Jun 21, 1988 - enhancer and the x light chain gene promoter individually by ... which are activated at a specific stage during differentiation ..... Harbor, NY.
The EMBO Journal vol.7 no.10 pp.3093-3098, 1988
Both immunoglobulin promoter and enhancer sequences are targets for suppression in myeloma -fibroblast hybrid cells Steffen Junker, Viggo Nielsen, Patrick Matthias' and Didier Picard' 2 Institute of Human Genetics, University of Aarhus, DK-8000 Aarhus C, Denmark, 'Institut fur Molekularbiologie II der UniversitSt Zurich, Honggerberg, CH-8093 Zurich, Switzerland and 2Department of Biochemistry and Biophysics, University of California Medical Center, San Francisco, CA 94143-0448, USA Communicated by W.Schaffner
When immunoglobulin (Ig)-producing B cells are fused with fibroblastic cells, expression of Igs is suppressed by a mechanism that selectively abolishes transcription of Ig genes. The suppression is also maintained in proliferating hybrids. We have used gene transfer followed by cell fusion to study this phenomenon further. Here we report that expression of a rearranged Ig heavy chain gene, stably integrated into a myeloma genome, is completely suppressed upon fusion with fibroblasts by a mechanism that is equally active on the endogenous myeloma X light chain gene. To define regulatory sequences within the Ig transcriptional unit that are involved in this down-regulation, we examined the transcriptional contributions of the IgH chain gene enhancer and the x light chain gene promoter individually by linking them to a heterologous reporter gene. Mouse myeloma cells were stably transformed with such test constructs and subsequently fused with mouse fibroblasts. To avoid any significant loss of chromosomes, hybrid cells were isolated shortly after fusion by fluorescence-activated cell sorting, and proliferating hybrids were harvested within 2-3 weeks. On the basis of RNase protection mapping of cytoplasmic RNA, and of nuclear run-on assays we showed that both the x light chain promoter and the IgH chain enhancer contain regulatory information that is made redundant or is suppressed in the hybrid environment. Key words: cell h ybrid/immunoglobulin gene/enhancer/
promoter/suppression
Introduction Tissue-specific and developmentally regulated gene expression appears to be mediated by the interaction of specific trans-acting factors with cis-acting DNA sequences (for a review see McKnight and Tjian, 1986; Maniatis et al., 1987). One of the best characterized examples of tissuespecific gene expression is that of immunoglobulin (Ig) genes which are activated at a specific stage during differentiation of lymphoid cells, and whose expression is restricted to B cells. Gene transfer experiments have led to the identification of at least two distinct sequence elements required for Ig gene regulation: the enhancer and the promoter. In the case of the IgH gene the contribution of some intragenic sequences (Grosschedl and Baltimore, 1985) that presumably affect (IRL Press Limited, Oxford, England
transcript stability (Gerster et al., 1986; Kelley and Perry, 1986) have also been reported. The enhancer element resides in the large intron between the variable and constant regions of both heavy (Banerji et al., 1983; Gillies et al., 1983; Neuberger, 1983) and x light chain genes (Queen and Baltimore, 1983; Picard and Schaffner, 1984a). In transfection experiments these enhancer elements can increase the transcription of linked genes independently of their position or orientation relative to the promoter (for a review see Serfling et al., 1985). Ig enhancers are strictly tissuespecific and are active only in cells of the lymphoid lineage (reviewed in Matthias et al., 1987). The Ig enhancers consist of a modular array of sequence elements that are binding sites for trans-acting factors, presumably mediating the changes in transcriptional activity (Augereau and Chambon, 1986; Sen and Baltimore, 1986a; for a review see Maniatis et al., 1987). Some of these elements activate transcription in response to physiological signals that are cell-type-specific, while others may be involved in repressing transcription in the improper tissue (Kadesch et al., 1986; Wasylyk and Wasylyk, 1986; Imler et al., 1987; Kiledjian et al., 1988; Weinberger et al., 1988). Recent studies have shown that the promoter sequences of x light chain (Queen and Stafford, 1984; Picard and Schaffner, 1985) and IgH genes (Grosschedl and Baltimore, 1985; Mason et al., 1985) also display cell-type specificity and bind tissue-specific factors (Staudt et al., 1986; Landolfi et al., 1987; Scheidereit et al., 1987). Furthermore, enhancers and promoters appear to