- Email: [email protected]
REGULATORY ELEMENTS OF THE LEUKAEMIA INHIBITORY FACTOR (LIF) PROMOTER IN MURINE BONE MARROW STROMAL CELLS
Article No. cyto.1998.0475, available online at http://www.idealibrary.com on
REGULATORY ELEMENTS OF THE LEUKAEMIA INHIBITORY FACTOR (LIF) PROMOTER IN MURINE BONE MARROW STROMAL CELLS G. Go¨llner, G. Bug, B. Rupilius, C. Peschel, C. Huber, H. G. Derigs Leukaemia inhibitory factor (LIF) plays an important role as a haematopoietically active cytokine. As described earlier in a murine model, interleukin 1 (IL-1) induced LIF mRNA and protein expression. We utilized the murine cell line +/+-1.LDA11 to further define regulatory mechanisms of LIF expression in bone marrow stromal cells. The production of LIF mRNA is stimulated by IL-1â, TNF-á, and the cAMP analogue 8-bromoadenosine 3 :5 -monophosphate (8BrcAMP). LIF mRNA expression is controlled at the transcriptional level. Different fragments from 542 to 45 bp 5 upstream of the transcriptional start site of the murine LIF gene were fused to the luciferase gene. All LIF-promoter luciferase constructs exhibited constitutive luciferase activity under serum free conditions. The level of luciferase activity decreased with LIF-promoter constructs of less than 249 bp (pLIF249) in size. When tested with the 314 bp LIF-promoter construct, incubation of stromal cells with IL-1â (500 U/ml) resulted in a 1.57-fold stimulation, with TNF-á (500 U/ml) in 2.06-fold stimulation, and with 8BrcAMP (0.5 mM) in a 3.42-fold stimulation of luciferase activity. By testing different deletion mutants we could narrow the IL-1 and TNF-á responsive promoter areas to the region 249 to 145 bp and the 8BrcAMP responsive area from 145 to 82 bp. Mobility shift experiments revealed that nuclear proteins from stromal cells form a DNA-protein complex by binding to the region from 249 to 145 bp of the LIF promoter. 1999 Academic Press
Murine leukaemia inhibitory factor is a glycosylated protein of 58 kDa. This cytokine has pleiotropic actions on haematopoietic cells, inflammation, and embryonic development. LIF also affects other tissues such as bone, liver, and neurons.1,2 It was identified, characterized, purified, and cloned on the basis of its ability to induce the differentiation of the murine myeloid leukaemic cell line M1.3 On haematopoietic cells, LIF has several distinct functions. It has been found to support proliferation of the factordependent myeloid cell line DA-1.4 When used alone, LIF does not stimulate the formation of colonies of any haematopoietic lineage, but it acts synergistically From the Division of Hematology, III. Dept. of Medicine, Johannes Gutenberg University, Langenbeckstrasse 1, D-55131 Mainz, Germany Correspondence to: Dr H. G. Derigs, Division of Hematology, III. Department of Medicine, Johannes Gutenberg-Universita¨t, Langenbeckstr. 1, D-55101 Mainz, Germany Received 21 July 1998; received in revised form 30 November 1998; accepted for publication 23 December 1998 1999 Academic Press 1043–4666/99/090656+08 $30.00/0 KEY WORDS: bone marrow stromal cells/Leukaemia Inhibitory Factor (LIF)/promoter activity/transcriptional control 656
together with other growth factors on the proliferation of haematopoietic progenitor cells.5,6 LIF is produced by monocytes, fibroblasts, astrocytes, different tumour cell lines, and mesenchymal cells.7–10 The regulation of LIF expression is largely unidentified. It is known that the 3 untranslated region of LIF mRNA contains AU-rich motifs. These sequences have been shown in a number of other mRNA species to convey a destabilizing effect. We have previously described the transcriptional regulation of LIF mRNA by inflammatory cytokines IL-1 and TNF-á in murine bone marrow stromal cell lines.11 The first 300 bp of the 5 promoter region display a high degree of cross-species conservation thereby indicating importance.12 Within this region there are four TATA-like elements and a second transcriptional start-site.13 The 5 LIF-promoter region contains binding motifs for the nuclear transcription factors SP-1 and AP-2.13 In the present study, we proved LIF expression in IL-1 stimulated human bone marrow stromal cells. Furthermore, we could narrow the IL-1 and TNF-á responsive areas of the murine LIF promoter to the CYTOKINE, Vol. 11, No. 9 (September), 1999: pp 656–663
Murine LIF promoter regulation / 657
region 249 to 145 bp and the 8BrcAMP responsive area to the region from 145 to 82 bp. Mobility shift experiments revealed that nuclear proteins from stromal cells form a DNA-protein complex by binding to the region from 249 to 145 of the LIF 5 promoter.
A
LIF
28S RNA
RESULTS 18S RNA
Control
IL-1
100 B
LIF protein concentration (pg/ml)
Expression of LIF in human bone marrow stromal cells was assessed at the protein level in culture supernatants by a sensitive ELISA and at the mRNA level by Northern blot analysis. In unstimulated cultures, very low levels of LIF mRNA were detected. After incubation with IL-1 (100 U/ml) for 6 h, significant LIF-mRNA expression was induced (Fig. 1A). These RNA-data were confirmed at the protein level (Fig. 1B). Furthermore, incubation with the cyclic AMP analogue 8BrcAMP (0.5 mM) also induced LIF protein in human bone marrow stromal cell culture supernatants (data not shown). To further define LIF regulation in stromal cells, we used the murine bone marrow stromal cell line +/+-1.LDA11 as a model. Transcriptional induction of LIF mRNA by IL-1, TNF-á and 8BrcAMP had been demonstrated before in these cells.11 A series of different murine LIF promoter fragments were cloned upstream of the luciferase gene and reporter plasmids were transfected into stromal cells. Under serum free conditions, constitutive luciferase expression of different LIF promoter fragments is shown (Fig. 2). The negative control plasmid pLIF542as, which contains the LIF-promoter in antisense orientation, showed no luciferase expression. The construct pLIF314 had a 1.390.08 fold (meanSEM) LIF promoter activity relative to the longest fragment pLIF542, which was used as standard. The activity of pLIF249 was 1.620.11 fold higher than that of pLIF542. Shorter fragments showed a decreased activity relative to pLIF542. The smallest construct (pLIF45) showed only a 0.10 fold activity relative to the fragment pLIF542. In order to investigate the induction of LIF transcription by cytokine incubation, we exposed the transfected bone marrow stromal cells to different stimuli. Stimulation of +/+.1-LDA11 cells, transfected with pLIF314, with IL-1â, TNF-á and the cAMP analogue 8BrcAMP showed an increase of luciferase activity (Fig. 3). Incubation of +/+.1-LDA11 transfected with pLIF314 with IL-1 (500 U/ml) alone resulted in a 1.570.09 fold activity. TNF-á (500 U/ml) stimulated luciferase activity 2.060.12 fold and 8BrcAMP (0.5 mM) 3.420.17 fold above medium control (Fig. 3). Co-incubation of +/+.1-LDA11 with combinations of IL-1, TNF-á and 8BrcAMP resulted in additive
75
50
25
0 Figure 1. cultures.
Medium
IL-1
LIF expression in human bone marrow stromal cell
(A) Induction of LIF-mRNA in human bone marrow stromal cells. Confluent secondary stromal cultures were incubated for 6 h in medium alone or with the addition of IL-1 (100 U/ml). Total cellular RNA was prepared and subjected to Northern-blot analysis with cDNA probe for human LIF. The 18 and 28S rRNA from ethidium bromide stained gel is shown as loading control. A representative of two independent experiments is shown. (B) Confluent secondary stromal cultures were incubated in medium alone or with the addition of IL-1 (100 U/ml). After incubation for 24 h, LIF concentrations in the supernatants were assessed using a specific ELISA. The mean (SEM) of 4 independent experiments is shown.
stimulation of luciferase expression (Fig. 3). Co-incubation with combinations of two stimuli from IL-1, TNF-á or 8BrcAMP resulted in a two- to seven-fold increase of luciferase activity, and co-incubation of all three stimuli resulted in a 10.650.93 fold increase. These data reflect earlier published results from Run On experiments in the same system (Fig. 3). To narrow the LIF promoter region, which is necessary for the stimulatory response, we incubated
658 / Go¨llner et al.
CYTOKINE, Vol. 11, No. 9 (September, 1999: 656–663)
from stromal cells incubated with IL-1â, TNF-á and the cAMP analogue 8-bromoadenosine 3 ,5 monophosphate (8BrcAMP) showed a DNA protein complex with both the 249 and the 145 bp LIFpromoter fragments (data not shown). To investigate the role of the AP-2 consensus sequence of the LIF-promoter further, which was located from 185 to 171 bp upstream of the transcription start site, we used the radiolabelled AP-2 oligonucleotide to detect a specific AP-2 in +/+-1.LDA11 protein extracts. With unstimulated and stimulated protein extract from +/+-1.LDA11 stromal cells, there was no DNA–protein complex detectable. In addition, the DNA–protein complex from radiolabelled fragment LIF249 with nuclear extract from stromal cells was not abolished by addition of a 30-fold excess of the AP-2 oligonucleotide (Fig. 6).
