DETAILED DESCRIPTION

[0025] This invention provides an effective means for increasing the local number of cytotoxic T cells (CTL) and/or natural killer (NK) cells. Lymphotactin, when administered, has been observed to increase the abundance of these cell types. This increase may result from attraction of said cell types, from proliferation of exisitng cells, or a combination of both. Antagonists will also block such effect on such cells in circumstances, e.g., in MHC mismatched contexts, including tissue or organ rejection and graft vs. host disease, where such responses are undesireable.

[0026] In addition, lymphotactin may have positive effects on antiviral processes, whether directly by blocking a chemokine receptor used as a viral coreceptor, or to attract the NK or CTL cells attracted to sites of virally infected cells.

[0027] Lymphotactin also exhibits an activity of inducing cell cycle quiescence in hematopoietic stem cells. As such, when various drugs are administered which specifically act on proliferating cells, these stem cells are largely unaffected. Such drugs include chemotherapy reagents, e.g., nucleotide or nucleoside analogs, or radiation therapy, which mutagenizes cells which have a slowed repair process.

[0028] Mammalian lymphotactin (Itn) has been well described, e.g., in U.S. Ser. No. 08/329,704 and related cases, and Kelner, et al. (1994) Science 266:1395-1399. The preferred mammalian lymphotactin is a natural human lymphotactin.

[0029] Lymphotactin is produced in a variety of ways, both conventional and less conventional. A general review of recombinant protein production can be found, e.g., Sambrook, et al. (1989) Molecular Cloning: A Laboratory Manual Cold Spring Harbor Press, N.Y., or in Ausubel, et al. (Eds.; 1989 and Suppl.) Current Protocols in Molecular Biology, Green/Wiley, NY, N.Y. Natural sources of protein may be used as a source, and protein isolated therefrom.

[0030] Another method for preparing protein or peptides includes polypeptide synthesis methods. See, e.g., Merrifield (1988) Science 232:341-347; and Dawson, et al. (1994) Science 266:776-779.

[0031] In antagonist embodiments, the antagonist can be an antibody or fragment specific for binding to lymphotactin, e.g., which interferes with the chemokine binding to its receptor. Anti-lymphotactin reagents are produced in a variety of ways, both conventional and less conventional. A general review of antibody production can be found, e.g., in Harlow and Lane (1988) Antibodies: A Laboratory Manual Cold Spring Harbor Press, N.Y., or in Colligan, et al. (Eds.; 1991 and Suppl.) Current Protocols in Immunology, Green/Wiley, NY, N.Y. Antibodies can be polyclonal mixture or monoclonal. Antibodies can be intact immunoglobulins or fragments thereof, derived from natural sources or from recombinant sources. Antibodies also include the immunoreactive portions of intact immunoglobulins, e.g., antigen binding portions such as Fab, Fv, etc.

[0032] In brief, methods to obtain anti-lymphotactin antibodies involve administering an amount of antigen, e.g., a fragment, sufficient to induce a humoral response in a mammal. The antibodies are either collected from the mammal's sera or lymphocytes removed, immortalized, and those cell clones secreting the desired antibodies isolated and cultured for harvest of the desired antibodies.

[0033] The quantities of reagents necessary for effective therapy will depend upon many different factors, including means of administration, target site, physiological state of the patient, and other medicants administered. Thus, treatment dosages should be titrated to optimize safety and efficacy. Typically, dosages used in vitro may provide useful guidance in the amounts useful for in situ administration of these reagents. Animal testing of effective doses for treatment of particular disorders will provide further predictive indication of human dosage. Various considerations are described, e.g., in Gilman, et al. (eds.) (1990) Goodman and Gilman's: The Pharmacological Bases of Therapeutics (8th ed.) Pergamon Press; and (1990) Remington's Pharmaceutical Sciences (17th ed.) Mack Publishing Co., Easton, Pa. Methods for administration are discussed therein and below, e.g., for oral, intravenous, intraperitoneal, or intramuscular administration, transdermal diffusion, and others. Pharmaceutically acceptable carriers will include water, saline, buffers, and other compounds described, e.g., in the Merck Index, Merck & Co., Rahway, N.J. Dosage ranges would ordinarily be expected to be in amounts lower than 1 mM concentrations, typically less than about 10 μM concentrations, usually less than about 100 nM, preferably less than about 10 μM (picomolar), and most preferably less than about 1 fM (femtomolar), with an appropriate carrier. Slow release formulations, or a slow release apparatus will often be utilized for continuous administration.

[0034] Thymokines, fragments thereof, and antibodies to it or its fragments, antagonists, and agonists, may be administered directly to the host to be treated or, depending on the size of the compounds, it may be desirable to conjugate them to carrier proteins such as ovalbumin or serum albumin prior to their administration. Therapeutic formulations may be administered in any conventional dosage formulation. While it is possible for the active ingredient to be administered alone, it is preferable to present it as a pharmaceutical formulation. Formulations typically comprise at least one active ingredient, as defined above, together with one or more acceptable carriers thereof. Each carrier should be both pharmaceutically and physiologically acceptable in the sense of being compatible with the other ingredients and not injurious to the patient. Formulations include those suitable for oral, rectal, nasal, or parenteral (including subcutaneous, intramuscular, intravenous and intradermal) administration. The formulations may conveniently be presented in unit dosage form and may be prepared by any methods well known in the art of pharmacy. See, e.g., Gilman, et al. (eds.) (1990) Goodman and Gilman's: The Pharmacological Bases of Therapeutics (8th ed.) Pergamon Press; and (1990) Remington's Pharmaceutical Sciences (17th ed.) Mack Publishing Co., Easton, Pa.; Avis, et al. (eds.) (1993) Pharmaceutical Dosage Forms: Parenteral Medications Dekker, N.Y.; Lieberman, et al. (eds.) (1990) Pharmaceutical Dosage Forms: Tablets Dekker, N.Y.; and Lieberman, et al. (eds.) (1990) Pharmaceutical Dosage Forms: Disperse Systems Dekker, N.Y. The therapy of this invention may be combined with or used in association with other chemotherapeutic or chemopreventive agents.