interact specifically as indicated by the observation that these two elements act synergistically to control the level of Ig expression (Garcia et al., 1986). Clearly, to comprehend the tissue-specificity of Ig gene expression it is necessary to understand not only the activity of Ig genes in B cells, but also their inactivity in other cell types. One approach of potential use to further elucidate the latter is that of somatic cell hybridization. By fusing dissimilar cell types the existence of trans-acting factors capable of activating or repressing the expression of several tissue-specific genes has been demonstrated (for a review see Ringertz and Savage, 1976). Early experiments showed that expression of, for example, Ig genes in lymphoid cells can be down-regulated upon fusion with fibroblasts (Periman, 1970; Coffino et al., 1971; Zeuthen and Nilsson, 1976; Junker, 1982). Suppression is rapid and is mediated through a mechanism that selectively abolishes transcription of Ig genes (Nishikura et al., 1984; Junker and Pedersen, 1985; Greenberg et al., 1987; Junker et al., 1988). Undoubtedly, the identification and characterization of the target sequences and factors involved in this suppression are crucial steps towards understanding the underlying mechanism. We have therefore combined gene transfer with cell fusion to test the x light chain gene promoter and the IgH gene enhancer individually for their contributions to the down-
3093
S.Junker et at.
regulation of Ig genes in a hybrid environment. We chose to fuse cloned transformants of mouse myeloma cells with mouse fibroblasts to be able to study the transcriptional activity of both recombinant and endogenous genes before and after cell fusion. Furthermore, since chromosome losses can seriously hamper genetic studies on hybrid cells, we applied a technique based on fluorescence-activated cell sorting (Junker, 1982) to isolate hybrids in quantities sufficiently large to be analysed as early as 2-3 weeks after fusion, i.e. before chromosome segregation was significant. Here we report that expression of an exogenously introduced IgH gene, stably integrated into a myeloma genome, gets suppressed in a tissue-specific manner upon fusion with fibroblasts. Moreover, we show that both the promoter and the enhancer, linked to a heterologous reporter gene, are targets for that down-regulation. Thus the suppression mechanism seems to act through the same tissuespecific regulatory sequences identified hitherto in Ig genes.
Fig. 1. Expression of experimentally introduced and of endogenous Ig genes in myeloma cells is selectively abolished upon fusion with fibroblasts. Slot-blot hybridization analysis of cytoplasmic RNA prepared from J558-JG5, cl. ID, and their 12-day-old hybrids. The filters were hybridized with probes specific for either -y2bheavy chain, X light chain or tk transcripts. To compensate for the relative concentration of specific transcripts per cell 5 jig of RNA from parental cells, and 10 jtg of RNA from hybrid cells were applied per slot.
Results Expression of a newly introduced IgH gene is down-regulated upon fusion with fibroblasts First we determined whether expression of a rearranged Ig gene, experimentally introduced into myeloma cells, would be regulated in the same manner as the endogenous Ig genes after fusion with fibroblasts. For that purpose we fused mouse cl. ID fibroblasts with the mouse myeloma cell line J558-JG5 which had been derived from the X light chain gene expressing myeloma J558L by stable transformation with a functionally rearranged IgH (7Y2b) gene (Gillies et al., 1983). Heterokaryons containing one genome equivalent from each parent cells (is:ls) were isolated using the FITCFACS technique (Junker, 1982). Twelve days after fusion, cytoplasmic RNA was prepared from proliferating hybrids and analysed by a slot-blot hybridization assay. Since the recipient J558L cells have lost their ability to express the endogenous IgH gene, but continue to synthesize X light chains, steady-state levels of X transcripts before and after fusion serve as an internal control. In the hybrids expression of both the endogenous X light chain gene and the introduced Cy2b heavy chain gene was completely abolished. In contrast, the level of transcripts of the housekeeping gene coding for thymidine kinase, in which cl. ID cells are deficient, was virtually unaffected upon fusion (Figure 1). These findings demonstrate that, apparently independently of its chromosomal location, the intact, transfected IgH gene contains all the necessary regulatory sequences to be subjected to down-regulation in a tissue-specific and selective manner upon fusion with fibroblasts.
