By TSUNEO KATO AND TOMONORI MINAGAWA* Department of Microbiology, Hokkaido University School of Medicine, Sapporo 060, Japan

Transcription

1 J. gen. ViroL (1981), 54, Printed in Great Britain 293 Cellular Source of Interferon Induced in Human Peripheral Blood Mononuelear Leukocytes by Mumps Virus or by Tumour Cells Persistently Infected with Mumps Virus By TSUNEO KATO AND TOMONORI MINAGAWA* Department of Microbiology, Hokkaido University School of Medicine, Sapporo 060, Japan (Accepted 26 January 1981) SUMMARY Human leukocyte interferon (IFN-~) was induced in cultures of human peripheral blood mononuclear leukocytes (PBML) from seropositive healthy donors infected by mumps virus, or by a human lung carcinoma cell line (Pc-10) persistently infected with mumps virus (Pc-10/MpV). Adherent cells (M~), non-t, non-b cells and B cells all produced IFN in response to mumps virus or Pc-10/MpV cells. The non-t, non-b cells co-cultured with Pc-10/MpV cells produced high-titred IFN. However, T cells did not produce any antiviral factors even in the presence of M~. Furthermore, T cells activated with phytohaemagglutinin (PHA-P), concanavalin A or treated with u.v.-irradiated mumps virus, or Pc-10/MpV cells pretreated with mitomycin C did not produce any antiviral factors in response to either mumps virus or Pc-10/MpV cells. INTRODUCTION At present, it is recognized that the human interferons (IFN) have been classified into three groups, x, fl and 7 (Interferon Nomenclature, 1980). Acid-stable IFN can be divided antigenically into leukocyte IFN-u and fibroblast IFN-fl (Havell et al., 1977). Viruses usually induce IFN-~ in normal human peripheral blood mononuclear leukocytes (PBML), probably by the non-t lymphocytes (Yamaguchi et al., 1977; Kirchner et al., 1979). However, acid-labile IFN-7 was induced in T lymphocyte-macrophage mixtures obtained from patients convalescent with herpes labialis by stimulation with herpes simplex virus (Valle et al., 1975), and also from patients previously immunized with vaccinia virus by stimulation with heat-inactivated vaccinia virus (Epstein et al., 1972). On the other hand, human PBML in contact with tumour cell lines produced IFN (Trinchieri et al., 1977), especially cell lines persistently infected with viruses (Ito et al., 1974; Santoli et al, 1978; Kadish et al., 1980). We have established and maintained a human lung carcinoma cell line persistently infected with mumps virus (Pc-10/MpV) which continuously released infectious virus but no antiviral factors (T. Kato & T. Minagawa, unpublished data). However, a mixed culture of Pc-10/MpV cells and human PBML produced substantial amounts of IFN. In this study, we have investigated the major cellular source and the type of IFN produced in human PBML from seropositive donors in response to mumps virus or Pc-10/MpV cells. METHODS Cell lines. A human lung carcinoma cell line (Pc-10), provided by Dr Y. Hayata, Department of Surgery, Tokyo Medical College, Tokyo, Japan and Pc-10 cells persistently infected with mumps virus (Pc-10/MpV), were grown in RPMI 1640 medium supplemented with 10% heat-inactivated foetal bovine serum (Gibco). The Pc-10/MpV cells, which had /81/ $ SGM

