WO1988005083A1

WO1988005083A1 - Method of inducible gene expression

Info

Publication number: WO1988005083A1
Application number: PCT/US1987/003428
Authority: WO
Inventors: David Baltimore; Ranjan Sen
Original assignee: Whitehead Institute For Biomedical Research
Priority date: 1986-12-24
Filing date: 1987-12-23
Publication date: 1988-07-14

Abstract

Inducible gene expression can be obtained by placing a gene of interest under regulation of an enhancer sequence. Expression of the gene can be induced by stimulating production of enhancer binding factor.

Description

METHOD OF INDUCIBLE GENE EXPRESSION

Background

Techniques for expression of a desired protein in eukaryotic cells generally do not allow for transient expression of a protein. In eukaryotic expression systems, the gene of interest is placed under the regulation of a strong promoter, which provides for continuous expression of the gene. For expression of genes encoding immunoglobulin chains, appropriate enhancer sequences are often included.

In certain circumstances it can be desirable to transiently rather than continuously express a protein in the cell. For example, when a cell is transfected to produce a protein which is cytotoxic, it would be advantageous to be able to determine when a cell would express the protein. In this way, cells can be grown to high density and then induced to produce the toxic protein.

Summary of the Invention The invention pertains to a method of providing for inducible expression of a gene. The method is based upon the discovery that NF-KB, a transcriptional regulatory factor which interacts with a defined site in the K immunoglobulin enhancer and which is constitutively produced only in eukaryotic cells that transcribe light immunoglobulin chains, can be induced in eukaryotic cells where it is not constitutively present. For example, the factor can be induced in pre-B cells by stimulation with bacterial lipopolysaccharide (LPS). The induction involves a post-translational activation and the combined action of LPS and cycloheximide generates a superinduction. An active phorbol ester also induces this factor, (via protein kinase C). The kinetics of induction by an active phorbol ester are more rapid than those for LPS stimulation. Moreover, phorbol ester-mediated induction of NF-KB was observed in a T cell line (Jurkat) and a non-lymphoid cell line (HeLa) and is therefore not restricted to B-lymphoid cells.

According to the method of this invention, the enhancer sequence or at least the portion of the enhancer sequence to which the NF-KB factor binds (i.e. the "B" site, See Baltimore and Sen, Cell 1986) is linked to a promotor-structural gene of interest so that transcription of the gene is placed under the influence of the enhancer sequence. This provides an inducible gene construct. A eukaryotic cell is transfeeted with the gene construct. At a desired time the transfected cell is exposed to the appropriate inducer of NF-KB, such as a mitogen (e.g., LPS or PHA) or a protein kinase C activator (e.g., a phorbol ester such as phorbol 12-myristate-13-acetate, PMA) or combinations thereof.

The method of this invention can be used to turn on enhanced production of a protein in a genetically transfected cell when desired. To optimize the production of proteins which are cytotoxic, such as products of certain oncogenes, as mentioned above, it may be desirable to overexpress the protein in the cell only transiently. Cells transfected with genes under regulation of the NF-K B factor can be induced to express the gene at an appropriate time upon exposure to an inducer of NF-KB.

Brief Description of the Figures Figure 1 shows the electrophoretic mobility shift analysis of (A) extracts derived from 10Z/3 cells before and after simulation with bacterial lipopolysaccharide (LPS) and (B) extracts derived from PD, an Abelson murine leukemia virus transformed pre-B cell line before and after stimulation with LPS.

Figure 2 shows the effect (A) of cycloheximide on LPS stimulation of 70Z/3 cells and (B) of anisemycin on LPS stimulation of 70Z/3 cells. Figure 3 shows the effect of phorbol

12-myristate-13-acetate (PMA) on NF-KB in 70Z/3 cells.

Figure 4 shows the induction of NF-KB in human T lymphoma cells and in HeLa cells.

