ACTIVITY

Reference to Government Grant

The invention described herein was supported in part by National Institutes of Health grants AI28526, DK41298 and HL46547. The Federal government has certain rights in the invention. Cross-Reference to Related Applications

This application claims priority from U.S. provisional patent application No. 60/001,329 filed Juy 21, 1995.

Field of the Invention

The invention relates to the screening of candidate drugs for activity against lipopolysaccharide endotoxin (LPS) and application in the treatment of LPS- induced endotoxemia. The invention further relates to murine LPS response gene

DNA and to preparation of cell lines and animals transformed with inactivated LPS response genes, for use in the screening method.

Background of the Invention Lipopolysaccharide endotoxin (LPS) is a complex macromolecule that comes from the cell walls of certain bacteria, some of which cause diseases like typhoid fever, dysentery, and urinary tract infections and from other bacteria which are common inhabitants of animal and human intestinal tracts but ordinarily do not cause disease. All of these bacteria have in common the same type of cell wall and are classified as Gram-negative (staining properties).

LPS induces the production and release of immunologically active cytokines and other mediators of the animal's inflammatory response. Cytokines are very active substances produced by various cells such as the mononuclear phagocyte and lymphocytes which stimulate other cells as well as each other. The

endothelial cells and granulocytes are also primary targets of endotoxins. Specific receptor molecules and signal transduction pathways in the cells of host animals are being intensively studied at present to gain a clearer understanding of how the endotoxin works. There are many pathophysiological effects of LPS, one of which is endotoxemia or septic shock which results from large amounts of endotoxin in the blood (40% fatality). The majority of the cases of septic shock are a consequence of Gram-negative bacteria (bacteria in the blood). However, the septic shock syn¬ drome can be induced by other organisms including Gram-positive bacteria and fungi. Nevertheless, extensive investigations into the causes of septic shock have established that a key factor in its development is the release of LPS from Gram- negative bacteria and the subsequent effects of the endotoxin on various cells in the body which become highly activated. As a result, the host is overwhelmed with many cell substances that lead to circulatory failure, shock and death. The cellular responses to endotoxin are under genetic control as shown by extensive studies in the mouse. The underlying genetic basis for the multiple response of the host to LPS was initially defined with the discovery of the C3H/HeJ mutant mouse strain in 1968 (Sultzer, B.M. Nature 219, 1253-1254 (1968)). The strain is hyporesponsive to the immunostimulatory and pathophysiological effects of the lipid A component of LPS and, therefore, is considered to have a fundamental deficiency in its reactions to LPS as compared to closely related responder strains.

This defect is specific and expressed in a variety of ways. For example, C3H/HeJ B cells do not proliferate or differentiate when exposed to LPS and the proliferation of their thymocytes induced by concanavalin A is not enhanced by LPS (Sultzer, B.M. and Nilsson, B.S. Nature New Biology (London) 240, 198- 200 (1972); Sultzer, B.M. Abst. Am. Soc. Microbiol. P. 85, M69 (1973); Tanabe, MJ., and M. Nakano, Microbiol. Immunol. 23, 1097-1104 (1979); Morrison, D.C. and J.L. Ryan, Adv. Immunol. 28, 294-312 (1979); Sultzer BM in "Beneficial effects of Endotoxin", Ch 11, Plenum, NY (1983); Sultzer BM, Castagna R, Bandekar J & Wong PMC Immunobiology 187, 257-271 (1993); Glode, L.M. and

D.L. Rosenstreich, J. Immunol 117, 2061, 2068 (1976); Sultzer, B.M. Infection and Immunity 13, 1579-1584 (1976)). Macrophage phagocytosis and cytotoxicity are not stimulated by LPS and the cytokines such as Tumor necrosis factor (TNF), interferon (IFN), colony stimulating factor (CSF), interleukin-l (IL-1) and the prostaglandins stimulated by LPS in responder cells are not induced in C3H/HeJ macrophages (Morrison, D.C. and J.L. Ryan, Adv. Immunol. 28, 294-312 (1979)). Furthermore, the C3H/HeJ mouse is highly resistant to lethal endotoxin shock as compared to normal endotoxin responder strains (Sultzer, B.M. Nature 219, 1253- 1254 (1968)). The failure of C3H/HeJ cells to respond to LPS is not due to the absence of helper cells or the presence of suppressor cells or for that matter the deficient binding of LPS to otherwise immunocompetent cells (Sultzer BM in "Beneficial effects of Endotoxin", Ch 11, Plenum, NY (1983). Rather, the explanations offered focus on a deficient trigger receptor or a failure somewhere in signal transduction after the initial interaction of the cells with LPS (Sultzer BM, Castagna R, Bandekar J & Wong PMC, Immunobiology 187, 257-271 (1993)).

From the results of classical type breeding experiments with the C3H/HeJ strain and various responder strains, it has been found that the mitogenic response to LPS is governed by a single locus composed of co-dominant alleles (Morrison, D.C. and J.L. Ryan, Adv. Immunol. 28, 294-312 (1979); Sultzer BM in "Beneficial effets of Endotoxin", Ch 11, Plenum, NY (1983); Sultzer BM, Castagna R, Bandekar J & Wong PMC, Immunobiology 187, 257-271 (1993); Glode, L.M. and D.L. Rosenstreich, /. Immunol 117, 2061-2068 (1976); Sultzer, B.M., Infection and Immunity 13, 1579-1584 (1976)). Watson determined that the locus (Lps ⁿ ) was on chromosome 4 and linked to the major urinary protein locus (Mup- 1), but downstream from Mup-1 and the Lyb2;2;4;6 genes which control B cell activation (Watson, J., K. Kelly, M. Largey & Taylor B.A., J. Immunol. 120, 422- 429 (1979)). The C3H/HeJ mouse strain contains the mutant Lps ^d allele.

Sultzer B.M., Castagna R., Bandekar J. & Wong P.M.C., Immunobiology 187, 257-271 (1993) describe the preparation of a cDNA library consisting of six sublibraries from C3H/OuJ mouse spleen cells. The introduction

of one sublibrary consisting of 2 x IO ⁴ independent clones into C3H/HeJ LPS non- responder spleen cells produced, after exposure to LPS in culture, approximately a four-fold increase in plaque-forming cells reactive to sheep red blood cells.

Vogel et al., Infec. Immun. 62, 4454-4459 (1994) describe development of a congeneic BALB/c mouse strain that contains a segment of chromosome 4 including the Lps ^d allele of the C3H/HeJ mouse strain.

Despite these findings, the purported Lps" gene has not been located.

Identification of the Lps" gene would permit a clearer understanding of the mechanism of LPS at the molecular cellular and host levels. This would permit the design of drugs to interfere with the mechanism and thereby prevent patients from dying from endotoxemia.

There is a need for mutant cell lines harboring homozygous Lps" null mutations, which may be used as important negative controls in the screening of potential anti-LPS drugs. There is a further need for nonhuman transgenic animals harboring such mutations. Such transgenic host animals lacking functional endogenous Lps" loci (i.e., having an Lps ⁿ knockout background) may serve as hosts for transplanted LPS-responsive cells, for use in the screening of potential anti-LPS drugs.

Summary of the Invention

According to one embodiment of the invention, a heterologous cloning vector is provided comprising the DNA SEQUENCE ID NO: 1. Such a vector is plasmid ATCC NO. 97225.

According to another embodiment of the invention, purified and isolated DNA is provided comprising DNA having the nucleotide sequence SEQ ID NO: 1.

According to a further embodiment of the invention, a host cell line transformed by the aforesaid heterologous cloning vector is provided. The host cell line expresses the DNA from the cloning vector to produce Lps" protein. According to yet a further embodiment, a process for preparing Lps" protein is provided. The process comprises culturing a host cell line hosting the aforesaid heterologous cloning vector.

In yet another embodiment, the invention is a mutant cell line derived from a lipopolysaccharide endotoxin responder parent cell line, the mutant cell line comprising a sub-strain of the parent cell line. The mutant cell line contains a homozygous Lps" null mutation generated by disruptive gene targeting of the Lps" gene locus such that the mutant line is substantially nonresponsive to lipopolysaccharide endotoxin stimulation.

In another embodiment, the invention is a nonhuman animal, or a stem cell comprising a diploid genome, comprising a homozygous Lps" null mutation generated by disruptive gene targeting of the Lps" gene locus whereby the animal or stem cell is rendered substantially nonresponsive to lipopolysaccharide endotoxin stimulation.