1.8 *** 1.6 ***
Relative luciferase activity
1.4
1.2
1
0.8 *** 0.6 *** 0.4
0.2
***
Figure 2.
45 IF pL
IF
82
5 pL
IF
14
9 pL
pL
IF
24
4 31 pL
IF
54 IF pL
pL
IF
54
2a
2
s
0
Unstimulated LIF-promoter fragment activity.
Relative luciferase activities in +/+-1.LDA11 stromal cells transfected with various LIF-promoter fragments. 48 h after start of transfection, cells were harvested and luciferase activity and â-galactosidase activity were determined in cytoplasmic extracts as described in Materials and Methods. Normalization for transfection efficacy was done by â-galactosidase activity. Mean (MW) and standard error (SEM) from twelve independent experiments are shown. The luciferase activity of the different fragments were plotted relative to the plasmid pLIF542. ***Significantly different luciferase activity from pLIF542, P <0.0001).
stromal cells, transfected with different reporter gene constructs, and used different stimuli. LIF-promoter fragments longer than 145 bp showed an increase of luciferase activity after IL-1 incubation (Fig. 4). Similar results were obtained after TNF-á incubation. In contrast, incubation with 8BrcAMP also stimulated LIF-promoter constructs larger than 82 bp. Gel mobility shift assays using radiolabelled LIFpromoter fragments LIF249 and LIF145 were performed to detect proteins that specifically bind to LIF-promoter regions. Under unstimulated conditions, one DNA–protein complex was detected with the fragment LIF249 (Fig. 5). The complex was abolished by the addition of the same unlabelled LIF-promoter fragment but not by a 30-fold excess of an unrelated oligonucleotide proving the specificity of the DNA binding. With fragment LIF145 no DNA-protein complex under unstimulated conditions was detectable (Fig. 5). Gel mobility shift assays with nuclear extract
DISCUSSION LIF is a cytokine with various actions on haematopoietic progenitor cells. The LIF gene is constitutively expressed at very low levels in the normal adult mouse.14 The expression of LIF in different cell types is induced by inflammatory mediators such as TNF-á and IL-1.7,15,16 Miyagi et al.17 and Anegon et al.18 reported recently that post-transcriptional mechanisms play a role in the induction of LIF expression in LPS-stimulated human monocytes and TPAstimulated human fibroblasts. Endothelial cells seem to be the main source of LIF expression in the bone marrow microenvironment.19 In the present study we proved LIF induction by IL-1 at the protein and mRNA level in human long term bone marrow culture cells. This data implies a physiological role of LIF in haematopoietic regulation. Our group showed previously that LIF mRNA expression is controlled at the transcriptional level in murine bone marrow stromal cells. Furthermore, LIF mRNA was shown to be highly unstable, with a half life of only 30 min.11 In the current study, we delineated active LIF promoter elements by using a sensitive reporter gene assay in a murine bone marrow stromal cell line. Different LIF-promoter fragments showed activity in unstimulated bone marrow stromal cells. Stahl and Gough19 described negative and positive control elements within the promoter region in the LIF gene, when transfected with different CAT constructs in human STO fibroblasts. The negative control element was located between 360 and 249 bp, and the positive regulatory element was located between 860 and 660 bp. In contrast to this data, we could not locate a negative regulatory element in the murine LIF promoter. In our system fragments spanning the
Murine LIF promoter regulation / 659
Run on
Luciferase assay -fold stimulation
LIF
IL-6
CHO-B
}
+/+ –1.LDA11
0
5
10
15
Control
IL-1
TNF
8BrcAMP
IL-1 + 8BrcAMP
TNF + 8BrcAMP
IL-1 + TNF
IL-1 + TNF + 8BrcAMP Figure 3.
Induction of LIF-transcription.
6
4
2
Figure 4.
45 IF pL
IF
82
5 pL
IF
14
9 pL
pL
IF
24
4 31 IF pL
IF
54
2
0
pL
region 314 to 249 bp of the LIF promoter had greater luciferase activity than the control plasmid pLIF542 (542 to +36 bp). Hsu and Heath20 showed in a stably transfected murine fibroblast cell line that a regulatory enhancer element is also located in the far 5 region of the LIF gene between 3200 and 1200 bp.28 Incubation of stromal cells with IL-1, TNF-á, and 8BrcAMP, which were transfected with LIF-promoter constructs increased the luciferase expression. These inducing effects were shown to be additive. These data confirm earlier published nuclear Run On results.11 Of note is the fact that in the Run ON experiments IL-1 resulted in the strongest induction of LIF expression, while in reporter gene assays 8BrcAMP resulted in the strongest induction (Fig. 4). The cause for this discrepancy might be the difference in incubation time for the two assays. Co-incubation with different stimuli resulted, in both assay systems, in more than additive
Relative luciferase activity
Comparison of nuclear transcription rates and luciferase activity of the LIF-promoter luciferase plasmid pLIF314 in unstimulated and stimulated +/+-1.LDA11 stromal cells. Stromal cells were incubated for 2 h (Run-on) or 24 h (Luciferase assay) with IL-1 (500 U/ml), TNF-á (500 U/ml) and 8BrcAMP (0.5 mM) or with combinations of these stimuli. Nuclear extracts were collected and Run On experiments were performed as described in Materials and Methods. The housekeeping gene CHO-B is shown as a positive control and IL-6 is shown as a differently regulated cytokine. Normalization for transfection efficacy in the reporter assay was done by â-galactosidase activity. Mean and standard error (SEM) from six independent experiments are shown in this figure. The luciferase activity of the different stimuli were plotted relative to the unstimulated plasmid pLIF314, which was assigned a value of 1.
Stimulation of LIF-promoter fragments.
Stimulation of the LIF-promoter luciferase plasmids by: (– –) IL-1; (– –), TNF-á; (– –), 8BrcAMP; and (— —), medium. +/+-1.LDA11 stromal cells were incubated for 24 h with IL-1 (500 U/ml), TNF-á (500 U/ml) and 8BrcAMP (0.5 mM). Normalization for transfection efficacy was done by â-galactosidase activity. Mean (MW) and standard error (SEM) from six independent experiments are shown in this figure. The luciferase activity is shown relative to the unstimulated plasmid pLIF542.
660 / Go¨llner et al.
CYTOKINE, Vol. 11, No. 9 (September, 1999: 656–663)
A
B LIF249 1
Figure 5.
2
LIF145 3
4
1
2
3
4
Nuclear protein binding to different LIF-promoter fragments in unstimulated bone marrow stromal cells.
Electrophoretic mobility shift analysis of nuclear extracts prepared from unstimulated +/+-1.LDA11 stromal cells. Gel mobility shift using radiolabelled LIF-promoter fragments (A) LIF249 and (B) LIF145. Lane 1, labelled LIF-promoter DNA but no extract; lane 2, labelled LIF-promoter DNA and nuclear extracts prepared from unstimulated +/+-1.LDA11 stromal cells; lane 3, same as lane 2 but unlabelled LIF-promoter DNA (30-fold molar excess of unlabelled DNA to labelled); lane 4, same as lane 2 but unlabelled oligonucleotide Oct-1 (30-fold molar excess of unlabelled DNA to labelled).
stimulation. Similar synergistic stimulatory effects of IL-1 and TNF-á are known from the expression of other cytokines, such as GM-CSF.21,22 By using reporter plasmids with different promoter fragments in stimulated stromal cells the IL-1 and TNF-á responsive area could be narrowed to the region from 249 to 145 bp of the LIF gene while the 8BrcAMP responsive area was located from 145 to 82 bp. An AP-2 consensus region is located in the region of the TNF-á and IL-1 responsive elements. However, mobility shift assays neither demonstrated AP-2 in protein extracts from the murine stromal cell line nor could the binding of proteins to the active LIF promoter region be competed by an AP-2 consensus oligonucleotide. Thus it seems unlikely that AP-2 plays an active role in LIF activation in murine bone marrow stromal cell lines. Further studies are currently underway to identify the protein which interacts with the LIF-promoter in murine bone marrow stromal cells.
MATERIAL AND METHODS Materials rh-IL-1á was kindly provided by Hoffmann La Roche (Nutley, NJ, USA). Murine IL-1â was a gift from Imunex (Seattle, Washington). Murine Tumor-Necrosis Factor-á was purchased from Genzyme (Kent, England). 8-bromoadenosine 3 :5 -monophosphate (8BrcAMP) was purchased from Sigma (Mu¨nchen, Germany). The AP-2 binding oligonucleotide and the AP-2 containing protein extract were purchased from Promega (Madison, WI, USA). The plasmid pLIF20.2 was kindly provided by Donald Metcalf (Walter and Eliza Hall Institute, Melbourne, Australia).