Detailed Description of the Invention

The NF-KB factor interacts with a site in the immunoglobulin gene enhancer (the B site; Sen and Baltimore, Cell, 1986) and its presence in nuclear extracts had previously only been evident in B cells and plasma cells. It has now been discovered that NF-KB induction can take place in cells where it is not constitutively present. (i) NF-KB factor can be induced by the mitogen lipopolysaccharide (LPS) in two cell lines representing a pre-B stage of B cell differentiation. (ii) Induction of this factor involves a post-translational modification of a pre-existing protein because the induction takes place even in the presence of translational inhibitors like cycloheximide and anisomycin. (iii) These translational inhibitors by themselves can at least partially induce NF-KB and synergize with LPS to produce a superinduction. (iv) An active phorbol ester like PMA can induce NF-KB by itself, and the time-course of this activation is more rapid than that with LPS alone. (v) It is also possible to induce this factor in cell lines other than those having a pre-B phenotype by means of an appropriate stimulus (e.g. in the human T cell line, Jurkat, by PHA and/or PMA or in HeLa cells with PMA). Thus , B cells and plasma cells appear to support constitutive presence of this factor, whereas in other cell types it can be induced transiently by an appropriate stimulus. The method of this invention is based upon the inducibility of the NF-KB factor in eukaryotic cells. This phenomenon can be exploited to provide for the transient overexpression of a gene product produced by a transfected gene in a eukaryotic cell at a chosen time. According to the method of this invention, a gene of interest is placed under the influence of the K-enhancer sequence containing the binding site for NF-KB. The entire enhancer sequence or a portion containing at least the NF-KB binding site can be employed. The K-enhancer sequence is linked to a structural gene of interest to provide an inducible gene regulatable by NF-KB. A gene construct is thus provided comprising i) a K-enhancer sequence or a portion of the K-enhancer sequence containing at least the sequence to which the factor NF-KB binds; ii) a promoter; and iii) a structural gene of interest.

Conventional recombinant DNA techniques can be used to prepare the construct. The K -enhancer sequence can be obtained from lymphoid cells which express the K-light chain. The K-enhancer can also be obtained from clones containing the sequence. See e.g., U.S. Patent Application Serial No. 817,441, filed January, 9, 1986. The construct can be prepared in or inserted into a transfection vehicle such as a plasmid.

The structural gene can be any gene or gene segment which encodes a useful protein for which transient overexpression is desired. Such proteins would normally be those that are damaging to cells when produced constitutively.

The structural gene can be used with its endogenous promoter or other eukaryotic promoter. Cells for transfection can be any eukaryotic cells used for the expression of eukaryotic proteins. Transfection procedures, such as the calcium precipitation technique and electroporation, are well known in the art.

At the desired time, the transfected cells can be stimulated with the appropriate inducer in an amount sufficient to induce production of NF-KB factor and consequently induce expression of the structural gene. The preferred inducer is a phorbol ester which acts rapidly and directly to activate protein kinase C and induces production of NF-KB. If the transfected cell is a lymphoid cell (e.g., B cell) which is responsive to a mitogen such as LPS or PHA, the mitogen may be used alone or in combination with phorbol ester. The expressed gene product can then be isolated/purified by standard techniques.

The invention is illustrated further by the following exemplification.

