The EMBO Journal vol.5 no.4 pp.719-724, 1986
The lysozyme enhancer: cell-specific activation of the chicken
lysozyme gene by a far-upstream DNA element
Manfred Theisen, Aribert Stief and Albrecht E.Sippel
Zentrum fur Molekulare Biologie der Universitiit Heidelberg (ZMBH), Im
Neuenheimer Feld 282, D-6900 Heidelberg 1, FRG
Communicated by Peter Gruss
The chicken lysozyme gene is constitutively expressed in
macrophages and controlled by steroid hormones in the
oviduct. We have investigated the influence of the 5' non-
coding region of this gene on its cell-specific transcriptional
activation. In transient transfection experiments we have iden-
tified a far-upstream cell-specific enhancer element 6.1 kb
5' to the transcriptional start site of the lysozyme gene.
Transcription from the lysozyme gene promoter is induced
by this element in a position- and orientation-independent
manner in lysozyme-producing myeloid cells (HBCI), but not
in non-producing chicken embryo fibroblasts (CEF-38). The
enhancer region correlates with a DNase-hypersensitive
chromatin site which is only detectable in cels of tissues in
which the lysozyme gene is transcribed. We suggest that this
far-upstream element is involved in the tissue-specific con-
trol of lysozyme gene activity.
Key words: chicken lysozyme gene/cell-specific expression/far
upstream cellular enhancer/DNase hypersensitivity/transient ex-
One of the key problems in understanding differentiation is the
regulation of tissue-specific expression of eukaryotic genes. The
regulation of the cell-specific activity of the chicken lysozyme
gene is an attractive model for the study of differential expres-
sion of genes in different tissues of the same organism. In tubular
gland cells ofthe chicken oviduct the lysozyme gene is transcribed
only when induced by steroid hormones (Schuitz et al., 1978),
whereas in macrophages it is constitutively expressed and not
dependent on steroid hormones (Hauser et al., 1981; Sippel et
al., 1986a). Although the level of lysozyme mRNA molecules
per cell differs by two orders of magnitude (Hauser et al., 1981),
initiation of transcription occurs at the same multiple start sites
with identical relative proportions (Sippel et al., 1986a, b and
Figure la). The various modes of regulation ofthe lysozyme gene
correlate with the appearance of different sets ofDNase I-hyper-
sensitive sites in the 5' chromatin of the gene domain (Fritton
et al., 1983, 1984; Sippel et al., 1986a, b). It is generally believed
that DNase-hypersensitive sites mark the position of regulatory
elements in the chromatin domain ofeukaryotic genes where trans
acting protein factors interact with the DNA (recent review;
Lubbe and Schaffner, 1985). A correlation between viral or
cellular enhancer elements and DNase-hypersensitive chromatin
regions has for example been demonstrated for the SV40 enhancer
(Saragosti et al., 1980; Jongstra et al., 1984), the immunoglobulin
kappa gene enhancer (Parslow and Granner, 1982) and the
glucocorticoid-dependent enhancer element ofMMTV (Zaret and
Yamamoto, 1984). A well-understood example that demonstrates
IRL Press Limited, Oxford, England
the correlation between differences in chromatin structure and
the action of cell-specific DNA binding proteins is the chicken
f-globin gene. In this case the DNase-hypersensitive structure
can be restored in vitro by incubation ofDNA with extracts from
erythrocytes that contain a specific trans-acting factor (Emerson
and Felsenfeld, 1984).
Gene transfer experiments have become an important tool to
test the function of regulatory sequences in cell-specific activa-
tion of eukaryotic genes in vivo. With this method enhancer se-
quences in the immunoglobulin genes (Queen and Balitmore,
1983; Banerji et al., 1983; Gillies et al., 1983; Picard and Schaff-
ner, 1984) have been identified. Regulatory sequences within the
first several hundred nucleotides of the 5' non-coding region,
that are responsible for tissue-specific and developmentally
regulated expression have been found in the case of the insulin
and chymotrypsin genes (Walker et al., 1983, Edlund et al.,
1985), the rat albumin gene (Ott et al., 1984), the muscle actin
gene (Melloul et al., 1984) and the mouse immunoglobulin genes
(Foster et al., 1985; Mason et al., 1985; Grosschedl and Balti-
more, 1985; Picard and Schaffner, 1985).
Previous DNA microinjection experiments had shown that the
lysozyme promoter region itself is inactive when introduced in-
to chicken bone marrow derived macrophages (Renkawitz et al.,
1982). We have therefore examined the influence of 5' upstream
sequences of the chicken lysozyme gene on the cell-specific
transcriptional activation of its promoter in a transient expres-
sion system. With these experiments we are able to demonstrate
that, in contrast to what has been found in many other eukaryotic
genes, a control element involved in the cell-specific activation
of transcription is located in the far-upstream gene region. The
activating DNA element, which is located 6.1 kb 5' to the
transcriptional start site, correlates with a tissue-specific DNase-
hypersensitive site in chromatin (Fritton et al., 1983; Sippel et
al., 1986a, b), which contains two binding sites for the TGGCA
protein, a ubiquitous nuclear DNA-binding protein (Nowock and
Sippel, 1982; Borgmeyer et al., 1984; Nowock et al., 1985).