Cell culture Human adherent bone marrow stromal cell cultures were essentially obtained as previously described.23 Bone marrow mononuclear cells, separated by centrifugation over Ficoll-Hypaque, were incubated at a cell density of 1106/
Murine LIF promoter regulation / 661
1
2
3
4
5
6
7
8
9
10
A TATA-Box
AP-2 TATA- 2.TSS
TATA-
TATA-
SP-1 CAAT
Box
Box
Box
–540 –260
–220 –190 –170
–120
–100
–80
B
–60
–40
1.TSS
–20
+1
pLIF542as
–542 pLIF542 –542 pLIF314 –314 pLIF249 –249 pLIF145 –145 pLIF82 –82 pLIF45 Figure 6. Nuclear protein binding to different LIF-promoter fragments in stimulated bone marrow stromal cells. Electrophoretic mobility shift analysis of nuclear extracts prepared from +/+-1.LDA11 stromal cells. Gel mobility shift using radiolabelled LIF-promoter fragment (A) LIF249, (B) LIF145. Lane 1, labelled LIF-promoter DNA but no extract; lane 2, labelled LIFpromoter DNA and nuclear extracts prepared from unstimulated +/+-1.LDA11 stromal cells; lane 3, labelled LIF-promoter DNA and nuclear extracts prepared from IL-1-stimulated (500 U/ml) +/+1.LDA11 stromal cells; lane 4, labelled LIF-promoter DNA and nuclear extracts prepared from TNF-á-stimulated (500 U/ml) +/+1.LDA11 stromal cells; lane 5 labelled LIF-promoter DNA and nuclear extracts prepared from 8BrcAMP-stimulated (0.5 mM) +/+1.LDA11 stromal cells; lane 6, same as lane 2 but unlabelled LIF-promoter DNA (30-fold molar excess of unlabelled DNA as labelled).
ml in 25 cm2 tissue culture flasks at 33C in culture medium consisting of RPMI 1640 supplemented with 10% fetal calf serum (FCS), 10% horse serum, 1.0 mM hydrocortisone (Sigma), and the additives as described.23 At weekly intervals, cultures were fed by replacing 75% of the culture medium. When cultures were covered more than 80% by adherent cells, the primary cultures were treated with trypsinEDTA (Biochrom, Berlin, Germany). Detached stromal cells were pooled and expanded in new culture flasks at a surface ratio of 1:5 in 75-cm2 tissue flasks. The adherent cells were incubated under the same culture conditions until the cultures became confluent again. By this culture method, homogeneous stromal cell layers were obtained that were morphologically and functionally comparable in different culture flasks. Hydrocortisone was removed from these cultures at least 3 days before use for RNA and protein analysis. LIF was measured in the culture supernatants of stromal cells after incubation for 24 h with various factors, as described in the Result section. The continuous clonal marrow adherent cell line +/+1.LDA11 was kindly provided by Dr Scott Boswell (IU Medical School, Indianapolis, IN). The cell line was estab-
–45 Figure 7.
LIF-promoter.
(A) Regulatory elements of the murine 5 -LIF promoter (TSS=transcription start site). (B) Design of the LIF-promoter constructs. DNA fragments comprising different portions of the 5 -flanking region of the LIF gene were subcloned in the luciferase expression vector PxP2.
lished from haematopoietically inactive long term bone marrow culture of WCB6F1-+/+ mice as has previously been described.24 The cell line was maintained in McCoy’s 5A medium supplemented with 10% FCS, 100 U/ml penicillin, and 100 ìg/ml streptomycin (Serva, Heidelberg, Germany).
Plasmids The LIF plasmids LIF249 and LIF145 were constructed by cloning polymerase chain reaction (PCR) products into a plasmid vector (Invitrogen, San Diego, CA, USA). All LIF luciferase-promoter fragments were constructed by subcloning fragments into HindIII/XhoI sites of the PxP2 vector.25 pLIF542 contains bp 542 to +35 of the 5 LIF promoter in the luciferase expression vector PxP2. The negative control plasmid PLIF542as contains the same fragment in antisense orientation in the luciferase expression vector PxP1. Plasmid pLIF314 contains bp 314 to +35, the plasmid pLIF249 contains bp 249 to +35 and the plasmid pLIF145 contains bp 145 to +35 of the 5 LIF promoter. The plasmid pLIF82 was constructed by a restriction digest with the enzyme BamHI and self-ligation of the vector. This Plasmid contains bp 82 to +35. The plasmid pLIF45 contains bp 45 to +35 of the 5 LIF promoter. Figure 7 shows the different LIF promoter fragments in relation to known function elements or consensus sequences. The plasmid RSVluc contained the luciferase cDNA controlled by the constituitively active promoter of the Rous Sarcoma Virus (RSV).
662 / Go¨llner et al.
CYTOKINE, Vol. 11, No. 9 (September, 1999: 656–663)
The plasmid RSVâ-gal contains a 5.6 kb galactosidase insert which is controlled by a RSV promoter.
A 1E + 006
Northern blot analysis Total cytoplasmic RNA was prepared using the single step method of guadinium/phenol-chloroform extraction as described earlier.27 10–15 ìg of RNA (depending on the least yield obtained in each experiment) were subjected to electrophoresis on a 1% agarose-formaldehyde gel and transferred onto nylon membrane (Hybond-N, Amersham Buchler). Blots were hybridized to á32P-labelled cDNA probes using random primer DNA labelling kit (Boehringer, Mannheim, Germany), washed and exposed to Cronex-4 autoradiography films (DuPont) at 70C. hu-LIF cDNA, a 600-bp BamHI-HindIII fragment cloned into pIC2ORvector, was obtained from Dr Mo¨vva (Sandoz, Basel).
Enzyme-linked immunoabsorbent assay (ELISA) Concentrations of LIF were measured using Biotra Human Cytokine Immunoassays (Amersham) according to the manufacturer’s instructions.
Reporter gene assays 2105 cells in 6-well plates were transfected with 1 ìg luciferase-plasmid, 0,25 ìg RSVâ-gal plasmid and 3 ìl of Lipofectamine reagent (Life Technologies, Inc., Gaitherburg, Germany) in 1 ml serum free medium (HL1; Ventrex, Portland, ME). After incubation for 3 h cells were washed with 1 phosphate-buffered saline (PBS) and 2 ml McCoy’s 5A medium with 10% FCS were added for 20 h. At 24 h after the start of transfection, the medium was replaced and cells were cultured under serum free conditions for 24 h. Fortyeight hours after the start of transfection cells were harvested for luciferase- and â-galactosidase assays. Cells were washed with 1PBS and were lysed for 15 min at room temperature with 200 ìl/well 1 reporter lysis buffer (Promega, Madison, WI). Cells were scraped with a rubber policeman and centrifuged at 14 000 g for 2 min at 4C. 20 ìl cell lysate was added
100 000
10 000
1000
100
10 pUC 18
RSVluc
pLIF542
pLIF542as
Control
IL-1
TNF
8BrcAMP
B 4.5 4 Relative luciferase activity
Run-on experiments were performed according to a modified method of Weber et al.26. Briefly, confluent cell layers (108 cells per treatment) were washed with ice cold PBS and the nuclei isolated by lysis in 0.5% NP-40 containing buffer. 108 nuclei were resuspended in 100 ml of 40% glycerol containing buffer and incubated in an equal volume of reaction buffer containing 5 mM MgCl2, 100 mM KCl, 1 mM of each ATP, CTP, GTP and 100 ìCi [á-32P]UTP (800 Ci/mM, Amersham Life Science, UK) at 26C for 30 min. The reaction was stopped by the addition of 200 ìl HSB buffer (0.5 M NaCl, 50 mM MgCl2, 2 mM CaCl2, 10 mM Tris, pH 7.4) and 2 ìl DNase I, 30 ìl RNase inhibitor and 24 ìg t-RNA. This mixture was allowed to incubate at 37C for 30 min. The 32P-labelled, nuclear RNA elongated in vitro was isolated by the method of Chomczynski and Sacchi27 and an equal number of cpm from each probe was hybridized at 65C against slot blots of denatured plasmids containing cDNA for cytokines of interest or control plasmids for 48 h. Slot blots were washed in 2SSC at 65C and then treated with RNase at 10 mg/ml and autoradiographed.
Luciferase activity (RLU)
Nuclear run-on transcription assay
3.5 3 2.5 2 1.5 1 0.5 RSVluc
Figure 8.
Sensitivity and specificity of the luciferase assay.
