Biological characterization of guinea pig lymphotoxin production

Biological Characterization of Guinea Pig Lymphotoxin Production John O. Rundell and Charles H. Evans Abstract: The generation of serum-flee lymphotoxin by antigen-stimulatedovalbumin immune strain 2/N guinea pig peritoneal leukocytes in vitro is dependent upon a number of common variables in cell culture. Lymphotoxin production varies directly with antigen concentration, duration of antigen exposure and leukocyte culture, leukocyte density, and medium volume. The maximum concentration of lymphotoxin generated in the medium has a limitingvalue that is independent of antigen concentTation, leukocyte density, and duration of antigen exposure or leukocyte culture, suggesting feedback control of lymphokine production. Optimal conditions for the reproducible maximum generation of guinea pig lymphotoxin were 107 oil-induced peritoneal leukocytes from immune animals cultured for 12-24 hr in 20 ml RPMi 1640 medium containing 100 p.g ovalbumin/ml throughout the period of culture.

Key Words: Lymphokine; Lymphotoxin

INTRODUCTION Lymphotoxin, macrophage activating factor, and macrophage inhibitory factor are lymphokines, secretion products of lymphocytes. Immunoregulatory roles have been described for macrophage activating factor and macrophage inhibitory factor (Bloom, 1969; Pick, 1977; Waksman, 1976; Papermaster, 1978). The physiological function of lymphoto×in, however, remains unclear. Lymphotoxin is a protein(s) with a molecular size and isoelectric charge in the albumin range (Coyne et al., 1973; Gately and Mayer, 1974; Rosenau and Tsoukas, 1976). Lymphotoxin is produced by antigen- or mitogen-stimulated leukocytes in vitro and exhibits a variety of cytotoxic activities for tumor and sometimes various allogeneic and xenogeneic nontumorigenic cells (Coyne et al., 1973; Holzman et al., 1973; Rosenberg et al., 1973; Tsoukas, 1976; Evans et al., 1977; Pick, 1977; Rosenau and Youdim, 1977). Lymphotoxin cytotoxic activities include: inhibition of cell division (Ruddle and Waksman, 1968), inhibition of DNA synthesis (Williams and Granger, 1973; Smith et al., 1977), radionuclide release and cytolysis (Trivets et al., 1976), and colony inhibition (Holzman et al., 1973, Evans et al., 1975). Phytohemagglutinin-induced guinea pig lymphotoxin has a selective cytotoxicity for in vitro neoplastically transformed cells, but not syngeneic nontumorigenic cells, and is a useful biological probe to study the carcinogenic process (Evans et al., 1977). Received June 6, 1979. From the Tumor Biology Section, Laboratory of Biology, Divisionof Cancer Cause and Prevention, National Cancer Institute, Bethesda, Maryland. Address requests for reprints to: John O. Rundell, Ph.D., Laboratory of Biology, Building 37, Room 2A-19, Bethesda, MD 20205. lmmunopharmacology2, 19-27 ( 1 9 7 9 )

0162-3109/79/02001909500.00

19

20

J.O. Rundell and C. H. Evans

Lymphotoxin also has been shown by passive transfer to have in vivo antitumor activity (Papermaster et al., 1976a,b; Youdim, 1977). The biological and physicochemical characterization of lymphotoxin is therefore of interest to understanding immunoregulatory mechanisms as well as fundamental tumor biology, including lymphotoxin's possible role in host physiology during carcinogenesis and tumor surveillance. The present investigation characterizes some of the variables of major functional significance affecting lymphotoxin production and describes the conditions for the reproducible generation of antigen-induced lymphotoxin in serum-free culture medium. MATERIALS AND METHODS

Immunization of Guinea Pigs Inbred strain 2/N guinea pigs from the colony at the National Institutes of Health were immunized to ovalbumin as described by Gately and Mayer (1974). Briefly, animals were inoculated with 500 ~ ovalbumin (Miles Laboratories, Elkhart, Ind.) in Freund's complete adjuvant (Grand Island Biological Co., Grand Island, N.Y.). The 1 rnl inoculum was divided between the footpads (0.05 m] each) and the subcutaneous dorsal neck. Seven to ten days later, each animal was boosted by intraderma] injection of 50/~g ovalbumin in 0.1 ml 0.15 M NaCI.