In another embodiment, the invention is a process for screening an agent for activity in inhibiting host response to lipopolysaccharide endotoxin comprising: (a) contacting said agent with a first cell line which is responsive to stimulation by lipopolysaccharide endotoxin, and inducing lipopolysaccharide endotoxin stimulation;

(b) determining the level of the response to lipopolysaccharide endotoxin stimulation of the first cell line in the presence of the agent; and (c) comparing the level of the response of the first cell line to lipopolysaccharide endotoxin in the presence of the agent with the level of response of a second cell line to lipopolysaccharide endotoxin in the presence of the same agent, the second cell line comprising a mutant subline of the first cell line comprising a homozygous Lps" null mutation generated by disruptive gene targeting of the Lps" gene locus such that the line is substantially nonresponsive to lipopolysaccharide endotoxin stimulation.

According to yet another embodiment, the invention is a process for screening an agent for activity in inhibiting host response to lipopolysaccharide endotoxin comprising: (a) administering said agent to a first non-human animal strain which is responsive to stimulation by lipopolysaccharide endotoxin, and inducing lipopolysaccharide endotoxin stimulation;

(b) determining the extent of response to lipopolysaccharide endotoxin stimulation of the first strain upon administration of the agent; and

(c) comparing the level of the response of the first strain to lipopolysaccharide endotoxin upon administration of the agent with the level of response of a second strain to lipopolysaccharide endotoxin upon administration of the same agent, the second strain comprising a mutant sub-strain of the first strain. The second strain contains a homozygous Lps" null mutation generated by disruptive gene targeting of the Lps" gene locus such that the strain is substantially nonresponsive to lipopolysaccharide endotoxin stimulation. In a further embodiment, the invention is a process for screening an agent for activity in inhibiting host response to lipopolysaccharide endotoxin comprising:

(a) administering said agent to a nonhuman host animal comprising a homozygous Lps" null mutation generated by disruptive gene targeting of the Lps" gene, said animal harboring transplanted human lipopolysaccharide endotoxin responding cells, and

(b) assaying the response of the animal to lipopolysaccharide endotoxin stimulation.

The term "agent" is used herein to denote a chemical compound, a mixture of chemical compounds, a biological macromolecule, or an extract made from biological materials such as bacteria, plants, fungi or animal (particularly mammalian) cells or tissues.

The term "LPS" as used herein means lipopolysaccharide endotoxin.

The term "Lps"" means the lipopolysaccharide endotoxin response gene, without regard to the species of origin. Thus, unless indicated to the contrary, "Lps"" includes not only the murine gene described herein, but also the various corresponding human and nonhuman forms including but not limited to

TC4/Ran (in humans and dog), Ju93 (in chickens) and Spil (in yeast).

The term "Lps" protein" as used herein means the polypeptide that is encoded by the Lps" gene.

Description of the Figures

Fig. 1 is a graph of the results of a modified Jerne's plaque-forming cell ("PFC") assay carried out on C3H/HeJ mouse spleen cells enriched for plaque- forming B-cells following introduction of C3H/HeOuJ DNA. C3H/HeJ mouse cells were electroporated with plasmid DNA from a cDNA pCD-plasmid library derived from LPS-stimulated C3H/HeOuJ spleen cells. Data from sublibrary LB2.2 were normalized based on background values from plaque ratios of 2.5/no DNA (*). For consistency, the data from one experiment that included sublibrary 2.2.DJ .B.3 was normalized to the data from another experiment that included 2.2. D J.B, 2.2.DJ.B.3 and 2.2.DJ.B.3.S. Therefore, these two ratios (**) are normalized based on the 2.2. D J.B sublibrary.

Fig. 2, upper panel, represents the results of a hybridization assay from various Xhol -digested and fractionated pCD sublibraries made from LPS- stimulated C3H/HeOuJ spleen cells. Fig. 2, lower panel, represents probing of the fractionated DNA with a 0.8 kb BamHl/Pstl DNA fragment specific to the Lps" cDNA. A 1.5 kb band is present in all PFC-positive sublibraries but not in all PFC -negative sublibraries, and the intensity of the band increased progressively as the results of the PFC assay became more positive.

Fig. 3 A comprises the nucleotide sequence of the Lps" mouse cDNA and its deduced amino acid sequence. Fig. 3B is a comparison of the deduced amino acid sequences of mouse Lps", human TC4/Ran, dog TC4/Ran, chicken Ju93 and yeast Spil. The complete amino acid sequence predicted for murine Lps" is shown on the top line by the single letter amino acid code. Dots indicate identical residues to those of Lps"; letters below the top line are different amino acid residues in the corresponding positions; the five boxes shown are the conserved functional domains shown to be involved in the binding and hydrolysis of guanine nucleotides.

Fig. 4 is a northern blot analysis of RNA from C3H/HeJ and

C3H/HeOuJ organs probed with a 0.8 kb BamHl/Pst DNA fragment specific to the

Lps" cDNA (top panel). The hybridization filter was stripped of Lps"-specific probe, and was hybridized with a GAPDH-specific probe (lower panel). After normalizing the intensity of the GAPDH band, the amount of Lps" RNA

(C3H/HeOuJ) does not appear to differ significantly from that of Lps ^d RNA (C3H/HeJ).

Detailed Description of the Invention Generally, the nomenclature used hereafter and the laboratory procedures in cell culture, molecular genetics, and nucleic acid chemistry and hybridization described below are those well known and commonly employed in the art. Standard techniques are used for recombinant nucleic acid methods, polynucleotide synthesis, cell culture, and transgene incorporation (e.g., electroporation, microinjection, lipofection). Generally enzymatic reactions, oligonucleotide syntheses, and purification steps are performed according to the manufacturer's specifications. The techniques and procedures are generally performed according to conventional methods in the art and various general references which are provided throughout this document. The procedures therein are believed to be well known in the art and are provided for the convenience of the reader. All the information contained therein is incorporated herein by reference.

The Lps" Gene Described herein is in the isolation of a murine gene, designated

Lps", that controls the response of cells, most notably B cells to LPS. The 1166 nucleotide cDNA (Fig. 3A; SEQ ID NO: 1) includes a 648 nucleotide open reading frame encoding a 216 amino acid protein (Fig. 3B; SEQ ID NO. 2). It is believed this gene encodes for a GTP-binding protein, which is a protein in cell membranes which is part of a signal transduction pathway needed for the cell to be activated to divide and make antibody. It is believed that this gene will be found not only in B cells, but also in other cell types which respond to LPS and in cells as well which produce cytokines and other factors that in large amounts can be deleterious to the host. The isolation of the Lps" gene has permitted for the first time the engineering of cell lines and transgenic nonhuman animals which contain an Lps" null allele, particularly cell lines and nonhuman animals which are homozygous

with rc _; εct to the Lps" null allele. Such homozygous null mutants or Lps" "knockout" mutants, are useful as reagents in the screening of agents for activity in inhibiting LPS stimulation.

The Lps" gene encodes a functional gene product which is present in polyclonally activated antibody producing B cells from LPS-responder mice such as the C3H/HeOuJ. Introduction of this cDNA (Lps") into C3H/HeJ B cells is sufficient to restore their responsiveness to LPS stimulation, resulting in the increase of antibody producing plaque-forming cells (PFCs) to a responder level. This was demonstrated by first extracting mRNA from LPS-stimulated C3H/HeOuJ spleen cells after the removal of erythrocytes and T cells. From these mRNA, a cDNA pCD-plasmid library composed of 6 sublibraries (LB2J - LB2.6) was constructed using the Okayama-Berg expression vector (Okayama, H. and P. Berg, Mol. Cell. Biol. 2, 161-169 (1982)). The library consisted of six sublibraries, ranging in size from 1 - 2 x IO ⁴ independent clones. Electroporation was used to introduce the pCD library DNA into

C3H/HeJ B cells. Because the efficiency of gene transfer using this technique is highly variable, standardization experiments using immortalized pre-B 70Z/3 lymphoid cells were performed. Under standardized optimal electroporation conditions, the efficiency of gene transfer was determined to be about 1 in 10,000 cells electroporated. Sublibraries of the plasmid DNA were electroporated into C3H/HeJ spleen cells enriched with plaque-forming B cells in a modified Jerne's plaque-forming assay.

After testing positive on the plaque assay, sublibrary LB2.2, containing a size of 2 x IO ⁴ independent clones, but not other sublibraries, was subdivided further into four sub-sublibraries, designated LB2.2.A-D. Each sub- sublibrary consisted of 6,000 independent clones. From those, LB2.2D was positive and was therefore divided further into ten sublibraries, LB2.2DJ-10, each of which consisted of 600 independent clones. LB2.2D.1 was tested positive and was subdivided into six sublibraries, LB2.2.DJ.A-F, with 100 pooled clones. Among them, LB2.2.DJ.B was positive and was then divided into 10 pools of sublibraries with 10 independent clones per sublibrary . From these , LB2.2. D J . B .3 was tested positive.