Influence ofthe S'-flanking regionon the cell-specificactivation
of the chicken lysozyme promoter
Transient expression of the chloramphenicol-acetyl-transferase
(CAT) gene fused to the lysozyme promoter was measured in
the two chicken cell lines HBCI (Beug et al., 1979) and CEF-38
(LSCC-H32, Kaaden et al., 1982). HBCI is a myeloid cell
transformed by the retrovirus MC29. CEF-38 was established
from spontaneously transformed embryo fibroblasts. Figure 1
shows an SI mappling analysis (Weaverand Weissman, 1979)
oftotal cellular RNA extracted from both cell lines.Approximate-
ly 100 lysozyme-specific transcripts can be detected perHBCI
cell, that is one-third the amount found in mature macrophages
(compare Figure 1, lanes 3, 5 and 6). In contrast nolysozyme-
specific transcripts could be detected bySI mapping analysisof
excess total cellular RNA from CEF-38 cells(Figure 1,lane7).In
M.Theisen, A.Stief and A.E.Sippel
Fig. 1. S1 mapping analysis of lysozyme gene specific transcripts. Total
cellular RNA was hybridized to a 5'-labelled BstNI fragment spanning the
promoter region of the lysozyme gene (b) and digested with 250 units SI
nuclease at 30°C for 2.5 h. (a) S 1-resistant DNA fragments were separated
on a 16% polyacrylamide 8 M urea sequencing gel after hybridization with
blood-derived chicken macrophages (lane 3), 30itgHBCI RNA (lane 5),
601gHBCI RNA (lane 6) and 150AgCEF-38 RNA (lane 7). Lane 1 is
without RNA. Numbers in parenthesis indicate the position of the multiple
start sites of lysozyme transcripts (Grez et al, 1981). Lane 4 contains
radioactively labelled Hpall restricted pBR322 DNA (fragment lengths
indicated in bp at the left margin). (b) Diagram of the promoter region of
the chicken lysozyme gene. The open box indicates the first exon (E).
Arrows indicate start site and direction of promoter specific transcripts and
the numbers give bp of respective restriction enzyme cutting sites.
tg laying hen oviduct total cellular RNA (lane 2), 20AgRNA from
a first approach we have generated a series ofplasmids contain-
ing different lengths of 5' flanking DNA sequences of the
lysozyme gene fused to the promoterless CAT gene of pCAT
3M (Bohnlein et al., 1985) (Figure 2b). A RsaI fragment cover-
ing the lysozyme gene promoter region from position -161 to
+ 14 (Figure 2a) was cloned into the unique BglII site ofpCAT
3M. This plasmid was named pLYSCAT1000. In plasmids
pLYSCAT2000, 3000 and 4000 the lysozyme gene upstream
DNA sequences were extended to positions -579, -1206 and
Logarithmically growing cultures of HBCI and CEF-38 cells
were transfected with these plasmids. Figure 3 shows the time
course of [14C]choramphenicol conversion by extracts from
transfected HBCI and CEF-38 cells. As a positive control
pSV2CAT (Gorman et al., 1982a), that contains the SV40 early
promoter/enhancer region, and which efficiently stimulates CAT
activity in both cell lines was used. pLYSCAT4000, which con-
tains 6.4 kb oflysozyme gene upstream DNA, is the only plasmid
that significantly enhances CAT activity, while pLYSCAT100O,
2000 and 3000 which contain 161, 579 and 1206 bp of 5' DNA,
Fig. 2. Map of pLYSCAT plasmids. (a) Shows the 5' region of the chicken
lysozyme gene. Positions of DNase I-hypersensitive sites (HS) are indicated
by vertical arrows. The numbers refer to their positions in kb relative to the
transcriptional start site at Exon
fragment covering HS -6.1 is indicated as a thick black arrow. The wavy
arrow shows the direction of transcription. Only those restriction sites
relevant for cloning are indicated: BamHI (B), BglII (Bg), Hinfl (Hf), RsaI
(R), SacI (S) and Sau3A (Sa). (b)
pLYSCAT plasmids. The CAT gene is symbolized by an open arrow;
sequences derived from the lysozyme gene region appear as open bars.
Numbers indicate the 5' end of the lysozyme inserts in bp relative to the
transcriptional start site. The black arrow symbolizes the position and
orientation of the 562-bp BamII-Sau3A fragment covering sequences of
HS -6.1. The BK virus HaeIII enhancer fragment (Rosenthal et al., 1983)
is indicated by a black box in pLYSCAT2500. Vector DNA is shown as a
1 (E). The 562-bp BamiHI-Sau3A
is a schematic representation of
do not stimulate conversion of chloramphenicol as compared to
pCAT 3M. The enhancement ofCAT activity by pLYSCAT4000
is specific for the lysozyme-producing cell line HBCI. This in-
dicates that cell-specific activating sequences are located more
than 1200 bp 5' to the transcriptional start site of the gene.