(A) Luciferase activity, measured in relative light units (RLU), was determined in +/+-1.LDA11 stromal cells transfected with various vectors as labelled. pUC18 was used as a negative control vector without luciferase cDNA. pRSVluc was used as a positive control. pLIF542 contained the LIF promoter sequence in sense orientation and pLIF542 as the same sequence in antisense orientation 5 of luciferase cDNA. 48 h after transfection cells were harvested and luciferase activity was determined in cytoplasmic extracts as described in Materials and Methods. (B) Luciferase activity of pRSVluc transfected +/+-1.LDA11 stromal cells after incubation for 24 h with either medium alone (control) or with addition of IL-1 (500 U/ml), TNF-á (500 U/ml) or 8BrcAMP (0.5 mM).
to 100 ìl luciferase assay buffer (Promega, Madison, WI) and light emission was measured in a luminometer (Berthold, Wildbad, Germany) for 10 s. â-galactosidase assays were performed using the Galacto-light reaction kit from Tropix (Bedford, MA). In short, 20 ìl cell lysate was incubated with 200 ìl of Galacton subtrate (Tropix, Bedford, MA) for 1 h at room temperature. After adding 300 ìl Accelerator (Tropix, Bedford, MA) the light emission was measured in a luminometer (Berthold, Wildbad, Germany). Figure 8 demonstrates the high sensitivity and specificity of this assay system.
Murine LIF promoter regulation / 663
The luciferase free control vector pUC18 only showed background activity . The positive control vector pRSVluc as well as the vector pLIF542 induced a significant luciferase activity measured in relative light units (RLU). In contrast, the control vector pLIF542as, which contained the LIF promoter fragment in antisense orientation, induced a 2 log lower luciferase activity (Fig. 8A). For control of non-specific effects of cytokine incubation on expression of reporter constructs we incubated stromal cells, which were transfected with the positive control vector pRSVluc with different cytokines. As can be seen from Figure 8B, there was no significant non-specific stimulatory effect from IL-1, TNF-á or 8BrcAMP. This rules out a post-transcriptional effect of these stimuli on the luciferase expression.
Electrophoretic mobility shift assays (EMSA) Preparation of nuclear extracts was performed as a modification of the method by Dignam et al.28. Protein concentrations were determined using the Biorad-kit.29 EMSA were performed by incubation of 1 ng 32P-labelled LIF-promoter DNA with 2 ìg nuclear extract in the presence of 1 ìg poly(dI-dC).poly(dI-dC), 10% glycerol and 0.05% NP-40 in 20 ìl binding buffer (10 mM Tris-HCl, pH 7.5), 50 mM NaCl and 0.5 mM DTT) for 20 min at room temperature. After incubation, the samples were loaded onto a 5% polyacrylamide gel and run at a current of 18 mA in gel buffer (7 mM Tris-HCl, pH 7.5, 3 mM sodium acetate and 1 mM EDTA). Gels were dried and autoradiography was performed at 70C.
Acknowledgements This study was supported by a research grant from the ‘‘Deutsche Forschungsgemeinschaft’’ (De 404/2-1).
REFERENCES 1. Gough NM (1992) Molecular genetics of leukemia inhibitory factor (LIF) and its receptor. Growth Factors 7:175–179. 2. Metcalf D (1992) Leukemia inhibitory factor-a puzzling polyfunctional regulator. Growth Factors 7:169–173. 3. Tomida M, Yamamoto Yamaguchi Y, Hozumi M (1984) Purification of a factor inducing differentiation of mouse myeloid leukemic M1 cells from conditioned medium of mouse fibroblast L929 cells. J Biol Chem 59:10978–10982. 4. Gascan H, Godard A, Ferenz C, Naulet J, Praloran V, Peyrat MA, Hewick R, Jacques Y, Moreau JF, Soulillou JP (1989) Characterization and NH2-terminal amino acid sequence of natural human interleukin for DA cells: Leukemia inhibitory factor. J Biol Chem 264:21509–21515. 5. Wang C, Lishner M, Minden MD, McCulloch EA (1990) The effects of leukemia inhibitory factor (LIF) on blast stem cells of acute myeloblastic leukemia. Leukemia 4:548–552. 6. Metcalf D, Hilton D, Nicola N A (1991) Leukemia inhibitory factor can potentiate murine megakaryocyte production in vitro. Blood 77:2150–2153. 7. Lu¨bbert M, Mantovani L, Lindemann A, Mertelsmann R, Herrmann F (1991) Expression of leukemia inhibitory factor is regulated in human mesenchymal cells. Leukemia 5:361–365. 8. Wesselingh SL, Gough NM, Finlay Jones JJ, McDonald PJ (1990) Detection of cytokine mRNA in astrocyte cultures using the polymerase chain reaction. Lymphokine Res 9:177–185. 9. Anegon I, Moreau JF, Godard A, Jacques Y, Peyrat MA, Hallet MM, Wong G, Soulillou JP (1990) Production of human
interleukin for DA cells (HILDA)/leukemia inhibitory factor (LIF) by activated monocytes. Cell Immunol 130:50–65. 10. Gascan H, Anegon I, Praloran V, Naulet J, Godard A, Soulillou JP, Jacques Y (1990) Constitutive production of human interleukin for DA cells/leukemia inhibitory factor (LIF) by human tumor cell lines derived from various tissues. J Immunol 144:2592– 2598. 11. Derigs HG, Boswell HS (1993) LIF mRNA expression is transcriptionally regulated in murine bone marrow stromal cells. Leukemia 7:630–634. 12. Willson TA, Metcalf D, Gough NM (1992) Cross-species comparison of the sequence of the leukaemia inhibitory factor gene and its protein. Eur J Biochem 204:21–30. 13. Stahl J, Gearing DP, Willson TA, Brown MA, King JA, Gough NM (1990) Structural organization of the genes for murine and human leukemia inhibitory factor. Evolutionary conservation of coding and non-coding regions. J Biol Chem 265:8833–8841. 14. Brown MA, Metcalf D, Gough NM (1994) Leukemia inhibitory factor and interleukin 6 are expressed at very low levels in the normal adult mouse and are induced by inflammation. Cytokine 6:300–309. 15. Hamilton JA, Waring PM, Filonzi EL (1993) Induction of leukemia inhibitory factor in human synovial fibroblasts by IL-1 and tumor necrosis factor-á. J Immunol 150:1496–1502. 16. Wetzler M, Estrov Z, Talpaz M, Kim KJ, Alphonso M, Srinivasan R, Kurzrock R (1994) Leukemia inhibitory factor in long-term adherent layer cultures: increased levels of bioactive protein in leukemia and modulation by IL-4, IL-1 beta, and TNFalpha. Cancer Res 54:1837–1842. 17. Miyagi T, Akashi M, Yamato K, Miyoshi I, Koeffler HP (1991) D-Factor: modulation of expression in fibroblasts. Leukemia Res 15:441–451. 18. Anegon I, Grolleau D, Soulillou JP (1991) Regulation of HILDA/LIF gene expression in activated human monocytic cells. J Immunol 147:3973–3980. 19. Stahl J, Gough NM (1993) Delineation of positive and negative control elements within the promoter region of the murine leukemia inhibitory factor (LIF) gene. Cytokine 5:386–393. 20. Hsu LW, Heath JK (1994) Identification of two elements involved in regulating expression of the murine leukeamia inhibitory factor gene. Biochem J 302:103–110. 21. Derigs HG, Burgess GS, Klingberg D, Nahreini TS, Mochizuki DY, Williams DE, Boswell HS (1990) Role for cyclic AMP in the postreceptor control of cytokine-stimulated stromal cell growth factor production. Leukemia 4:471–479. 22. Seelentag WK, Mermod JJ, Montesano R, Vasalli P (1987) Additive effects of interleukin 1 and tumour necrosis factor-á on the accumulation of the three granulocyte and macrophage colonystimulating factor mRNAs in human endothilial cells. EMBO J 6:2261–2265. 23. Aman MJ, Rudolf G, Goldschmitt J, Aulitzky WE, Lam C, Huber C, Peschel C (1993) Type I interferons are potent inhibitors of interleukin-8 production in hematopoetic and bone marrow stromal cells. Blood 82:2371–2378. 24. Derigs HG, Reifel-Miller A, Kaushansky K, Hromas RA, Boswell HS (1994) Granulocyte-macrophage colony-stimulating factor expression is regulated at transcriptional and posttranscriptional levels in a murine bone marrow stromal cell line. Exp Hematol 22:924–932. 25. Nordeen SK (1988) Luciferase reporter gene vectors for analysis of promoters and enhancers. Biotechniques 6:454–458. 26. Weber BJ, Horiguchi R, Luebbers R, Sherman M, Kufe D (1989) Posttranscriptional stabilization of c-fms mRNA by a labile protein during human monocytic differentiation. Mol Cell Biol 9:769–775. 27. Chomczynski P, Sacchi N (1987) Single-step method of RNA isolation by guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159. 28. Dignam JD, Lebovitz RM, Roeder RG (1983) Accurate transcription initiation by RNA polymerase II in a soluble extract from mammalian nuclei. Nucleic Acids Res 1:1475–1489. 29. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254.