Assay system to map target sequences To identify sequence elements in the transcriptional unit of Ig genes involved in that down-regulation, we tested the enhancer and promoter elements individually for their contribution to Ig gene suppression in hybrid cells. The following four constructs (Figure 2) were tested: the two control recombinants, p,BGN and pflGSN, contained the rabbit 13-globin gene without or with the SV40 enhancer inserted downstream of it, respectively. Recombinant pflGHN was obtained by inserting the 1 kb XbaI fragment
3094
,3-globin
Neo
EZZ
p,3GN
IgH
Neo'
P,3GHN SV40
pOGSN SV40
Neo' *PX -400
pxjN
Fig. 2. Schematic maps of plasmid constructs introduced into X63Ag8 myeloma cells. Wavy line, bacterial plasmid sequences; thick line, rabbit flanking sequences. An arrow indicates the direction of transcription of the ,B-globin gene. The position and relative orientation of the TK/Neo cassette is indicated by an open box. The IgH and SV40 enhancer are both in opposite orientation relative to their natural orientation. All plasmids were linearized before transfection.
from the J - C intron, containing the IgH enhancer, downstream of the f-globin gene. In pxfN, we replaced the 3-globin gene promoter by the x light chain gene promoter and inserted the SV40 enhancer downstream of this chimeric reporter gene, since the activity of the x gene promoter alone is low even in myeloma cells (Picard and Schaffner, 1985). In addition, all plasmids contained as a selection marker the neomycin resistance gene under the control of the thymidine kinase promoter from herpes simplex virus. The thymidine kinase promoter was presumably shielded from the enhancer influence by bacterial plasmid sequences on one side, and the fl-globin gene on the other. The strategy in all the subsequent experiments was to transform mouse myeloma X63Ag8 cells with linearized recombinants (i) to promote a more predictable integration mode and hence topology of the various regulatory elements, and (ii) to maximize the chance of integrating both the 3-globin test gene and the neo gene in an intact form. Cytoplasmic RNA was isolated from single clones of transformants and analysed for ,(-globin and neo transcripts by a slot-blot hybridization assay. Furthermore, the number of copies of integrated genes was determined in some of the clones before a suitable one was selected for fusion with cl. 1D fibroblasts.
Targets for Ig suppression in cell hybrids
-(D.
nfca.C:
xJ _*
z
> >
z
-._f
x
_i
f-
nec
-
04
_
_ - 1
*0*O
rt
-
-* AX
-
i
w
4
h
.
I
Fig. 3. Expression of the ,B-globin gene linked to Ig regulatory sequences is down-regulated upon fusion to fibroblastic cells. RNase protection mapping of cytoplasmic RNA from X63Ag8 transformants, cl. ID cells, and of their 12-14-day-old hybrids selected in HAT or in G418 medium (1 mg/ml). The homogeneously labelled j3-globin probe, synthesized with the SP6 RNA polymerase system contains the first 13-globin intron and therefore yields two protected fragments (Picard and Schaffner, 1985): one of 145 nt (jlinit), representing the first exon of correctly initiated transcripts, and one fragment of 210 nt (/2+rt) representing the major part of the second fl-globin exon. The latter band is generated by correctly initiated transcripts, but also by read-through transcripts, initiated far upstream of the promoter and spliced to the second exon. Therefore, only the band termed j3linit should be considered. Correctly initiated ,B-globin transcripts from pxjN yield just one protected fragment (X/) of 196 nt after RNase mapping. RNase protection mapping with the neo probe yields a fragment of 315 nt. Twenty tzg of hybrid RNA and 10 Ag of parental RNA were used per analysis.
hybr'Id
H
p
%A
rid
...d..
0
neo
0
*
0
0
0
*
3-Globi n
MHC I
0
3-actin
0
0
*
S.0 0
pGJC M
13
p.-, Y.
ybrid
'
-3 iGN
hytr.
1D
0
rleo 3S-Globin,
*0 *
MHC I 3-actin
9
0S
*
*
*
0
*0 *
*
0
@0
*
0
pUC M
13
Fig. 4. Nuclear run-on assays on cl. ID cells, X63Ag8 transformants, and 18-20-day-old hybrid cells. Duplicate samples of DNA probes immobilized on filters, and twice the amount of labelled run-on transcripts from hybrid cells compared to parental cells were used in the hybridization reactions.