Lymphotoxin Production Lymphotoxin was obtained from guinea pig peritoneal leukocytes as previously described by Evans et al. (1977), except that in the current investigations antigen stimulation replaced mitogen stimulation. Peritoneal exudate cells (PEC) were induced by intraperitoneal (ip) injection of 35 ml sterile mineral oil (Drakeol 6VR, Pennsylvania Refining Co., Butler, Pa.) into unimmunized or immunized animals 2 - 8 weeks after ovalbumin immunization. Three days later, the animals were killed by cervical dislocation and the PEC collected by peritoneal lavage with Hank's balanced salt solution at room temperature, sedimented at 180 x g for I0 min at room temperature, resuspended in serum-free RPMI 1640 medium at 37°C, and resedimented at 400 x g for 5 min at room temperature. The washed PEC pellet was again resuspended in serum-free RPM[ 1640 and the nucleated cells enumerated in a hemacytometer. The yield of PEC from ovalbumin-immune or unimmunized guinea pigs was the same and averaged 3.3 x 108 leukocytes per animal (range 1.8-3.7 x 10~). The cell populations consisted of approximately 60% monocytes-macrophages, 20% polymorphonuclear leukocytes, 20% lymphocytes, and 0 . 1 - I % erythrocytes. The PEC were diluted to 0 . 5 - 2 5 x I0 ~ nucleated cells/ml in 37°C RPMI 1640 with 0 - 2 5 0 / ~ ovalbumin/ml without fetal bovine or other serum unless specifically noted, 10 ml plated in 100 mm tissue culture dishes (Falcon Plastics, Oxnard, Calif.), and incubated at 37°C in a 5% CO2 in air, water saturated, atmosphere. After incubation for 1 - 7 2 hr, peritoneal exudate cell culture supematant solutions (PES) were collected, pooled, and cen~fuged at 1000 x g for 10 min at room temperature to remove PEC. The resultant cell-free PES was filtered through a sterile 0.22 membrane filter and stored at -40°C.

Lymphotoxin Assay The lymphotoxin colony inhibitory activity of PES preparations was assayed with a reproducibility of - 5 % as previously described by Evans et al. (1977). The benzo(a)pyrene neoplastically transformed strain 2/N guinea pig cell line I 0 4 C I was utilized throughout as the Abbreviations. PEC; peritonealexudatecells; PES: peritonealexudate cell culture supernatant solution; OA: ovalbumin.

Guinea Pig Lymphotoxin

21

target cell. Single cell suspensions of the target cells were obtained by mild trypsinization of exponentially growing monolayer I04CI cultures in RPMI 1640 medium supplemented with 10% fetal bovine serum and diluted in the same medium and serum. Lymphotoxin, as PES, was added to a final concentration of 0, 0.I, 1.0, or 10% (v/v) and 4 ml of the target celllymphotoxin mixture containing 100 cells were added to each 60 mm tissue culture dish. After 7 days incubation, the medium was removed, the colonies washed with Dulbecco's phosphate-buffered saline, fixed in methanol, stained with 10% aqueous Giemsa (Fisher Scientific, Fairlawn, N.J.), and colonies greater than 0.2 mm in diameter counted. RESULTS

Influence of Ovalbumin Concentration upon the Production of Lymphotoxin Ovalbumin induced the generation of lymphotoxin by ovalbumin immune but not by nonimmune PEC. Typical results are recorded in Table 1. The ovalbumin-induced production of lymphotoxin varied directly with the ovalbumin concentration in the PEC culture medium. As illustrated in Figure 1, the lymphotoxin colony inhibitory activity of PES after 24 hr incubation of PEC with 1.25 to 250/~g ovalbumin/ml PEC medium exhibited 16 to 86% inhibition of 104C1 colony formation when assayed at 10% lymphotoxin (v/v) in the target cell medium. Approximately 50% inhibition of colony formation was routinely obtained with ovalbumin stimulation concentrations of 12.5-25 /~g/ml PEC medium when PES lymphotoxin was assayed at 10% v/v in the target cell culture medium.