After further division into 10 single clones, among them,

LB2.2.DJ.B.3.S was tested positive. The LB2.2.DJ.B.3.S plasmid preparation contained a single species of plasmid DNA; its introduction and expression in

C3H/HeJ spleen cells led to an increase in the numbers of PFCs. C3H/HeJ cells can be stimulated by LPS after expression of this cDNA.

The LB2.2.DJ .B.3.S plasmid has been given the name "pCD-Lps"" . The plasmid was deposited in the American Type Culture Collection, Brooklawn Drive, Bethesda, MD, on July 19, 1995. It has been assigned the accession number 97225. At the amino acid level, Lps" is identical to the Ran/TC4 GTPase found in humans and dogs, and has 98.6% and 82.8% homology to the Ju93 and Spil proteins in chicken and yeast, respectively (Drivas G.T., Shih A., Coutavas E. Rush M.G. & D'eustachio P., Mol Cell Biol 10, 1793-1798 (1990); Dupree P. Olkkonen V.M. & Chavrier P., Gene 120, 325-326 (1992); Trueb J. & Trueb B., FEBS 306, 181-184 (1992); Matsumoto T. & Beach D., Cell 66, 347-360 (1991)). The entire disclosures of each of Drivas et al. , Dupree et al. , Trueb & Trueb and Matsumoto & Beach are incorporated herein by reference. At the DNA level, the homology of Lps" to human TC4/Ran, dog TC4/Ran, chicken Ju93 and yeast Spil 89.7%, 90.7%, 85.4% and 68.7%, respectively. Id. Therefore, the Lps" molecule is highly conserved throughout evolution, particularly the five domains of the molecule, G1-G5 (Figure 3B), which have been shown to be involved in the binding and hydrolysis of guanine nucleotides. The Ran/TC4 G protein is the expression product of the Lps" gene.

The Lps" gene may be cloned by any of the known procedures which permit the large scale isolation of plasmid DNA. Replication of plasmid DNA may be carried out according to conventional techniques, such as described by Sambrook et al , Eds., Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, 1989, particularly Chapter 1 thereof, p. 1.1-1.110, incorporated herein by reference. One suitable plasmid cloning vector for this purpose is the widely used pBR322 plasmid (Bolivar et al, Gene 2:95, 1977), as described by Sambrook et al, supra, p. 1.12. The plasmid described herein

containing the Lps" cDNA was transformed into DH5α bacteria for preparation of large amounts of plasmid DNA. The aforesaid plasmid may also be conveniently used to transform HB101 bacteria.

Expression of Lps" Protein

The nucleic acid molecules that encode Lps" protein may be inserted into known vectors for use in standard recombinant DNA techniques for producing recombinant Lps" protein. Standard recombinant DNA techniques include those techniques as described by Sambrook et al. , Eds., Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, 1989 and by Ausubel et al., Current Protocols in Molecular Biology, J. Wiley & Sons, New York, NY (1991). The vectors may be circular or non-circular. The host nay be prokaryotic or eukaryotic. The preferred prokaryotic host comprises E. coli. Preferred eukaryotic hosts include yeasts, insect and mammalian cells. Preferred mammalian cells include, primate cells such monkey cells transformed by simian viruses (e.g., COS cells), human cells, and Chinese hamster ovary (CHO) cells.

Insertion of the Lps" gene into an appropriate expression vector is easily accomplished when the requisite DNA sequences and cloning vector have been cut with the same restriction enzyme or enzymes, since complementary DNA termini are thereby produced. If this cannot be accomplished, it may be necessary to modify the cut ends that are produced by digesting back single-stranded DNA to produce blunt ends, or by achieving the same result by filling in the single-stranded termini with an appropriate DNA polymerase such as the Klenow fragment of DNA Polymerase I. In this way, blunt-end ligation with an enzyme such as T4 DNA ligase may be carried out.

For insertion of the Lps" gene into a vector, any site desired could also be produced by ligating nucleotide sequences (linkers) onto the DNA termini. Such linkers may comprise specific oligonucleotide sequences that encode restriction site recognition sequences. The cleaved vector and the modified Lps" gene may also be modified by homopolymeric tailing, as described by Morrow, Methods in Enzymology 68:3 (1979).

According to one embodiment of recombinant Lps" production, recombinant Lps" protein is obtained via bacterial expression vectors. One such useful vector, pDS56-6XHIS (Hochuli et al, Biotechnology 6, 7351-7367, 1988), contains an initiator codon followed by six histidine residues at the amino terminus to facilitate purification by chromatography on a nickel matrix. Similar vectors suitable for this purpose are commercially available under the tradename "The QIA Expressionist" from Quiagen Inc., Chatsworth, CA. Methods of protein purification by nickel matrix chromatography are known (Reddy et al, Oncogene 7, 2085-2092, 1992). Modified versions of vector pDS56-6XHIS may be substi- tuted, such as versions using the natural methionine of the cDNA but containing six histidines at the C-terminal end of the coding region. The Lps" gene is engineered into such vectors by site-directed mutagenesis, such as according to the procedure of Higuchi et al, Nucleic Acids Res. 16, 7351-6367 (1988), and the recombinant vector is used to transform bacterial cells to produce recombinant Lps" protein. According to one embodiment for recombinant Lps" expression in host bacteria, bacteria are transformed with an appropriate plasmid, e.g. pDS56- 6XHIS, which contains Lps" DNA and which permits subsequent purification of recombinant Lps" protein by nickel affinity matrix chromatography. Transformation of host bacteria is carried out, for example, according to the teachings of Reddy et al, Oncogene 7, 2085-2092 (1992). An overnight culture of bacteria transformed with the appropriate plasmid is induced by the addition of isopropyl- 3-D- thiogalactopyranoside to a final concentration of 1 mM and further grown for 2 hours. The bacteria are pelleted and lysed in 15 ml of 6M GuHCl, pH 8.0. The clarified supernatants are then passed through a nickel chelate column which has been equilibrated with 6M GuHCl. Following extensive washing with loading buffer, the recombinant protein is eluted with 6M GuHCl (pH 5.0), and the eluate dialyzed stepwise in 1.0M and 0JM GuHCl (pH 7.6). The final dialysis is performed against a buffer containing 20mM Hepes (pH 7.6), 7% glycerol, 70mM NaCl and 1 mM dithiothreitol. The purity of the recombinant protein may be determined by SDS-PAGE according to the method of Lammeli, Nature 227, 680- 685 (1970).

Advantageously, the Lps" protein is a biologically pure or isolated preparation meaning that it has undergone some purification away from other proteins and/or non-proteinaceous materials. The purity of the preparation may be represented as at least 40% Lps", preferably at least 60% Lps", more preferably at least 75% Lps", even more preferably at least 85% Lps", and most preferably at least 95% Lps", relative to non-Lps" material as determined by weight, activity, amino acid similarity, antibody reactivity or other conventional means.

Recombinant Lps" protein can be used to generate Lps"-specific antibodies. The antibodies are useful in inhibiting the Lps-induced response in cultured cells and animal hosts, which may find use as negative controls in the screening of potential agents which inhibit the response.

Knockout Lps" Mutant Cell Lines and Transgenic Nonhuman Animals Chimeric targeted mice are derived according to Hogan, et al,

Manipulating the Mouse Embryo: A Laboratory Manual, Cold Spring Harber Laboratory (1988) and Teratocarcinomas and Embryonic Stem Cells: A Practical Approach, EJ. Robertson, ed., IRL Press, Washington, D.C, (1987) which are incorporated herein by reference. Embryonic stem cells are manipulated according to published procedures (Teratocarcinomas and Embryonic Stem Cells: A Practical Approach, E. J. Robertson, ed., IRL Press, Washington, D.C. (1987); Zjilstra et al., Nature 342:435-438 (1989); and Schwartzberg et al, Science 246:799-803 (1989), each of which is incorporated herein by reference). Oligonucleotides can be synthesized on an Applied Bio Systems oligonucleotide synthesizer according to specifications provided by the manufacturer.

According to the practice of the invention, the endogenous Lps" alleles of a cell line or nonhuman animal are functionally disrupted so that expression of endogenously encoded Lps" gene is suppressed or eliminated. In general, polynucleotide constructs are employed for this purpose. Methods for accomplishing this result are described in detail in WO 95/11968 with respect to

transgenic non-human animals and mammalian cells hosting a transgene encoding an amyloid precursor protein (APP) and "knock out" mutants thereof. The entire disclosure of WO 95/11968 is incorporated herein by reference. Similar methods of preparing transgenic non-human animals and mammalian cells characterized by knock-out mutations are described in WO 94/06908, WO 94/28122 and WO 94/28123, the entire disclosures of which are incorporated herein by reference.