To obtain cell hybrids whose genomes would be as homogeneous and as intact as possible, fluorescenceactivated cell sorting (Junker, 1982) was applied a few hours
were
after fusion to isolate Is: is hybrids that were then grown for no more than 3 weeks. RNase protection mapping was used for the determination of steady-state levels of 3095
S.Junker et al.
cytoplasmic j3-globin and neo transcripts in parental and hybrid cells. To compensate for the relative concentration of specific transcripts per cell, 20 Ag of hybrid RNA, but only 10 jtg of parental RNA, was used per analysis. The IgH
gene
enhancer and the
x
light chain
gene
promoter are targets for suppression
We first analysed the effect of fusion with cl. ID cells on expression of the 3-globin and neo genes. For that purpose hybrids between cl. ID and independent clones of X63Ag8, containing the ,B-globin gene with its own promoter, but without an enhancer (p,BGN), were selected in HAT medium or in G418 medium. Figure 3b shows that steady-state levels of neither the neo nor the (3-globin transcripts appear to be significantly altered in 12- 14-day-old hybrids irrespective of the selection pressure applied. Thus both the neo and the (3-globin genes can serve as suitable standards. To investigate the possible contribution of the IgH enhancer to down-regulation, the IgH-containing construct p3GHN was compared to the construct pjGSN. The latter contains the SV40 enhancer and should work equally well in both myeloma and fibroblast cells. The autoradiograph of a representative RNase protection mapping is shown in Figure 3a. Steady-state levels of j-globin and neo transcripts were not significantly reduced when p3GSN transformants were fused with cl. ID cells. But remarkably, fusion with p,BGHN transformants resulted in an 8- to 11-fold reduction of (-globin transcripts compared to that in unfused cells, while expression of the neo gene was reduced no more than 2- to 4-fold as measured by densitometry. The results were essentially identical when independent clones of transformants originating from different transformation experiments were tested, and when hybrids were maintained under different selection pressures. To examine whether the Ig promoter also contributes to down-regulation of Ig transcripts in a hybrid environment, X63Ag8 pxfN transformants, in which the 1-globin gene is transcribed from the x gene promoter, were tested in a similar fashion. Indeed, the steady-state level of chimeric ,B-globin transcripts drops at least 30-fold upon fusion with cl. ID cells (Figure 3c). These results suggest that both the 1 kb XbaI fragment carrying the IgH enhancer and the x light chain gene promoter region including 400 bp upstream from the x mRNA cap site contain regulatory information that is made redundant or suppressed upon fusion with fibroblasts. Suppression of the ,l-globin gene linked to Ig regulatory sequences is at the transcriptional level The observation that down-regulation of ,3-globin transcripts occurred only with recombinants containing the IgH enhancer or the x light chain gene promoter suggests that the rate of transcription rather than RNA stability is affected. That hypothesis was directly tested by performing nuclear run-on transcription assays with nuclei prepared from transformants containing the various ,B-globin gene constructs mentioned in the preceding section, from their hybrids with cl. ID harvested 18-20 days after fusion, and from cl. ID cells (Figure 4). The endogenous x light chain gene and the integrated neo gene were used as internal controls. In hybrids containing the complete genomic contributions from each parent cell, transcription of the x gene is down-regulated (Junker et al., 3096
1988), whereas transcription of the neo gene remains largely unaffected. In hybrids containing the 3-globin gene linked to the IgH enhancer (p,BGHN), ,B-globin transcription was hardly detectable, whereas the magnitude of 3-globin transcription in hybrids containing the p,BGSN or pfGN constructs was essentially unaffected. Furthermore, transcription of the f-globin gene directed by the x gene promoter was barely detectable in hybrid cells. These results confirm the data from the RNase protection experiments and suggest that both Ig promoter and enhancer sequences contribute to transcriptional down-regulation in hybrid cells.