DNA sequences covering a hypersensitive region 6.1 kb 5' to the
gene are responsiblefor thecell-specificenhancement oflysozyme
Since pLYSCAT4000 is the only plasmid containing DNA se-
quences covering the DNase hypersensitive site at -6.1 kb (HS
-6.1, Figure 2a) that is strictly correlated with the active state
of the lysozyme gene (Sippel et al., 1986a, b), we tested whether
this far-upstream DNA region by itself is able to enhance
lysozyme promoter CAT activity. For this purpose we have con-
structed plasmids pLYSCAT2100, 2200, 2300 and 2400 (Figure
2b). A 562-bp BamHI-Sau3A fragment covering HS -6.1 was
ligated in sense and antisense orientation either into the BglH
at position -579 ofpLYSCAT2000 to generate pLYSCAT2100
pLYSCAT2000, downstream of the CAT gene to generate
pLYSCAT2300 and pLYSCAT2400. In the circular molecules
used for transfection this corresponds to a position approximately
3.3 kb 5' to the lysozyme promoter. To test whether lysozyme
promoter-derived expression of the CAT gene can generally be
stimulated by enhancer sequences, we have constructed plasmid
pLYSCAT2500, containing the 216-bp HaelI enhancer fragment
of BK virus (Rosenthal et al., 1983) cloned into the Bgm
at position -579 of pLYSCAT2000.
Transfection with pLYSCAT2500 leads to the same high level
of stimulation of CAT activity in both HBCI and CEF-38 cells
(Figure 3). This general enhancement of CAT activity occurs
the unqiue BamHI
Cell-specific enhancement of chicken lysozyme gene transcription
Fig. 3 Cell-specific enhancement of CAT gene expression after transfection with pLYSCAT plasmids. The time course is shown for conversion of
chloramphenicol by extracts from HBCI and CEF-38 cells transfected with the indicated plasmids. Cell extracts were prepared 48 h after transfection. 20 jd
aliquots were incubated in a total volume of 180I1 as described in Materials and methods. The incubations were terminated after 15, 30 or 45 min by
extraction with ethyl acetate. The indicated values of chloramphenicol conversion represent an average of at least two independent transfections.
irrespective of the activity of the endogenous lysozyme gene in
the two transfected cell types and demonstrates the absence of
tissue specificity for the lysozyme promoter function in these
cells. In contrast, transfection ofboth cell lines with the -6.1-kb
lysozyme DNA-containing plasmids pLYSCAT2100, 2200, 2300
and 2400 leads to a stimulation ofCAT activity only in the HBCI
cells, the cell type which naturally transcribes the lysozyme gene.
The enhancement is relatively independent oforientation and posi-
tion ofthe 562-bp insert, even though pLYSCAT2300 and 2400,
which contain the insert 2.7 kb further upstream to the promoter,
are less active than pLYSCAT2100 and 2200.
To ascertain whether the cell-specific stimulation ofCAT ac-
tivity by the 562-bp BamHI-Sau3A fragment is due to enhanc-
ed transcription from the lysozyme promoter, we analysed total
cellular RNA from transfected HBCI cells by SI mapping. A
1490-bp HindIll -EcoRI fragment (Figure 4b), labelled at the
EcoRI site of pLYSCAT3000 was used as probe for specific
transcripts derived from transfected lysozyme promoter-CAT
gene constructs. A 280 nucleotides-long SI-resistant DNA frag-
ment can be detected with RNA from HBCI cells transfected with
pLYSCAT2500, 2100 and 2200 (Figure 4a, lanes 5-7). This
band corresponds to CAT-specific transcripts initiated at posi-
tions + 1/-2 of the lysozyme promoter. No CAT-specific band
can be detected after transfection with pLYSCAT2000, which
lacks the upstream element (Figure 4a, lane 4). Endogenous
lysozyme gene-specific transcripts were mapped as an internal
control (Figure 4a). The simultaneous SI mapping results prove
BamHI-Sau3A fragment, as well as the BK enhancer control,
is due to cis-activation of transcription from the transfected
lysozyme promoter. From this we conclude that DNA sequences
covering the DNase-hypersensitive chromatin site at -6.1 kb
function as a tissue-specific cellular enhancer.