Exemplification

Experimental Procedures Cell lines and Extracts: 70Z/3 and PD cells were grown in RPMI 1640 medium supplemented with 10% inactivated fetal calf serum, 50 uM β-mercaptoethanol and penicillin and streptomycin sulfate (pen-strep) antibiotics. LPS (GIBCO) stimulation was carried out with 10-15 ug/ml. For experiments using protein synthesis inhibitors and LPS, cell cultures were treated with inhibitors approximately 20 minutes prior to addition of LPS. Cycloheximide (Sigma) was used to 10 ug/ml, which causes greater than 95% inhibition of protein synthesis in 70Z/3 cells (Wall, R., et al., Proc. Natl. Acad. Sci. USA 83:295-298, (1986). Anisomycin (Sigma) was used at 10 ug, which causes approximately 99% inhibition of protein synthesis in HeLa cells (Grollman, A. P., J. Biol. Chem. 242:3226-3233, 1967). Phorbol ester activation of 70Z/3 cells was carried out using the active ester phorbol 12-myristate-13-acetate (PMA) or the inactive ester phorbol 12, 13-didecanoate at a concentration of 25 ng/ml for the times indicated in the text. All treatments were carried out at cell densities varying between 5X10⁵-10⁶ cells/ml. Jurkat cells were grown in RPMI 1640 medium with 10% inactivated fetal calf serum and pen-strep antibiotics. Phytohemagglutinin (PHA) treatment was done at 5 ug/ml and PMA treatment at 50 ng/ml. HeLa cells were grown in MEM medium with 5% horse serum and pen-strep antibiotics. Phorbol ester (PMA treatment was at 50 ng/ml with cell density varying between 7X10⁵-10⁶ cells/ml.

Nuclear extracts were generated essentially according to the protocol of Dignam, J.D. et al., Nucl. Acids Res. 11:1475-1489 (1983) and protein concentration was determined using a Bradford assay with serum albumin standards.

Gel Binding Analysis: Gel binding analyses were carried out as described earlier using a radioactive Ddel to Haelll fragment (K-3) derived from the K-enhancer (Sen and Baltimore, 1986). Levels of NF-KB induced by various stimuli were normalized to total protein present in the extracts. Further, analysis with a different fragment that contains a binding site for the ubiquitous factor NF-A, shows that this nuclear protein remains at approximately constant levels in all of the extracts reported here. Thus, the modulation of NF-KB activity is not a reflection of variability of nuclear factors in general under these conditions. For competition experiments, the specific and non-specific competitors DNA's were included in the mixture (in amounts shown in Figure 4C) prior to addition of the protein. The competitor fragments u300, u400, KE and SV40E which have been described earlier (Sen and Baltimore, 1986) were isolated from low melting point agarose gels and quantitated by spotting onto ethidium bromide-containing agarose plates.

NF-KB can be induced in pre-B cell lines with bacterial lipopolysaccharide

To examine whether NF-KB factor might be inducible in 70Z/3 cells, cells were stimulated with LPS for 20 hours and nuclear extracts derived from these cells were assayed for the presence of NF-kB factor, using the electrophoretic mobility shift assay described previously. Singh, H. et al., Nature, 319:154-158 (1986). U.S. Patent Application Serial No. 817,441, the teachings of which are incorporated by reference herein. To assay for NF-KB, a DNA fragment containing its binding site (K3 fragment; Sen and Baltimore, supra,) was end-labelled and incubated with extracts derived from either unstimulated 70Z/3 (11s (Figure 1A, lanes 2,3) or LPS-stimulated 70Z/3 cells (Figure 1A, lanes 4,5) in the presence of increasing amounts of the carrier poly d(IC). Binding reactions were carried out for 15-30 minutes at room temperature in a final volume of 15 ul containing 9 ug of total protein, 3.5 ug (lanes 2,4) or 4.5 ug (lanes 3,5) of nonspecific carrier DNA poly d(IC) and 0.2-0.5 ug of probe. Reaction products were fractionated by electrophoresis through low ionic strength polyacrylamide gels and visualized by autoradiography. Lane 1: free DNA fragments; lane 6: nucleoprotein complex generated by interaction of NF-KB with the fragment K-3 in a nuclear extract derived from the B cell line WEHI 231.