Discussion Expression of Igs in lymphoid cells ceases shortly after fusion with, for example, fibroblasts (Junker and Pedersen, 1985). The arrest of Ig expression is maintained in proliferating hybrids through a mechanism that selectively abolishes transcription of Ig genes. By combining techniques of gene transfer with somatic cell fusion we have identified sequences within the Ig transcriptional unit which are targets for that mechanism. We introducted test genes into the genome of mouse myeloma cells, fused cloned stable transformants with mouse cl. ID fibroblasts, and subsequently isolated hybrids by fluorescence-activated cell sorting. The advantages of that approach are 3-fold: (i) transcription rates of both introduced and endogenous genes can be compared directly; (ii) the approach provides the possibility of examining simultaneously steady-state levels of transcripts from recombinant test genes as well as endogenous Ig genes; and (iii) the hybrids can be analysed shortly after fusion before segregation of chromosomes becomes significant. We first established that an IgH gene, newly introduced into a myeloma genome, is subjected to the same selective down-regulation imposed on an endogenous Ig (i.e. X light chain) gene upon fusion with fibroblasts. We then analysed the expression of the heterologous ,3-globin gene linked to lymphocyte-specific regulatory elements. Remarkably, if an IgH enhancer was inserted 2.5 kb downstream of the test gene, or if the (3-globin promoter was replaced with the x light chain gene promoter, steady-state levels of (3-globin transcripts became greatly reduced as a consequence of transcriptional down-regulation upon fusion with cl. ID cells. In contrast, transcription of the 3-globin gene, driven by its own promoter, was not significantly affected in the hybrid environment irrespective of the presence of an SV40 enhancer. Since we compared single transformants, our data cannot reveal whether transcription of, for example, the ,B-globin gene linked to the IgH enhancer dropped below the basal level of transcription of the j-globin gene alone. However, we consider it likely that the activity of the Ig regulatory sequences in hybrid cells assumed the characteristics of that in fibroblast cells, that is inactivity of the IgH enhancer, and low residual activity of the x gene promoter. Thus we conclude that both the IgH enhancer and the x promoter are deactivated upon fusion with cl. ID cells, most likely through the same mechanism(s) that down-regulates transcription of the endogenous Ig genes. Furthermore, it is reasonable to assume that it is the combined effect of the down-regulation of both the enhancer and the promoter
Targets for Ig suppression in cell hybrids
that leads to complete suppression of expression of the endogenous Ig genes. Whether the same mechanism is responsible for both immediate and long-term suppression in hybrid cells is presently unknown. What could be the mechanism(s) of suppression? A simple gene dosage effect and thus dilution of positive factors upon fusion seems very unlikely given that x gene suppression cannot be relieved or prevented by increasing the input of the myeloma genome relative to that of the fibroblast partner (Junker et al., 1988). Several models can at present explain the fast kinetics and sequence-specificity of suppression. (i) The fibroblast cell could contribute specific repressors which would bind directly to sequences within the enhancer and the promoter, and somehow interfere with lymphocytespecific activation. In fact, several regions within the IgH enhancer have been shown to have negative regulatory activity upon a heterologous promoter in non-B cells (Kadesch et al., 1986; Wasylyk and Wasylyk, 1986; Imler et al., 1987; Kiledjian et al., 1988; Weinberger et al., 1988). Since this effect could only be demonstrated when the IgH enhancer was upstream and very close to the test promoter, the significance of these findings, in particular to our system, remains unclear. Also it has been reported that blocking protein synthesis with cycloheximide in mouse fibroblast cells containing an integrated copy of the IgH gene results in stimulation of expression of that gene, suggesting the existence of a labile repressor molecule (Ishihara et al., 1984). However, when we treated myeloma-fibroblast hybrids with cycloheximide, we were unable to detect reactivation of the suppressed x light chain gene (Junker et al., 1988). Therefore this putative labile repressor seems unlikely to be involved in the suppression we observe. (ii) Fusion to the fibroblast cell could lead to inactivation of positive activators contributed by the myeloma cell. One such candidate is the lymphoid-specific form of the octamer binding factor (Staudt et al., 1987; Landolfi et al., 1987), whose central role in B cell specificity has been demonstrated by a variety of in vivo and in vitro experiments (Dreyfus et al., 1987; Gerster et al., 1987; Lenardo et al., 1987; Scheidereit et al., 1987; Wirth et al., 1987). Indeed, both the x promoter and the IgH enhancer contain this octamer sequence element. Inactivation of factors can be envisaged as a result of repression at the level of synthesis of activators or as a result of post-translational inactivation. According to the latter hypothesis crucial lymphocyte-specific modifications of one or several activator proteins would be altered very rapidly upon introduction of the fibroblast modification system by fusion. In fact, an active form of the x enhancer factor NFxB is presumably maintained in B cells by some physiological stimulus, which is not operative in, for example, HeLa cells. In those cells the factor is present in an inactive form, but can, however, be induced by phorbol esters (Sen and Baltimore, 1986b). In conclusion, we think it is hardly a coincidence that the target sequences for suppression map to the same regulatory elements which have been defined earlier as being required for lymphocyte-specific transcription of Ig genes. Obviously, precise characterization of the target sequences as well as the factors is needed to substantiate the notion that suppression in hybrids on the one hand and lymphocytespecific activity on the other may merely be two different manifestations of the same underlying mechanism of tissuespecific gene expression. Work along these lines is now in progress.