Gene targeting, which is a method of using homologous recombination to modify a mammalian genome, can be used to introduce genetic changes into cultured cells. By targeting a gene of interest in embryonic stem (ES cells, these changes can be introduced into the germlines of laboratory animals to study the effects of the modifications on whole organisms, among other uses. The gene targeting procedure is accomplished by introducing into tissue culture cells a DNA targeting construct that has a segment homologous to a target locus and which also comprises an intended sequence modification (e.g., insertion, deletion, point mutation). The treated cells are then screened for accurate targeting to identify and isolate those which have been properly targeted. A common scheme to disrupt gene function by gene targeting in ES cells is to construct a targeting construct which is designed to undergo a homologous recombination with its chromosomal counterpart in the ES cell genome. The targeting constructs are typically arranged so that they insert additional sequences, such as a positive selection marker, into coding elements of the target gene, thereby functionally disrupting it. Targeting constructs usually are insertion-type or replacement-type constructs (Hasty et al. (1991) Mol. Cell. Biol. 11: 4509).

The invention encompasses production of mutant cell lines and nonhuman animals that have the endogenous Lps" gene inactivated by gene targeting with a homologous recombination targeting construct. Typically, an Lps" gene sequence is used as a basis for producing PCR primers that flank a region that will be used as a homology clamp in a targeting construct. The PCR primers are then used to amplify, by high fidelity PCR amplification (Mattila et al. (1991) Nucleic Acids Res. 19: 4967; Eckert, K.A. and Kunkel, T.A. (1991) PCR Methods and Applications: 1 : 17; U.S. Patent 4,683,202; which are incorporated herein by

reference), a genomic sequence from a genomic clone library or from a preparation of genomic DNA, preferably from the cell line or strain of nonhuman animal that is to be targeted with the targeting construct. The amplified DNA is then used as a homology clamp and/or targeting region. Thus, homology clamps for targeting an Lps" gene may be readily produced on the basis of nucleotide sequence information available in the art and/or by routine cloning. General principles regarding the construction of targeting constructs and selection methods are reviewed in Bradley et al. (1992) Bio/Technology 10: 534, incorporated herein by reference. It may be appreciated that genomic DNA corresponding to the Lps" cDNA identified herein and the prior art may be derived using an appropriate cDNA fragment as a probe to identify and isolate genomic Lps" from an appropriate genomic DNA library.

Targeting constructs can be transferred into pluripotent stem cells, such as murine embryonal stem cells, wherein the targeting constructs homologously recombine with a portion of an endogenous Lps" gene locus and create mutation(s) (i.e., insertions, deletions, rearrangements, sequence replacements, and/or point mutations) which prevent the functional expression of the endogenous Lps" gene.

One method is to delete, by targeted homologous recombination, essential structural elements of the endogenous Lps" gene. For example, a targeting construct can homologously recombine with an endogenous Lps" gene and delete a portion spanning substantially all of one or more exons to create an exon-depleted allele, typically by inserting a replacement region lacking the corresponding exon(s). Cell lines and transgenic animals homozygous for the exon-depleted allele (e.g., by breeding of heterozygotes to each other) produce cells which are essentially incapable of expressing a functional endogenous Lps" polypeptide. Similarly, homologous gene targeting can be used, if desired, to functionally disrupt an Lps" gene by deleting only a portion of an exon.

Targeting constructs can also be used to delete essential regulatory elements of an endogenous Lps" gene, such as promoters, enhancers, splice sites, polyadenylation sites, and other regulatory sequences, including cis-acting sequences that occur upstream or downstream of the Lps" structural gene but which

participate in endogenous Lps" gene expression. Deletion of regulatory elements is typically accomplished by inserting, by homologous double-crossover recombination, a replacement region lacking the corresponding regulatory element(s). An alternative method is to interrupt essential structural and/or regulatory elements of an endogenous Lps" gene by targeted insertion of a polynucleotide sequence, and thereby functionally disrupt the endogenous Lps" gene. For example, a targeting construct can homologously recombine with an endogenous Lps" gene and insert a nonhomologous sequence, such as a neo expression cassette, into a structural element (e.g., an exon) and/or regulatory element (e.g., enhancer, promoter, splice site, polyadenylation site) to yield a targeted Lps" allele having an insertional interruption. The inserted sequence can range in size from about 1 nucleotide (e.g., to produce a frameshift in an exon sequence) to several kilobases or more, as limited by efficiency of homologous gene targeting with targeting constructs having a long nonhomologous replacement region.

Targeting constructs can also be employed to replace a portion of an endogenous Lps" gene with an exogenous sequence (i.e., a portion of a targeting transgene); for example, a first exon of an Lps" gene may be replaced with a substantially identical portion that contains a nonsense or missense mutation.

Inactivation of an endogenous murine Lps" locus may be achieved by targeted disruption of the Lps" gene by homologous recombination in mouse embryonic stem cells. For inactivation, any targeting construct that produces a genetic alteration in the target Lps" gene locus resulting in the prevention of effective expression of a functional gene product of that locus may be employed. If only regulatory elements are targeted, some low-level expression of the targeted gene may occur (i.e., the targeted allele is "leaky"), however the level of expression may be sufficiently low that the leaky targeted allele is functionally disrupted. A targeting construct may be transferred by electroporation of microinjection into a totipotent embryonal stem (ES) cell line, such as the murine AB-1 or CCE lines. The targeting construct homologously recombines with

endogenous sequences in or flanking of Lps" gene locus and functionally disrupts at least one allele of the Lps" gene. Typically, homologous recombination of the targeting construct with endogenous Lps" locus sequence will result in integration of a nonhomologous sequence encoding and expressing a selectable marker, such as neo, usually in the form of a positive selection cassette. The functionally disrupted allele is known as an Lps" "null" allele. ES cells having at least one Lps" null allele are selected for by propagating the cells in a medium that permits the preferential propagation of cells expressing the selectable marker. Selected ES cells are examined by PCR analysis and/or Southern blot analysis to verify the presence of a correctly targeted Lps" allele. Breeding of nonhuman animals which are heterozygous for a null allele may be performed to produce nonhuman animals homozygous for said null allele, so-called "knockout" animals (Donehower et al (1992) Nature 256: 215; Science 256: 1392, incorporated herein by reference). Alternatively ES cells homozygous for a null allele having an integrated selectable marker can be produced in culture by selection in a medium containing high levels of the selection agent (e.g., G418 or hygromycin). Heterozygosity and/or homozygosity for a correctly targeted null allele can be verified with PCR analysis and/or Southern blot analysis of DNA isolated rom an aliquot of a selected ES cell clone and/or from tail biopsies. Gene targeting techniques which have been described, include but are not limited to: co-electroporation, "hit-and-run", single-crossover integration, and double-crossover recombination (Bradley et al. (1992) Bio/Technology 10: 534). The preparation of the homozygous Lps" null mutants can be practiced using essentially any applicable homologous gene targeting strategy known in the art. The configuration of a targeting construct depends upon the specific targeting technique chosen. For example, a targeting construct for single-crossover integration or "hit-and-run" targeting need only have a single homology clamp linked to the targeting region, whereas a double-crossover replacement-type targeting construct requires two homology clamps, one flanking each side of the replacement region.

For example and not limitation, a targeting construct comprises, in order: (1) a first homology clamp having a sequence substantially identical to a

sequence within about 3 kilobases upstream (i.e., in the direction opposite to the translational reading frame of the exons) of an exon of an endogenous Lps" gene, (2) a replacement region comprising a positive selection cassette having a pgk promoter driving transcription of a neo gene, (3) a second homology clamp having a sequence substantially identical to a sequence within about 3 kilobases downstream of said exon of said endogenous Lps" gene, and (4) a negative selection cassette, comprising a HSV tk promoter driving transcription of an HSV tk gene. Such a targeting construct is suitable for double-crossover replacement recombination which deletes a portion of the endogenous Lps" locus spanning said exon and replaces it with the replacement region having the positive selection cassette. The deleted exon is one which is essential for expression of a functional Lps" gene product. Thus, the resultant exon-depleted allele is functionally disrupted and is termed a null allele.

Targeting constructs comprise at least one Lps" homology clamp linked in polynucleotide linkage (i.e., by phosphodiester bonds) to a targeting region. A homology clamp has a sequence which substantially corresponds to, or is substantially complementary to, an endogenous Lps" gene sequence of a cell line or a nonhuman host animal, and may comprise sequences flanking the Lps" gene.

