gene (Bark et al., 1987), suggesting that specific upstream sequences can be functionally utilized by both RNA polymerases II and III. Moon and Krause (1991) have demonstrated the existence of a common upstream element for ~011,II and III genes that contributes to their transcriptional efficiency. (c) The 130-hp repetitive enhancer element (RE) of rat rDNA cannot modulate polII-mediated
transcription in
vivo
Finally, a comparison of the enhancer element studied here with the enhancer activity of the repeat sequences in rDNA is in order. Since the discovery that frog rDNA repeat sequence (60/81-bp element) can enhance ~011 transcription in vivo, attempts have been made to identify similar elements in other species. We identified such an element in the rat rDNA and showed that the intergenic spacer region that contains 113 repeats of the 130-bp sequence can stimulate ~011 transcription in vitro (Dixit et al., 1987). Recently, we demonstrated the enhancer activity of this repeat sequence in vivo as well (Ghosh et al., 1993). It was of interest to investigate whether this RE element can also stimulate ~0111 transcription. To test this possibility experimentally, three constructs, peat, pRrEn130SV40Pcat and pSV2cat (for details, see Fig. 1A, e, 1 and m) were used for transfection into CHO cells. CAT activity measurements after transfection showed that the ~011 RE had no effect on the ~0111 promoter activity (Table I, compare 1 with e). Based on these observations, it is reasonable to propose the existence of two types of enhancers at least for the rat rRNA gene, a polIspecific element comprising a repetitive sequence proximal to the core promoter and the other consisting of a single copy element far upstream from the core promoter, which appears to have a general enhancer function.
ACKNOWLEDGEMENTS
This work was supported by USPHS grant CA 31894 from the National Cancer Institute. We thank Barbara J. Worrell and Joanne Schwerman for assistance in the preparation of this manuscript.
REFERENCES Bark, C., Weller, P., Zabielski, J., Janson, L. and Pettersson, U.: A distant enhancer element is required for polymerase III transcription of a U6-RNA gene. Nature 328 (1987) 3566359. Bell, S.P., Jantzen, H.M. and Tjian, R.: Assembly of alternative multiprotein complexes directs rRNA promoter selectively. Genes Dev. 4(1990)943-954.
Dixit, A., Garg, L.C., Chao, W. and Jacob, S.T.: An enhancer element in the far upstream spacer region of rat ribosomal RNA gene. J. Biol. Chem. 262 (1987) 11616-11622. Dixit, A., Garg, L.C. and Jacob, ST.: A c&acting sequence within the rat ribosomal DNA enhancer region can modulate RNA polymerase II directed transcription of the metallothionein I gene in vitro. DNA 8 (1989) 311-320. Ehrlich, R., Maguire, J.E. and Singer, D.S.: Identification of negative and positive regulatory elements associated with a class I major histocompatibility complex gene. Mol. Cell. Biol. 8 (1988) 695-703. Garg, L.C., Dixit, A. and Jacob, S.T.: A 37-bp pair element in the far upstream spacer region can enhance transcription of rat rDNA in vitro and can bind to the core promoter binding factor(s). J. Biol. Chem. 264 (1989) 220-224. Gorman, C.M., Moffat, L.F. and Howard, B.H.: Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol. Cell. Biol. 2 (1982) 10441051. Ghosh, A.K., Hoff, C.M. and Jacob, S.T.: Characterization of the 130-bp repeat enhancer element of the rat ribosomal gene: Functional interaction with transcription factor E,BF. Gene 125 (1993) 217-222. Hoff, C.M. and Jacob, S.T.: Characterization of the factor EiBF from a rat hepatoma that modulates ribosomal RNA gene transcription and its relationship to the human Ku autoantigen. Biochem. Biophys. Res. Commun. 190 (1993) 747-753. Hoff, C.M., Ghosh, A.K., Prabhakar, B.S. and Jacob, S.T.: E,BF, Ku-related protein, is a positive regulator of RNA polymerase I transcription initiation. Proc. Natl. Acad. Sci. USA 91 (1994) 762-766. Jacob, S.T., Zhang, J., Garg, L.C. and Book, C.B.: Multiple functional enhancer motifs of rat ribosomal gene. Mol. Cell. Biochem. 104 (1991) 155-162. Johnson, S.P. and Warner, J.R.: Unusual enhancer function in yeast rRNA transcription. Mol. Cell. Biol. 9 (1989) 498664993. Learned, R.M., Learned, T.K., Haltiner, M.M. and Tjian, R.T.: Human rRNA transcription is modulated by the coordinate binding of two factors to an upstream control element. Cell 45 (1986) 847-857. Larch, Y., Lue, N.F. and Kornberg, R.D.: Interchangeable RNA polymerase I and II enhancers. Proc. Natl. Acad. Sci. USA 87 (1990) 8202-8206. Moon, I.L. and Krause, M.O.: Common RNA polymerase I, II, III upstream elements in mouse 7SK gene locus revealed by the inverse polymerase chain reaction. DNA Cell Biol. 10 (1991) 23-32. Moss, T.: A transcriptional function for the repetitive ribosomal spacer in Xenopus heuis. Nature 302 (1983) 223-228. Paule, M.R.: Polymerase I transcription termination and processing: a meeting review. Gene Expr. 3 (1993) l-9. Pikaard, C.S., Pape, L., Henderson, S., Ryan, K., Pwalman, M., Lopata, M., Reeder, R.H. and SollnerWebb, B.: Enhancers for polymerase I in mouse ribosomal DNA. Mol. Cell. Biol. 10 (1990) 4816-4825. Reeder, R.H.: Enhancers and ribosomal gene spacers. Cell 38 (1984) 3499351. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual, Vol. 1. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989, pp. 7.10-7.11. Serfling E., Jasin M. and Schaffner, W.: Enhancer and eukaryotic gene transcription. Trends Genet. 1 (1985) 224-230. Zhang, J. and Jacob, S.T.: Purification and characterization of a novel factor which stimulates rat ribosomal gene transcription in vitro by interacting with enhancer and core promoter elements. Mol. Cell. Biol. 10 (1990) 5177-5286. Zhang, X.-K., Dong, J.-M. and Chiu, J.-F.: Regulation of a-fetoprotein gene expression by antagonism between AP-1 and the glucocorticoid receptor at their overlapping binding site. J. Biol. Chem. 266 (1991) 8248-8254.