Enhancement of the cellular immune response using carbohydrate primed DTH effector cells expressing the CD5+/CD8- phenotype

Abstract

DTH-Effector cells are primed with carbohydrate antigens and used to enhance the cellular immune response. Tumors have been inhibited by DTH-Effector cells primed with epiglycanin and with synthetic T and Tn antigens. Either the DTH-Effector cells, or these tumor-associated carbohydrate antigens directly, may be used for tumor prophylaxis and therapy.

Description

BACKGROUND OF THE INVENTION

Vertebrates have two basic immune responses: humoral or cellular. Humoral immunity is provided by the special class of cells produced by B lymphocytes. These cells produce antibodies which circulate in the blood and lymphatic fluid. On the other hand, cell mediated immunity is provided by the T cells of the lymphatic system.

The cellular immune response is particularly effective against fungi, parasites, intracellular viral infections, cancer cells and foreign matter, whereas the humoral response primarily defends against the extracellular phases of bacterial and viral infections.

Containment of antigen at its point of entry is accomplished by walling off the area by local inflammation. Acute inflammation is characterized by the influx of plasma proteins and polymorphonuclear leukocytes. Chronic inflammation is characterized by the infiltration of T-lymphocytes and macrophages. When acute (antibody induced) and chronic (T cell induced) inflammations occur in the o skin, they are called immediate and delayed type hypersensitivity reactions respectively. ITH peaks at 24 hours, and subsides in 48 hours DTH appears in 24-48 hours and peaks at 48-72 hours. The subset of T cells involved in DTH reactions are called here DTH-Effector cells.

Epiglycanin (epi) is the major cell surface glycoprotein produced by the mammary adenocarcinoma transplantable cell line TA3Ha. Friberg, Jr., J.N.C.I., 48:1463 (1972); Codington, et al., Canc. Res., 43:4373 (1983). The TA3Ha carcinoma cells are covered by a mucin-like glycocalix composed mainly of epiglycanin. Codington, et al., J.N.C.I., 60:811 (1978); Miller, et al., J.N.C.I., 68:981 (1982). Epi is mainly carbohydrate in composition, and expresses multiple T and Tn determinants. T and Tn are general carcinoma autoantigens. Springer, Science, 224:1198 (1984).

Synthetic T and Tn antigens have been prepared, and used in DTH diagnostics. Lemieux, Ep Appl. 44, 188. However, their therapeutic or prophylactic use has not been disclosed.

Bretscher, Eur. J. Immunol., 9:311-316 (1979) cultured DTH-Effector cells in vitro and injected them with the sensitizing antigen (burro erythrocytes) into the foot pads of mice. We have used cancer-associated, well characterized carbohydrate antigens to sensitize effector cells and have taught a therapeutic use for them.

It is known that tumors may be treated with a mixture of IL-2 and IL-2-activated killer cells. See Fortune, 16-21 (Nov. 25, 1985), reporting on the work of Steven Rosenberg. The problem with the Rosenberg work is that the therapy induces a population of cells known as LAK (lymphokine activated killer) cells. LAK cells are not antigen-specific; therefore normal tissue may be attacked. Also, since the LAK cells are non-specific, relatively large doses of IL-2 are required to induce the required anti-tumor activity. Finally, LAK cells are not thought to confer memory to the immune system. In contrast, DTH effector cells are antigen specific; therefore after stimulation much smaller quantities of lymphokine may be required to induce proliferation. Also because of their specificity, DTH effector cells are less likely to attack self. Moreover, DTH effector cells are though to have memory.

SUMMARY OF THE INVENTION

We have found that the cell-mediated immune response to cancer cells is enhanced by parenteral administration of natural and synthetic cancer-associated carbohydrate antigens or by parenteral administration of DTH effector cells primed by such antigens. This enhancement is likely to be advantageous for both therapy and prophylaxis when the body's primary defense against a challenge is a cellular immune response, e.g., against tumors. The use of other specific carbohydrate antigens, including polysaccharides, glycolipids and glycoproteins, may be useful in protecting a o subject against other threats controlled by T cells. It is known that parasites, viruses, bacteria and fungi bear surface carbohydrate determinants. Lymphokines may be used in conjunction with the specific antigens or primed DTH-Effector cells of this invention to enhance the specific CMI response.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows that epi-primed splenocytes induce local DTH swellings in the presence of epi, while unprimed spleen cells, irradiated tumor cells, and primed cells in the absence of antigen do not.