[0035] These methods require adequate sources of lymphotactin as antigens. The antigens can either be intact lymphotactin or immunoreactive peptides. Recombinant expression of lymphotactin is a convenient means for obtaining lymphotactin for use as antigens. For a general review of the applicable recombinant technology see Sambrook, et al. (1989) Molecular Cloning—A Laboratory Manual, Cold Spring Harbor Press, CSH, NY Specific techniques for expressing and purifying lymphotactin are known. Expression of lymphotactin is described in PCT/US03554(WO/91/00349) and in Malefyt, et al. (1992) Curr. Opin. Immunology 4:314-320. Alternatively, peptide synthesis may be used to obtain intact or immunoreactive portions of lymphotactin.

[0036] The antibodies for use in this invention are preferably autologous for the patient thereby minimizing further immunological problems. Immunodeficient individuals will tend to be less reactive to non-self antibodies, and thus non-self antibodies derived from cells of the same species are also useful. Antibodies of different species are useful but means to control possible adverse immunoreactions must be undertaken. For example, humanized rat antibodies can minimize immune responses in human patients.

[0037] The antibodies for use in this invention are typically neutralizing antibodies and will preferably have binding constants which are greater than or approximates the affinity of lymphotactin for its natural receptor. Antibodies having a binding constant 100-fold less than these cytokines for their corresponding receptors are less preferred. Binding comparisons are carried out using standard equilibrium methods. The basic technology is described in Chpt 25 of Vol. 1: Immunochemistry, Ed. D. M. Weir, 4th Ed. 1986, Blackwell Scientific Publ. 25.1-25.30. Alternatively, one can use an assay for determining the molar excess of antibody which neutralizes a defined amount of IL-10 in a standard in vitro bioassay. Examples of such assays are found in Mosmann and Fong (1989) J. Immunol. Methods 116:151 (IL-4) and Fiorentino et al., 1989, J. Exp. Med. 170:2081. A reasonable range are those antibodies which neutralize a given amount of lymphotactin in a 10 to 1,000 fold excess.

[0038] The means of administration of the antagonists, e.g., αlymphotactin, are typically parenteral, preferably intravenous. The antagonists are infused into the patient using standard intravenous techniques. The antagonists are first suspended into a sterile, physiologically-compatible media, such as phosphate buffered saline. Pharmaceutically acceptable excipients such as lecithin, glucose, dextrose, antibiotics may also be included with the antagonists.

[0039] When the antagonists are antibodies, they are administered in an amount which provides circulating levels of anti-lymphotactin at about 1 to 150 μg/ml and preferably 10 to 100 μg/ml of sera for each antibody. The antibodies typically will have a 2-7 day half-life and repeated administration is necessary when levels of anti-lymphotactin are below these levels. Total amount of anti-lymphotactin applied per administration are between 1 and 10 mg/kg of body weight for each antibody.

[0040] The broad scope of this invention is best understood with reference to the following examples, which are not intended to limit the invention to specific embodiments.

EXPERIMENTAL

EXAMPLE 1

[0041] In vivo Effects: Methodology

[0042] Abbreviations used herein: Lptn, lymphotactin; mLptn, murine lymphotactin; hLptn, human lymphotactin; CT-hLptn, carboxy-terminal truncated human lymphotactin; IPTG, isopropyl β-D-thiogalactopyranoside; HRP, horse radish peroxidase; AP, alkaline phosphatase; RT, room temperature; min, minute.

[0043] Injection of lymphotactin (Lptn) into the peritoneum caused an influx of lymphocytes at 24 hours. Phenotypic analysis of the cellular influx showed that a large proportion of these cells were T lymphocytes, however, a large number of natural killer (NK) cells were also present. This effect of mLptn was specific since the cellular influx was blocked with a mLptn-specific monoclonal antibody (mAb). Similar results were observed when Lptn was injected subcutaneously and the tissue analyzed by immunohistochemistry using an anti-CD3_ε mAb. Microchemotaxis assays confirmed that murine NK cells respond to mLptn and also showed human NK clones to be similarly responsive to recombinant human Lptn (hLptn). Immunohistochemical analysis of IL-2 activated murine NK cells and Northern analysis of human NK clones revealed that these cells also produce Lptn, suggesting that a self-regulatory migration mechanism exists in NK cells. Together these data confirm, in vivo, the lymphocyte-specificity of Lptn previously observed in vitro and extend its chemotactic effects to the NK cell lineage. The functional consequences of truncating the carboxy terminus of hLptn (CT-hLptn) were also investigated. This truncated molecule (which is missing the carboxy-terminal 22 amino acids of hLptn) had no detectable activity on human peripheral blood lymphocytes. In addition, while hLptn was found to attract murine splenocytes in vitro, the carboxy-terminal truncated hLptn was again inactive on murine splenocytes. This observation indicates the presence of structural features in the carboxy terminus of Lptn that are necessary for its biological activity.