2 294 T. KATO AND T. MINAGAWA been established and maintained in our laboratory, had mumps antigens detectable on the cell surface and in the cytoplasm of each cell. These produced approx. 104 p.f.u./ml infectious virus in the culture fluid which was harvested every 3 days (T. Kato & T. Minagawa, unpublished data). A human amnion cell line (FL) used for interferon assay was grown in RPMI 1640 medium supplemented with 5 % foetal bovine serum. Viruses. Mumps virus (MpV) which had been isolated in our laboratory from the cerebrospinal fluid of a patient suffering from mumps viral meningitis, was propagated twice in Vero cells and stored at -70 C until use. The titre of the virus was p.f.u./ml. Vesicular stomatitis virus (Indiana strain), which had been grown in mouse embryo fibroblasts, was used for the assay of interferon (2 106 p.f.u./ml). Separation of human lymphocytes. Human peripheral blood was obtained by venipuncture from 20 healthy volunteers whose ages ranged from 20 to 30 years. All the volunteers had anti-mumps virus-neutralizing antibody titres ranging from 1:32 to 1:128. The PBML were isolated from heparinized blood by sedimentation on a Ficoll-Paque gradient (Pharmacia). The interphase cells were washed once with M-phosphate-buffered saline, twice with MEPS [154 mm-nac1, 10 mm-sodium EDTA, 8 mm-kh 2 PO 4 and 0.04 % methyl cellulose (w/v)] to deplete platelets and finally twice with RPMI 1640 medium. PBML (3 107 cells in 3 ml medium) were incubated in a 65 x 15 mm plastic dish (Nunc) at 37 C for 90 min, and the non-adherent cells removed and incubated in another plastic dish at 37 C for 30 min to completely deplete the adherent cells. The non-adherent cells obtained were washed twice with medium and then used for the separation of lymphocyte subpopulations. Almost all the adherent cells were considered to be monocytes (M~) as judged by Giemsa and peroxidase staining. Separation of T and non-t cells. The non-adherent cells were fractionated by the E-rosette formation technique (Wybran et al., 1973). Cells were mixed with sheep red blood cells (SRBC) at an SRBC :cell ratio of 50 : 1. The mixture was incubated at 37 C for 15 min, the cells were spun down, and then further incubated at 4 C for 60 min. The cells were resuspended and centrifuged on a FicoU-Paque gradient. The SRBC rosette-forming (T) cells were obtained in a pellet, and the SRBC were lysed with 0.85 % NH4CI. Non-T cells were taken from the interphase. Both T cells and non-t cells were washed three times with medium. The T cells comprised approx. 60 to 80 % of the non-adherent cells. Fraetionation of cells by nylon wool column. The B cell-rich fraction was obtained by using a nylon wool column (Julius et al., 1973). Non-T cells in 2 ml medium were slowly loaded on to a nylon wool column which was packed with 0.6 g (dry wt.) nylon wool, then sealed and incubated at 37 C for 90 min. Surface immunoglobulin-positive (slg +) cells comprised approx. 80 % of the nylon wool-adherent (B) cells and <2 % of the nylon wool non-adherent (non-t, non-b) cells. Separation of activated T cells. T cells were activated by the following methods. T cells (1 107) were stimulated with 50 #g phytohaemagglutinin (PHA-P; Difco) or with 2 #g concanavalin A (Con A; Wako, Osaka, Japan) and, 3 days later when lymphocyte transformation was obvious by microscopy, cells were washed twice with medium. Similarly, 1 10 T T cells were co-cultured with 1 x 106 Pc-10/MpV cells which had been treated with mitomycin C (25 ~tg/ml) for 30 min to stop cell growth and then washed four times with medium to remove mitomycin C, or were inoculated with p.f.u, equivalent of u.v.-inactivated mumps virus. On day 5, when lymphocyte transformation was obvious, T cells were separated from the mixtures and washed twice with medium. DNA synthesis in activated T cells. The T cells ( cells/ml medium) stimulated by the above methods were immediately labelled with 3H-thymidine (2/tCi/ml) for 1 h and were then treated with 5% trichloroacetic acid (TCA) and placed in an ice bath for 30 min. The TCA-insoluble fraction was collected on glass fibre filters (GF/C, Whatman) and the radioactivity was counted Downloaded by a liquid from scintillation counter (Horiba LS-551). by