Materials and methods Plasmid constructions All clones were made according to standard techniques (Maniatis et al., 1982) and are based on the following genes: the rabbit haemoglobin ,31 gene (Maniatis et al., 1978; van Ooyen et al., 1979), and the neomycin resistance gene from Tn5 transcribed from the thymidine kinase promoter of herpes simplex virus (recombinant pVV2; Meneguzzi et al., 1984). p,BGN: the BamHI-HindIII NeoR fragment was ligated with the PstI-HindIll (natural XbaI) ,B-globin fragment and with pSP65 cleaved with PstI and BamHI. Before transfection the plasmid was linearized with Sacl. p,BGHN, p,BGSN: the SacI-BamHI fragment containing the neo gene and the 3' half of the thymidine kinase gene was ligated into pI3G (Banerji et al., 1983) cleaved with SacI and BamHI. The IgH enhancer as a 1 kb XbaI fragment and the SV40 enhancer as a 196 bp fragment (spanning SV40 nucleotides 99-295) were inserted into the single XbaI site of the resulting clone to give p,BGHN and p,BGSN, respectively. Both enhancers are thus located at the same 3' position relative to the f3-globin gene. Prior to transfection the plasmids were linearized with Sall. pxI3N: the PvuIl-BamHI fragment containing the neoR gene was cloned into px(3 +6 (Picard and Schaffner, 1985) at the NruI site. The plasmid was linearized with KpnI.
Cell culture, transformation and fusion The mouse myeloma cell line X63Ag8 (Kohler and Milstein, 1975) and cl. 1D which is a clonal derivative of the mouse L fibroblastic cell line (Kit et al., 1963), were grown in RPMI 1640 supplemented with 10% fetal calf serum. The mouse myeloma cell line J558-JG5 (Gillies et al., 1983), obtained from Dr S.D.Gillies, had been transformed with the plasmid pSV-y2bVC containing the Ecogpt gene as a selection marker, and in addition a 9 kb Bgtll fragment that included the VDJ and Cy2b exons and the 3.5 kb intron between the VDJ and C segments. The cell line was maintained in RPMI 1640 supplemented with hypoxanthine (15 yg/ml), xanthine (250 /ig/ml) and mycophenolic acid (6 jig/ml). For transformation, 107 X63Ag8 cells (in 10 cm Petri dishes) were washed with TBS buffer (25 mM Tris-HCI, 137 mM NaCl, 5 mM KCI, 0.7 mM CaCl2, 0.6 mM Na2HPO4, pH 7.4; Kimura and Dulbecco, 1972), and 1.0 ml of TBS containing DNA at 16 Ag/ml and DEAE-dextran (Pharmacia; Mr 5 x 105) at 0.2 mg/ml was added. After 30 min at room temperature, the mixture was aspirated, and 2 ml of 25% (v/v) dimethylsulphoxide in TBS was added for 4 min. The cells were then washed three times with TBS and incubated with fresh medium. The selection was initiated 48 h after transformation by adding geneticin (G418) at a final concentration of 2 mg/mi. At 8-10 days after transfection, individual clones of transformants were identified under the microscope, hand-picked using a capillary tube and transferred to microtitre plates. Cell fusion was done in suspension using 50% polyethylene glycol (mol. wt 6000) as previously described (Junker and Pedersen, 1981), and the FITC-FACS technique (Junker, 1982) was applied for isolation of hybrids. Briefly, X63Ag8 cells that partially grow in suspension were labelled with the fluorochrome fluorescein isothiocyanate (FITC) and then fused with cl. ID cells that grow in monolayers. At 6-8 h after fusion all floating cells were washed off. The remaining attached cells comprising fluorescent hybrids and non-fluorescent cl. ID cells were harvested, and a cell sorter was used to isolate hybrids and subpopulations thereof. The hybrid system allows one to apply different selection pressures, because the parental cells contain multiple markers that can be selected for or against: (i) X63Ag8 cells grow in suspension, cl. ID and most hybrids grow attached to plastic; (ii) cl. ID is killed at concentrations of G418 of 1 mg/ml, X63Ag8 transformants are not; (iii) X63Ag8 is deficient in hypoxanthine phosphoribosyl transferase, and cl. ID in thymidine kinase; both cells are therefore killed in HAT medium. Thus hybrid cells can be maintained either as (i) monolayer cells resistant to G418 (1 mg/ml); (ii) monolayer cells resistant to HAT; or (iii) monolayer cells resistant to both HAT and G418.