Time Course of Ovalbumin-lnduced Lymphotoxin Production The generation of lymphotoxin by PEC was dependent upon the concentration and the time of exposure to ovalbumin. As shown in Figure 2, when 12.5 or 125/~g ovalbumin/ml PEC culture medium was added at the time of PEC plating (zero time) and PES collected 1 through 24 hr later, lymphotoxin activity in PES reached a maximum after about 6 hr of PEC culture when ovalbumin was 12.5 /~g/ml, or after 12 hr when ovalbumin was 125 pg/ml. This ovalbumin dose-dependent maximum lymphotoxin activity was maintained unaltered through at least 72 hr of continuous PEC culture (Table 2). No additional PES lymphotoxin activity was produced when the PEC were refed after 24 or 48 hr of culture as shown in the table. Table 1 Oualbumin-induced generation of guinea pig lymphotoxin from immune leukocytes in serum-free medium RPMI 1640 PEC culture medium supplements PEC-cultured leukocytes Nonimmune Ovalbumin immune Ovalbumin immune Ovalbumin immune Ovalbumln immune

% FBS 0 0 10 10 0

t~ ovalbumin/ml

Lymphotoxin actiuity of PES~ (target cell colonies, %)

I00 0 0 I00 I00

i00 86 87 59 39

" PES was obtained from 24 hr cultures of 107 peritoneal exudate leukocytes incubated in 10 ml RPMI 1640 mediunVl00 mm tissueculture dish. The lymphotoxin activity of PES was assayedby determining the colony formation of 104C1 target cells following 7 days incubation with 10% v/v PES in the target cell culture medium. Target cell colonies is the ralio of the number of colonies formed in the presence of PES to the number formed in medium without PES expressedas a percent.

100

A

(73 ku

8O 70

Z

0 50 II 0 ¢.) 50 W

40 30

er"

2O 10

-o I

125 12.5

I

I

I

62.5 125 OVALBUMIN (pg/ml) IN PEC MEDIUM

250

Figure 1 Influence of antigen concentration upon lymphotoxin production. Ovalburninimmune PEC were cultured at 10 ~ leukocytes per 100 m m dish in 10 ml serum-free RPMI 1640 medium supplemented throughout the duration of culture with antigen at the indicated concentrations. After 24 hr incubation, PES were collected and assayed for lymphotoxin colony inhibitory activity against I04CI target cells.

Figure 2 Time course of lymphotoxin production. Ovalbumin-immune PEC were cultured at 10 ~ leukocytes per 100 m m dish in 10 ml serum-free RPMI 1640 medium supplemented throughout the duration of culture with 12.5 or 125 /J.g ovalbumin/ml PEC culture medium. After incubation for the indicated times, PES were collected and assayed for lymphotoxin colony inhibitory activity against I04CI target cells. I00 9O

v

(.)

5° MJ

¢J

P,,

4O 30

• - 12.5 pg OAJml o - 125 pg OA/ml

20 10 I

I

I

I

I

2 4 6 12 18 24 INCUBATION TIME AFTER ADDITION OF OVALBUMIN (HRS) 22

Table 2

Duration of antigen-induced lymphotoxin production in serum-free medium

Culture condition~

Time of leukocyte culture (hr)

Lymphotoxinactivity generated (target cell colonies %)b

Standard

0-24

46

0-48 0 - 72

51 49

24-48 48-72 24-72

98 97 97

Refed

" Ovalbumin-immune PEC were plated at 107 leukocytes/100 mm dish in 10 ml serum-free RPMI 1640 medium supplemented with 100 pg ovalbumin/ml. Standard PES was collected after 24, 48, or 72 hr incubation. Refed PES was obtained by resuspending the centrifuged pellet of PEC removed from the standard 24 hr PES in 10 ml RPMI 1640 supplemented with 100 p,g ovalumin/ml and returning the cells to the adherent leukocytes in the original culture dishes. 24 hr later, the 2 4 - 4 8 hr PES was collected from half the refed dishes and the nonadherent cells returned to the dishes in fresh medium-ovalbumin for collection of 4 8 - 7 2 hr PES 24 hr later. At that time, the 2 4 - 72 hr PES was also collected from the other half of the refed dishes. The ratio of the number of colonies formed in the presence of PES to the number formed in medium without PES expressed as a percent.

Figure 3 Influence of delay in antigen stimulation of cultured leukocytes upon the generation of lymphotoxin. Ovalbumin-immune ,DEC were cultured at 107 leukocytes per 100 ram dish in 10 rn/ serum-free RPMI 1640 medium. Immediately after plating (zero time) and at 3, 6, 24, and 48 hr postplating, ovalbumin was added to 100 pJg/m/ medium. 24 hr after antigen addition PES were collected and assayed for lymphotoxin colony inhibitory activity against 104C1 target cells.