FIG. 2 shows that this reaction is specific to cancer cell lines bearing the primer as a surface antigen.

FIG. 3 shows that the reaction was with the T alpha, T beta and Tn determinants of epi.

FIG. 4 shows that the phenotype of the anticarbohydrate DTH effector cells is Thy-1+, Lyt-1+, Lyt-2-.

FIG. 5 shows systemic transfer of DTH by a DTH reaction to irradiated tumor cells after IV administration of Effectors.

FIG. 6 shows inhibition of tumor growth after local transfer of DTH-Effector cells mixed with tumor cells.

FIG. 7 shows the importance of the priming dose.

FIG. 8 shows that at lower doses of antigen the cellular response predominated.

DETAILED DESCRIPTION OF THE INVENTION

Example 1

Use of s-TAGs to Enhance Cellular Immune Response to Tumors

Synthetic tumor-associated glycoproteins (S-TAGS) and other carbohydrate antigens are known in the art and may be prepared by any convenient technique. T and Tn antigens are preferred. For synthetic methods, see Kaifu and Osawa, Carbohydr. Res., 58:235 (1977); Ratcliffe, et al., 93:35 (1981); Paulsen, et al., 104:195 (1982); Bencomo and Sinay, 116:69 (1983).

Soluble S-Tags or their aggregates are mixed with an adjuvant for intradermal, intramuscular or intraperitoneal administration. It is conceivable that intravenous administration will also be successful.

Groups of five mice at a time were primed with specific doses of T-alpha (TF), T-beta (asialo-GMI disaccharide) and Tn S-Tags. After 7-10 days, they were challenged with 3.times.10.sup.3 live TA3Ha cells. Markedly increased survival was noted in mice receiving about 1 ug of synthetic antigen. Thus, for T-beta, the following results were observed (Table I):

TABLE I______________________________________Dose Survival (days)______________________________________0.2 ug 20, 27, 27, 29, 311.0 ug 27, 38, 43, 43, 492.0 ug 16, 21, 24, 3010 ug 20, 21, 28, 3320 ug 21, 21, 28, 2850 ug 16, 18, 31100 ug 20, 20, 21, 27, 29PBS 23, 27, 29Norm 13, 13, 18, 27, 28______________________________________

It should be noted that TA3Ha is very deadly. The normal post-transplantation life expectancy of a mouse is only 15-20 days.

Example 2

Use of Cultured DTH Effector Cells as Tumor Inhibitors

We obtained mouse primed DTH effector cells by the following procedure. CAFl/J (Balb x A/J) mice from our animal unit were immunized with 30 ug Epi (or 2 ug of any synthetic antigen), in 50% complete Freund's adjuvant, intraperitoneally, boosted one week later with a similar injection. 7-10 days after the last injection the animals were sacrificed, their spleens removed and passed through a nylon mesh to make a single cell suspension. The cells were washed and cultured in a costar well at 1.5.times.10.sup.7 cells/8 ml. (RPMI+P.S.+10% FCS)/well. Each well received 10 ug epi (or 1-2 ug of any of the synthetic antigens). The cells o were cultured at 37.degree. C. and 5% CO.sub.2 for 6 days and then harvested by gently dissociating them from the plastic with a rubber policeman. The primed cells were then tested for DTH reactivity with a number of antigens, and for their ability to inhibit the growth of TA3Ha cell line in vivo.

To obtain human primed DTH Effector cells, we would use a modification of the method of Rosenberg. Lymphocytes are removed from blood of a living subject by plasmapheresis. The lymphocytes are cultured with a specific antigen, such as cancer-associated glycoprotein (epiglycanin or a T or Tn S-TAG), or the combination of a specific antigen with IL-2, rather than with IL-2 alone. The human lymphocytes are cultured in suitable media for 6 days at a preferred concentration of about 10.sup.6 cells/ml. The cells are washed to remove antigen. The cells may then be returned to the patient, intravenously, for therapeutic or prophylactic purposes.