Although no lower or upper size boundaries for recombinogenic homology clamps for gene targeting have been conclusively determined in the art, the best mode for homology clamps is believed to be in the range between about 50 basepairs and several tens of kilobases. Consequently, targeting constructs are generally at least about 50 to 100 nucleotides long, preferably at least about 250 to 500 nucleotides long, more preferably at least abut 1000 to 2000 nucleotides long, or longer. Construct homology regions (homology clamps) are generally at least about 50 to 100 bases long, preferably at least about 100 to 500 bases long, and more preferably at least about 750 to 2000 bases long. It is believed that homology regions of about 7 to 8 kilobases in length are preferred with one preferred embodiment having a first homology region of about 7 kilobases flanking one side of a replacement region and a second homology region of abut 1 kilobase flanking the other side of said replacement region. The length of homology (i.e., substantial identity) for a homology region may be selected at the discretion of the

practitioner on the basis of the sequence composition and complexity of the endogenous Lps" gene target sequence(s) and guidance provided in the art. Targeting constructs have at least one homology region having a sequence that substantially corresponds to, or is substantially complementary to, an endogenous Lps" gene sequence (e.g., an exon sequence, an enhancer, a promoter, an intronic sequence, or a flanking sequence within about 3-20 kb of an Lps" gene). Such a targeting transgene homology region serves as a template for homologous pairing and recombination with substantially identical endogenous Lps" gene sequence(s). In targeting constructs, such homology regions typically flank the replacement region, which is a region of the targeting construct that is to undergo replacement with the targeted endogenous Lps" gene sequence. Thus, a segment of the targeting construct flanked by homology regions can replace a segment of an endogenous Lps" gene sequence by double-crossover homologous recombination. Homology regions and targeting regions are linked together in conventional linear polynucleotide linkage (5' to 3' phosphodiester backbone). Targeting constructs are generally double-stranded DNA molecules, most usually linear.

Typically, targeting constructs used for functionally disrupting endogenous Lps" genes will comprise at least two homology regions separated by a nonhomologous sequence which contains an expression cassette encoding a selectable marker, such as neo (Smith and Berg (1984) Cold Spring Harbor Symp. Ouant. Biol. 49: 171; Sedivy and Sharp (1989) Proc. Natl. Acad. Sci. (U.S.A.) 86: 227; Thomas and Capechi (1987), Cell 51: 503. However, some targeting transgenes may have the homology region(s) flanking only one side of a nonhomologous sequence. Targeting transgenes of the invention may also be of the type referred to in the art as "hit-and-run" or "in-and-out" transgenes (Valancius and Smithies (1991) Mol. Cell. Biol. 11: 1402; Donehower et al. (1992) Nature

356: 215; (1991) J.NIH Res. 3: 59; which are incorporated herein by reference).

The positive selection expression cassette encodes a selectable marker which affords a means for selecting cells which have integrated targeting transgene sequences spanning the positive selection expression cassette. The negative selection expression cassette encodes a selectable marker which affords a

means for selecting cells which do not have an integrated copy of the negative selection expression cassette. Thus, by a combination positive-negative selection protocol, it is possible to select cells that have undergone homologous replacement recombination and incorporated the portion of the transgene between the homology regions (i.e., the replacement region) into a chromosomal location by selecting for the presence of the positive marker and for the absence of the negative marker.

Preferred expression cassettes for inclusion in the targeting constructs encode and express a selectable drug resistance marker and/or a HSV thymidine kinase enzyme. Suitable drug resistance genes include, for example: gpt (xanthine-guanine phosphor ibosytltransf erase), which can be selected for with mycophenolic acid; neo (neomycin phosphotransferase), which can be selected for with G418 or hygromycin; and DFHR (dihydrofolate reductase), which can be selected for with methotrexate (Mulligan and Berg (1981) Proc. Natl. Acad. Sci. (U.S.A.) 78: 2072; Souther and Berg (1982) J. Mol Appl. Genet. 1: 327; which are incorporated herein by reference).

Selection for correctly targeted recombinants will generally employ at least positive selection, wherein a nonhomologous expression cassette encodes and expresses a functional protein (e.g., neo or gpt) that confers a selectable phenotype to targeted cells harboring the endogenously integrated expression cassette, so that, by addition of a selection agent (e.g. , G418 or mycophenolic acid) such targeted cells have a growth or survival advantage over cells which do not have an integrated expression cassette.

It is preferable that selection for correctly targeted homologous recombinants also employ negative selection, so that cells bearing only nonhomologous integration of the transgene are selected against. Typically, such negative selection employs an expression cassette encoding the herpes simplex virus thymidine kinase gene (HSV tk) positioned in the transgene so that it should integrate only by nonhomologous recombination. Such positioning generally is accomplished by linking the HSV tk expression cassette (or other negative selection cassette) distal to the recombinogenic homology regions so that double-crossover replacement recombination of the homology regions transfers the positive selection

expression cassette to a chromosomal location but does not transfer the HSV tk gene (or other negative selection cassette) to a chromosomal location. A nucleoside analog, ganciclovir, which is preferentially toxic to cells expressing HSV tk, can be used as the negative selection agent, as it selects for cells which do not have an integrated HSV tk expression cassette. FIAU may also be used as a selective agent to select for cells lacking HSV tk.

In order to reduce the background of cells having incorrectly integrated targeting construct sequences, a combination positive-negative selection scheme is typically used (Mansour etal. , Nature 336: 348-352 (1988) incorporated herein by reference).

Generally targeting constructs preferably include: (1) a positive selection expression cassette flanked by two homology regions that are substantially identical to host cell endogenous Lps" gene sequences, and (2) a distal negative selection expression cassette. However, targeting constructs which include only a positive selection expression cassette can also be used. Typically, a targeting construct will contain a positive selection expression cassette which includes a neo gene linked downstream (i.e., towards the carboxy-terminus of the encoded polypeptide in translational reading frame orientation) of a promoter such as the HSV tk promoter or the pgk promoter. More typically, the targeting transgene will also contain a negative selection expression cassette which includes an HSV tk gene linked downstream of a HSDV tk promoter.

It is preferred that targeting constructs have homology regions that are highly homologous to the predetermined target endogenous DNA sequence(s), preferably isogeneic (i.e., identical sequence). Isogeneic or nearly isogeneic sequences may be obtained by genomic cloning or high-fidelity PCR amplification of genomic DNA from the strain of nonhuman animals which are the source of the ES cells used in the gene targeting procedure.

A targeting construct based on the design employed by Zjilstra, et al , Nature 342: 435-438 (1989) for the successful disruption of the mouse β2- microglobulin gene can be used for disrupting the murine Lps" gene. The neomycin resistance gene (neo), from the plasmid pMclNEO insert uses a hybrid

viral promoter/enhancer sequence to drive neo expression. Tis promoter is active in embryonic stem cells. Therefore, neo can be used as a selectable marker for integration of the knock-out construct. The HSV tk gene is added to the end of the construct as a negative selection marker against random insertion events (Zjilstra et al, supra).

Vectors containing a targeting construct are typically grown in E. coli and then isolated using standard molecular biology methods, or may be synthesized as oligonucleotides. Direct targeted inactivation which does not require prokaryotic or eukaryotic vectors may also be done. Targeting transgenes can be transferred to host cells by any suitable technique, including microinjection, electroporation, lipofection, biolistics, calcium phosphate precipitation, and viral- based vectors, among others. Other methods used to transform mammalian cells include the use of Polybrene, protoplast fusion, and others (see, generally, Sambrook et al. Molecular Cloning: A Laboratory Manual, 2d ed., 1989, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., which is incorporated herein by reference).

For making transgenic non-human animals (which include homologously targeted non-human animals), embryonal stem cells (ΕS cells) are preferred. Murine ΕS cells, such as AB-1 line grown on mitotically inactive SNL76/7 cell feeder layers (McMahon and Bradley (1990) Cell 62: 1073) essentially as described (Robertson, ΕJ. (1987) in Teratocarcinomas and Embryonic Stem Cells: A Practical Approach. ΕJ. Robertson, ed. (Oxford: IRL Press), p. 71-112) may be used for homologous gene targeting. Omer suitable ΕS lines include but are not limited to, the Ε14 line (Hooper et al. (1987) Nature 326: 292-295), the D3 line (Doetschman et al (1985) J. Embryol. Exp. Morphj. 87: 27- 45), and the CCE line (Robertson et al. (1986) Nature 323: 445-448). The success of generating a mouse line from ES cells bearing a specific targeted mutation depends on the pluripotence of the ES cells (i.e., their ability, once injected into a host blastocyst, to participate in embryogenesis and contribute to the germ cells of the resulting animal). The blastocysts containing the injected ES ells are allowed to develop in the uteri of pseudopregnant nonhuman females and are born as

chimeric mice. The resultant transgenic mice are chimeric for cells having inactivated endogenous Lps" loci and are backcrossed and screened for the presence of the correctly targeted transgene(s) by PCR or Southern blot analysis on tail biopsy DNA of offspring so as to identify transgenic mice heterozygous for the inactivated Lps" locus. By performing the appropriate crosses, it is possible to produce a transgenic nonhuman animal homozygous for functionally disrupted Lps" alleles. Such transgenic animals are substantially incapable of making an endogenous Lps" gene product, and can be used as host for transplanted LPS responder cells. The Lps" homozygous null mutant transgenic animals will typically comprise rats or mice, but nonmurine species such as dogs and nonhuman primates, for example, may be utilized.