++'il

i+

40

i+

3O

10 I I 36

I 24

TIME OF OVAI.BUMIN ADDITION AFTER PLATING PEC (HRS) 23

I 48

J.O. Rundell and C. H. Evans

24 100-

o - 1.25 ~g OA/ml • - 12.5/~g OA/ml

90

/'.- 125pg OA/ml

8O

t, 86o o

o

40

10 I 1

I I 2 3 PEC/PLATE (xl0-')

I 4

Figure 4 Effect of leukocyte density and antigen concentration in culture on lyrnphotoxin production. Ovalbumin-immune ,DEC were cultured at the densities and antigen concentrations indicated per 100 mm dish in 10 ml serum-free RPMI 1640 medium. After 24 hr, PES were collected and assayed for lymphotoxin colony inhibitory activity against 104C1 target cells.

Ability of Ovalbumin to Induce Lymphotoxin Production at Various Times after PEC Culture

The capacity of peritoneal leukocytes to generate lymphotoxin in response to ovalbumin stimulation diminished progressively during culture in the absence of antigen. For example, when PEC were incubated with I00 ~g ovalbumin/ml commencing 0, 3, 6, 24, or 48 hr after plating and the PEC cultured for 24 hr after the addition of antigen, lymphotoxin production declined by 10% with 3 hr, 20°70 with 6 hr, 60% with 24 hr, and 85% with 48 hr incubation of PEC before ovalbumin addition as illustrated in Figure 3. No diminuition in ovalbumininduced lymphotoxin production occurred, however, when the PEC culture medium was supplemented with 10% fetal bovine serum. Influence of Leukocyte Number and Culture Medium Volume on Lymphotoxin Production The generation of lymphotoxin activity varied in direct production to the number of PEC plated. Figure 4 demonstrates that a linear increase in lymphotoxin production occurred with 1.25 or 12.5 /~g ovalbumin/m] PEC culture medium and a PEC density between 1 and 4 × l 0 T leukocytes per 100 mm tissue culture dish. When 125 p.g ovalbumin/ml were employed, however, no significant increase in lymphotoxin production resulted with increasing numbers of PEC. The effect of PEC culture medium volume on lymphotoxin activity production is illustrated in Figure 5. When PEC exposed to 50/~g ovalbumin/ml were cultured in 10 or 20 ml of medium, similar lymphotoxin activities were obtained per unit volume PES. The total lymphotoxin activity produced, however, increased in direct proportion to the increase in culture medium volume. This proportionality was independent of PEC density in that

25

Guinea Pig Lymphotoxin

1009O

m

,,=,7o o, 6o

85o
o-10 ml •-20 ml

5 10 PEC/PLATE (xl0 "=)

50

Figure 5 Lymphotoxin production as a function of leukocyte culture density and medium volume. Ovalbumin immune PEC were plated in 10 or 20 ml serum-free RPMI 1640 medium supplemented with 50 tgj ovalbumin/ml. PES were collected after 24 hr incubation and assayed [or lymphotoxin colony inhibitory activity against 104C1 target cells.

similar lymphotoxin activities were obtained for PEC culture medium volumes of 10 and 20 ml at PEC densities ranging from I x 106 to 5 × 10 r leukocytes per plate. DISCUSSION Several laboratories have demonstrated that antigen- or mitogen-induced lymphotoxin preparations produced by populations of cultured leukocytes from several anatomic sites and species are active against selected neoplastic cells in vitro (Coyne et al., 1973; Holzman et al., 1973; Rosenberg et al., 1973; Rosenau and Tsoukas, 1976; Evans et al., 1977; Pick, 1977; Youdim, 1977). This laboratory has previously shown that guinea pig PES and peripheral blood lymphocyte lymphotoxin has colony inhibiting activity for syngeneic neoplastic cells, but does not inhibit colony growth of their nonneoplastic counterparts (Evans et al., 1975, 1977). This specificity of lymphotoxin may reflect important, possible unique, qualities of neoplastic cells. The utilization of lymphotoxin as a tool for studies aimed at describing the biological basis for the differential sensitivity of neoplastic versus nonneoplastic cells toward lymphotoxin requires larger quantities of more purified lymphokine than have been available. The results described in the present article characterize the conditions of guinea pig leukocyte culture resulting in routine optimal and reproducible serum-free lymphotoxin production. Ovalbumin was selected as the lymphotoxin-inducing antigen because of its successful use in the production of guinea pig lymphotoxin (Gately and Mayer, 1974), and its lack of toxicity or interference with various target cells in colony inhibition assays (RuncleU and