[0044] The chemokines are a large family of chemoattractant cytokines whose members have been subdivided into three subfamilies on the basis of the positions of four invariant cysteines (reviewed in Baggiolini, et al. (1994) Adv. in Immun. 55:97-179; and Schall (1994) pp. 419-460, in Thomson (ed.) The Cytokine Handbook Academic Press, New York, N.Y.). The first two cysteines of the C—X—C or α chemokine subfamily are separated by a single amino acid residue, while the C—C or β chemokine subfamily members all have the first two cysteines immediately adjacent to one another. We recently described the cloning of murine Lymphotactin (Lptn) (Kelner, et al. (1994) Science 266:1395-1399) a third type of chemokine. This protein, designated the C or γ chemokine, lacks two of the four invariant cysteines (C₁ and C₃) normally found in chemokines (Kelner, et al. (1994); Kennedy, et al. (1995) J. Immunol. 155:203-209; Müller, et al. (1995) Eur. J. Immunol. 25:1744-1748; and Yoshida, et al. (1995) FEBS Lett. 360:155-159) and thus can form only one of the two intrachain disulfide bonds found in the α and β chemokines. A human form of Lptn was subsequently described (Kennedy, et al. (1995) J. Immunol. 155:203-209; Müller, et al. (1995) Eur. J. Immunol. 25:1744-1748; and Yoshida, et al. (1995) FEBS Lett. 360:155-159) which also lacks two cysteine residues. Lptn also has an extended carboxy terminus which is unusual among chemokines and further distinguishes it structurally. The gene coding for Lptn has been localized to chromosome 1 in both mouse and human, which is a different location from either the C—C or C—X—C chemokines. Lptn also differs functionally from other chemokines since it has no activity on either macrophages or neutrophils, and appears instead to be specific for lymphocytes. The in vitro chemotactic activity of Lptn has been demonstrated on both CD4⁺ and CD8⁺ T cells of mouse and human. Murine thymocytes, including the CD4⁻CD8⁻, CD4⁺, and CD8⁺ populations, also respond to Lptn. This finding is supported, e.g., by Dorner, et al. (1997) J. Biol. Chem. 272:8817-8823. A recent report has suggested that Lptn could be used in conjunction with IL-2 to enhance the antitumor activity of T lymphocytes. See, e.g., Dilloo, et al. (1996) Nature Med. 2:1090-1095. An exception are the CD4⁺CD8⁺ thymocytes which do not respond to Lptn. Binding of Lptn to either murine or human T cells has been shown to produce an intracellular Ca⁺⁺ flux similar to that observed in response to other chemokines. Finally, Lptn is also unusual in that it does not bind to the duffy antigen receptor for chemokines which can bind many C—C and C—X—C chemokines (Szabo, et al. (1995) J. Biol. Chem. 270:25348-25351.).

[0045] Expression of Lptn is essentially restricted to activated murine CD4⁻CD8⁻CD3⁻CD25⁺CD44⁺ thymocytes (pro-T cells), from which it was originally cloned and to activated, class-I-MHC-restricted T cells, including CD8⁺ T cells from both mouse and human and TCRαβ⁺CD4⁻CD8⁻ T cells in mouse. A low level of expression is detectable in human CD4⁺ T cells though it is likely that this signal arises from a human equivalent to the murine NK1.1⁺CD4⁺ T cells, which also express Lptn.

[0046] Here are reported studies of the in vivo activity of Lptn. The results of these studies confirm the previous in vitro studies in which Lptn was found to be a potent chemoattractant for T lymphocytes. In addition, it is demonstrated that Lptn attracts not only T cells, but also natural killer cells. It is shown that both murine NK cells and human NK clones respond to and produce Lptn in vitro, and that the carboxy terminus of Lptn is necessary for its activity. It is also shown that murine NK cells and human NK clones produce Lptn. Finally, the production and characterization of a neutralizing anti-mLptn mAb is described.

[0047] Cells and cell lines: Purified murine NK cells were obtained from the spleens of Rag-1^−/− mice by Fluorescence Activated Cell Sorting (FACS) for NK1.1⁺Gr-1⁻ cells utilizing a FACstar flow cytometer (Becton Dickinson, Mountain View, Calif.). The purity of NK cells obtained by this method was routinely >97%. Human NK clones were the generous gift of Dr. J. Phillips (DNAX). The A3.2 T cell hybridoma was produced by fusion of BW5147 with purified αβTCR⁺CD4⁻CD8⁻ thymocytes using standard techniques. The resulting hybridomas were subcloned and selected for high IL-4 production, a characteristic of αβTCR⁺CD4⁻CD8⁻ T cells. Vicari and Zlotnik (1996) Immunol. Today 17:71-76.

[0048] Mice: Female CB6F1/Sim BR mice were obtained from Simonsen Laboratories (Gilroy, Calif.). Rag-1^−/− knockout mice were obtained from The Jackson Laboratory (Bar Harbor, Me.) and have subsequently been bred and housed in a specific pathogen-free facility. Animals used were between six and ten weeks of age.

[0049] Recombinant murine lymphotactin (mLptn): The coding region of murine Lptn was subcloned into the pET-3a expression vector (Novagen, Madison, Wis.). Expression plasmids were then used to transform the BL21 (DE3) strain of E. coli (Novagen) containing pLys S. The transformed cells were grown in Luria broth until the O.D. at 560 nm had reached 1.0, at which point they were induced with 0.4M IPTG for an additional 4 h. The cells were then harvested by centrifugation and resuspended in 50 mM Tris (pH 7.5), 5 mM EDTA, 1 mM Pefa Bloc. Soluble protein was purified by chromatography on Q sepharose (pH 8.5; Pharmacia), SP sepharose (pH 8.5; Pharmacia), and POROS R2/M. Fractions containing mLptn were identified by migration of protein on SDS-PAGE. Fractions containing highly pure mLptn were then pooled and quantitated. The identity of the purified protein was confirmed by amino-terminal amino acid sequencing. The recombinant mLptn produced in this manner has an amino terminus sequence of MVGTEVLEQS-. The endotoxin content was <6 EU/mg of purified protein. The FLAG-murine Lptn (FLAG-mLptn) was produced as previously described in Kelner, et al. (1994) Science 266:1395-1399.

[0050] Recombinant human lymphotactin: Full-length recombinant hLptn was obtained from Genzyme Diagnostics (Cambridge, Ma.). The carboxy terminal truncated version of hLptn (CT-hLptn) was produced as follows: a synthetic gene encoding residues 1-71 of human Lptn was constructed with codons optimized for expression in E. coli, and ligated into Xba1 and Kpn1 sites of the T7-driven expression vector pCC101. CT-hLptn was expressed in a BL21 (DE3) strain of E. coli containing pLys S. Transformed cells were grown to an O.D., at 600 nm, of between 0.7 and 1.0, at which time expression was induced with IPTG (0.5 mM). Cells were grown for an additional three hours, harvested by centrifugation, and frozen at −80° C.