Cell-specific and position-independent stimulation of the SV40
early promoter by the lysozyme enhancer
Enhancers are able to stimulate transcription not only from
homologous but also from heterologous promoters (Yaniv, 1982;
Banerji et al., 1983; Khoury and Gruss, 1983). We have ex-
amined the influence ofthe lysozyme enhancer on the SV40 early
promoter fused to the CAT gene. The 562-bp cellular enhancer
fragment was cloned into pAIOCAT2 (Laimins et al., 1983) to
replace the SV40 72-bp enhancer at various positions and orien-
tations (Figure 5a). pLElCAT and pLE2CAT contain the
lysozyme enhancer fragment ligated into the unique BglII site
of pAlOCAT2 close to the SV40 early promoter in sense and
antisense orientation. pLE3CAT and pLE4CAT contain the
enhancer fragment in sense and antisense orientation in the BamHI
site ofpAIOCAT2, downstream ofthe CAT gene. Since we have
transfected circular molecules, this is equivalent to a position of
the enhancer 3.8 kb 5' to the SV40 promoter. Figure Sb sum-
marises the results obtained after transfection of HBCI and
CEF-38 cells. The lysozyme enhancer stimulates SV40 promoter-
derived expression of the CAT gene in the same orientation-,
and position-independent and cell-specific way, as demonstrated
with the lysozyme promoter.
We have identified a far-upstream cellular enhancer element in
the chicken lysozyme gene region by transient gene transfer ex-
periments. The enhancer is necessary to stimulate transcription
from the homologous lysozyme promoter or the heterologous
SV40 early promoter in lysozyme-producing myeloid cells. It
is completely inactive in chicken fibroblasts, which do not ex-
press the lysozyme gene, and in mouse Ltk- cells (data not
M.Theisen, A.Stief and A.E.Sippel
Fig. 4. SI mapping analysis of CAT gene specific transcripts in total
cellular RNA of HBCI cells transfected with pLYSCAT plasmid DNA.
(a) CAT gene specific transcripts were mapped in HBCI RNA 48 h after
transfection with pLYSCAT2000 (lane 4), pLYSCAT2500 (lane 5),
pLYSCAT2100 (lane 6) and pLYSCAT2200 (lane 7). 150yg of RNA were
hybridized to a 1490-bp HindI-EcoRl
EcoRI site that had been isolated from pLYSCAT3000 (Figure 2b). Hybrids
were digested with 100 units SI nuclease at 25°C for 3 h. Endogenous
lysozyme-specific transcripts were determined simultaneously as an internal
control using in addition the BstNI hybridization probe (Figure lb). After
separation on a 7% polyacrylamide 8 M urea sequencing gel an S1-resistant
DNA fragment of 280 nucleotides in length indicates the CAT gene-specific
transcripts that are initiated at position + 1/-2 of the transfected lysozyme
gene promoter. The SI-resistant DNA bands corresponding to the
endogenous lysozyme gene specific transcripts are indicated as in Figure la.
Lanes 1 and 2 show aliquots of the labelled BstNI and the HindlII-EcoRI
fragments used as hybridization probes, respectively. Lane 3 contains
pBR322 DNA digested with HpaII as a size marker, with fragment lengths
indicated at the left margin. (b) Diagram of the 5' region on the lysozyme
promoter-CAT fusion gene. The open box indicates the CAT gene part of
the construct. Arrows indicate the direciton of transcription from the
lysozyme promoter and the numbers give bp of the respective restriction
fragment (see b) labeled at the
As could be shown for other enhancer elements (Parslow and
Granner, 1982; Jongstraet al., 1984; Zaret and Yamamoto,
1984; Saragostietal., 1980), the lysozyme enhancer correlates
with aDNase-hypersensitive site, which is present only in
chromatin of cellsnaturally expressing the lysozyme gene (Sip-
peletal., 1986a, b).This indicates the interaction of regulatory
factors with these sequences. Our group identified a nuclear
DNA-binding protein,the TGGCA protein, that specifically binds
to twosymmetricalDNA sites within this region (Nowock and
Sippel, 1982; Borgmeyeret al., 1984). The chicken TGGCA
protein,which is functionally related to HeLa cell nuclear fac-
tor I(Leegwateretal., 1986)ispresentin all chicken cells tested
(Borgmeyeretal., 1984; Nowock et al., 1985) including HBCI
and CEF-38 cells (A.W.Piischel and A.E.Sippel, unpublished
results).Theubiquitous presenceof this factor indicates that the
bindingofonlythisproteinis insufficient to explain the tissue
specificityof thelysozymeenhancer. On the other hand the
TGGCAprotein recognizesDNA sequences within several viral
enhancers, forexamplethe enhancers of BK virus (Nowock et
al., 1985),JC virus and simiancytomegalovirus (Hennighausen
etal., 1985).In thisrespectit is also worth mentioning that the
NFI/TGGCAproteinis a DNA binding protein which functions
asstimulatingfactor of in vitro adenovirus DNA replication
(Nagata et al., 1984;Rawlins etal., 1984; Guggenheimer et al.,
1984; Leegwater et al., 1985).