We have demonstrated that co-injection of a mammal with epiglycanin-primed DTH Effector cells and antigen (irradiated tumor cells) induced a DTH reaction (FIG. 1). Spleen cells from mice primed and boosted with 30 ug and 20 ug of epi extracted and purified from TA3Ha ascites fluid and CFA (intraperitoneally administered one and two weeks beforehand) were cultured at 1.5.times.10.sup.7 cells/well with: (a) 10 ug epi/well and (b) no antigen. In addition, unprimed spleen cells were cultured with (o) 10 ug epi/well and (d) no antigen. Cultures were harvested on the 6th day and 10.sup.7 cells were injected subcutaneously into the foot pad of unimmunized mice together with 10.sup.6 irradiated TA3Ha cells. As a control (e), 10.sup.6 irradiated TA3Ha cells were injected alone. All the DTH values are presented as net DTH swellings, i.e., [swelling of primed cells +Ag.]-[swelling of primed cells alone]. The large differential responses demonstrated that immunization with a cancer mucin can induce a DTH reaction to those cancer cells which have the mucin at the cell surface.

Subsequently, we showed that this reaction was tumorspecific (FIG. 2). A DTH effector cell population was generated as in Example I and tested with (a) TA3Ha; (b) Epi-M (epi linked to a sepharose microsphere); (c) mKSA; and (d) no antigen. All the DTH values are presented as net DTH swellings, i.e., [swelling of primed cells +Ag.]-[swelling of primed cells alone]. The result was that the immunization with the epi mucin induced a specific DTH response to the mucin and to the cells which bear the mucin (TA3Ha) but not against those that do not bear the mucin (mKSA).

We also explored the antigenic specificity of the DTH reaction in greater detail (FIG. 3). 10.sup.7 T cells were primed with epi and were co-injected with a series of antigens (TA3Ha, Tn, T-alpha, T-beta, Epi, Fucose ("Fuc"), HSA, BSA, sepharose microspheres, and no antigen). All soluble antigens were linked to sepharose microspheres and used at 12 ug/foot pad except epi which was used at 50 ug/foot pad. All DTH results are net DTH measurements. The experiment shows that the DTH measured by the injection of epi reacts with the carbohydrate determinants on S-Tags which are also know to be epi determinants but not those carbohydrates determinants not represented on epi. (eg. Fucose).

SUMMARY OF SPECIFICITY OF DTH RESPONSE

______________________________________ TRIGGERING Ag:IMMUNIZING Ag: Epi Tn Ta T.sub.b Fuc BSA______________________________________Epi: + + + + - -T.sub.a + + + + - -T.sub.B: + + + + - -Tn: + + + + - -None: - - ND ND ND______________________________________

To confirm that the reactions we were witnessing were indeed DTH reactions we designed an experiment to check for the cell surface phenotype of DTH effector cells. (FIG. 4) An effector cell population was prepared as before and treated with (a) no treatment; (b) anti-Thy 1.2 antibodies plus complement; (c) anti-Lyt 1.2 plus complement; (d) anti-Lyt 2.2 plus complement; and (e) complement alone. The antigen alone was also used as a control. All the values presented are net DTH. Treatment of the DTH effector cells with anti-Thy or anti-Lyt-1 antibodies with complement kills the DTH effector cells but treatment with anti-Lyt-2 plus complement did not kill the DTH effector cells; thus, the phenotype of the specific anti-carbohydrate DTH effector cells is Thy-1+, Lyt-1+, Lyt-2.sup.- --the phenotypic pattern shown by other DTH effector cells reacting against proteins. See Fujiwara, J. Immunol, 135:2187 (September, 1985).

Thy-1 is an antigen found on all mouse T cells. See Roitt, et al., IMMUNOLOGY FIG. 2.5 (2985). Lyt-1 is a mouse helper T cell antigen of 67KD believed to be equivalent to the human antigen detected by the Ortho monoclonal antibody T1. Similarly, Lyt 2 (Ly2) is equivalent to the human T cell antigen recognized by Ortho monoclonal antibody T8 or Becton Dickinson monoclonal antibody Leu2a.

The designation CD5 is the preferred scientific term for cell types marked by the antibodies T1/Leu1, or Lyt-1. The designation CD8 is the preferred scientific term for the cell types marked by T8/Leu2 or Lyt-2,3. Thus, the DTH effector cells of the present invention are preferably described as T cells of the CD5 phenotype and not of the CD8 phenotype. See "Note on Nomenclature, Nature, 325:660 (Feb. 19, 1987), adopting the CD nomenclature, Shaw, "Characterization of Human Leukocyte Differentiation Antigens, " Immunology Today, 8:1 (1987); and Leucocyte Typing (Bernard, ed., 1984) and Leucocyte Typing II (Reinherz, ed., 1986).

