6.25 Bovine natural killer activity against virally infected cells inhibited by monoclonal antibodies

Veterinary Immunology and Immunopathology, 39 (1993) 269-274 0165-2427/93/$06.00 © 1993 - Elsevier Science Publishers B.V. All fights reserved

269

6.25 Bovine natural killer activity against virally infected cells inhibited by monoclonal antibodies Gary Splitter*, Sang Hoon Choi Department of Animal Health and Biomedical Sciences, Universityof Wisconsin-Madison, Madison, W153706, USA

Abstract

Forty-four monoclonal antibodies (mAbs) were evaluated for their ability to alter natural killer (NK) cell lysis of virally infected target cells. Six of the mAhs inhibited the lysis of the target cells, while one of the mAhs enhanced lysis. Four of the inhibitory mAbs, CACT26A, CACT16A, CACTB45A and MUC76A, had very marked activity. These mAbs with inhibitory or enhancing activity recognized ( 1 ) WC2 molecules (CACTB44A, CACT 16A, CACT26A) present on putative NK cells, (2) molecules on granulocytes, monocytes, a subpopulation of lymphocytes (CACTB45A and TH2A), (3) CD1 la (MUC76A), and a protein ofCD3 (MM1A).

Introduction

Natural killer (NK) cells are a heterogeneous cluster of non-T, non-B lymphoid cells that lyse certain vitally infected and tumour cells without prior antigenic stimulation. The frequency of NK cells in animals correlates to the degree of host resistance to certain viral infections or tumours. NK cells in humans and mice have been identified by the expression of CD 16 and CD56 (Herberman et al., 1986); however, to date these molecules have not been identified on bovine NK cells. Also, several NK surface molecules in humans and mice, e.g. CD 11a, have been associated with NK-target cell interaction (Werfel et al., 1991 ). The participation of accessory molecules in bovine NK cell function has not been established. Therefore, identifying mAbs that bind bovine NK cells would be useful in understanding the participation of these cells in the immune response to viral pathogens. NK cells have been described that lyse bovine herpesvirus-1 infected cells (Cook et al., 1989). Specific viral glycoproteins participate in the NK-target cell interaction (Palmer et al., 1990). In this study, a panel of 44 mAbs to bovine lymphoid cells was analyzed for their effect on a NK cell lysis of bo*Corresponding author.

270

G. Splitter, Sang Hoon Choi / Vet. lmmunol. Immunopathol. 39 (1993) 269-274

vine herpesvirus- 1 infected target cells. The results indicate that certain mAbs can decrease or increase the lysis of virally infected target cells. Materials and methods

Effector cells Bovine peripheral blood cells (PBMC) were obtained by lymphoprep density gradient centrifugation from two Guernsey cattle 3 years of age. To remove adherent cells, PBMC were cultured on plastic overnight at 37 °C under 5% CO2 in RPMI medium supplemented with 10% fetal calf serum. After incubation, cells were gently agitated and resuspended in fresh medium for use as enriched effector cells in cytotoxicity studies.

Target cells The canine osteosarcoma, D I 7 (ATCC no. CCL183, American Type Culture Collection, Rockville, MD), was cultured in minimum essential medium supplemented with 10% fetal calf serum and maintained at 37 °C under 5% CO2.

Antibodies (mAbs) Antibodies used in this study were part of the workshop. Antibodies were dialysed against phosphate-buffered saline prior to use to remove sodium azide, or a control sample of 0.2% azide in medium was used to compare the effect on target cell lysis by N K cells, mAbs were added to each assay at a final concentration of 1 0 / t g / m l - 1. In addition, mAbs of the same isotype but irrelevant specificities were used to control for isotype effects in the assays. Each mAb was assessed once for an effect on the cellular assay. Those altering N K cell cytolytic activity were evaluated in one to four additional experiments.