REGULATORY ELEMENTS OF THE LEUKAEMIA INHIBITORY FACTOR (LIF) PROMOTER IN MURINE BONE MARROW STROMAL CELLS G. Go¨llner, G. Bug, B. Rupilius, C. Peschel, C. Huber, H. G. Derigs Leukaemia inhibitory factor (LIF) plays an important role as a haematopoietically active cytokine. As described earlier in a murine model, interleukin 1 (IL-1) induced LIF mRNA and protein expression. We utilized the murine cell line +/+-1.LDA11 to further define regulatory mechanisms of LIF expression in bone marrow stromal cells. The production of LIF mRNA is stimulated by IL-1â, TNF-á, and the cAMP analogue 8-bromoadenosine 3 :5 -monophosphate (8BrcAMP). LIF mRNA expression is controlled at the transcriptional level. Different fragments from 542 to 45 bp 5 upstream of the transcriptional start site of the murine LIF gene were fused to the luciferase gene. All LIF-promoter luciferase constructs exhibited constitutive luciferase activity under serum free conditions. The level of luciferase activity decreased with LIF-promoter constructs of less than 249 bp (pLIF249) in size. When tested with the 314 bp LIF-promoter construct, incubation of stromal cells with IL-1â (500 U/ml) resulted in a 1.57-fold stimulation, with TNF-á (500 U/ml) in 2.06-fold stimulation, and with 8BrcAMP (0.5 mM) in a 3.42-fold stimulation of luciferase activity. By testing different deletion mutants we could narrow the IL-1 and TNF-á responsive promoter areas to the region 249 to 145 bp and the 8BrcAMP responsive area from 145 to 82 bp. Mobility shift experiments revealed that nuclear proteins from stromal cells form a DNA-protein complex by binding to the region from 249 to 145 bp of the LIF promoter. 1999 Academic Press
Murine leukaemia inhibitory factor is a glycosylated protein of 58 kDa. This cytokine has pleiotropic actions on haematopoietic cells, inflammation, and embryonic development. LIF also affects other tissues such as bone, liver, and neurons.1,2 It was identified, characterized, purified, and cloned on the basis of its ability to induce the differentiation of the murine myeloid leukaemic cell line M1.3 On haematopoietic cells, LIF has several distinct functions. It has been found to support proliferation of the factordependent myeloid cell line DA-1.4 When used alone, LIF does not stimulate the formation of colonies of any haematopoietic lineage, but it acts synergistically From the Division of Hematology, III. Dept. of Medicine, Johannes Gutenberg University, Langenbeckstrasse 1, D-55131 Mainz, Germany Correspondence to: Dr H. G. Derigs, Division of Hematology, III. Department of Medicine, Johannes Gutenberg-Universita¨t, Langenbeckstr. 1, D-55101 Mainz, Germany Received 21 July 1998; received in revised form 30 November 1998; accepted for publication 23 December 1998 1999 Academic Press 1043–4666/99/090656+08 $30.00/0 KEY WORDS: bone marrow stromal cells/Leukaemia Inhibitory Factor (LIF)/promoter activity/transcriptional control 656
together with other growth factors on the proliferation of haematopoietic progenitor cells.5,6 LIF is produced by monocytes, fibroblasts, astrocytes, different tumour cell lines, and mesenchymal cells.7–10 The regulation of LIF expression is largely unidentified. It is known that the 3 untranslated region of LIF mRNA contains AU-rich motifs. These sequences have been shown in a number of other mRNA species to convey a destabilizing effect. We have previously described the transcriptional regulation of LIF mRNA by inflammatory cytokines IL-1 and TNF-á in murine bone marrow stromal cell lines.11 The first 300 bp of the 5 promoter region display a high degree of cross-species conservation thereby indicating importance.12 Within this region there are four TATA-like elements and a second transcriptional start-site.13 The 5 LIF-promoter region contains binding motifs for the nuclear transcription factors SP-1 and AP-2.13 In the present study, we proved LIF expression in IL-1 stimulated human bone marrow stromal cells. Furthermore, we could narrow the IL-1 and TNF-á responsive areas of the murine LIF promoter to the CYTOKINE, Vol. 11, No. 9 (September), 1999: pp 656–663
Murine LIF promoter regulation / 657
region 249 to 145 bp and the 8BrcAMP responsive area to the region from 145 to 82 bp. Mobility shift experiments revealed that nuclear proteins from stromal cells form a DNA-protein complex by binding to the region from 249 to 145 of the LIF 5 promoter.
A
LIF
28S RNA
RESULTS 18S RNA
Control
IL-1
100 B
LIF protein concentration (pg/ml)
Expression of LIF in human bone marrow stromal cells was assessed at the protein level in culture supernatants by a sensitive ELISA and at the mRNA level by Northern blot analysis. In unstimulated cultures, very low levels of LIF mRNA were detected. After incubation with IL-1 (100 U/ml) for 6 h, significant LIF-mRNA expression was induced (Fig. 1A). These RNA-data were confirmed at the protein level (Fig. 1B). Furthermore, incubation with the cyclic AMP analogue 8BrcAMP (0.5 mM) also induced LIF protein in human bone marrow stromal cell culture supernatants (data not shown). To further define LIF regulation in stromal cells, we used the murine bone marrow stromal cell line +/+-1.LDA11 as a model. Transcriptional induction of LIF mRNA by IL-1, TNF-á and 8BrcAMP had been demonstrated before in these cells.11 A series of different murine LIF promoter fragments were cloned upstream of the luciferase gene and reporter plasmids were transfected into stromal cells. Under serum free conditions, constitutive luciferase expression of different LIF promoter fragments is shown (Fig. 2). The negative control plasmid pLIF542as, which contains the LIF-promoter in antisense orientation, showed no luciferase expression. The construct pLIF314 had a 1.390.08 fold (meanSEM) LIF promoter activity relative to the longest fragment pLIF542, which was used as standard. The activity of pLIF249 was 1.620.11 fold higher than that of pLIF542. Shorter fragments showed a decreased activity relative to pLIF542. The smallest construct (pLIF45) showed only a 0.10 fold activity relative to the fragment pLIF542. In order to investigate the induction of LIF transcription by cytokine incubation, we exposed the transfected bone marrow stromal cells to different stimuli. Stimulation of +/+.1-LDA11 cells, transfected with pLIF314, with IL-1â, TNF-á and the cAMP analogue 8BrcAMP showed an increase of luciferase activity (Fig. 3). Incubation of +/+.1-LDA11 transfected with pLIF314 with IL-1 (500 U/ml) alone resulted in a 1.570.09 fold activity. TNF-á (500 U/ml) stimulated luciferase activity 2.060.12 fold and 8BrcAMP (0.5 mM) 3.420.17 fold above medium control (Fig. 3). Co-incubation of +/+.1-LDA11 with combinations of IL-1, TNF-á and 8BrcAMP resulted in additive
75
50
25
0 Figure 1. cultures.
Medium
IL-1
LIF expression in human bone marrow stromal cell
(A) Induction of LIF-mRNA in human bone marrow stromal cells. Confluent secondary stromal cultures were incubated for 6 h in medium alone or with the addition of IL-1 (100 U/ml). Total cellular RNA was prepared and subjected to Northern-blot analysis with cDNA probe for human LIF. The 18 and 28S rRNA from ethidium bromide stained gel is shown as loading control. A representative of two independent experiments is shown. (B) Confluent secondary stromal cultures were incubated in medium alone or with the addition of IL-1 (100 U/ml). After incubation for 24 h, LIF concentrations in the supernatants were assessed using a specific ELISA. The mean (SEM) of 4 independent experiments is shown.
stimulation of luciferase expression (Fig. 3). Co-incubation with combinations of two stimuli from IL-1, TNF-á or 8BrcAMP resulted in a two- to seven-fold increase of luciferase activity, and co-incubation of all three stimuli resulted in a 10.650.93 fold increase. These data reflect earlier published results from Run On experiments in the same system (Fig. 3). To narrow the LIF promoter region, which is necessary for the stimulatory response, we incubated
658 / Go¨llner et al.
CYTOKINE, Vol. 11, No. 9 (September, 1999: 656–663)
from stromal cells incubated with IL-1â, TNF-á and the cAMP analogue 8-bromoadenosine 3 ,5 monophosphate (8BrcAMP) showed a DNA protein complex with both the 249 and the 145 bp LIFpromoter fragments (data not shown). To investigate the role of the AP-2 consensus sequence of the LIF-promoter further, which was located from 185 to 171 bp upstream of the transcription start site, we used the radiolabelled AP-2 oligonucleotide to detect a specific AP-2 in +/+-1.LDA11 protein extracts. With unstimulated and stimulated protein extract from +/+-1.LDA11 stromal cells, there was no DNA–protein complex detectable. In addition, the DNA–protein complex from radiolabelled fragment LIF249 with nuclear extract from stromal cells was not abolished by addition of a 30-fold excess of the AP-2 oligonucleotide (Fig. 6).