The protocol for this comparison was as follows. A primed cell population was obtained, washed, counted, and divided into separate tubes for treatment. The cells were gently pelleted and resuspended in Leibowitz media containing (1) anti Thy 1.2 antibody (at 1/1000 diln., obtained from NEN), (2) anti Lyt 1.2 antibody (at 1/1000 diln., obtained from NEN) or (3) anti Lyt 2.2 antibody (at 1/1000 diln., obtained from NEN) at 2.times.10.sup.7 cells/ml , and incubated at 4.degree. C. for 45 min. The cells were then spun down and resuspended in guinea pig complement (1/10 diln.) in Leibowitz media at 2.times.10.sup.7 cells/ml., and incubated for 30 min. in a 37.degree. C. water bath. After incubation, cells were washed with Leib +10% FCS and counted.

A comparative study was undertaken to determine whether the reaction was of the delayed type. DTH Effector cells were injected together with TA3Ha and the reaction was compared with reactions in other mice injected with only the DTH Effector Cells or only the TA3Ha. It was found that the reaction peaked in 24 hours and had disappeared by 96 hours. No DTH (emphasis on delayed) was found with the two controls.

As further confirmation that the reaction observed was a DTH, a mononuclear infiltrate was observed the site of swelling.

The foregoing experiments all related to local transfer of DTH. We also investigated the possibility that DTH Effector cells would have a systemic effect.

DTH effector cells were administered intravenously to the host animal. Four hours later, irradiated TA3Ha cells were injected into the footpad. At 6, 24, 48 and 72 hours, the DTH reaction was measured (open boxes) and compared with a negative control (plusses) (FIG. 5).

The irradiated tumor cells used in the above-described experiments will not proliferate. In the next experiments, however, untreated tumor cells were employed.

DTH primed cells were induced as described above. 5.times.10.sup.7 primed cells injected IV plus 2.5.times.10.sup.5 live tumor cells were injected s.c. into the foot pad. The control mice were injected with normal spleen cells plus tumour cells. The foot pad swelling was typically measured at 12, 24, 48, and 72 hours and every two days thereafter. The determine the component of the swelling which was due to a DTH reaction, as opposed to a combination of DTH and tumour growth, mice were injected with primed or normal cells plus irradiated tumor cells. The swelling was attributed only to the DTH reaction. Tumor growth was measured every 2 days starting 48 hours after the peak DTH swelling. The DTH and the tumor size were measured using calipers in order to estimate the increase in foot pad thickness.

FIG. 6 shows that the primed cells inhibited the growth lo of the TA3Ha tumor.

Other experiments have shown that the nature of the reaction to the antigen is dependent on the dose of primer administered to the DTH-Effector cells. At lower doses, cell-mediated immunity was predominant, while at higher doses, the reverse was true.

A dose dependence was also found when splenocytes are primed with Tn in culture (FIG. 7).

We have further shown that the DTH effector cells stimulated by epi will "home in" on a tumor and inhibit tumor growth. DTH-Effector cells were primed with epi as previously described. One day prior to intravenous injection of the epi-primed DTH effector cells, the TA3Ha tumor was transplanted into the footpads of mice. Mice with tumors transplanted at the same time were used as controls and tumor growth was determined by measuring the swelling of the footpads (in mm) with calipers. The size of the footpad of each mouse at the times stated is given in the tables below:

______________________________________EXPERIMENT 1 Day 6. post iv Day 8. post iv______________________________________Systemic Transfer 10, 12, 12, 10, 15 50, 15, 80, 80, 50Effector cellsControl Animals 140, 25, 35 190, 200, 220(3 mice)EXPERIMENT 2 Day 5 Day 6 Day 8 Day 10 post iv post iv post iv post iv______________________________________Systemic 10, 20, 55, 20, 100, 140, 80, 60 300, 360, 340, 380Transfer 15, 10 15, 20of DTHEffectorcells(4 mice)Control 60, 75, 170, 160, 340, 300, 320, greater thanAnimals 60, 110 170, 110 340 500 (3 mice), 410(4 mice)______________________________________

As may be seen from the above data, the IV-administered, epi-primed DTH-Effector cells exerted a tumor-inhibitory effect at the footpad site of tumor transplantation.