Cytotoxicity studies Cytotoxicity assays were performed by 5iCr release as previously described (Cook and Splitter, 1989). Briefly, adherent target cells were removed with trypsin-EDTA, diluted to 1 × 105 cells ml-1 in complete medium containing 3.0/tCi Na 51CrO4 m1-1, then 100/zl were pipetted into 96 well round-bottomed plates. After overnight incubation, target cells were washed three times to remove unincorporated 51Cr and, when appropriate, live bovine herpesvirus-1 (Cooper strain ATCC VR 864) was added for 1 h in 100/tl at a multiplicity of infection of ten. Target cells were washed three times and effector cells were added in quadruplicate to target cells at selected ratios. Enriched

G. Splitter, Sang Hoon Choi / Vet. Immunol. Immunopathol. 39 (1993) 269-274

271

effector cells, suspended in 100/zl, were added to 51C-labelled target cells. After a 19 h coincubation of effector and target cells, 100 #1 of supernatant were collected from each well and counted by gamma spectrophotometry. Cytotoxicity was calculated using the formula: % cytotoxicity= (experimental c p m - spontaneous cpm) / (maximum c p m - spontaneous cpm) × 100. Lysis of non-virally infected target cells was not observed, as previously reported (Cook and Splitter, 1989; Palmer et al., 1990). Results Forty-four mAbs were added separately to each NK assay, and the extent of inhibition or augmentation evaluated and compared with effector-target cells cultured without mAbs or with control mAbs. Six mAbs exhibited an inhibitory effect on N K cell cytolysis. Four of these dramatically inhibited lyric activity. Representative assays of these mAbs are illustrated in Fig. 1. One mAb, MM 1A, augmented lytic activity in five out of six assays (Fig. 2 ). The remaining 37 mAbs failed to alter NK cell lysis of virally infected target cells. A summary of the findings is shown in Table 1. A

B

30

50 M~jn(~76A

I

I Contl

40

CACTI~5A

20

30

e

2O 10 !

•

20:1

;r

•

10:1

T

0

|

2:1

.

|

20:1

C

.

|

10:1

2:1

D

3O

I

50' 40"

------I---

Contl

CACT26A

m 20, 30 ,,J

20 10.

10 •

20:1 Ef fect

-

=

10:1 or:Ta

•

•

2:1 rg et R==tlo

0

2o:1 l O : 1

" 2;1

Effector:TargetRatlo

Fig. 1. mAbs that markedly inhibited NK cell lysis of bovine herpesvirus infected D 17 cells, MUC76A (panel A), CACTB45A (panel B), CACTI6A (panel C) and CACT26A (panel D). Bovine non-adherent cells cultured with viraUy infected D 17 cells in the presence of an isotype control mAb were used as the control (Contl).

272

G. Splitter, Sang Hoon Choi / Vet. lmmunol, lmmunopathol. 39 (1993) 269-274 50,

Contl 8

40-

MM1A

30.J 2o

20:1

10:1

2:1

Effeclor:TartetRRtlo

Fig. 2. mAb that stimulates N K cell lysis of bovine herpesvirus infected D17 cells. Control (Contl) was similar to Fig. 1.

Table 1 Monoclonal antibodies assayed for functional alteration of NK cell lysis of bovine herpesvirus infected D 17 cells Monoclonal antibody

Effect

Monoclonal antibody

Effect

CACTB44A MUC76A TH2A CACT26A CACTB45A CACT 16A MM 1A BAS21A CACTB 15B BAQI5A ACTG5A MUC2A BAG 18A BAQ 15A TH 14B CACT 100A TH 18A CACT77A BAQ89A G55A CAPP2A CACTB 1A TH1A

Inhibitory Inhibitory Inhibitory Inhibitory Inhibitory Inhibitory Stimulatory 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

BAQ86A BAT 15A 12H 1A BAG25A BAQ72A CACT 101A GC50A 1 BAGB27A CACTB6A B18A BAQ 153A CACT 18A CH59B GC6A CACT I 14A CACT 108A CACT 137A GX 18A CAGB 12A GC65A BAQ91A BAQ30A