Thelysozymeenhancer appears as an open chromatin struc-
ture in all celltypes transcribing the lysozyme gene (Sippel et
al., 1986a, b), independentof the level of RNA detectable in
them. We thereforesuggestthat it is a far-upstream control ele-
ment which has thegeneralfunction of determining the active
state of the chromatin domain. Steroid regulation, as in chicken
oviduct cells (Renkawitz et al., 1984) and modulation of the
transcriptionalrate in mature macrophages, as compared with
myeloidHBCIcells, could then be associated with other DNase-
hypersensitivesites(Figure 2a)in the lysozyme gene region (Frit-
ton etal., 1984).
In the case of theimmunoglobulin genes (Foster et al., 1985;
Mason etal., 1985; Grosschedl and Baltimore, 1985; Picard and
Schaffner, 1985) it has been demonstrated that promoter se-
quences display cell-type specificity independent of the cell-
specificenhancer. The chicken lysozyme promoter region, which
appearsin anopenchromatin structure in all cells and in which
thelysozyme geneis naturally transcribed (Fritton et al., 1983,
1984; Sippeletal., 1986b), does not seem to have this type of
specificityin the cells used for our gene transfer experiments.
The combination of a 579-bp DNA fragment of the lysozyme
promoter regionwith the BK virus enhancer in pLYSCAT2500
leads tocomparableCAT gene activities in both lysozyme-
producingHBCI and non-producing CEF-38 cells.
Our resultsstrengthenthe view that stage and cell-specific ex-
pressionofeukaryotic genesrests on the action of cellular con-
trol elements which can be well separated from the promoter
region. Upto now this concept has been based on results from
averyfewexamples (Banerjiet al., 1983; Picard and Schaff-
ner, 1984; Brent and Ptashne, 1984; Prochownik, 1985; Edlund
etal., 1985).The control elements regulating cell specificity of
transcriptionare located in the promoter proximal or internal
regionsof thesegenes.In contrast to that, cell-specific expres-
sion of thelysozyme geneis determined by an enhancer element
locatedapproximately6 kb upstream from the promoter region
of the gene. This result indicates, that at least for somegenes,
Cell-specific enhancement of chicken lysozyme gene transcription
1020 30 40506070 8090100
Fig. 5. Cell-specific and position-independent activation of the SV40 earlry promoter by the lysozyme enhancer. (a) Gives a schematic representation of the
pLECAT plasmids. The lysozyme enhancer fragment (black arrow, compare Figure 2) was ligated into the unique BglIl
5' to the SV40 early promoter of pAlOCAT2 (Laimins et al., 1983) in either orientation. Since we have transfected circular plasmid molecules the latter is
equivalent to a position 160 bp 3' of the poly(A) point of the indicator gene. The SV40 promoter region is indicated as an open box. The SV40 72-bp
enhancer repeats in pSV2CAT (Gorman et al., 1982a) are indicated by a black box. Cell-specific enhancement by the lysozyme enhancer is demonstrated in
(b). The indicated plasmids were introduced into HBCI cells (H) and CEF-38 cells (C). CAT activity was determined by a 60 min incubation of cell extracts
as described in Figure 3 and Materials and methods. The indicated values of conversion of chloramphenicol represent an average of at least two independent
site 130 bp or the BamHI site 3.8 kb
larger parts of gene-flanking DNA must be considered for gene
transfer experiments in order to ensure correct cell- and stage-
Materials and methods
Construction ofplasmids and DNA preparation
All plasmids were constructed and isolated by standard recombinant DNA techni-
ques (Maniatis et al., 1982) using subcloned DNA fragments of the chicken
lysozyme gene region containing recombinant phages Xlys3O and Xlys3l (Linden-
maier et al., 1979).
pLYSCATXO00. A RsaI fragment covering the lysozyme gene promoter region
from position - 161 to + 14 was isolated from plasmid pBRPO.8 (Jung et al.,
1980) and ligated into the unique BgllI site ofpCAT 3M (Bohnlein et al., 1985)
using BglII linkers. The correct integration of the insert was tested by sequenc-
ing the plasmid DNA (Maxam and Gilbert, 1977).
pLYSCA72000. To construct pLYSCAT2000, pLYSCAT1000 was digested with
EcoRI and Hinfl and the 335-bp EcoRI-Hinfl fragment covering 260 bp ofCAT
gene sequence and lysozyme promoter sequences to position -55 was isolated
and ligated to a Bgl l-Hinfl fragment containing sequences from position -579
to -55 of the lysozyme gene region. Ligated fragments were then inserted into
the pCAT 3M vector DNA that was linearized by a partial digest with EcoRI
and recut with Bgll.
pLYSCAT3000. pLYSCAT2000 was linearized by a partial digest with Bgll and
a BgllI fragment covering positions -1206 to -579 of the lysozyme gene region
was inserted into the BgtH site at position -579.
pLYSCAT4000. A 5570-bp Sacl fragment from position -6400 to -830 of the
lysozyme gene region was cloned into the SacI site at position -830 of
pLYSCA72100, 2200. A 562-bp BamHI-Sau3A fragment covering HS -6.1
was isolated from plasmid pHB596 (Borgmeyer et al., 1984) and cloned into
the BgtII site at position -579 of pLYSCAT2000, that had been linearized by
a partial digest with BgtI. pLYSCAT2100 contained the insert in sense, whilst
pLYSCAT2200 contained it in antisense orientation. The orientation of the in-
sert was determined by restriction enzyme mapping.