RNA analyses Cytoplasmic RNA was prepared according to the method of Favaloro et al. (1980). RNA slot-blots were made essentially as described by Edwards and Adamson (1986), and nick translation and hybridization as described previously (Junker and Pedersen, 1985). The Cy2b probe was described in Gillies et al. (1983), the X probe in Picard and Schaffner (1984b), and the thymidine kinase (tk) probe in Bradshaw and Deininger (1984). RNase protection mapping was done essentially as described by Melton et al. (1984) with slight modifications as specified in Picard and Schaffner (1985). The SP6 polymerase-generated,B-globin RNA probes are described in Picard and Schaffner (1985). The neo probe is a Pvull fragment from pVV2 (Meneguzzi et al., 1984) cloned into pSP64.
3097
S.Junker et al. Nuclear run-on assays Nuclear run-on assays were done essentially as described in Gerster et al. (1986) except that [a-32P]UTP was used for in vitro pulse labelling of nascent transcripts. The probe for x gene transcripts was described in Picard and Schaffner (1984b), the MHC class I and the (-actin probes in Gerster et al. (1986), and the neo probe in Meneguzzi et al. (1984). The j3-globin probe was a 508 bp PvuIl-BglI fragment from the (-globin cDNA, cloned into Ml3mp8 (coding strand).
Acknowledaements We are grateful to Professor W.Schaffner for hosting some of the experiments. We thank Dr S.D.Gillies for making available to us the J558-JG5 cell line and the Cy,2b probe, and Dr P.L.Deininger for the tk probe. We also acknowledge the proficient technical assistance of Uffe B.Jensen during various phases of this work. This project was supported by the Danish Cancer Society through grant no. 86-024 to S.J. During part of this work, P.M. was the recipient of an EMBO long-term fellowship.
References Augereau,P. and Chambon,P. (1986) EMBO J., 5, 1791-1797. Banerji,J., Olson,L. and Schaffner,W. (1983) Cell, 33, 729-740. Bradshaw,H.D., Jr and Deininger,P.L. (1984) Mol. Cell. Biol., 4, 2316-2320. Coffino,P., Knowles,B., Nathenson,S.G. and Scharff,M.D. (1971) Nature, 231, 87-90. Dreyfus,M., Doyen,N. and Rougeon,F. (1987) EMBO J., 6, 1685-1690. Edwards,S.A. and Adamson,E.D. (1986) Exp. Cell Res., 165, 473 -480. Favaloro,J., Treisman,R. and Kamen,R. (1980) Methods Enzymol., 65, 727-728. Garcia,J.V., Stafford,J. and Queen,C. (1986) Nature, 322, 383-385. Gerster,T., Picard,D. and Schaffner,W. (1986) Cell, 45, 45-52. Gerster,T., Matthias,P., Thali,M., Jiricny,J. and Schaffner,W. (1987) EMBO J., 6, 1323-1330. Gillies,S.D., Morrison,S.L., Oi,V.T. and Tonegawa,S. (1983) Cell, 33, 717-728. Greenberg,A., Ber,R., Kra-Oz,Z. and Laskov,R. (1987) Mol. Cell. Biol., 7, 936-939. Grosschedl,R. and Baltimore,D. (1985) Cell, 41, 885-897. Imler,J.