The Lps" homozygous null mutant transgenic animals are preferably immunoincompetent, so that they may serve as hosts for transplanted LPS- responder cells without host rejection of the transplanted cells. The host may be rendered immunoincompetent by whole body irradiation or drugs. These methods have been used for immunosuppression of several species permitting xenograft growth. Other "artificial" methodologies for immunosuppression to permit xenografting include thymectomy and anti-thymocyte serum (ATS). These and other such methods are described by Langdon and Smyth, "Types of Immunodeficiency in Mice", Chapter 2 in The Nude Mouse in Oncology Research, E. Boven and B. Winograd, eds., CRC Press, Inc., Boa Raton FL, 1991, p. 11-23, incorporated herein by reference.

More preferably, the null Lps" host animal will comprise a species including any of the spontaneously arising or engineered mutations inducing immunoincompetence. The so-called "nude" mice strains contain the nu mutation which gives rise to hypoplasia. The nu mice, and other strains of immunodeficient mice are described by Langdon and Smyth, supra.

The test agents screened for activity in inhibiting LPS stimulation can be any molecule, compound, or other substance which can be added to a cell culture or administered to the test animal without substantially interfering with cell or animal viability. Suitable test agents may be small molecules, biological

polymers, such as polypeptides, polysaccharides, polynucleotides, and the like. The test compounds will typically be administered to transgenic animals at a dosage of from about 1 ng/kg to 10 mg/kg, usually from about 10 μg/kg to 1 mg/kg. The test agents may be administered to cell cultures at a dosage in the range of from about 1 ng/ml to 1 mg/ml, usually from about 1 μg/ml to 100 μg/ml.

Test compounds which are able to inhibit LPS stimulation in cell culture or transgenic animals are considered as candidates for further determinations of the ability to block LPS stimulation in animals and humans.

In one embodiment of the invention, the test agent is a molecule which is known to affect the signal transduction pathway mediated through GTP- binding protein. It is believed that, based upon the sequence homology of the herein-described murine Lps" cDNA to the Ran/TC4/Spil molecules, that agents which affect this pathway are the best candidates for screening for ability to block LPS stimulation. Functionally, Ran/TC4/Spil molecules are known to regulate cell-cycle progression (Marsumoto R. and Beach D., Cell 66, 347-360 (1991); Matsumoto T. and Beach D., Molec. Biol. Cell 4, 337-345 (1993)) and mRNA transport (Belhumeur P. et al , Mol. Cell Biol. 13, 2152-2161 (1993)). They have been shown to act as a GTPase switch (Marsumoto R. and Beach D., Cell 66, 347- 360 (1991)), which can prevent the premature initiation of mitosis when the Ran/TC4 is associated with a protein called RCC1 exchange factor (Nishitani H. et al , EMBO J 10, 1555-1564 (1991); Uchida S. et al. , Mol Cell Biol 10, 577-584 (1990)), or which can enhance their GTP hydrolysis activity when they are associated with a GTPase-activating protein (GAP).

In an in vitro test for inhibition of LPS response, the potential LPS response-inhibiting agent, typically a drug, is contacted with a suitable LPS- responder cell line, which is then stimulated by LPS. The cell line is typically a lymphocyte cell line or macrophage cell line. LPS stimulation results in a response in such cells which is typically manifested as, for example, the occurrence of cell proliferation, antibody production, or cytokine release. Various types of human cell lines, particularly B cell lines and macrophage cell lines, are commonly available for this purpose. An example of an

LPS responder cell line is WEHI-23, which is an immature B-lymphoma. An LPS responder macrophage cell line believed useful in the practice of the invention is P388D1. Typically, the individuals from whom these cell lines have been derived have responded to LPS and have undergone endotoxin shock. The degree of stimulation induced by LPS challenge in the drug-treated cells is determined with resort to any one of the known assays which provide either qualitative or quantitative data of LPS-stimulated biological response. Assays include plaque- forming assays which indirectly measure antibody production by B cells against target hemolytic cells, typically sheep red blood cells (SRBC); cell proliferation assays which measure the extent of DNA synthesis in response to LPS stimulation; and cytokine release assays which measure the extent of release of one or more cytokines (typically from macrophages) in response to LPS stimulation.

The plaque forming assay relies on B cells producing antibodies against sheep red blood cells (SRBC), in response to polyclonal activation by LPS. Cells producing antibody to SRBCs are cultured with a suspension of SRBC as indicator cells under condition of LPS stimulation, and dilutions of drug-containing medium, in a well-known chambered apparatus for conducting hemolytic plaque assays. A plaque or zone of red cell lysis in a chamber represents an antibody- producing cell. The extent of plaque formation is an indicia of the degree of LPS stimulation in the presence of the test drug. A hemolytic plaque assay for determining the extent of B cell response to LPS stimulation is described in detail in Example 1 below.

A cell proliferation assay relies on the use of B cell lines, purified B cells or mixtures of cells combining B cells, e.g. spleen cells. The cells are incubated with drug at various concentrations under LPS stimulation. The cells are pulsed with, e.g.[ ³ H]-thymidine. The level of radiolabel uptake, which is proportional to the extent DNA synthesis and therefore of cell proliferation, is determined by liquid scintillation counting. The extent of cell proliferation is an indicia of the degree of LPS stimulation in the presence of the test drug. A particularly useful B cell proliferation assay of the present invention is described by Sultzer, Infec. Immun. 13, 1579-1584 (1976), the entire disclosure of which is incorporated herein by reference. The uptake of [ ³ H]-

thymidine is used to measure the increase of DNA synthesis, and hence cell proliferation, in spleen cells cultured with LPS.

According to a cytokine release assay, drug induced suppression of LPS response is determined indirectly in an assay which measures the proliferative response of a cytokine dependent cell line, e.g. an IL-1 dependent cell line or tissue necrosis factor dependent cell line. Drug inhibition of cell activation is observed as the suppression of cytokine release from the cytokine producer clone, which is assayed as the suppression of cytokine-driven proliferation of the cytokine- dependent cell line. The extent of proliferation of the latter is an indicia of the degree of LPS stimulation in the presence of the test drug.

In each of the foregoing assays, a control is employed according to the practice of the present invention comprising a mutant cell line derived from the LPS-responding parent cell line used as a reagent to screen the candidate drug for activity in blocking LPS stimulation. The mutant cell line comprises a sub-strain of the parent cell line, and contains a homozygous Lps" null mutation generated as described herein by disruptive gene targeting of the Lps" gene locus. The mutant cell line is substantially nonresponsive to LPS stimulation. The level of the response of the parent cell line and mutant line to LPS stimulation in the presence of the same test drug is then compared. Use of the homozygous Lps" null mutant cell line as a control ensures that the drug action on blocking LPS stimulation of the parent cell line is due specifically to the action of the wild-type Lps" gene present in the parent line, as the sole difference in the genetic makeup of the parent and mutant lines lies in the presence or absence of a functioning Lps" gene. The parent and mutant cell lines may comprise cell lines arising from any LPS-responsive organism. Human and murine cell lines, particularly mouse cell lines, are preferred. The cDNA nucleotide sequence of the human Lps" cDNA is disclosed by Drivas et al, supra, incorporated herein by reference. The murine Lps" cDNA is disclosed herein as SEQ ID NOJ . In an in vivo test for inhibition of LPS response, the potential LPS response-inhibiting agent, typically a drug, is administered to a nonhuman host transgenic animal which contains a homozygous Lps" null mutation, which further

harbors transplanted LPS -responding cells. The drug is administered under conditions of LPS stimulation. The response of the host animal, which is a function solely of the transplanted LPS-responding cells, is assayed.

LPS may be administered to me test animal by the intravenous or intraperitoneal route. The latter is preferred. Test drugs may be given prior to, at the time of, or subsequent to the administration of LPS challenge. For lethality measurements of endotoxin shock, LD50 measurements are routinely used. A range of dosages is therefore first tested to establish an LD50 statistic.

Death within 72-96 hours is used as the endpoint for endotoxin shock. Alternatively, animals can be tested for cytokines in their blood after LPS stimulation by standard tests as an indicator of LPS stimulation. The doses of LPS used in this instance may be sublethal or in the lethal range.

The transplanted cells are preferably human cells, most preferably human hematopoietic stem cells (HSCs) or LPS-responding cells such as B cells or monocytes. Human HSCs have the capacity to give rise to a large number of functional B cells and monocytes. Introduction of such human cells into and LPS- nonresponding transgenic mouse strain provides an excellent in vivo system for testing since the background LPS stimulation signal will be very low; therefore, the chance of detecting an LPS response signal is higher than in another system in which the background host cells are also sensitive to LPS stimulation.