pLYSCA72300, 2400. The 562-bp BamHI-Sau3A fragment was ligated to
pLYSCAT2000 that had been linearized with BamHI. The insert is in sense orien-
tation in pLYSCAT2000 and in antisense orientation in pLYSCAT2400.
pLYSCAT2500. The 216-bp HaeIII enhancer fragment of BK virus (Rosenthal
et al., 1983) was ligated to pLYSCAT2000 that had been linearized at the Bgll
site at position -579.
pLElCAT, pLE2CAT. pAlOCAT2 (Laimins et al., 1983) was linearized at its
BglII site. The 562-bp BamHI-Sau3A fragment was ligated into this vector in
sense orientation (pLElCAT) and antisense orientation (pLE2CAT).
pLE3CAT, pLE4CAT. The BamHI-Sau3A enhancer fragment was cloned into
the unique BamHI site of pAlOCAT2 in sense (pLE3CAT) and antisense
DNA samples for transfection experiments were purified by centrifugation in
two successive CsCl gradients.
HBCI cells (Beug et al., 1979) and CEF-38 cells (Kaaden et al., 1982) were
grown in 150 cm2 tissue culture flasks (Costar) in Iscove's Modified Dulbecco's
Medium (IMDM, Gibco) supplemented with 8% FCS and 2% chicken serum
at 37°C, 5% CO2. In continuous culture, cells were harvested by trypsinization,
diluted 1:5 and fed with 30 ml of fresh medium twice a week.
Primary macrophage cultures were obtained from cultured white blood cells.
Blood was collected by heart puncture and the buffy coat was separated by cen-
trifugation at 800 g for 10 min through lymphocyte separation medium, density
1.077 (Flow). 1 x 108 cells per 150 cm2 flask were kept in IMDM, 8% FCS,
2% chicken serum at 5% C02, 40°C. The adherent cell fraction was used for
RNA preparation after 5 days.
Cell transfection and CAT assay
DNA was introduced into the cells by the calcium phosphate co-precipitation
method (Graham and van der Eb, 1973; Gorman et al., 1982a, b) with the follow-
ing modifications: 24 h before transfection 3 x 106 HBCI cells or 1.5 x 106
CEF-38 cells were plated on a 10 cm tissue culture dish (Falcon) with 10 ml
IMDM. 4 h before transfection cells were fed with 10 ml fresh medium. 25jig
of plasmid DNA were suspended in 438IA 10 mM Tris-HCI, pH 7.6, follow-
ed by the addition of 62A1 of 2 M CaCl2. After mixing, the DNA-CaCI2 solu-
tion was added dropwise to 500,1 of 2 x HBS (2 x HBS = 280 mM NaCl,
1.5 mM Na2HP04*2H20, 50 mM Hepes, pH 7.13). 30 min later the
DNA-CaPO4 suspension was added to the cells. Cells were incubated for 4 h
at 37°C, 5% CO2. Thereafter cells were shocked with 3 ml 15% glycerol, 140 mM
NaCl, 0.75 mM Na2HPO4*2H20, 25 mM Hepes, pH 7.13 for 2.5 min (Scholer
and Gruss, 1984), supplemented with fresh medium and incubated for 48 h at
37°C, 5% CO2. Fresh medium was added after 24 h. Cells were harvested in
40 mM Tris-HCI, pH 7.5, 1 mM EDTA, 150 mM NaCl, resuspended in 150 1l
0.25 M Tris-HCI, pH 7.8 and lysed by sonification. The lysate was cleared
in an Eppendorf centrifuge and the cell extract was stored at -200C.
CAT activity was measured by incubating 20 td of extract in a total volume
of 180 1l containing 0.25 M Tris-HCI, pH 7.8, 4 mM acetyl-CoA, 0.2,Ci
[14C]chloramphenicol (sp. act. 58 mCi/mmol). The reaction was terminated by
extraction of chloramphenicol with ethyl acetate and the acetylated derivatives
were separated from unmodified chloramphenicol by t.l.c. on silica gel (Machery
and Nagel) with chloroform-methanol (95:5). The conversion value was
calculated after determining the portion of acetylated and unmodified chloram-
phenicol by scintillation counting.
Preparation ofRNA and SI mapping analysis
Total cellular RNA was prepared by lysis ofHBCI cells, CEF-38 cells and chicken
M.Theisen, A.Stief and A.E.Sippel
macrophages in 5 M guanidinium thiocyanate, 50 mM Tris-HCI, pH 7.6, 10 mM
EDTA, 5% f3-mercaptoethanol (Chirgwin et al., 1979). Oviduct tissue was
homogenized in the same buffer using a Polytron homogenizer (Kinematica
GmbH). RNA was pelleted through a cushion of 5.7 M CsCI, 0.1 M EDTA
in a Beckman SW40 rotor at 29 000 r.p.m., 20°C for 22 h (Kaplan et al., 1979).