3 IFN induced in human PBML by mumps virus 295 Interferon (IFN) induction. Whole or fractionated PBML were infected with mumps virus at a multiplicity of infection (m.o.i.) of 5. After adsorption of the virus at 37 C for 1 h the cells were washed twice with medium and 1 x 106 cells in 1.5 ml medium were cultivated in a Nunc multiplate dish. Similarly, cells of whole or fractionated PBML in 1-5 ml medium were co-cultured with 1 x 105 cells of Pc-10 or Pc-10/MpV cells. IFN assay. The culture fluid was harvested and centrifuged at rev/min for 2 h. Half of the supernatant was dialysed against 0.2 M-KC1-HC1 buffer ph 2 at 4 C for 48 h followed by dialysis against medium. IFN activity of the acid-treated and untreated sample was assayed by the method described by Armstrong (1971), using the FL cell line and vesicular stomatitis virus as a challenge virus. The IFN titre of the sample was corrected with the NIH leukocyte IFN standard (G ) and expressed as International Units (IU)/ml. Neutralization oflfn by anti-ifn sera. Anti-IFN-a and anti-ifn-fl rabbit sera were kindly provided by Drs S. Yonehara and S. Kobayashi of the Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan. Anti-IFN-a serum had been prepared by injecting rabbits with highly purified (> 108 IU/mg protein) human leukocyte IFN induced in Namalwa lymphoblastoid cells. Anti-IFN-fl serum had been prepared by injecting rabbits with highly purified (> 107 IU/mg protein) human fibroblast IFN induced in human foreskin fibroblasts by poly(ri), poly(rc). These sera were sufficiently diluted to neutralize 100 IU/ml standard IFN-a and IFN-fl (obtained from the Blood Centre of Hokkaido, Sapporo, and the Torey Basic Research Institute, Kamakura, Japan respectively). The IFN samples containing approx. 100 IU were serially diluted twofold and each dilution was mixed with the equal volume of antiserum. After incubation at 37 C for 30 min the residual IFN activities were assayed. The neutralization of IFN by anti-ifn serum was expressed as a percentage of the control using normal rabbit serum. RESULTS Production of IFN-a in whole PBML by mumps virus or Pc-l O/Mp V cells Cells (1 10 6) of whole PBML were inoculated with p.f.u, of mumps virus or co-cultured with 1 x 105 cells of Pc-10 or Pc-IO/MpV cells. As shown in Fig. 1, IFN activity was detected in the culture fluids 4 h after the addition of the virus or Pc-10/MpV cells, and the maximum titre of IFN was reached by 24 to 48 h. The titres of IFN produced by stimulation with Pc-10/MpV cells were usually fourfold higher than those by mumps virus. As the IFN titres induced by the Pc-10 cells were very low (< 100 IU/ml), IFN induction in the PBML cells could have been due to mumps virions or mumps antigens on the cell surface of Pc-i0/MpV cells. All IFN samples were acid-stable and not neutralized with anti-ifn-fl serum. Spontaneous IFN production in whole PBML was not observed during 5 days cultivation of PBML without mumps virus or Pc-10/MpV cells. Production of lfn in fractionated PBML by mumps virus or Pc-I O/MpV cells PBML, obtained from three individuals, were fractionated into adherent cells (M~), non-adherent cells, SRBC rosette-forming (T) cells and non-t cells. IFN production in these fractionated cells by mumps virus or Pc-10/MpV cells was then examined. The T cells did not produce any IFN with either inducer. Titres of IFN induced by Pc-10/MpV cells were much higher than those produced by mumps virus. The non-t cells produced high titre IFN in response to Pc-10/MpV cells (Table 1). All these IFN samples were acid-stable. The percentages of the residual activity of IFN samples treated with anti-ifn-a serum or with anti-ifn-fl serum were <4% and 100% respectively. Non-T cells were then further fractionated into nylon wool-non-adherent (non-t, non-b) cells and -adherent (B) cells. There was no difference between Downloaded the non-t, from non-b cells and the B ceils in by the production of IFN by

4 296 T. KATO AND T. MINAGAWA I I I (> Z <11 ~ 7" f Incubation (h) Fig. 1. Production of IFN in human PBML by mumps virus or by Pc-10/MpV cells. Cells (1 10 6) of whole PBML were inoculated with p.f.u, of mumps virus (0) or co-cultured with 1 x 10 5 cells of Pc-10 (A) or Pc-10/MpV (O). Spontaneous release of IFN ([3) was also measured. Table 1. Production of lfn in fractionated PBML by mumps virus or Pc-IO/MpV cells IFN titre (IU/ml) after stimulation with.& Cell Donor Pc- 10 MpV Pc- 10/MpV Whole PBML T.K H.I K.K. ND* Adherent T.K H.I K.K. ND 240 ND Non-adherent T.K H.I K.K. ND T T.K. < 10 < 10 < 10 H.I. < 10 < 10 < 10 K.K. < 10 < 10 < 10 Non-T T.K H.I K.K. ND * ND, Not done. Table 2. Production of lfn in non-t, non-b cells by mumps virus or Pc-IO/MpV cells IFN titre (IU/ml) after stimulation with F Cell Donor MpV Pc- 10/MpV slg (%)* Non-T T.K H.I Non-adherentS" T.K (Non-T, non-b) H.I Adherent (B)~" T.K H.I * Percentage of cells with surface immunoglobulin. I" Cells did not adhere to nylon wool.