Unstimulated 70Z/3 cell extracts lacked a major band evident with B cell extracts. (Figure 1A, lane 6; arrow). This nucleoprotein complex band was induced in the 70Z/3 cells after LPS treatment for 20 hours. The band was not competed away even with 4.5 ug of poly d(IC) (lane 5). This induction phenomenon was not restricted to the 70Z/3 cell line; another pre-B cell line, PD (Lewis S., et al., Cell, 30:807-816 (1982)), was weakly positive for the factor prior to induction (Figure 1B, lane 3; Sen and Baltimore, 1986) but was strongly induced by LPS (Figure 1B lanes 5,6). A number of other minor bands could be seen in the binding assay. Some of these were inducible and others were not. The major inducible band comigrated with the major band produced by B cell and plasma cell extracts (typified by WEHI 231 extracts in Figure 1A, lane 6 and Figure 1B, lane 2). This band had been characterized earlier by competition experiments and the binding site of the factor had been localized by methylation interference experiments defining the band as one produced by interaction of the NF-KB factor with the B site within the enhancer (a site containing the sequence GGGGACTTTCC). Thus, two pre-B cell lines, one with a rearranged K gene (70Z/3) and the other in the process of undergoing rearrangement (PD), are clearly inducible by LPS for NF-KB activity.

Induction of NF-KB by LPS does not require protein synthesis Recently it has been reported that induction of transcription in 70Z/3 does not require new protein synthesis. Nelson, K.J. et al., Proc. Nat. Acad. Sci. USA, 82:5305-5309 (1985). Thus, induction of gene expression was evident in cells pretreated (10 minutes) with the translation inhibitors cycloheximide or anisomysin followed by stimulation with LPS. Further, Wall et al. reported that expression could be induced in the presence of cycloheximide alone. As a result, they argued in favor of a labile repressor blocking the activation of genes in this cell line. See Wall, R. et al. Proc. Nat. Acad. Sci. USA, 83:295-298 (1986). To determine if these characteristics of transcriptional activation were paralleled by changes in the levels of NF-KB, extracts derived from 70Z/3 cells which had been treated with LPS alone, with a translation inhibitor alone or with both together were analyzed. To make it possible to make direct correlations with the published reports concerning the effects of translational inhibitors on expression in pre-B cells, a 4 hour time point in these experiments was examined, although maximal stimulation of expression by LPS takes 14-20 hours. Binding reactions were carried out as described above for induction by bacterial lipopolysaccharide and contained 2.5, 3.5 or 4.5 ug poly d(IC) with each set of extracts. End-labelled K-3 fragment was the probe (lane 1) and was incubated with 9-11 ug of protein from extracts derived from: untreated 70Z/3 cells (lanes 2,3,4), 70Z/3 cells treated for 4 hours with 10 ug/ml of LPS (lanes 5,6,7), 70Z/3 cells treated for 4 hours with 10 ug/ml of LPS and 10 ug/ml cycloheximide (lanes 8,9,10); 70Z/3 cells treated with 10 ug/ml of cycloheximide alone (lanes 11,12,13) and WEHI 231 cells (lane 14). The characteristic nucleoprotein complex is indicated by the arrow. In accord with the transcriptional analyses, uninduced 70Z/3 cells were negative for NF-KB (Figure 2A, lanes 2-4), and treatment with either LPS alone (Figure 2A, lanes 5-7) or with cycloheximide alone (Figure 2A, lanes 11-13) for 4 hours induced the factor. Unexpectedly, stimulation of 70Z/3 with LPS in the presence of cycloheximide for 4 hours gave a superinduction of NF-KB (Figure 2A, lanes 8-10) , increasing it to a level above that seen after a 20 hour induction. Qualitatively, the same result was observed when anisomycin was used as a translation inhibitor (Figure 2B). Binding reactions were as described above, using 2.5 and 3.5 ug of poly d(IC) and protein from untreated 70Z/3 cells (lanes 2,3); 70Z/3 cells after induction with LPS alone (lanes 4,5); 70Z/3 cells with LPS induction in the presence of anisomycin (lanes 6,7); 70Z/3 cells treated with anisomycin by Itself (lanes 8,9) and the B cell WEHI 231 as a positive control (lane 10). The characteristic nucleoprotein complex is indicated by the arrow. Thus, the presence of anisomycin (10 uM) during a 4 hour stimulation with LPS gave a superinduction of NF-K B (Figure 2B, lanes 6,7) relative to either LPS alone (Figure 2B, lanes 4,5) or anisomycin alone (Figure 2B lanes 8-9). Once again, prior to LPS treatment there was no detectable NF-KB activity in 70Z/3 (Figure 2B lanes 2,3). Although treatment of 70Z/3 with cycloheximide alone or with LPS alone gave approximately equivalent amounts of NF-KB (Figure 2A compare lanes 5-7 with lanes 11-13), the level of NF-KB induced with anisomycin alone appeared to be much less (Figure 2B, compare lanes 8,9 with lanes 4,5). This is probably due to drug toxicity because, even after a short exposure to anisomycin, the cells looked quite unhealthy. Presumably this also accounts for a lesser degree of superinduction seen with LPS and anisomycin. Thus, the K-enhancer binding factor NF-KB appears to be inducible in 70Z/3 cells in the absence of protein synthesis. Further, it appears to be inducible by either of 2 different translation inhibitors alone and is superinduced when the cells are stimulated with LPS and the inhibitor.