[0051] The cell pellets were sonicated in 50 ml of buffer (25 mM sodium acetate, 1 mM EDTA, pH 5.0) and centrifuged (10,000 rpm, 15 min) to pellet the protein. The protein pellet was resuspended in a minimal amount of 6 M guanidine HCl and diluted into 50 volumes of sonication buffer containing 6 M urea. Soluble protein was then loaded onto a 30 ml SP Sepharose Fast Flow column, washed with 2-3 bed volumes of the loading buffer, and eluted with a 0-0.4 M gradient of NaCl. Protein-containing fractions were pooled and dialyzed against the sonication buffer at 4° C. to remove the urea. Oxidation of the single disulfide bond in the protein was accomplished by 1:1 dilution into a redox buffer (100 mM Tris, 1 mM EDTA, 2 mM oxidized glutathione, 0.2 mM reduced glutathione pH=8.0) with stirring overnight at 4° C. Final purification was achieved by reversed-phase HPLC on a Vydac C₄ semi-prep column. Mass spectral analysis confirmed oxidation of the cysteine residues and retention of the amino-terminal methionine.

[0052] Preparation of monoclonal anti-Lptn antibody: A male Lewis rat was immunized intraperitoneally with 25 μg of FLAG-mLptn in 1.0 ml of complete Freund's adjuvant. The rat was immunized three more times, at two week intervals, with 25 μg of FLAG-mLptn in incomplete Freund's adjuvant. Serum was collected after the third and fourth immunizations and assayed for anti-FLAG-mLptn reactivity by ELISA (described below). Four days after the last immunization the animal was euthanized and its splenocytes fused with cells, e.g., as described by Chrétien, et al. (1989) J. Immunol. Meth. 117:67-81. Hybridomas were initially screened against FLAG-mLptn by ELISA followed by a secondary screening against mLptn (non-FLAG). A third screening was performed to identify those hybridomas with reactivity to mLptn, as judged by Western blotting. The 4D8 anti-mLptn hybridoma was selected for cloning by these criteria and the mAb produced by this hybridoma was further characterized as having an IgG2a isotype.

[0053] In vivo experiments: Mice were injected intraperitoneally with 10 μg of purified mLptn in 200 μl of PBS (pH 7.4). Alternatively, mice were injected with 200 μl of PBS containing 10 μg of mLptn that had been preincubated with 500 μg of purified anti-mLptn mAb 4D8 or 500 μg of isotype control for 30 min on ice. Control animals were initially injected with either 200 μl PBS or 200 μl of PBS containing 10 pg of LPS (equivalent to the amount of endotoxin present in the mLptn preparation) as an endotoxin control. No differences were observed between the PBS and PBS+LPS injected animals; therefore PBS+LPS was subsequently used as a negative control. Animals were euthanized after 24 h or 72 h and the peritoneal cavity was then washed with 5 ml of ice-cold PBS. Cell counts were obtained microscopically and adjusted for the amount of PBS recovered according to the formula: adjusted cell number=total cell number×(vol. injected/vol. recovered). Cell viability was consistently above 97%.

[0054] A group of mice was injected subcutaneously in the hind footpads. Each animal was injected with 20 μl of PBS containing 1 μg mLptn in one footpad and with a PBS control containing 1 pg of LPS (an amount equivalent to the endotoxin present in the mLptn preparation) in the opposite footpad. The footpads and ankles of the mice were examined and measured with Vernier calipers twice daily for 3 days to monitor for inflammation and/or swelling. For immunohistochemical analysis, some animals were euthanized after 20-24 h and the footpads were removed and frozen in O.C.T. compound (Baxter Diagnostics Inc., McGraw Park, Ill.

[0055] Antibodies and flow cytometric analyses: Cells obtained from peritoneal lavage were washed once in ice-cold PBS (2% FBS) and then resuspended in the same buffer for staining with the appropriate dilution of CD19-FITC, NK1.1-PE, and CD3ε-biotin, followed by streptavidin-Tricolor™. For some experiments, varying combinations of CD4-FITC, CD8-PE, Mac-1-PE, and Gr-1-biotin were also used. Flow cytometric analyses were performed on a FACScan II (Becton Dickinson) and analyzed using CellQuest software (Becton Dickinson). Antibodies used for flow cytometry were obtained commercially (PharMingen, San Diego, Calif.) as was the streptavidin-tricolor™ conjugate (Caltag, So. San Francisco, Calif.).

[0056] ELISA and western blotting: For ELISA, 50 ng of either mLptn or BSA, suspended in PBS (pH 7.2) was added to each well of a 96-well, PVC U-bottom plate (Dynatech, Chantilly, Va.) and the plate was incubated for 2 h at 37° C. The plate was then washed 3× with PBS (pH 7.2)/0.005% Tween-20 in a Dynatech plate washer and the supernatant from the 4D8 hybridoma was added in serial dilution. RPMI+10% FCS was used as a medium control. Following an incubation of one hour at room temperature (RT), the plate was washed again and 50 μl of horseradish peroxidase (HRP)-conjugated mouse anti-rat IgG (1:5000 dilution in PBS (pH 7.2) containing 0.005% Tween-20 and 0.1% BSA; Jackson ImmunoResearch, West Grove, Pa.) was added to each well. The plate was incubated for one hour at RT, washed again, and the assay developed using TMB-peroxidase (Kirkegaard and Perry, Gaithersburg, Md.) as a substrate. Optical density was read at 450 nm.

[0057] For Western blotting, 3 μg of either mLptn or hLptn (Genzyme) were suspended in Laemmli's buffer (containing 2% β-mercaptoethanol), boiled 5 min, and loaded onto a Tris-Glycine polyacrylamide gel with a 4-20% gradient (Novex, San Diego, Calif.). The gel was electrophoresed and protein transferred onto Immobilon-IPVH (Millipore, Bedford, Mass.) via electroblotting. The resulting protein blot was blocked overnight at 4° C. in wash buffer (0.01 M Tris ,pH 7.4, 0.155 M NaCl, 0.02% Tween-20) containing 5% w/v nonfat dry milk. The blot was subsequently incubated with the 4D8 rat-anti-mLptn mAb (1:500 dilution of hybridoma supernatant) for 1.5 h, at RT. The blot was then washed 3× (5 min each wash) in wash buffer and incubated with a 1:5000 dilution of HRP-conjugated goat anti-Rat Ig (Amersham, Little Chalfont, UK) for one hour at room temp. The blot was again washed (6×, 5 min each wash) and developed with the ECL chemiluminescence detection system (Amersham) according to the manufacturer's instructions.