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

G. Splitter, Sang Hoon Choi / Vet. ImmunoL Immunopathol. 39 (1993) 269-274

273

Discussion

Little is known about the ceU surface molecules which distinguish bovine NK cells from other lymphoid cells. Consequently, the NK cell surface molecules that recognize, adhere to or lyse target cells are currently unidentified. The present study has identified seven mAbs which recognize molecules present on NK cells---non-T, non-B lymphoid cells as they have been recognized at the time of submission to the workshop. Those mAbs were CACTB44A (workshop ~145), CACT26A (#166), CACT16A (~142), TH2A (#161), CACTB45A (#187), MUC76A (~176) and MM1A (~138). Six of the seven mAbs interfere with the function of NK cells--CACTB44A, CACT26A, CACT 16A, TH2A, CACTB45A and MUC76A. The findings reported here are consistent with results in other species which point to a distinct characteristic of NK cells; that is, NK cells lack the classic antigen recognition molecules found on T cells or B cells. However, it has been shown that NK cells in several species share surface molecules characteristic of other immune cell populations, e.g. CD 16, CD45 and the p75 component of the IL-2 receptor (Erber, 1990; Baume et al., 1992). Additionally, two mAbs, TH2A and CACTB45A, have been previously shown to recognize cell surface molecules on bovine granulocytes, monocytes and a subpopulation of lymphocytes as submitted to the workshop (W. Davis, unpublished findings, 1992). In this study, those same two mAbs have been shown to affect bovine NK cell activity. Two of the 44 mAbs investigated in this study, MUC76A and MM1A appear to be mAbs that recognize surface molecules shared by a number of cell types, including NK cells. The mAb MUC76A reportedly recognizes CD 11 a, an integrin adhesion molecule found on many cell types, whose ligand is ICAM-1 and functions in cell-cell interaction. The fact that MUC76A has now been shown to inhibit bovine NK cell activity as well, provides evidence that CD 11 a plays a role in NK cell-target cell interaction. Lastly, one mAb augmented NK activity, MM 1A, which is putatively specific for CD3. Although the increased NK activity was not dramatic, the presence of the mAb consistently elevated the lytic activity observed. Others have reported that NK cells express the zeta component of the CD3 complex of proteins and that the zeta protein functions in NK cell activation once the target cell has been engaged (Vivier et al., 1991; Moingeon et al., 1992). Our results with MM 1A and bovine NK cells would be consistent with these previously reported findings. It is possible that NK surface proteins recognized by the six mAbs discussed above were also recognized by other members of the panel of 44 mAbs we tested. This discrepancy in functional effect between two such groups of mAbs may be explained by epitope recognition differences. A functionally affective mAb must react with an epitope on the NK cell surface protein which is crit-

274

G. Splitter. Sang Hoon Choi / Vet. lmmunoL ImmunopathoL 39 (1993) 269-274

ical to the biological activity of that molecule. A second mAb may react with the same surface protein; however, its cognate epitope may be on a region of the molecule that is unimportant to biological activity. Further studies will be required to identify the specific N K cell surface molecules that were recognized and functionally affected by the seven mAbs discussed here. Some of the mAbs may share epitopes. Although the findings in the present study are not complete, they are preliminary evidence that a series of mAbs influence the activity of bovine N K cells. We now have the basis for further characterization of N K cells and their function.

References Baume, D.M., Robertson, M.J., Levine, H. Manley, T.J., Schow, P.W. and Ritz, J., 1992. Differential responses to interleukin 2 define functionally distinct subsets of human natural killer cells. Eur. J. Immunol., 22( 1 ): 1. Cook, C.G. and Splitter, G.A., 1989. Characterization of bovine mononuclear cell populations, with natural cytolytic activity against bovine herpesvirus 1-infected cells. Cell. Immunol., 120: 240. Cook, C.G., Letchworth, G.J. and Spitter, G.A., 1989. Bovine naturally cytolytic cell activation against bovine herpesvirus type 1-infected cells does not require late viral glycoproteins. Immunology, 66: 565. Erber, W., 1990. Human leukocyte differentiation antigens: review of the CD nomenclature. Pathology, 22: 61. Herberman, R.B., Reynolds, C.W. and Ortaldo, J.R. 1986. Mechanism of cytotoxicity by natural killer cells. Annu. Rev. Immunol., 4:651. Moingeon, P., Lucich, J.L., McConkey, D.J., Letourneur, F., Malissen, B., Kochan, J., Chang, H.C., Rodewald, H.R. and Reinherz, E.L., 1992. CD3 zeta dependence of the CD2 pathway of activation in T lymphocytes and natural killer cells. Proc. Natl. Acad. Sci. USA, 89 (4): 1492. Palmer, L.D., Leary, T.P., Wilson, D.M. and Splitter, G.A., 1990. Bovine natural killer-like cell responses against cell lines expressing recombinant bovine herpesvirus type-I glycoproteins. J. Immunol., 145: 1009. Werfel, T., Witter, W. and Gotze, O., 1991. CD1 lb and CD1 lc antigens are rapidly increased on human natural killer cells upon activation. J. Immunol., 147: 2423. Vivier, E., Morin, P.M., O'Brien, C., Schlossman, S.F. and Anderson, P., 1991. CD2 is functionally linked to the zeta-natural killer receptor complex. Eur. J. Immunol., 21 (4): 1077.