1.8 *** 1.6 ***
Relative luciferase activity
1.4
1.2
1
0.8 *** 0.6 *** 0.4
0.2
***
Figure 2.
45 IF pL
IF
82
5 pL
IF
14
9 pL
pL
IF
24
4 31 pL
IF
54 IF pL
pL
IF
54
2a
2
s
0
Unstimulated LIF-promoter fragment activity.
Relative luciferase activities in +/+-1.LDA11 stromal cells transfected with various LIF-promoter fragments. 48 h after start of transfection, cells were harvested and luciferase activity and â-galactosidase activity were determined in cytoplasmic extracts as described in Materials and Methods. Normalization for transfection efficacy was done by â-galactosidase activity. Mean (MW) and standard error (SEM) from twelve independent experiments are shown. The luciferase activity of the different fragments were plotted relative to the plasmid pLIF542. ***Significantly different luciferase activity from pLIF542, P <0.0001).
stromal cells, transfected with different reporter gene constructs, and used different stimuli. LIF-promoter fragments longer than 145 bp showed an increase of luciferase activity after IL-1 incubation (Fig. 4). Similar results were obtained after TNF-á incubation. In contrast, incubation with 8BrcAMP also stimulated LIF-promoter constructs larger than 82 bp. Gel mobility shift assays using radiolabelled LIFpromoter fragments LIF249 and LIF145 were performed to detect proteins that specifically bind to LIF-promoter regions. Under unstimulated conditions, one DNA–protein complex was detected with the fragment LIF249 (Fig. 5). The complex was abolished by the addition of the same unlabelled LIF-promoter fragment but not by a 30-fold excess of an unrelated oligonucleotide proving the specificity of the DNA binding. With fragment LIF145 no DNA-protein complex under unstimulated conditions was detectable (Fig. 5). Gel mobility shift assays with nuclear extract
DISCUSSION LIF is a cytokine with various actions on haematopoietic progenitor cells. The LIF gene is constitutively expressed at very low levels in the normal adult mouse.14 The expression of LIF in different cell types is induced by inflammatory mediators such as TNF-á and IL-1.7,15,16 Miyagi et al.17 and Anegon et al.18 reported recently that post-transcriptional mechanisms play a role in the induction of LIF expression in LPS-stimulated human monocytes and TPAstimulated human fibroblasts. Endothelial cells seem to be the main source of LIF expression in the bone marrow microenvironment.19 In the present study we proved LIF induction by IL-1 at the protein and mRNA level in human long term bone marrow culture cells. This data implies a physiological role of LIF in haematopoietic regulation. Our group showed previously that LIF mRNA expression is controlled at the transcriptional level in murine bone marrow stromal cells. Furthermore, LIF mRNA was shown to be highly unstable, with a half life of only 30 min.11 In the current study, we delineated active LIF promoter elements by using a sensitive reporter gene assay in a murine bone marrow stromal cell line. Different LIF-promoter fragments showed activity in unstimulated bone marrow stromal cells. Stahl and Gough19 described negative and positive control elements within the promoter region in the LIF gene, when transfected with different CAT constructs in human STO fibroblasts. The negative control element was located between 360 and 249 bp, and the positive regulatory element was located between 860 and 660 bp. In contrast to this data, we could not locate a negative regulatory element in the murine LIF promoter. In our system fragments spanning the
Murine LIF promoter regulation / 659
Run on
Luciferase assay -fold stimulation
LIF
IL-6
CHO-B
}
+/+ –1.LDA11
0
5
10
15
Control
IL-1
TNF
8BrcAMP
IL-1 + 8BrcAMP
TNF + 8BrcAMP
IL-1 + TNF
IL-1 + TNF + 8BrcAMP Figure 3.
Induction of LIF-transcription.
6
4
2
Figure 4.
45 IF pL
IF
82
5 pL
IF
14
9 pL
pL
IF
24
4 31 IF pL
IF
54
2
0
pL
region 314 to 249 bp of the LIF promoter had greater luciferase activity than the control plasmid pLIF542 (542 to +36 bp). Hsu and Heath20 showed in a stably transfected murine fibroblast cell line that a regulatory enhancer element is also located in the far 5 region of the LIF gene between 3200 and 1200 bp.28 Incubation of stromal cells with IL-1, TNF-á, and 8BrcAMP, which were transfected with LIF-promoter constructs increased the luciferase expression. These inducing effects were shown to be additive. These data confirm earlier published nuclear Run On results.11 Of note is the fact that in the Run ON experiments IL-1 resulted in the strongest induction of LIF expression, while in reporter gene assays 8BrcAMP resulted in the strongest induction (Fig. 4). The cause for this discrepancy might be the difference in incubation time for the two assays. Co-incubation with different stimuli resulted, in both assay systems, in more than additive
Relative luciferase activity
Comparison of nuclear transcription rates and luciferase activity of the LIF-promoter luciferase plasmid pLIF314 in unstimulated and stimulated +/+-1.LDA11 stromal cells. Stromal cells were incubated for 2 h (Run-on) or 24 h (Luciferase assay) with IL-1 (500 U/ml), TNF-á (500 U/ml) and 8BrcAMP (0.5 mM) or with combinations of these stimuli. Nuclear extracts were collected and Run On experiments were performed as described in Materials and Methods. The housekeeping gene CHO-B is shown as a positive control and IL-6 is shown as a differently regulated cytokine. Normalization for transfection efficacy in the reporter assay was done by â-galactosidase activity. Mean and standard error (SEM) from six independent experiments are shown in this figure. The luciferase activity of the different stimuli were plotted relative to the unstimulated plasmid pLIF314, which was assigned a value of 1.
Stimulation of LIF-promoter fragments.
Stimulation of the LIF-promoter luciferase plasmids by: (– –) IL-1; (– –), TNF-á; (– –), 8BrcAMP; and (— —), medium. +/+-1.LDA11 stromal cells were incubated for 24 h with IL-1 (500 U/ml), TNF-á (500 U/ml) and 8BrcAMP (0.5 mM). Normalization for transfection efficacy was done by â-galactosidase activity. Mean (MW) and standard error (SEM) from six independent experiments are shown in this figure. The luciferase activity is shown relative to the unstimulated plasmid pLIF542.
660 / Go¨llner et al.
CYTOKINE, Vol. 11, No. 9 (September, 1999: 656–663)
A
B LIF249 1
Figure 5.
2
LIF145 3
4
1
2
3
4
Nuclear protein binding to different LIF-promoter fragments in unstimulated bone marrow stromal cells.
Electrophoretic mobility shift analysis of nuclear extracts prepared from unstimulated +/+-1.LDA11 stromal cells. Gel mobility shift using radiolabelled LIF-promoter fragments (A) LIF249 and (B) LIF145. Lane 1, labelled LIF-promoter DNA but no extract; lane 2, labelled LIF-promoter DNA and nuclear extracts prepared from unstimulated +/+-1.LDA11 stromal cells; lane 3, same as lane 2 but unlabelled LIF-promoter DNA (30-fold molar excess of unlabelled DNA to labelled); lane 4, same as lane 2 but unlabelled oligonucleotide Oct-1 (30-fold molar excess of unlabelled DNA to labelled).
stimulation. Similar synergistic stimulatory effects of IL-1 and TNF-á are known from the expression of other cytokines, such as GM-CSF.21,22 By using reporter plasmids with different promoter fragments in stimulated stromal cells the IL-1 and TNF-á responsive area could be narrowed to the region from 249 to 145 bp of the LIF gene while the 8BrcAMP responsive area was located from 145 to 82 bp. An AP-2 consensus region is located in the region of the TNF-á and IL-1 responsive elements. However, mobility shift assays neither demonstrated AP-2 in protein extracts from the murine stromal cell line nor could the binding of proteins to the active LIF promoter region be competed by an AP-2 consensus oligonucleotide. Thus it seems unlikely that AP-2 plays an active role in LIF activation in murine bone marrow stromal cell lines. Further studies are currently underway to identify the protein which interacts with the LIF-promoter in murine bone marrow stromal cells.
MATERIAL AND METHODS Materials rh-IL-1á was kindly provided by Hoffmann La Roche (Nutley, NJ, USA). Murine IL-1â was a gift from Imunex (Seattle, Washington). Murine Tumor-Necrosis Factor-á was purchased from Genzyme (Kent, England). 8-bromoadenosine 3 :5 -monophosphate (8BrcAMP) was purchased from Sigma (Mu¨nchen, Germany). The AP-2 binding oligonucleotide and the AP-2 containing protein extract were purchased from Promega (Madison, WI, USA). The plasmid pLIF20.2 was kindly provided by Donald Metcalf (Walter and Eliza Hall Institute, Melbourne, Australia).