Phorbol ester can induce NF- B in 70Z/3 The tumor promoting phorbol ester, phorbol 12-myristate-13-acetate (PMA), has been shown to induce surface immunoglobulin in 70Z/3, presumably via activation of K transcription and transport of complete immunoglobulin to the cell surface (Rosoff P.M. et al., J. Biol. Chem., 259:7056-7060 1984;

Rosoff, P.M. and Cantley, L.C., J. Biol. Chem., 260 9209-9215, (1985). To determine if this activation is reflected in an increase of NF-KB, extracts derived from 70Z/3 cells after a 4 hour stimulation with PMA at 50 ng/ml were analyzed. Binding reactions using K -3 as a probe (lane 1) were carried out as detailed in Fig. 1A legend with protein from untreated 70Z/3 cells (lane 2) or 70Z/3 cells that had been treated with PMA at 50 ng/ml for 4 hour (lanes 3,4). Lane 5 is the positive control for NF-KB in extracts from WEHI 231. There was a striking induction of NF-KB activity in these extracts (Figure 3A, compare lanes 3,4 with lane 2). Thus an active phorbol ester by itself is capable of inducing NF-KB activity in 70Z/3 cells, implicating protein kinase C as a possible intermediate in the post-translational modification reaction that produces NF-KB in these cells [(Bell, R.M. Cell, 45:631-632 (1986); Nishizuka, Y., Nature, 308:693-697, (1984)]. An inactive phorbol ester (phorbol 12, 13 didecanoate) did not cause induction of NF-KB under similar conditions (data not shown).

Time course of activation of NF-KB by LPS and PMA are different LPS-mediated stimulation of surface Ig expression of mRNA accumulation reaches a maximum after at least one cell cycle, i.e., in 14-18 hours. Recent work has shown that LPS stimulation of RNA synthesis, as measured by nuclear run on assays [Nelson, K.J., et al., Proc. Natl. Acad. Sci USA,