[0058] Immunohistochemistry: Tissue was sectioned into 5 μm slices which were thaw mounted onto organosilicone subbed slides (American Histology Reagent Co., Stockton, Calif.) and fixed by acetone immersion (5 min, −20° C.). Sections were then rinsed 3× in 0.01 M Hepes-buffered Hank's Balanced Salt Solution (Hepes/HBSS) for 5 min, each wash. Sections were subsequently blocked with 1% hydrogen peroxide, 0.2 M sodium azide in Hepes/HBSS for 20 min at RT. An additional blocking step was carried out using 10% normal goat serum in Hepes/HBSS (10 min, RT).

[0059] Following fixation and blocking, sections were incubated with hamster-anti-mouse CD3_ε (PharMingen) at a concentration of 5 μg/ml in Hepes/HBSS for 2 hours at RT. The sections were washed as before and bound antibody detected with biotin-conjugated, goat-anti-hamster IgG (2.0 μg/ml in Hepes/HBSS, 1 h, RT; Vector Laboratories, Burlingame, Calif.). The sections were washed as before and final detection carried out using the Vectastain Elite ABC kit (Vector Laboratories) according to the manufacturer's instructions. Slides were then washed with Hepes/HBSS (no saponin) and developed with 3,3′-diaminobenzidine tetrahydrochloride (DAB, 0.5 mg/ml in 0.05 M Tris (pH 7.4) containing 0.0075% H₂O₂; Sigma Chemical Co., St. Louis, Mo.) as substrate. The reaction was quenched with distilled water and the sections were dehydrated in graded ethanol, immersed in xylene, and finally mounted with Permount (Fisher Scientific, Springfield, N.J.) for examination and photography.

[0060] Intracellular detection of lymphotactin: Activated A3.2 T cells were obtained by incubation of cells with RPMI 1640/10% FBS in tissue culture flasks precoated with CD3ε (PharMingen; 10 μg/ml in PBS) for 3 h. Purified murine NK cells were activated by incubation with IL-2 (500 U/ml in RPMI 1640/10% FBS) for 5 days. Following activation, cells were washed 2× in DME (no serum) and 3× in Hepes/HBSS. Cell concentration was adjusted to 1×10⁶/ml in Hepes/HBSS and 50 μl of cell suspension (5×10⁴ cells) pipetted into each of the wells of an adhesion slide (Bio-Rad Laboratories, Hercules, Calif.). Following adhesion (20 min, RT), cells were fixed in 3% formalin-Hepes/HBSS (15 min, RT) and washed 2× in Hepes/HBSS. The cells were then blocked sequentially with avidin D (2 drops/ml of avidin blocking solution; Vector Laboratories) for 15 min, followed by a biotin block (2 drops/ml of biotin blocking solution; Vector Laboratories) for 15 min, a sodium azide-peroxide block (0.2 M NaN₃, 0.1% H₂O₂) for 20 min, and finally normal rabbit serum (10%) for 10 min. Blocking solutions were made in permeabilization buffer (Hepes/HBSS/0.1% saponin). The 4D8 anti-mLptn mAb was then added (5 μg/ml in permeabilization buffer) and the slides allowed to incubate for one hour at RT. The slides were washed in permeabilization buffer (3×, 5 min each wash) and incubated with a biotin-conjugated rabbit-anti-rat Ig (1:300 dilution in permeabilization buffer; Vector Laboratories) for 30 min at RT. Bound antibody was detected using the Vectastain Elite ABC kit (Vector Laboratories) and the slides developed with DAB as before.

[0061] In vitro microchemotaxis assay: The microchemotaxis assays were carried out using a modified 48-well Boyden chamber migration assay. See Bacon, et al. (1988) Br. J. Pharmacol. 95:966-974. Serial dilutions of mLptn or hLptn were made into serum free RPMI. Duplicate wells of the lower half of the microchemotaxis chamber (Neuro Probe Inc., Cabin John, Md.) were filled with the appropriate dilutions and the upper chambers of the assembly were filled with 40 μl, of the appropriate cell suspension (1-2×10⁶ cells/ml). Data was obtained by counting five non-overlapping high-power microscope fields (400×magnification at eyepiece) from each of the wells. Cells were considered to have chemotaxed if the chemotactic index (Cl=# cells migrating in experimental well/# cells migrating in media only) was greater than two.

[0062] Expression of Lptn message: Expression of Lptn mRNA by human NK clones was determined by Northern blotting. Total RNA was extracted from each clone using RNAsol (Tel-Test Inc. Friendswood, Tex.) according to the manufacturer's instructions. Northern blotting was performed by standard methods using 10 μg of total RNA from each clone.

EXAMPLE 2

[0063] Characterization of an Anti-mLptn mAb

[0064] Anti-Lptn mAbs were raised in Lewis rats by immunization with purified recombinant FLAG-mLptn. Hybridomas produced from the splenocytes of immunized animals were then screened by ELISA for reactivity with FLAG-mLptn. The hybridomas which were positive in this initial screen were then tested for ELISA reactivity against non-FLAG mLptn in order to eliminate any hybridomas which were reactive with the FLAG peptide. Those hybridomas with ELISA reactivity to mLptn were then screened for their ability to recognize mLptn in Western blotting. One hybridoma designated 4D8, identified through this screening procedure, exhibited strong reactivity with mLptn by ELISA but did not score with the Bovine Serum Albumin protein control (FIG. 1A). Furthermore, 4D8 reacted with purified recombinant mLptn by Western blotting but was unreactive with full-length hLptn (FIG. 1B). Taken together, these data indicate that the 4D8 mAb specifically recognizes mLptn.