The practice of the invention is illustrated by the following noniimiting examples.

Example 1 Lps" cDNA

A. Preparation of C3H/HeOuJ cDNA Library

(1) Isolation of mRNA

Messenger RNA was extracted from LPS-stimulated C3H/HeOυJ spleen cells according to the method of Sultzer et al, Immunobiology 187, 257-271

(1993). Accordingly, LPS responder cells from C3H/HeOuJ mice were processed to remove erythrocytes and then suspended at 2.5 million cells per ml in RPMI

culture media supplemented with 5% fetal calf serum. Purified LPS from Salmonella typhi 0-901 was added to the cell suspension at 10 μg per ml and the cell cultures were incubated at 37°C in an atmosphere of 10% CO ₂ for 24 h. A total of 15 x 10 ⁸ cultured cells were collected, washed in phosphate-buffered saline and lysed in 5M guanidine thiocyanate solution. The lysates were then spun over a 100 w/v cesium chloride solution at a speed of 32,000 rpm using a SW42 rotor and a Beckman ultracentrifuge for 16 hours. Cellular RNA at the bottom of the tubes were collected and to obtain messenger RNA, they were passed through an oligo-dT cellulose column at least twice.

(2) Construction of cDNA Library

The mRNA were used to construct a cDNA library according to the method of Okayama and Bert, Mol. Cell. Biol. 2, 161-169 (1982), utilizing the pCD expression vector plasmid. The method generates cDNAs from full-length transcripts at high efficiency. The library consisted of six sublibraries (LB2J- LB2.6) ranging in size from 1-2 x IO ⁴ independent clones.

B. Electroporation of C3H/HeOuJ cDNA into C3H/HeJ LPS- nonresponder Spleen Cells

(1) Electroporation Standardization

Standardization experiments using immortalized preB 70Z/3 lymphoid cells under optimal electroporation conditions indicated that the efficiency of gene transfer according to this method was about 1 in 10,000 cells electroporated.

(2) Preparation of C3H/HeJ LPS-nonresponder Spleen Cells Enriched with Plaque-forming B Cells

C3H/HeJ spleens were removed and placed in cytotoxic medium (CyM, Cedarlane Labs), and red blood cells were lysed by treating the cells with

0.75 % NH ₄ C1 twice. T cells were removed by treatment with anti-Thy 1.2 antibody

(Sigma) at 1: 10,000 dilution in the presence of a LOW TOX-M rabbit complement

(Accurate Chemical Co.) at 37°C. The cells were then washed with R5 (RPMI with 5% FCS) and resuspended at 3 x IO ⁷ viable cells per ml.

(3) Electroporating and Assav of cDNA Sublibraries The plasmid DNA of five of the six sublibraries obtained was electroporated (BioRad GENEPULSER) into the C3H/HeJ Lps-nonresponder spleen cells, and a modified Jerne's plaque-forming assay was carried out, as follows. One milligram of C3H/HeOuJ sublibrary plasmid cDNA was added to 1 ml of resuspended cells and transferred to an ice cold 0.4 cm GENEPULSER curvette. The mixmre was electroporated at 250V and 960uF with a BioRad GENEPULSER and incubated on ice for 10-30 minutes. These electroporated cells were then stimulated with 100 mg/ml of S. typhi lipopolysaccharide and incubated for 72 hours in an incubator containing 7% O ₂ , 10% CO ₂ and balanced with N ₂ . Each day, the cells were fed with R5. After 72 hours of incubation, the cells were harvested and washed in MSS buffer solution. 200,000 viable cells were mixed with 10% sheep red blood cells (Organon Teknika; Cappel) and liquified 0.7 agarose in MSS buffer and plated on a solidified 0.7% agarose plate. These plates were then overlayered with guinea pig complement at 1: 15 dilution and incubated in 7% O ₂ , 10% CO ₂ , and balanced with N ₂ for 2-3 hours, after which, the numbers of plaques were recorded.

Since a large number of cells were needed, electroporation of spleen cells with eacn sublibrary plasmid DNA was done separately at different times. Because of that, negative control (no DNA, but still stimulated with LPS) and positive control (no DNA, but stimulated with EP) were incorporated each time. The results are shown in Table 1. EP= endotoxin associated protein which activates C3H/HeJ cells (Sultzer B.M., Castagna R., Bandekar J. & Wong P.M.C., Immunobiology 187, 257-271 (1993)). TMTC=Too Many To Count.

Table 1. LPS-stimulated Plaque-Forming Cells after Electroporation of pCD library DNA.

DNA Electro. Stimulant PFC/ 10 Ratio dishes (+ve/Ctl)

LB2.3 + LPS 7 0.78

None + LPS 9

None - EP TMTC

LB2.4 + LPS 11 1.38

None + LPS 8

None - EP TMTC

LB2.5 + LPS 7 0.58

None + LPS 12

None - EP TMTC

LB2.2 + LPS 22 5.5

None + LPS 4

None - EP TMTC

The results in Table 1 suggest that electroporation of spleen cells using plasmid DNA from sublibrary LB2.2 gave higher number of plaques compared to those with other sublibrary DNA or with no plasmid DNA at all.

C. Subdivision of Positive Sublibrary

Sublibrary LB2.2, containing 2 x IO ⁴ independent clones, was subdivided into four sub-sublibraries, LB2.2.A-D, each having 6,000 independent clones. From those, LB2.2.D was positive and was therefore divided further into ten sublibraries, LB2.2.DJ-10, each having 600 independent clones. LB2.2.DJ was tested positive and was subdivided into six sublibraries, LB2.2.DJ.A-F, with

100 pooled clones. Among them, LB2.2.DJ.B was positive and was then divided into ten pools of sublibraries with ten independent clones per sublibrary. From these, LB2.2.DJ.B.3 was tested positive. After further division into 10 single clones, among them, LB2.2.D.1.B.3.S was tested positive. This plasmid has been given the name "pCD-Lps"". The cDNA is named "Lps"". The plasmid preparation contained a single species of plasmid DNA; its introduction and expression in CeH/HeJ spleen cells led to an increase in the numbers of PFCs.

The ratio of PFCs from the "positive" over those of "negatives" increased as the number of independent colonies in each sublibrary decreased (Figure 1). These data are consistent with the idea that the enrichment of one particular clone of plasmid DNA accounts for the increase in the numbers of PFC detected.

To explain the data more consistently, the data in Fig. 1 were normalized based on the background values from the plaque ratios of 2.5/no DNA. These two experiments (Fig. 1, *) were done with the same batch of mice. For consistency, the data from one experiment that included the 2.2.DJ.B.3 sublibrary was normalized to the data from another experiment that included 2.2. D J.B, 2.2.DJ.B.3 and 2.2.DJ.B.3.5. Therefore the two ratios (Fig. 1, **) are normalized based on the 2.2. D J.B sublibrary.

D. DNA Sequencing

The LB2.2.D.1.B.3.S cDNA sequence was analyzed using a GCYG program (Devereux J., Haeberli P., Smithies O., Nucl Acid Res 12, 387-395 (1984)) and found to contain an open reading frame of 648 bp, which predicts a protein of 216 amino acids with a molecular weight of 24,423 daltons (Figure 3A).

Example 2 Hybridization of Positive and Negative pCD Sublibraries with pcD-Lps" cDNA To further confirm that the expression of the Lps" cDNA resulted in LPS-stimulated plaque formation by electroporated C3H/HeJ spleen cells, plasmid preparations from various cDNA sublibraries were reanalyzed. These

plasmids were digested with Xhol, an enzyme whose recognition sites flank all cDNA inserts at both ends in the pCD vector. The Xhol recognition sequence is rarely represented in the mammalian genome, and therefore, the frequency of Xhol site within the cDNA is low. Plasmid DNA were extracted from bacteria of various pCD sublibraries made from LPS-stimulated C3H/HeOuJ spleen cells. 5-10ug of DNA were digested with Xhol, which has two unique sites flanking all cDNA inserts in the pCD vector. After digestion, the DNA were fractionated by agarose gel, and stained with ethidium bromide. The results appear in Fig. 2, upper panel. The DNA in the gel were then transferred to a nylon filter, fixed by UV cross-linking, hybridized with a ³² P-dCTP labeled probe from a 0.8kb BamHI /PstI DNA fragment specific to the Lps" cDNA, and exposed to an X-ray film. The results appear in Fig. 2, lower panel.

"PFC negative" and "PFC positive" are the groups of plasmid DNA from sublibraries that were tested negative or positive by the plaque assay. LB2.2 was the PFC-positive sub-sublibrary, followed by LB2.2.D, followed by LB2.2.DJ, etc. All of the sublibrary plasmid DNA that were tested positive by the plaque assay produced a 1.5kb signal, whereas those tested negative by the plaque assay did not (Figure 2). In addition, the intensity of the signal was proportional to the estimated representation of pCD-Lps" (Figure 2). These data show that the expression of pCD-Lps" cDNA in C3H/HeJ mice accounts for the LPS-stimulated plaque formation by polyclonal activated B cells.