82:5305-5309 (1985); Wall et al. supra, (1986)] can be seen as early as 4 hours after stimulation and that the DNAse I hypersensitive site associated with the K enhancer can be detected as early as 1 hour post-stimulation. 70Z/3 cell extracts were generated after stimulation, either by LPS or PMA for varying lengths of time to examine the time-course of NF-KB induction. Analysis for NF-KB activity using the binding assay showed that the time course of activation of NF-KB by these two agents was quite different (Figure 3B). Binding reactions were carried out with extracts derived from 70Z/3 cells that had been treated with LPS at 10 ug/ml (lanes 3-7) or PMA at 25 ng/ml (lanes 8-12) for various lengths of time as shown above each lane in the figure. Lane 2 is a positive control for NF-KB in WEHI 231 extracts. With LPS alone, a nucleoprotein complex band reflecting the presence of NF-KB increased until 2 hour post-stimulation. Subsequently, a slight decrease occurred and then the level remained constant. By contrast, in PMA-stimulated cells, NF-KB was detected at maximal levels within 0.5 hours after stimulation, remained at this level for 2-3 hours and then began to drop off rapidly, such that by 8 hours it was barely detectable. Because prolonged exposure of cells to phorbol esters is known to result in desensitization of endogenous protein kinase C (Rodriquez-Pena, A. and Rozengurt, E., Biochem Biophys. Res. Comm., 120:1053-1009, 1984; EMBO J, 5:77-83 1986), a possible explanation for the rapid decline of NF-KB may be that its maintenance as a binding factor requires continuous activity of protein kinase C. A similar phenomenon has been described recently by Blemis and Erikson where S6 kinase activity assayed by phosphorylation of S6 protein) first rises and then falls during prolonged exposure to PMA. See Blemis, J. and Erikson, R.L., Proc. Natl. Acad. Sci. USA, 83:1733-1737 (1986). Although it has been reported that LPS may directly activate protein kinase C (Wightman, P.D. and Raetz, C.R.H., J. Biol. Chem., 259:10048-10052, 1984) the different kinetics of induction of NF-KB by LPS and PMA implies that these activators feed into a common pathway through distinguishable sites of activation.

Non pre-B cell lines can also be activated to produce NF-KB

Previous analysis had shown that NF-KB is present only in cell lines representing the B cell or plasma cell stages of B lymphoid differentiation, but was undetectable in a variety of non B cells, pre-B cells and T cells (Sen and Baltimore, 1986). However, as shown above, this factor may be induced to high levels in pre-B cells upon stimulation with LPS. To check if this inducibility was restricted to cells having a pre-B phenotype only or was a general characteristic of the other constitutively negative cell lines, representative examples of cell types (T cells and non lymphoid cells) were examined them for induction of NF-KB after appropriate stimulation.

The human T leukemia cell line, Jurkat, can be stimulated to produce interleukin-2 (IL-2) by the combined influence of phytohemagglutinin (PHA) and phorbol ester (PMA) (Gillis, S. and Watson, J., J. Exp. Med., 152:1709-1719, 1980; Weiss et al., J.

Immunol., 133:123-128, 1984). Nuclear extracts were prepared from Jurkat cells that had been stimulated with either PHA alone or PMA alone or both together and analyzed for the presence of NF-KB (Fig. 4A). The human T lymphoma Jurkat was stimulated with phytohemagglutinin (PHA) and phorbol 12-myristate- 13-acetate (PMA) individually or together for 20 hours. Nuclear extracts made after treatment were analyzed by the mobility shift assay using K-3 fragment as the labelled probe. Binding reactions typically contained 6 ug of protein, 2.5-3.5 ug of poly d(IC) and 0.3-0.5 ng of end-labelled DNA probe. Lane 1: no protein added; lane 2: WEHI 231 extract (positive control); lane 3: extract from uninduced Jurkat cells: lane 4: Jurkat cells stimulated with PIIA alone; lane 5: Jurkat cells stimulated with pHA and PMA; Lane 6: Jurkat cells stimulated with PMA alone. The arrow shows the position of the expected nucleoprotein complex generated by interaction of NF-KB with K-3 fragment. As originally observed, extracts derived from uninduced Jurkat cells were negative for NF-KB activity (Fig. 4A, lane 3). However, extracts made from Jurkat cells which had been stimulated either with PHA or PMA contained detectable levels of NF-KB (Fig. 4A, lanes 4,6) and the extracts from the co-stimulated cells showed higher levels of the factor (Fig. 4A, lane 5). Thus, a factor with the properties of NF-KB can be induced in a T cell line after appropriate activation.