-L., Lemaire,C., Wasylyk,C. and Wasylyk,B. (1987) Mol. Cell. Biol., 7, 2558-2567. Ishihara,T., Kudo,A. and Watanabe,T. (1984) J. Exp. Med., 160, 1937-1942. Junker,S. (1982) Exp. Cell Res., 139, 51-61. Junker,S. and Pedersen,S. (1981) Biochem. Biophys. Res. Comunun., 102, 977-984. Junker,S. and Pedersen,S. (1985) Exp. Cell Res., 158, 349-359. Junker,S., Nielsen,V., Matthias,P. and Picard,D. (1988) Exp. Cell Res., in press. Kadesch,T., Zerros,P. and Ruezinsky,D. (1986) Nucleic Acids Res., 14, 8209-8221. Kelley,D.E. and Perry,R.P. (1986) Nucleic Acids Res., 14, 5431-5447. Kiledjian,M., Su,L.-K. and Kadesch,T. (1988) Mol. Cell. Biol., 8, 145-152. Kimura,G. and Dulbecco,R. (1972) Virology, 49, 394-403. Kit,S., Dubbs,D.R., Piekarski,L.J. and Hsu,T.C. (1963) Exp. Cell Res., 31, 297-312. Kohler,G. and Milstein,C. (1975) Nature, 256, 495-497. Landolfi,N.F., Capra,J.D. and Tucker,P.W. (1986) Nature, 323, 548-551. Lenardo,M., Pierce,J.W. and Baltimore,D. (1987) Science, 236, 1573-1577. Maniatis,T., Hardison,R.C., Lacy,E., Lauer,J., O'Connell,C., Quon,D., Sim,G.K. and Efstratiadis,A. (1978) Cell, 15, 687-701. Maniatis,T., Fritsch,E.F. and Sambrook,J. (1982) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Maniatis,T., Goodbourn,S. and Fischer,J.A. (1987) Science, 236, 1237-1245. Mason,J.O., Williams,G.T. and Neuberger,M.S. (1985) Cell, 41, 479-487. Matthias,P., Gerster,T. and Schaffner,W. (1987) In Eckstein,F. and Lilley,D.M.J. (eds), Nucleic Acids and Molecular Biology. Springer Verlag, Berlin, pp. 221-240. McKnight,S. and Tjian,R. (1986) Cell, 46, 795-805. Melton,D.A., Krieg,P.A., Rebagliati,M.R., Maniatis,T., Zinn,K. and Green,M.R. (1984) Nucleic Acids Res., 12, 7035-7056.
3098
Meneguzzi,G., Binetruy,B., Grisoni,M. and Cuzin,F. (1984) EMBO J., 3, 365-371. Neuberger,M.S. (1983) EMBO J., 2, 1373-1378. Nishikura,K., Ar-Rushdi,A., Erikson,J., DeJesus,E., Dugan,D. and Croce,C.M. (1984) Science, 224, 399-402. Periman,P. (1970) Nature, 228, 1086-1087. Picard,D. and Schaffner,W. (1984a) Nature, 307, 80-82. Picard,D. and Schaffner,W. (1984b) EMBO J., 3, 3031-3035. Picard,D. and Schaffner,W. (1985) EMBO J., 4, 2831-2838. Queen,C. and Baltimore,D. (1983) Cell, 33, 741-748. Queen,C. and Stafford,J. (1984) Mol. Cell. Biol., 4, 1042-1049. Ringertz,N.R. and Savage,R.E. (1976) Cell Hybrids. Academic Press, New York. Scheidereit,C., Heguy,A. and Roeder,R.G. (1987) Cell, 51, 783-793. Sen,R. and Baltimore,D. (1986a) Cell, 46, 705-716. Sen,R. and Baltimore,D. (1986b) Cell, 47, 921-928. Serfling,E., Jasin,M. and Schaffner,W. (1985) Trends Genet., 1, 224-230. Staudt,L.M., Singh,H., Sen,R., Wirth,T., Sharp,P.A. and Baltimore,D. (1986) Nature, 323, 640-643. van Ooyen,A., van den Berg,J., Mantei,N. and Weissmann,C. (1979) Science, 206, 337-344. Wasylyk,C. and Wasylyk,B. (1986) Cell, 45, 553-560. Weinberger,J., Jat,P.S. and Sharp,P.A. (1988) Mol. Cell. Biol., 8, 988-992. Wirth,T., Staudt,L. and Baltimore,D. (1987) Nature, 329, 174-178. Zeuthen,J. and Nilsson,K. (1976) Cell Differ., 4, 355-368.
Received on May 13, 1988; revised on June 21, 1988