RNA pellets were resuspended in diethylpyrocarbonate-treated double-distilled
sterile water, EtOH precipitated and stored as aqueous solutions at -20°C.
For SI mapping analysis (Weaver and Weissman, 1979), RNA samples were
hybridized to an at least 40-fold molar excess of the appropriate DNA probe,
that had been 5' end-labelled with polynucleotide kinase (Maniatis et al., 1982)
in 10 1l of400 mM NaCl, 40 mM Pipes, pH 6.4, 1 mM EDTA, 80% formamide
at 43°C for 12 h. After hybridization, probes were diluted with 250 1I of 300 mM
NaCl, 30 mM sodium acetate, pH 4.5, 3 mM ZnSO4, 100 ig/ml denatured and
sonified salmon sperm DNA and digested with 250 units SI nuclease (Sigma)
at 30°C for 2.5 h to analyse endogenous lysozyme gene-specific transcripts, or
with 100 units SI nuclease at 25°C for 3 h to analyse CAT gene-specific
transcripts. S1-resistant DNA fragments were analysed on polyacrylamide urea
sequencing gels (Maniatis et al., 1982). Gels were exposed to Kodak XAR-5
films using intensifier screens (Dr Goos, Suprema).
We thank H.Beug and T.Graf for the cell lines HBCI and CEF-38, P.Gruss for
plasmids pSV2CAT, pAIOCAT2, pCAT3M and pBKVCAT and U.Strech-Jurk
for helping to prepare the macrophage cultures. This work was supported by grants
from the Deutsche Forschungsgemeinschaft (Si 165/4-1) and the Fonds der
Banerji,J., Olson,L. and Schaffner,W. (1983) Cell, 33, 729-740.
Beug,H., von Kirchbach,A., Doderlein,G., Conscience,J.F. and Graf,T. (1979)
Cell, 18, 375-390.
Bohnlein,E., Chowdhury,K. and Gruss,P. (1985) Nucleic Acids Res., 13,
Borgmeyer,U., Nowock,J. and Sippel,A.E. (1984) Nucleic Acids Res., 12,
Brent,R. and Ptashne,M. (1984) Nature, 312, 612-615.
Chirgwin,J.M., Przybyla,A.E., MacDonald,R.J., Rutter,J.M. (1979) Bio-
chemistry, 18, 5294-5299.
Edlund,T., Walker,M.D., Barr,P.J. and Rutter,W.J. (1985) Science, 230,
Emerson,B.M. and Felsenfeld,G. (1984) Proc. Natl. Acad. Sci. USA, 81, 95-99.
Foster,J., Stafford,J. and Queen,C. (1985) Nature, 315, 423-425.
Fritton,H.P., Sippel,A.E. and Igo-Kemenes,T. (1983) Nucleic Acids Res., 11,
Fritton,H.P., Igo-Kemenes,T., Nowock,J., Strech-Jurk,U., Theisen,M. and Sip-
pel,A.E. (1984) Nature, 311, 163-165.
Gillies,S.D., Morrison,S.L., Oi,V.T. and Tonegawa,S. (1983) Cell, 33, 717-728.
Gorman,C.M., Moffat,C.-F. and Howard,B.H. (1982a) Mol. Cell. Biol., 2,
Gorman,C.M., Merlino,G.T., Willingham,M.C., Pastan,I. and Howard,B.H.
(1982b) Proc. Natl. Acad. Sci. USA, 79, 6777-6781.
Graham,F.L. and van der Eb,A.J. (1973) Virology, 52, 456-457.
Grez,M., Land,H., Gisecke,K., Schutz,G., Jung,A. and Sippel,A.E. (1981) Cell,
Grosschedl,R. and Baltimore,D. (1985) Cell, 41, 885-897.
Guggenheimer,R.A., Stillman,B.W., Nagata,K., Tamanoi,F. and Hurwitz,J.
(1984) Proc. Natl. Acad. Sci. USA, 81, 3069-3073.
Hauser,H., Graf,T., Beug,H., Geiser-Wilke,I., Lindenmaier,W., Grez,M.,
Land,H., Giesecke,K. and Schutz,G. (1981) In Neth,R., Gallo,R.C., Graf,T.,
Mannweiler,K. and Winkler,K. (eds), Haematology and Blood Transfusion.
Springer, Berlin, Vol. 26, pp. 175-178.
Hennighausen,L., Siebenlist,U., Danner,D., Leder,P., Rawlins,D., Rosenfeld,P.
and Kelly,T. (1985) Nature, 314, 289-292.
Jongstra,J., Reudelhuber,T.L., Oudet,P., Benoist,C., Chae,C.-B., Jeltsch,J.-M.,
mathis,D.J. and Chambon,P. (1984) Nature, 307, 708-714.