The human HeLa cell line, which is constitutively negative for NF-KB (Sen and Baltimore, 1986), was used as an example of a non-lymphoid line. These cells were induced with PMA for 2 hours and extracts derived from treated and untreated cells were analyzed for NF-KB activity (Fig. 4B). HeLa cells were treated with PMA (50 ng/ml) for 2 hours and the extracts derived thereafter were analyzed for induction of NF-KB. Binding reactions contained 15-18 ug of protein, 3.5 ug of poly d(IC) and

0.3-0.5 ug of end-labelled DNA probe. Lane 1: K-3 fragment/no protein added; lane 2: K-3 fragment incubated with extracts derived from the human B lymphoma EW; lane 3: K-3 fragment incubated with uninduced HeLa cell nuclear extract; lane 4: K-50 fragment (derived from the K -heavy chain enhancer and containing a copy of the conserved octamer sequence ATTTGCAT) incubated with uninduced HeLa cell extracts; lane 5: K -3 fragment incubated with induced HeLa cell extracts. The untreated HeLa extract (Fig. 4B, lane 3) did not show a nucleoprotein complex which comigrated with the complex generated in B cell extracts. However treatment with PMA induced a factor that generated the characteristic DNA-protein complex produced by NF-KB (Fig. 4B, lane 5). As a control, both the uninduced and induced extracts showed equivalent levels of the ubiquitous NF-A1 DNA binding protein when analyzed using a probe containing the sequence ATTTGCAT (Fig. 4B, lanes 4,6; Singh, H. et al., Nature 319:154-158, 1986 ) . Therefore treatment of HeLa cells with PMA induces a factor that can form a nucleoprotein complex with the K-3 fragment.

To further characterize the DNA-protein complex formed in the PMA-treated HeLa cell extracts, competition experiments were carried out. Binding reactions were carried out using end-labelled K-3 fragment, 3.5 ug of poly d(IC) and 15-18 ug of nuclear extract in the present of 50 ug of unlabelled competitor DNA derived from various immunoglobulin and viral regulatory sequences (lanes 5-9). The complex generated in PMA-induced HeLa cell extracts (Fig. 4C, lane 4) was specifically competed away by the inclusion of 50 ng of unlabelled DNA in the binding reaction containing either the enhanced (Fig. 4C, lane 7) or the SV40 enhanced (Fig. 4c, lane 8) but was unaffected by two DNA fragments that together span the K-enhancer (Fig. 4C, lane 5,6), or by a 250 bp fragment containing the K promoter (Fig. 4C, lane 9). This pattern of competition exactly parallels the pattern observed earlier using the K-3 fragment in binding experiments with B cell derived extracts (Sen and Baltimore, 1986). These results further strengthen the conclusion that the NF-KB factor can be induced in non-lymphoid cells as well as lymphoid cells following appropriate stimulation.

Equivalent

Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.

Claims

1. A method of inducible gene expression, comprising the steps of: a. preparing a genetic construct comprising: i) a kappa enhancer sequence or a portion of the kappa enhancer sequence containing at least the sequence to which the transcriptional regulatory factor NF-KB binds; ii) a promoter; and iii) a structural gene of interest; b. transfecting a eukaryotic host cell with the genetic construct; and c. stimulating the transfected cell when desired with a substance which stimulates

NF-KB activation and binding to the enhancer sequence to induce expression of the gerie.

2. A method of Claim 1, wherein the structural gene encodes a cytotoxic protein.

3. A method of Claim 1, wherein the substance which stimulates NF-KB is an activator of protein kinase C.

4. A method of Claim 3, wherein the activation of protein kinase C is a phorbol ester.

5. A method of Claim 4, wherein the phorbol ester is phorbol 12-myristate-13-acetate.

6. A method of Claim 1, wherein the substance which stimulates NF-KB is a mitogen.

7. A method of Claim 6 wherein the mitogen is lipopolysaccharide or phytohemagglutinin.

8. A method of inducible gene expression, comprising: a. preparing a gene construct comprising: i) enhancer sequence for kappa light immunoglobulin chain; ii) a promoter; iii) structural gene b. transfecting a eukaryotic host cell with the gene construct; c. exposing transfected cells to PMA when desired to induce expression of the gene.