EXAMPLE 3

[0065] Intraperitoneal Injection of Lptn Causes an Influx of Cells

[0066] The in vitro chemotactic activity of mLptn on murine thymocytes and mature murine T lymphocytes has been described. In order to confirm this activity in vivo, mLptn was injected into the peritoneal cavity of female CB6F1 mice. Initially, doses of 1 μg and 5 μg of Lptn were used, however, no effect was observed following injection of 1 μg and the effects were marginal with 5 μg. A strong influx of cells into the peritoneum was observed, however, when 10 μg of Lptn were injected. This response was evident at 24 hours, but had largely disappeared by 72 hours post-injection. This cellular influx into the peritoneum was not observed in control animals injected with an endotoxin-matched (10 pg LPS) PBS control. Furthermore, this effect was abolished when 0.5 mg of the anti-mLptn mAb 4D8 was co-injected with 10 μg mLptn.

EXAMPLE 4

[0067] NK Cells and T Cells are Attracted by Lptn in Vivo

[0068] Cells recovered from the peritoneum of mice injected 24 hours previously with mLptn were analyzed by FACS to determine what type of cells were attracted. The cells recovered from the mLptn-injected mice showed a marked increase in the percentage of cells matching the known scatter profile of lymphocytes. This population was found to be significantly enriched for T cells (CD3⁺NK1.1⁻), as was expected. Surprisingly this population was also enriched for NK cells (NK1.1⁺CD3⁻). Although the percentage of B cells (identified as CD19⁺CD3⁻ cells) in the peritoneal lymphocyte population was reduced, the absolute number of B cells, based on the percentages obtained by FACS and the total number of cells recovered from the peritoneum, was not significantly affected. The absolute numbers of NK cells in the peritoneum of Lptn-injected mice increased about 5-fold while the absolute number of T cells increased about 2-fold. In some experiments the CD4:CD8 phenotype of the CD3⁺ cells was also examined, however, the relative ratio of CD4:CD8 (approx. 1.2:1) observed in the Lptn-injected mice as compared to control animals was not significantly different (see Table 1). In contrast, the absolute number of macrophages (Mac-1⁺ cells) present in the peritoneum did not show a significant change in mLptn-injected animals.

[0069] This finding is confirmed, e.g., in Giancarlo, et al. (1996) Eur. J. Immunol. 26:3238-3241; and Maghazaci, et al. (1997) FASEB J. 11:765-774.

EXAMPLE 5

[0070] Subcutaneous Injection of mLptn Attracts T Lymphocytes

[0071] We sought to confirm the ability of Lptn to attract T lymphocytes in vivo by utilizing a subcutaneous injection model. In this model, mice were injected in one footpad with a PBS control, while the other footpad received an injection of 1 μg of mLptn. The animals were sacrificed after 24 h, the footpads excised, and analyzed by immunohistochemistry. Serial sections from the PBS-injected footpads showed no significant staining with the anti-CD3_ε mAb used while the footpads receiving mLptn showed numerous cells with anti-CD3_ε reactivity. Interestingly, at no time did any animal show signs of inflammation in either the PBS- or mLptn-injected footpads.

EXAMPLE 6

[0072] Lymphotactin Attracts NK Cells in Vitro

[0073] In vitro microchemotaxis assays were performed on purified murine NK cells in order to confirm their in vivo Lptn-responsiveness that had been observed for these cells. Murine NK cells (NK1.1⁺Gr-1⁻) were isolated from Rag-1⁻ deficient mice in order to insure that contamination of the NK cells by other lymphocyte populations, particularly NK1.1⁺CD3⁺ T cells, would not be a concern. The freshly isolated NK cells were either tested immediately in the microchemotaxis assay or were activated with IL-2 for 5 days and then examined for chemotaxis to mLptn. The IL-2 activated NK cells were responsive to Lptn in vitro with a peak response around 10⁻⁸ M. Freshly isolated NK cells, however, were found to be unresponsive to mLptn in these assays. See also, e.g. Giancarlo, et al. (1996) Eur. J. Immunol. 26:3238-3241; and Maghazaci, et al. (1997) FASEB J. 11:765-774.

EXAMPLE 7

[0074] Chemotaxis of Human NK Cell Clones

[0075] Because murine NK cells were chemotactic in response to Lptn it was useful to determine whether human NK cells would be similarly responsive. To address this question, several human NK clones were examined for their ability to respond to hLptn in the microchemotaxis assay. Some NK clones were found to respond very well while others showed little or no response. The clones which did respond displayed a similar dose-response to that observed with murine NK cells, with a peak response around 10⁻⁸ M. Altogether, two of seven human NK clones examined showed significant Lptn responsiveness (i.e., a chemotactic index of 2 or more).

EXAMPLE 8

[0076] NK cells Produce Lymphotactin

[0077] Since NK cells were found to respond to Lptn, it was of interest to determine whether or not these cells could also make Lptn. In order to demonstrate that murine NK cells express Lptn, the 4D8 anti-mLptn mAb was used for intracellular staining. While no staining of IL-2 activated murine NK cells was detected with an tso isotype control antibody, intracellular staining with the 4D8 mAb was easily detected. Since murine Lptn mRNA had previously been found to be expressed in murine αβTCR⁺CD4⁻CD8⁻ thymocytes, the A3.2 αβTCR⁺CD4⁻CD8⁻ thymocyte hybridoma was examined as a control for expression of Lptn. A3.2 cells which had been activated on solid phase anti-CD3_ε for 3 hours showed a pattern of intracellular staining similar to that of the IL-2 activated murine NK cells. The isotype control for these cells was similarly negative. Neither freshly isolated NK cells nor unactivated A3.2 cells showed staining for Lptn.

[0078] Human NK clones were also examined for expression of Lptn. Since no antibody to hLptn is presently available, a panel of human NK clones were analysed for expression of Lptn message. Total RNA was extracted from a series of human NK clones and subjected to Northern blotting with a probe corresponding to the coding sequence of human Lptn. A signal of the appropriate size was observed in each of the clones examined, although the intensity of the signal varied from clone to clone. An actin probe, used as a control, demonstrated equivalent mRNA loading.