Example 3

Use of Epiglycanin as a Tumor Inhibitor

Epiglycanin (epi) was extracted from the ascites fluid of TA3Ha tumor-bearing mice.

Ascites from outbred TA3Ha ip tumors was collected and cells were removed by centrifugation. The samples were stored frozen. Before extraction, the ascites was thawed and incubated at 37.degree. C. for 2 hours, and any aggregates were removed by centrifugation. The freezing, thawing and clotting procedure was performed twice before PNA extraction.

25-40 ml. bed volume of PNA agarose (E. Y. Laboratories ) was washed with PBS, mixed with ascites and tumbled gently at room temperature for 24-48 hours. The slurry was poured into a column and extensively washed with PBS at 4.degree. C. Epi was eluted from PNA agarose using galactose. The eluted epi-galactose mixture was then dialyzed against PBS to remove the galactose. The sample was lyophilized and stored at -20.degree. C. in dessicator.

Groups of five mice at a time were primed with specific doses of epi. After 2-3 days they were challenged with 3.times.10.sup.3 live TA3Ha cells. Table 2 below shows the relationship between the priming dose and survival.

TABLE 2______________________________________ No. alivePriming dose: Survival: (days) ave. s.e. (at d. so)______________________________________0.2 ug ##STR1## 33.8 6.3 2/51 ug ##STR2## 44.2 4.8 4/52 ug ##STR3## 25.2 6.2 4/510 ug 20, 22, 22, 25 22.3 .9 0/420 ug ##STR4## 29.0 6.0 1/5100 ug ##STR5## 25.5 5.3 1/5PBS 19, 19, 20, 20 20 0.3 0/4______________________________________ Data calculated on day 49 i.e. these mice are still alive. These mice were sacrificed on day 88 to continue the experiment.

Thus, the preferred priming dose was 1 ug, and this dosage appeared to confer protective immunity against TA3Ha cells.

In another experiment, it was determined that the humoral response was negligible at that dosage (FIG. 8), suggesting that the tumor was inhibited by a mechanism other than antibodies, possibly CMI.

A summary of our experiences with Epi and T-alpha in tumor inhibition is given in Table 3 below:

______________________________________Dose of Ag Epi Ta Glu-BSA/HSA______________________________________0.1 ug 1/20 4/15 0/100/5 ug 5/20 7/15 0/101.0 ug 8/20 5/15 0/105.0 ug 8/15 6/15 0/1010.0 ug 7/20 3/15 0/10cont. 0/20 0/15 0/10______________________________________

Since Glu-BSA/HSA presents a carbohydrate which is not expressed on the surface of the TA3Ha tumor cells, administration of that substance did not provide protection. EPI, which is the predominant surface carbohydrate antigen on TA3Ha cells, and T-alpha, which is one of the EPI determinants, did provide protection.

Claims

1. A method of enhancing cellular immunity in a mammal to a predetermined carbohydrate antigen which comprises administering to said mammal a composition comprising an effective amount of isolated CD5 positive, CD8 negative T lymphocytes primed with an isolated or synthetic antigen bearing a predetermined carbohydrate determinant to exhibit DTH effector activity which is specific for such determinant and capable thereby of enhancing cellular immunity to an antigen bearing said carbohydrate determinant, wherein the composition is essentially free of T cells lacking DTH-effector activity specific for said carbohydrate determinant.

2. The method of claim 1 in which the DTH-Effector cells are primed by cultivating them in a medium containing an effective amount of a tumor associated carbohydrate antigen bearing said determinant.

3. The method of claim 2 in which the tumor-associated carbohydrate antigen is epiglycanin.

4. The method of claim 2 in which the antigen is a synthetic antigen.

5. The method of claim 2 in which the medium does not contain IL-2.

6. The method of claim 1 wherein the priming antigen is a mucin.

7. The method of claim 2 in which the tumor-associated carbohydrate antigen bears a T-alpha determinant.

8. The method of claim 2 in which the tumor-associated carbohydrate antigen bears a T-beta determinant.

9. The method of claim 2 in which the tumor-associated carbohydrate antigen bears a Tn determinant.