Example 3 Northern Blot Analysis of RNA from C3H/HeJ and C3H/HeOuJ Organs

RNA were extracted from various organs of C3H/HeJ and C3H/HeOuJ mice: bone marrow, spleen, lung, kidney and thymus. Ten micrograms of total RNA from each sample were loaded onto a 1 % formaldehyde gel, fractionated, transferred to a nylon filter and hybridized with a ³² P-dCTP labeled probe from a 0.8kb BamHl/Pstl DNA fragment specific to the Lps" cDNA, and the filter was subjected to phosphoimage analysis. A 1.2kb band was present

in all samples analyzed (Fig. 4, upper panel). To normalize the signal for comparison, the filter was stripped of Lps"-specific probe, and was subsequently hybridized with a GAPDH-specific probe (Fig. 4, lower panel). After normalizing the intensity of the GAPDH band, the amount of Lps" RNA (C3H/HeOuJ) does not appear to differ significantly from that of Lps ^d RNA (C3H/HeJ). Ou = C3H/HeOuJ; He = C3H/HeJ; BM = bone marrow; Spl = Spleen; Lu = Lung; Kid = Kidney; and Thy = Thymus. The results show that both the Lps" and Lps ^d genes are expressed and the steady level of gene expression is similar in both strains of mice. This data suggests that small genetic changes must have occurred within the coding sequences of Lps ^d . As a result, either a non-functioning protein is produced in the Lps ^d mutation or a post-translational abnormality has occurred. All references cited with respect to synthetic, preparative and analytical procedures are incorporated herein by reference.

The present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof and, accordingly, reference should be made to the appended claims, rather than to the foregoing specification, as indication the scope of the invention.

SEOUENCE LISTING (1) GENERAL INFORMATION:

(i) APPLICANTS: Temple University - Of The

Commonwealth System of Higher Education; The Research Foundation of State University of New

York; Peter M.C. Wong, Barnet M. Sultzer, Hong Chen, Anthony D. Kang and Raymond Castagna (ii) INVENTORS: Peter M.C. Wong, Barnet M.

Sultzer, Hong Chen, Anthony D. Kang, Raymond Castagna

(iii) TITLE OF INVENTION: LIPOPOLYSACCHARIDE ENDOTOXIN RESPONSE GENE AND USE THEREOF IN SCREENING AGENTS FOR ANTI- ENDOTOXEMIA ACTIVITY

(iv) NUMBER OF SEQUENCES: 2

(v) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Seidel, Gonda, Lavorgna & Monaco, P.C. (B) STREET: Suite 1800, Two Perm Center Plaza

(C) CITY: Philadelphia

(D) STATE: Pennsylvania

(E) COUNTRY: U.S.A.

(F) ZD?: 19102 (vi) COMPUTER READABLE FORM:

(A) MEDHJM TYPE: Diskette, 3.50 inch, 720 Kb

(B) COMPUTER: IBM PS/2

(C) OPERATING SYSTEM: MS-DOS

(D) SOFTWARE: WordPerfect 5.1 (vii) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER:

(B) FILING DATE:

(C) CLASSDJ1CATION: (viii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: 60/001,329 (B) FILING DATE: 21 July 1995

(ix) ATTORNEY/AGENT INFORMATION:

(A) NAME: Monaco, Daniel A.

(B) REGISTRATION NUMBER: 30,480

(C) REFERENCE/DOCKET NUMBER: 6056-209 PC (x) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: (215) 568-8383

(B) TELEFAX: (215) 568-5549

(C) TELEX: None (2) INFORMATION FOR SEQ ID NO:l: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1166 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear (ii) SEQUENCE DESCRTPTION: SEQ ID NO:l:

CCC CCC TCC GCG CGC CGG CGT CCG CTG CGT CTC CGG CAT TTG AAT CGC GTC 51.

CGC CAT CTT TCC AGC TCC AGT CGG ACA GGC GCG CAG ACT CTT CTG GAA GGA 102

TCC GCC GCG ATG GCC GCC CAG GGA GAG CCG CAG GTC CAG TTC AAG CTC GTC 153 CTG GTG GGC GAC GGC GGC ACC GGG AAG ACA ACC TTC GTG AAG CGC CAC TTG 204

ACG GGC GAG TTT GAG AAG AAG TAT GTA GCC ACC CTG GGC GTG GAG GTG CAC 255

CCG CTC GTC TTC CAT ACC AAC AGA GGA CCC ATC AAG TTC AAC GTG TGG GAC 306

ACG GCC GGC CAG GAG AAG TTC GGG GGC CTG CGC GAT GGC TAC TAC ATC CAA 357

GCC CAG TGT GCC ATT ATA ATG TTT GAT GTA ACC TCA AGA GTT ACT TAC AAG 408 AAT GTA CCT AAC TGG CAT AGA GAT CTG GTA CGA GTG TGT GAA AAC ATC CCC 459

ATT GTA TTG TGT GGC AAC AAA GTG GAT ATT AAA GAC AGG AAA GTG AAG GCA 510

AAA TCT ATT GTC TTC CAC CGG AAG AAG AAT CTT CAG TAC TAT GAC ATT TCT 561

GCC AAA AGT AAC TAC AAC TTT GAA AAG CCT TTC CTC TGG CTT GCC AGA AAG 612

CTC ATT GGA GAT CCT AAC TTG GAG TTT GTT GCC ATG CCT GCT CTT GCC CCA 663 CCT GAG GTG GTC ATG GAC CCA GCT TTG GCA GCA CAG TAC GAG CAT GAT TTA 714

GAG GTT GCT CAG ACG ACT GCT CTC CCA GAT GAG GAT GAT GAC CTG TGA GAA 765

AGT GAA GCT GGA TGC CCT GCG TCA GAA GTC TAG TTT TAT AGG CAA CTG TCC 816

TGT GAT GTC AAG CGG TGC AGC GCG TGT GCC ACC TTA TTT AGC TAA GCA GAT 867

CGT GTA CTT CAT TGG GAT GCT GAA GGA GAT GAA TGG GCT TCG AGT GAA TGT 918

GGC AGT TAA ACA TAC CTT CAT TTT TTG GAC TTG CAT ATT TAG CTG TTT GGA 969 ACA GAG TTG TTT CTT TTC TGA ATT TCA AAG ATA AGA CTG CTG CAG TCC CAT 1020 CGC AAT ATC CAG TGG GGA AAT CTT GTT TGT TAC TGT CAT TCC CAT TCT TTT 1071 CGT TAG AAT CAG AAT AAA GTT GTA TTT CAA ATA ATC TAA AAA AAA AAA AAA 1122 AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AA 1166

(3) INFORMATION FOR SEQ TD NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 216 amino acids (B) TYPE: amino acids

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) SEQUENCE DESCRTPTION: SEQ ID NO:2:

Met Ala Ala Gin Gly Glu Pro Gin Val Gin Phe Lys Leu Val

5 10

Leu Val Gly Asp Gly Gly Thr Gly Lys Thr Thr Phe Val Lys Arg His Leu

15 20 25 30

Thr Gly Glu Phe Glu Lys Lys Tyr Val Ala Thr Leu Gly Val Glu Val His 35 40 45

Pro Leu Val Phe His Thr Asn Arg Gly Pro Ile Lys Phe Asn Val Trp Asp

50 55 60 65

Thr Ala Gly Gin Glu Lys Phe Gly Gly Leu Arg Asp Gly Tyr Tyr Ile Gin

70 75 80

Ala Gin Cys Ala Ile Ile Met Phe Asp Val Thr Ser Arg Val Thr Tyr Lys 80 90 95

Asn Val Pro Asn Trp His Arg Asp Leu Val Arg Val Cys Glu Asn Ile Pro

100 105 110 115

Ile Val Leu Cys Gly Asn Lys Val Asp Ile Lys Asp Arg Lys Val Lys Ala 120 125 130

Lys Ser Ile Val Phe His Arg Lys Lys Asn Leu Gin Tyr Tyr Asp Ile Ser 135 140 145 150

Ala Lys Ser Asn Tyr Asn Phe Glu Lys Pro Phe Leu Trp Leu Ala Arg Lys

155 160 165

Leu lie Gly Asp Pro Asn Leu Glu Phe Val Ala Met Pro Ala Leu Ala Pro

170 175 180

Pro Glu Val Val Met Asp Pro Ala Leu Ala Ala Gin Tyr Glu His Asp Leu

185 190 195 200

Glu Val Ala Gin Thr Thr Ala Leu Pro Asp Glu Asp Asp Asp Leu

205 210 215