EXAMPLE 9

[0079] Carboxyl-terminal Truncated Lymphotactin Lacks Chemotactic Activity

[0080] As discussed above, Lptn has an unusually long carboxy-terminal tail extending approximately 22 amino acids beyond the last amino acid of most C—C chemokines. A truncated version of human Lptn, lacking the carboxy-terminal 22 amino acids (CT-hLptn), was produced as part of an effort to understand the structure of lymphotactin. We took advantage of the existence of this truncated molecule in order to investigate the role that the tail plays in the activity of Lptn. The activity of the CT-hLptn protein on human PBLs was compared with that of full length hLptn in the microchemotaxis assay. The truncated protein was found to lack detectable activity (FIG. 9A) when compared with full-length hLptn. Human and mouse Lptn have considerable similarity at the amino acid level, particularly in the carboxy-terminal 22 amino acids, so the ability of hLptn to attract murine lymphocytes was also investigated. For this series of microchemotaxis assays murine splenocytes were utilized which have previously been shown to respond to Lptn. Full length hLptn was found to attract murine splenocytes although the dose response was shifted, with a peak at 10⁻⁷ M hLptn, as compared to the 10⁻⁸ M peak obtained with mLptn. The CT-hLptn lacked chemotactic activity for murine splenocytes, in accordance with the findings obtained using human PBLs.

[0081] The ability of Lptn to attract lymphocytes in vitro has been previously reported. It is, however, important to study the effects of chemokines in vivo. Here, a novel activity for lymphotactin is described, namely the ability to attract NK cells, and confirm in vivo that Lptn is capable of recruiting T lymphocytes. Intraperitoneal injection of Lptn causes a significant influx of NK cells as well as T lymphocytes. This influx of T and NK cells was specific as it was inhibited by co-injection of the anti-mLptn mAb, 4D8. Similarly, subcutaneous injection of Lptn causes an influx of T cells, confirming that this protein has the T cell chemotactic properties in vivo that have been observed in vitro.

[0082] The observation that Lptn chemoattracted NK cells in vivo was an unexpected result. NK cell chemotaxis in vitro using microchemotaxis assays was demonstrated. Interestingly, freshly isolated NK cells fail to respond to mLptn in vitro, while activated NK cells were found to respond very well. NK cells recovered from the peritoneum of mLptn-injected mice were evaluated for expression of Ly49 and CD69, two markers associated with NK activation, but no significant expression of these molecules was observed, suggesting that these cells were not activated. Other investigators have obtained similar in vitro results with freshly isolated versus IL-2 activated NK cells. See, e.g., Allavena, et al. (1994) Eur. J. Immunol. 24:3233-3236; and Maghazachi, et al. (1994) J. Immunol. 153:4969-4977. Taub et al. (1995) J. Immunol. 155:3877-3888, however, have reported that fibronectin coating of the porous filters utilized in the microchemotaxis assays allows migration of freshly isolated NK cells. Another possible explanation for the increase in peritoneal NK cells is that Lptn is expanding a pool of NK cells already present in the peritoneum. No effect of Lptn on proliferation of NK, either in the presence or absence of exogenous IL-2, was observed, suggesting that this is unlikely. Taken together, these observations strongly suggest that Lptn recruits NK cells in vivo.

[0083] It was also investigated whether human NK cells might be similarly responsive to Lptn. Some, but not all, human NK clones were found to respond to hLptn in microchemotaxis assays. This further confirms the NK cell response to Lptn and demonstrates that human Lptn exhibits similar properties to murine Lptn. The varying ability of human NK clones to respond to lymphotactin may reflect clone-specific differences in receptor expression or in the relative mobility of particular clones. Alternatively, since these clones are maintained in culture by constant restimulation, their ability to respond to chemotactic stimuli may depend on how recently they had been stimulated.

[0084] Although a number of chemokines have been reported to attract NK cells in vitro, this may be the first report demonstration that a chemokine is capable of attracting NK cells in vivo.

[0085] In addition to the NK cell chemotactic response to lymphotactin, it was also found that both IL-2 activated murine NK cells and human NK cell clones express Lptn protein or Lptn message, respectively. This raises the possibility that an activated NK cell could recruit other NK cells or T lymphocytes to the site of an infection or into a tumor mass. Indeed, Lptn message has been reported to be rapidly up-regulated in activated T cells and here it is observed that protein can be expressed by the A3.2 thymocyte hybridoma within three hours of activation, indicating that Lptn protein is produced shortly after the appearance of its mRNA.

[0086] As previously mentioned, one of the structural features which distinguishes Lptn from other chemokines is the unusually long carboxy-terminus of Lptn, that is highly conserved (68.2% identity in the last 22 amino acids) between mouse and human proteins. It was investigated whether this extended “tail” was necessary for Lptn's function by examining the activity of a carboxy-terminal truncated version of the human protein. This truncation removes the carboxy-terminal 22 amino acids of Lptn and results in a protein that is similar in size to other chemokines. The ability of CT-hLptn to cause chemotaxis of human peripheral blood lymphocytes was evaluated. In contrast to the full-length hLptn, which showed clear chemotactic activity for human PBLs, the CT-hLptn was completely inactive. Similarly, while the full-length hLptn was active on murine splenocytes, the truncated version was once again inactive. The results of these experiments demonstrate that the carboxy terminus of Lptn is indeed critical to at least these functions. This is in contrast to the data obtained with the murine chemokine JE, which also possesses an extended C-terminus. Truncation of the extended tail of that protein does not significantly affect is biologic activity. One possible explanation for these differences is that the C-terminus is directly or indirectly involved in binding to the Lptn receptor. Another possibility is that because Lptn lacks two of the four cysteine residues normally found in chemokines (and must therefore also lack one of the two disulfide linkages present in most C—C and C—X—C chemokines) the extended carboxy-terminal tail is needed to stabilize the protein's structure. Both the C—C and C—X—C chemokines have been shown to possess a carboxy-terminal α-helix. See, e.g., St. Charles, et al. (1989) J. Biol. Chem. 264:2092-2099; Clore, et al. (1990) Biochemistry 29:1689-1696; and Baldwin, et al. (1991) Proc. Natl. Acad. Sci. U.S.A. 88:502-506. It is reasonable to believe that the truncation of the carboxy-terminal 22 amino acids disrupts the predicted carboxy-terminal o-helix of lymphotactin and similarly destroys its biologic activity.