Cell culture Human adherent bone marrow stromal cell cultures were essentially obtained as previously described.23 Bone marrow mononuclear cells, separated by centrifugation over Ficoll-Hypaque, were incubated at a cell density of 1106/
Murine LIF promoter regulation / 661
1
2
3
4
5
6
7
8
9
10
A TATA-Box
AP-2 TATA- 2.TSS
TATA-
TATA-
SP-1 CAAT
Box
Box
Box
–540 –260
–220 –190 –170
–120
–100
–80
B
–60
–40
1.TSS
–20
+1
pLIF542as
–542 pLIF542 –542 pLIF314 –314 pLIF249 –249 pLIF145 –145 pLIF82 –82 pLIF45 Figure 6. Nuclear protein binding to different LIF-promoter fragments in stimulated bone marrow stromal cells. Electrophoretic mobility shift analysis of nuclear extracts prepared from +/+-1.LDA11 stromal cells. Gel mobility shift using radiolabelled LIF-promoter fragment (A) LIF249, (B) LIF145. Lane 1, labelled LIF-promoter DNA but no extract; lane 2, labelled LIFpromoter DNA and nuclear extracts prepared from unstimulated +/+-1.LDA11 stromal cells; lane 3, labelled LIF-promoter DNA and nuclear extracts prepared from IL-1-stimulated (500 U/ml) +/+1.LDA11 stromal cells; lane 4, labelled LIF-promoter DNA and nuclear extracts prepared from TNF-á-stimulated (500 U/ml) +/+1.LDA11 stromal cells; lane 5 labelled LIF-promoter DNA and nuclear extracts prepared from 8BrcAMP-stimulated (0.5 mM) +/+1.LDA11 stromal cells; lane 6, same as lane 2 but unlabelled LIF-promoter DNA (30-fold molar excess of unlabelled DNA as labelled).
ml in 25 cm2 tissue culture flasks at 33C in culture medium consisting of RPMI 1640 supplemented with 10% fetal calf serum (FCS), 10% horse serum, 1.0 mM hydrocortisone (Sigma), and the additives as described.23 At weekly intervals, cultures were fed by replacing 75% of the culture medium. When cultures were covered more than 80% by adherent cells, the primary cultures were treated with trypsinEDTA (Biochrom, Berlin, Germany). Detached stromal cells were pooled and expanded in new culture flasks at a surface ratio of 1:5 in 75-cm2 tissue flasks. The adherent cells were incubated under the same culture conditions until the cultures became confluent again. By this culture method, homogeneous stromal cell layers were obtained that were morphologically and functionally comparable in different culture flasks. Hydrocortisone was removed from these cultures at least 3 days before use for RNA and protein analysis. LIF was measured in the culture supernatants of stromal cells after incubation for 24 h with various factors, as described in the Result section. The continuous clonal marrow adherent cell line +/+1.LDA11 was kindly provided by Dr Scott Boswell (IU Medical School, Indianapolis, IN). The cell line was estab-
–45 Figure 7.
LIF-promoter.
(A) Regulatory elements of the murine 5 -LIF promoter (TSS=transcription start site). (B) Design of the LIF-promoter constructs. DNA fragments comprising different portions of the 5 -flanking region of the LIF gene were subcloned in the luciferase expression vector PxP2.
lished from haematopoietically inactive long term bone marrow culture of WCB6F1-+/+ mice as has previously been described.24 The cell line was maintained in McCoy’s 5A medium supplemented with 10% FCS, 100 U/ml penicillin, and 100 ìg/ml streptomycin (Serva, Heidelberg, Germany).
Plasmids The LIF plasmids LIF249 and LIF145 were constructed by cloning polymerase chain reaction (PCR) products into a plasmid vector (Invitrogen, San Diego, CA, USA). All LIF luciferase-promoter fragments were constructed by subcloning fragments into HindIII/XhoI sites of the PxP2 vector.25 pLIF542 contains bp 542 to +35 of the 5 LIF promoter in the luciferase expression vector PxP2. The negative control plasmid PLIF542as contains the same fragment in antisense orientation in the luciferase expression vector PxP1. Plasmid pLIF314 contains bp 314 to +35, the plasmid pLIF249 contains bp 249 to +35 and the plasmid pLIF145 contains bp 145 to +35 of the 5 LIF promoter. The plasmid pLIF82 was constructed by a restriction digest with the enzyme BamHI and self-ligation of the vector. This Plasmid contains bp 82 to +35. The plasmid pLIF45 contains bp 45 to +35 of the 5 LIF promoter. Figure 7 shows the different LIF promoter fragments in relation to known function elements or consensus sequences. The plasmid RSVluc contained the luciferase cDNA controlled by the constituitively active promoter of the Rous Sarcoma Virus (RSV).
662 / Go¨llner et al.
CYTOKINE, Vol. 11, No. 9 (September, 1999: 656–663)
The plasmid RSVâ-gal contains a 5.6 kb galactosidase insert which is controlled by a RSV promoter.
A 1E + 006
Northern blot analysis Total cytoplasmic RNA was prepared using the single step method of guadinium/phenol-chloroform extraction as described earlier.27 10–15 ìg of RNA (depending on the least yield obtained in each experiment) were subjected to electrophoresis on a 1% agarose-formaldehyde gel and transferred onto nylon membrane (Hybond-N, Amersham Buchler). Blots were hybridized to á32P-labelled cDNA probes using random primer DNA labelling kit (Boehringer, Mannheim, Germany), washed and exposed to Cronex-4 autoradiography films (DuPont) at 70C. hu-LIF cDNA, a 600-bp BamHI-HindIII fragment cloned into pIC2ORvector, was obtained from Dr Mo¨vva (Sandoz, Basel).
Enzyme-linked immunoabsorbent assay (ELISA) Concentrations of LIF were measured using Biotra Human Cytokine Immunoassays (Amersham) according to the manufacturer’s instructions.
Reporter gene assays 2105 cells in 6-well plates were transfected with 1 ìg luciferase-plasmid, 0,25 ìg RSVâ-gal plasmid and 3 ìl of Lipofectamine reagent (Life Technologies, Inc., Gaitherburg, Germany) in 1 ml serum free medium (HL1; Ventrex, Portland, ME). After incubation for 3 h cells were washed with 1 phosphate-buffered saline (PBS) and 2 ml McCoy’s 5A medium with 10% FCS were added for 20 h. At 24 h after the start of transfection, the medium was replaced and cells were cultured under serum free conditions for 24 h. Fortyeight hours after the start of transfection cells were harvested for luciferase- and â-galactosidase assays. Cells were washed with 1PBS and were lysed for 15 min at room temperature with 200 ìl/well 1 reporter lysis buffer (Promega, Madison, WI). Cells were scraped with a rubber policeman and centrifuged at 14 000 g for 2 min at 4C. 20 ìl cell lysate was added
100 000
10 000
1000
100
10 pUC 18
RSVluc
pLIF542
pLIF542as
Control
IL-1
TNF
8BrcAMP
B 4.5 4 Relative luciferase activity
Run-on experiments were performed according to a modified method of Weber et al.26. Briefly, confluent cell layers (108 cells per treatment) were washed with ice cold PBS and the nuclei isolated by lysis in 0.5% NP-40 containing buffer. 108 nuclei were resuspended in 100 ml of 40% glycerol containing buffer and incubated in an equal volume of reaction buffer containing 5 mM MgCl2, 100 mM KCl, 1 mM of each ATP, CTP, GTP and 100 ìCi [á-32P]UTP (800 Ci/mM, Amersham Life Science, UK) at 26C for 30 min. The reaction was stopped by the addition of 200 ìl HSB buffer (0.5 M NaCl, 50 mM MgCl2, 2 mM CaCl2, 10 mM Tris, pH 7.4) and 2 ìl DNase I, 30 ìl RNase inhibitor and 24 ìg t-RNA. This mixture was allowed to incubate at 37C for 30 min. The 32P-labelled, nuclear RNA elongated in vitro was isolated by the method of Chomczynski and Sacchi27 and an equal number of cpm from each probe was hybridized at 65C against slot blots of denatured plasmids containing cDNA for cytokines of interest or control plasmids for 48 h. Slot blots were washed in 2SSC at 65C and then treated with RNase at 10 mg/ml and autoradiographed.
Luciferase activity (RLU)
Nuclear run-on transcription assay
3.5 3 2.5 2 1.5 1 0.5 RSVluc
Figure 8.
Sensitivity and specificity of the luciferase assay.