26

J.O. Rundell and C. H. Evans

Evans, unpublished observations, 1978). The results demonstrate that lymphotoxin production varies directly with antigen and leukocyte concentration, duration of leukocyte culture and exposure to antigen, and medium volume. The optimal conditions for maximum reproducible lymphotoxin production were observed to be 12-24 hr incubation of 107 immune peritoneal leukocytes in 20 ml serum-free RPMI 1640 medium containing I00 p.g ovalbumin/ml. The observation that the total lymphotoxin production varies directly with the serum-free medium volume, but that lymphotoxin concentration is independent of serum-free medium volume, limits the quantity of lymphotoxin that can be produced. It also suggests that lymphotoxin production is controlled by extracellular lymphokine concentration, perhaps by a feedback mechanism. The possibility of feedback control is further supported by the independence of the maximum lymphotoxin concentration generated with leukocyte density at high antigen concentrations. Although the immunoregulatory function and other physiological roles of lymphotoxin remain to be clarified feedback control mechanisms are common in enzymic and in hormonal metabolic pathways. REFERENCES

Bloom BR (1969) In Mediators of Cellular Immunity. Eds., HS Lawrence and M Landy. New York: Academic Press, pp. 249-319. Coyne JA, Remold HG, Rosenberg SA, David JR (1973) Guinea pig lymphotoxin (LT). II. Physicochemical properties of LT produced by lymphocytes stimulated with antigen or concanavalin A: its differentiation from migration-inhibitory factor (MIF).J Imrnunol i I0:1630. Evans CH, Cooney AM, DiPaolo JA (1975) Colony inhibition mediated by nonimmune leukocytes in vitro and skin reactivity in vivo as indices of tumorigenicity of guinea pig cultures transformed by chemical carcinogens. Cancer Res. 35:1045. Evans CH, Rabin ES, DiPaolo JA (1977) The susceptibility of guinea pig cells to the colonyinhibitory activity of lymphotoxin during carcinogenesis. Cancer Res 37:898. Gately MK, Mayer MM (1974) The molecular dimensions of guinea pig lymphotoxin. J Immunol 112:168. Holzman RS, Lebowitz AS, Valentine FT, Lawrence HS (1973) Preparalion and properties of cloning inhibition factor. I. Inhibition of HeLa cell cloning by stimulated lymphocytes and their culture supernatants. Cell Imrnunol 8:249. Papermaster BW (1978) The measurement of lymphokines in cancer patients and its implications in diagnostic testing. Antibiot Chemother 22:32. Papermaster BW, Holten-nan OA, Klein E, Djerasse I, Rosner D, Dau T, Costanzi Jl (1976a) Preliminary observations on tumor regressions induced by local administration of a lymphoid cell culture supernatant fraction in patients with cutaneous metastatic lesions. Clin Irnrnunol Irnrnunopathol 5:31. Papermaster BW, Holterman OA, Klein E, Pamett S, Dobkin D, Laudico R, Djerassi I (1976b) Lymphokine properties of a lymphoid cultured cell supematant fraction active in promoting tumor regression. Clin Imrnunol Imrnunopathol 5:48. Pick E (1977) Lymphokines: conllol and pharmacological modulation of their production and action. In Irnrnunopharrnacology. Eds., JW Hadden, RG Coffey and F Spreafico. New York: Plenum, pp. 163-202. Rosenau W, Tsoukas CD (1976) Lymphotoxin. A review and analysis. Arn J Pathol 84: 580. Rosenberg SA, Henrichon M, Coyne JA, David JR (1973) Guinea pig lymphotoxin (LT). I. In vitro studies of LT produced by antigen stimulation of lymphocytes. J Irnmuno/ I I0:1623. Ruddle NH, Waksman BH (1968) Cytotoxicity mediated by soluble antigen and lymphocytes in delayed hypersensitivity. III. Analysis of mechanism. J Exp Med 128:1237.

Guinea Pig Lymphotoxin

27

Smith ME, Laudico R, Papermaster BW (1977) A rapid quantitative assay for lymphotoxin. J Immunol Methods 14:243. Trivers G, Braungart D, Leonard EJ (1976) Mouse lymphotoxin. J Immunol 117:130. Waksman BH (1976) Immunoglobulins and lymphokines as mediators of inflammatory cell mobilization and target cell killing. Cell Irnmunol 27:309. Williams TW, Granger GA (1973) Lymphocyte in vitro cytotoxicity: mechanism of human lymphotoxin-induced target cell destruction. Cell Immunol 6:171. Youdim S (1977) Destruction of experimental malignant melanoma by mediators of cellular immunity. Cancer Res 37:572.