Jung,A., Sippel,A.E., Grez,M. and Schutz,G. (1980) Proc. Natl. Acad. Sci. USA,
Kaaden,O.R., Lange,S. and Stiburek,B. (1982) In Vitro, 18, 827-834.
Kaplan,B.B., Berstein,S.L. and Giaio,A.E. (1979) Biochem. J., 183, 181-184.
Khoury,G. and Gruss,P. (1983) Cell, 33, 313-314.
Laimins,L.A., Khoury,G., Gorman,C.M., Howard,B. and Gruss,P. (1983)Proc.
Natl. Acad. Sci. USA, 79, 6453-6457.
Leegwater,P.A.J., van Driel,W. and van der Vliet,P.C. (1985)EMBO J., 4,
Leegwater,P.A.J.,van der Vliet,P.C., Rupp,R.A.W., Nowock,J. and Sippel,A.E.
(1986) EMBO J., 5, 381-386.
Lindenmaier,W., Nguyen-Huu,C., Lurz,R., Stratmann,M., Wurtz,T., Hauser,
H.J., Sippel,A.E. and Schutz,G. (1979) Proc. Natl. Acad. Sci. USA, 76,
Lubbe,A. and Schaffner,W. (1985) Trends Neurosci., 8, 100-104.
Mason,J.O., Williams,G.T. and Neuberger,M.S. (1985) Cell, 41, 479-487.
Maniatis,T., Fritsch,E.F. and Sambrook,J. (1982) Molecular Cloning. A
Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Har-
Maxam,A. and Gilbert,W. (1977) Proc. NatI. Acad. Sci. USA, 74, 560-564.
Melloul,D., Aloni,B., Calvo,J., Yaffe,D. and Nudel,U. (1984) EMBO J., 3,
Nagata,K., Guggenheimer,R.A. and Hurwitz,J. (1984) Proc. Natl. Acad. Sci.
USA, 80, 6177-6181.
Nowock,J. and Sippel,A.E. (1982) Cell, 30, 607-615.
Nowock,J., Borgmeyer,U., Puschel,A.W., Rupp,R.A.W. and Sippel,A.E. (1985)
Nucleic Acids Res., 13, 2045-2061.
Ott,M.-O., Sperling,L., Herbomel,P., Yaniv,M. and Weiss,M.C. (1984) EMBO
J., 3, 2502-2510.
Parslow,T.G. and Granner,D.K. (1982) Nature, 299, 449-451.
Picard,D. and Schaffner,W. (1984) Nature, 307, 80-82.
Picard,D. and Schaffner,W. (1985) EMBO J., 4, 2831-2838.
Prochownik,E.V. (1985) Nature, 316, 845-848.
Queen,C. and Baltimore,D. (1983) Cell, 33, 741-748.
Rawlins,D.R., Rosenfeld,P.J., Wides,R.J., Challberg,M.D. and Kelly,T.J.,Jr.
(1984) Cell, 37, 309-319.
Renkawitz,R., Beug,H., Graf,T., Matthias,P., Grez,M. and Schutz,G. (1982)
Cell, 31, 167-176.
Renkawitz,R., Schutz,G., von der Ahe,D. and Beato,M. (1984) Cell, 37,
Rosenthal,N., Kreiss,M., Gruss,P. and Khoury,G. (1983) Science, 222,749-754.
Saragosti,S., Mayne,G. and Yaniv,M. (1980) Cell, 20, 65-73.
Scholer,H.R. and Gruss,P. (1984) Cell, 36, 403-411.
Schuitz,G., Nguyne-Huu,M.C., Giesecke,K., Hynes,N.E., Groner,B., Wurtz,T.
and Sippel,A.E. (1978) Cold SpringHarbor Symp. Quant. Biol., 42,617-624.
Sippel,A.E., Nowock,J., Theisen,M., Borgmeyer,U., Strech-Jurk,U., Bonifer,C.,
Igo-Kemenes,T. and Fritton,H.P. (1986a) In Clayton,R.M. and Truman,D.E.S.
(eds), Co-ordinated Regulation ofGene Expression. Plenum, New York, Lon-
don, in press.
Sippel,A.E., Fritton,H.P., Theisen,M., Borgmeyer,U., Strech-Jurk,U. and Igo-
Kemenes,T. (1986b) In Grodzicker,T., Sharp,P.A. and Botchan,M.R. (eds),
DNA Tumor Viruses: Gene Expressionand Replication. Cancer Cells 4/1986,
Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, in press.
Walker,M.D., Edlund,T., Boulet,A.M. and Rutter,W.J. (1983) Naure, 306,
Weaver,R.F. and Weissman,C. (1979) Nucleic Acids Res., 7, 1175-1193.
Yaniv,M. (1982) Nature, 297, 17-18.
Zaret,K.S. and Yamamoto,K.R. (1984) Cell, 38, 29-38.
Received on 1I December 1985; revised on 27 January 1985