(A) Luciferase activity, measured in relative light units (RLU), was determined in +/+-1.LDA11 stromal cells transfected with various vectors as labelled. pUC18 was used as a negative control vector without luciferase cDNA. pRSVluc was used as a positive control. pLIF542 contained the LIF promoter sequence in sense orientation and pLIF542 as the same sequence in antisense orientation 5 of luciferase cDNA. 48 h after transfection cells were harvested and luciferase activity was determined in cytoplasmic extracts as described in Materials and Methods. (B) Luciferase activity of pRSVluc transfected +/+-1.LDA11 stromal cells after incubation for 24 h with either medium alone (control) or with addition of IL-1 (500 U/ml), TNF-á (500 U/ml) or 8BrcAMP (0.5 mM).
to 100 ìl luciferase assay buffer (Promega, Madison, WI) and light emission was measured in a luminometer (Berthold, Wildbad, Germany) for 10 s. â-galactosidase assays were performed using the Galacto-light reaction kit from Tropix (Bedford, MA). In short, 20 ìl cell lysate was incubated with 200 ìl of Galacton subtrate (Tropix, Bedford, MA) for 1 h at room temperature. After adding 300 ìl Accelerator (Tropix, Bedford, MA) the light emission was measured in a luminometer (Berthold, Wildbad, Germany). Figure 8 demonstrates the high sensitivity and specificity of this assay system.
Murine LIF promoter regulation / 663
The luciferase free control vector pUC18 only showed background activity . The positive control vector pRSVluc as well as the vector pLIF542 induced a significant luciferase activity measured in relative light units (RLU). In contrast, the control vector pLIF542as, which contained the LIF promoter fragment in antisense orientation, induced a 2 log lower luciferase activity (Fig. 8A). For control of non-specific effects of cytokine incubation on expression of reporter constructs we incubated stromal cells, which were transfected with the positive control vector pRSVluc with different cytokines. As can be seen from Figure 8B, there was no significant non-specific stimulatory effect from IL-1, TNF-á or 8BrcAMP. This rules out a post-transcriptional effect of these stimuli on the luciferase expression.
Electrophoretic mobility shift assays (EMSA) Preparation of nuclear extracts was performed as a modification of the method by Dignam et al.28. Protein concentrations were determined using the Biorad-kit.29 EMSA were performed by incubation of 1 ng 32P-labelled LIF-promoter DNA with 2 ìg nuclear extract in the presence of 1 ìg poly(dI-dC).poly(dI-dC), 10% glycerol and 0.05% NP-40 in 20 ìl binding buffer (10 mM Tris-HCl, pH 7.5), 50 mM NaCl and 0.5 mM DTT) for 20 min at room temperature. After incubation, the samples were loaded onto a 5% polyacrylamide gel and run at a current of 18 mA in gel buffer (7 mM Tris-HCl, pH 7.5, 3 mM sodium acetate and 1 mM EDTA). Gels were dried and autoradiography was performed at 70C.
Acknowledgements This study was supported by a research grant from the ‘‘Deutsche Forschungsgemeinschaft’’ (De 404/2-1).
REFERENCES 1. Gough NM (1992) Molecular genetics of leukemia inhibitory factor (LIF) and its receptor. Growth Factors 7:175–179. 2. Metcalf D (1992) Leukemia inhibitory factor-a puzzling polyfunctional regulator. Growth Factors 7:169–173. 3. Tomida M, Yamamoto Yamaguchi Y, Hozumi M (1984) Purification of a factor inducing differentiation of mouse myeloid leukemic M1 cells from conditioned medium of mouse fibroblast L929 cells. J Biol Chem 59:10978–10982. 4. Gascan H, Godard A, Ferenz C, Naulet J, Praloran V, Peyrat MA, Hewick R, Jacques Y, Moreau JF, Soulillou JP (1989) Characterization and NH2-terminal amino acid sequence of natural human interleukin for DA cells: Leukemia inhibitory factor. J Biol Chem 264:21509–21515. 5. Wang C, Lishner M, Minden MD, McCulloch EA (1990) The effects of leukemia inhibitory factor (LIF) on blast stem cells of acute myeloblastic leukemia. Leukemia 4:548–552. 6. Metcalf D, Hilton D, Nicola N A (1991) Leukemia inhibitory factor can potentiate murine megakaryocyte production in vitro. Blood 77:2150–2153. 7. Lu¨bbert M, Mantovani L, Lindemann A, Mertelsmann R, Herrmann F (1991) Expression of leukemia inhibitory factor is regulated in human mesenchymal cells. Leukemia 5:361–365. 8. Wesselingh SL, Gough NM, Finlay Jones JJ, McDonald PJ (1990) Detection of cytokine mRNA in astrocyte cultures using the polymerase chain reaction. Lymphokine Res 9:177–185. 9. Anegon I, Moreau JF, Godard A, Jacques Y, Peyrat MA, Hallet MM, Wong G, Soulillou JP (1990) Production of human
interleukin for DA cells (HILDA)/leukemia inhibitory factor (LIF) by activated monocytes. Cell Immunol 130:50–65. 10. Gascan H, Anegon I, Praloran V, Naulet J, Godard A, Soulillou JP, Jacques Y (1990) Constitutive production of human interleukin for DA cells/leukemia inhibitory factor (LIF) by human tumor cell lines derived from various tissues. J Immunol 144:2592– 2598. 11. Derigs HG, Boswell HS (1993) LIF mRNA expression is transcriptionally regulated in murine bone marrow stromal cells. Leukemia 7:630–634. 12. Willson TA, Metcalf D, Gough NM (1992) Cross-species comparison of the sequence of the leukaemia inhibitory factor gene and its protein. Eur J Biochem 204:21–30. 13. Stahl J, Gearing DP, Willson TA, Brown MA, King JA, Gough NM (1990) Structural organization of the genes for murine and human leukemia inhibitory factor. Evolutionary conservation of coding and non-coding regions. J Biol Chem 265:8833–8841. 14. Brown MA, Metcalf D, Gough NM (1994) Leukemia inhibitory factor and interleukin 6 are expressed at very low levels in the normal adult mouse and are induced by inflammation. Cytokine 6:300–309. 15. Hamilton JA, Waring PM, Filonzi EL (1993) Induction of leukemia inhibitory factor in human synovial fibroblasts by IL-1 and tumor necrosis factor-á. J Immunol 150:1496–1502. 16. Wetzler M, Estrov Z, Talpaz M, Kim KJ, Alphonso M, Srinivasan R, Kurzrock R (1994) Leukemia inhibitory factor in long-term adherent layer cultures: increased levels of bioactive protein in leukemia and modulation by IL-4, IL-1 beta, and TNFalpha. Cancer Res 54:1837–1842. 17. Miyagi T, Akashi M, Yamato K, Miyoshi I, Koeffler HP (1991) D-Factor: modulation of expression in fibroblasts. Leukemia Res 15:441–451. 18. Anegon I, Grolleau D, Soulillou JP (1991) Regulation of HILDA/LIF gene expression in activated human monocytic cells. J Immunol 147:3973–3980. 19. Stahl J, Gough NM (1993) Delineation of positive and negative control elements within the promoter region of the murine leukemia inhibitory factor (LIF) gene. Cytokine 5:386–393. 20. Hsu LW, Heath JK (1994) Identification of two elements involved in regulating expression of the murine leukeamia inhibitory factor gene. Biochem J 302:103–110. 21. Derigs HG, Burgess GS, Klingberg D, Nahreini TS, Mochizuki DY, Williams DE, Boswell HS (1990) Role for cyclic AMP in the postreceptor control of cytokine-stimulated stromal cell growth factor production. Leukemia 4:471–479. 22. Seelentag WK, Mermod JJ, Montesano R, Vasalli P (1987) Additive effects of interleukin 1 and tumour necrosis factor-á on the accumulation of the three granulocyte and macrophage colonystimulating factor mRNAs in human endothilial cells. EMBO J 6:2261–2265. 23. Aman MJ, Rudolf G, Goldschmitt J, Aulitzky WE, Lam C, Huber C, Peschel C (1993) Type I interferons are potent inhibitors of interleukin-8 production in hematopoetic and bone marrow stromal cells. Blood 82:2371–2378. 24. Derigs HG, Reifel-Miller A, Kaushansky K, Hromas RA, Boswell HS (1994) Granulocyte-macrophage colony-stimulating factor expression is regulated at transcriptional and posttranscriptional levels in a murine bone marrow stromal cell line. Exp Hematol 22:924–932. 25. Nordeen SK (1988) Luciferase reporter gene vectors for analysis of promoters and enhancers. Biotechniques 6:454–458. 26. Weber BJ, Horiguchi R, Luebbers R, Sherman M, Kufe D (1989) Posttranscriptional stabilization of c-fms mRNA by a labile protein during human monocytic differentiation. Mol Cell Biol 9:769–775. 27. Chomczynski P, Sacchi N (1987) Single-step method of RNA isolation by guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159. 28. Dignam JD, Lebovitz RM, Roeder RG (1983) Accurate transcription initiation by RNA polymerase II in a soluble extract from mammalian nuclei. Nucleic Acids Res 1:1475–1489. 29. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254.