[0087] Previous work has also demonstrated that amino-terminus of the C—C and C—X—C chemokines is very sensitive to modification and that even minor changes can result in the loss of their chemotactic activity. See Zhang, et al. (1994) J. Biol. Chem. 268:15918-15924; and Gong and Clark-Lewis (1995) J. Exp. Med. 181:631-640. In contrast to this, the murine Lptn with a amino-terminal FLAG sequence is active. This contrasts with the C—C chemokine MCP-1, which loses biologic activity if an amino-terminal FLAG is added to its sequence. In this report another version of mLptn with an amino-terminal methionine residue is active in vitro and in vivo. These data suggest that the biologic activity of Lptn is retained even after alterations in its amino-terminus. The structures of the full length and carboxy-terminal truncated hLptn are currently being determined and this information will certainly shed some light on the structure-function relationship of lymphotactin.

[0088] Lymphotactin is an important chemokine, not only because of its unique structure and chromosomal localization, but also because it represents, with the possible exception of the monokine induced by interferon-γ (Mig) (Liao, et al. (1995) J. Exp. Med. 182:1301-1314) the only lymphocyte-specific chemokine. The present invention shows that the lymphocyte specificity of Lptn is observed in vivo. The characterization of Lptn in vivo has also uncovered a novel activity for Lptn, namely chemoattraction of NK cells, which in turn suggests the possibility for anti-viral and anti-tumor effects. The data discussed herein shows that the NK activity for Lptn observed in mice also applies to hLptn and human NK cells. The development of a neutralizing anti-Lptn mAb is also described. 1

EXAMPLE 10

[0089] Lymphotactin and Tissue Rejection; Graft vs. Host Disease

[0090] The expression pattern of Lptn is highly specific. The cells that produce Lptn include: NK cells, dendritic epidermal γδ T cells, and class I restricted T cells. Dendritic cells may produce Lptn as well. Of this list, by far the most important producers, in terms of numbers, are the class I restricted T cells. Furthermore, Lptn is a very early and important product of these cells following activation. These cells play a critical role in various immune responses, including immunity to tumors as well as organ rejection.

[0091] The cells recruited by Lptn in vivo have been studied. These include what are likely to be subsets of CD4 and CD8 T cells, but the biggest change (as a percentage) is observed in the NK compartment, where there is a 5-10 fold increase in the number of NK cells. Taken together, these observations imply that Lptn likely plays an important role in the modulation of these responses. As such, it is likely that: (a) Lptn constitutes an early recruiting signal for class I restricted T cells, as well as for NK cells. Thus, Lptn would be necessary for the development of these immune reponses, since failure to recruit these cells would abrogate them. (b) Interference with this normal early signaling mechanism would result in a significant reduction in the magnitude of these immune responses. One important process where early activation of class I-restricted T cells, as well as NK cells, is critical is in the process of tissue or organ rejection, or in graft vs. host disease. Thus, it is likely that neutralzation of Lptn, e;.g., using a monoclonal antibody antagonist or a mutein antagonist, will result in modulation of the fate of the transplant across MHC differences. (c) The presence of Lptn in a tumor would result in the recruitment into the tumor mass of cells capable of destroying the tumor cells, including CD4 and CD8 T cells as well as NK cells.

EXAMPLE 11

[0092] Hematopoietic Effects of Lymphotactin

[0093] Lin⁻ Sca-1⁺ rhodamine^lo c-kit⁺ stem cells were isolated. These cells were incubated for at 37° C. and 5% CO₂ in 1.5 ml eppendorf tubes at 400 cells per 400 μl. Growth factors, alone or in appropriate combinations, were added to the Iscove's Modified Dulbecco's Medium (IMDM)+15% fetal bovine serum (FBS)+penn-strep+HEPES at a final concentration of 50-100 ng/ml, in the 400 μl/tube.

[0094] At day 7, an aliquot of each culture which is ¼ of the input volume is withdrawn and the culture is refed with 200 μl of growth factors in IMDM+15% FBS. 5 μl is withdrawn from the sample and counted.

[0095] The remaining cells were diluted with IMDM+15% FBS and underlayed with FBS. The culture was spun at 1000 RPM for 5 min. The supernate is aspirated and the cells resuspended in 100 μl IMDM+15% FBS.

[0096] CFU-c assays were set-up to contain IL-3+IL-6+SCF+epo. The cells from each delta culture were added to a single set-up along with methylcellulose (final conc 0.8%). The contents of each set-up were mixed with a 3 ml syringe and 16 Ga needle and plated into three 35 mm pertri dishes. Cultures were incubated at 37° C. and 5% CO₂ for 7 days and colonies were enumerated.

[0097] Harvest of ¼ of the culture is repeated on day 14, day 21, and day 28.

[0098] The effects of lymphotactin, alone or in combination with MIP-1ε, Flt3 ligand, Stem Cell Factor, IL-6, and/or IL-3 are shown in FIGS. 10 (day 7), 11 (day 14), and 12 (day 21). These data show that lymphotactin, in combination with other factors, will have the effect of preventing hematopoietic stem cell from entering cycle and producing colonies at early time points, e.g., day 7, but will allow later colony formation, e.g., day 21. This data indicates the ability of lymphotactin to protect hematopoietic stem cells from damage by chemotherapeutic reagents by preventing their entry in cycle and the resultant cellular proliferation and differentiation for up to 21 days. The colony formation seen at day 21 indicates that the progenitors remained quiescent rather than dying during the initial culture period.

[0099] All references cited herein are incorporated herein by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

[0100] Many modification an variations of this invention can be made without departing from its spirit and scope, as will be apparent to those skilled in the art. The specific embodiments described herein are offered by way of example only, and the invention is to be limited only by the terms of the appended claims, along with the full scope of the equivalents to which such claims are entitled.

[0101] Although the present invention has been described in some detail by way of illustration and example for purposes of clarity and understanding, it will be recognized that certain changes and modifications may be practiced within the scope of the claims.