5 IFN induced in human PBML by mumps virus 297 Table 3. Production of lfn in T and M~ mixtures by mumps virus or Pc-lO/MpV eells r Expt. C ell Inducer 1 IFN titre (IU/ml) harvested on day "x II M~ (1 x 106) MpV ND* 160 ND Pc- 10/MpV ND 560 ND T (1 x 106) MpV < 10 < 10 ND ~ 10 NO Pc-10/MpV < 10 < 10 ND < 10 ND M~ + T MpV ND 160 ND Pc-10/MpV ND 600 ND M~ (2 x 105) MpV < 10 < 10 Pc- 10/MpV T (1 x 106) MpV < 10 < 10 < 10 < 10 < 10 Pc-10/MpV < 10 < 10 < 10 < 10 < 10 M~ + T MpV < 10 < 10 Pc- 10/MpV * ND, Not done. Table 4. Lack of lfn production in T cells activated with PHA-P, Con A, Pc-IO/MpV cells pretreated with mitomyein C or u.v.-irradiated mumps virus Time IFN titre Expt. Treatment (days) Ct/min Inducer (IU/ml) PHA-P (50/tg/ml) Control* 80 MpV 80 Pc-10/MpV 80 Con A (2 #g/ml) Control 120 MpV 120 Pc- 10/MpV 120 Control _+ 63 Control < 10 MpV < 10 Pc-10/MpV < 10 Pc- 10/MpV Control < 10 (mitomycin C) MpV < 10 Pc- 10/MpV < 10 U.v.-MpV Control < 10 MpV < 10 Pc- 10/MpV < 10 * Control medium (RPMI 1640). mumps virus. However, Pc-10/MpV cells induced the most IFN in non-t, non-b cells from two individuals (Table 2). Production of IFN in T cells and macrophage (T/M~) mixtures by mumps virus or Pc-10/Mp V cells T cells (1 x 106) were mixed with 2 x 105 or 1 x 106 M# cells and mumps virus or Pc-10/MpV cells (1 x 105) was added and the mixtures cultured for 6 days. The IFN from both inducers in these mixtures was due solely to their content of M~ cells, because all IFN samples obtained in this experiment were acid-stable (Table 3).

6 298 T. KATO AND T. MINAGAWA Lack of lfn induction in activated T cells by mumps virus or by Pc-I O/MpV cells No T ceils ever produced IFN by induction of mumps virus or Pc-10/MpV ceils, even in the presence of M~. The T cells were treated with 50/~g/ml PHA-P or 2 #g/ml Con A, which transformed the T cells and produced acid-labile IFN on day 3 after the addition of PHA-P or Con A. The T cells were then washed twice with medium and cultured with mumps virus or Pc-10/MpV cells for 24 h. In the culture fluids from these mixtures acid-labile IFN, not neutralized with anti-ifn-a and -8 sera, was detected. However, this IFN induction might have been due to the continuing effect of PHA-P or Con A, and not to the addition of mumps virus or Pc-10/MpV cells, because the transformed T cells spontaneously released the same level of IFN without mumps virus or Pc-10/MpV cells (Table 4). Transformation of T cells was observed 5 days after contact with mitomycin C-treated Pc-10/MpV cells or u.v.-irradiated mumps virus, but the T cells did not produce any antiviral factors. Furthermore, these transformed T cells did not produce any IFN when restimulated with mumps virus or Pc-10/MpV cells. These results indicate that T cells could be transformed by stimulation with PHA-P, Con A, mitomycin C-treated Pc-10/MpV cells or u.v.-irradiated mumps virus, but the transformed T cells did not produce any IFN in response to mumps virus or Pc-10/MpV cells. DISCUSSION It has been established that mumps virus can replicate in human PBML (Duc-Nguyen & Henle, 1966), and McFarland et al. (1980) reported that mumps virus stimulated virus-specific lymphocyte transformation of T cells obtained from seropositive donors. In the light of these findings, we initiated the present work to determine whether the induction of acid-labile IFN-7 in human PBML obtained from seropositive donors could be achieved in response to mumps virus or human lung carcinoma cells persistently infected with mumps virus and which do not produce IFN themselves. However, we have no evidence for the production of IFN-~, only for that of IFN-u. Pc-10/MpV cells induced four times more acid-stable IFN-tx than did mumps virus in whole PBML. Neither Pc-10/MpV nor Pc-10 cells produced IFN spontaneously and Pc-10 cells could induce little IFN in whole PBML. Free mumps virus did not seem to be a major inducer in this system, since Pc-10/MpV cells released less than 104 p.f.u./ml of mumps virus, and little IFN was induced by mumps virus even if PBML were infected with mumps virus at an m.o.i, of 5. Therefore, IFN produced by a mixture of Pc-10/MpV cells and whole PBML seemed to result from the interaction between whole PBML and mumps virus components on the cell surface of Pc-10/MpV cells. Alternatively, there might be lymphocyte subpopulations which produce IFN more efficiently in response to virus-carrier cells. The capacity to induce IFN in whole PBML on co-culture with Pc-10/MpV cells was due to their content of SRBC non-rosetting, nylon wool-non-adherent (non-t, non-b) cells, while mumps virus induced approx, equal amounts of IFN in all cell populations, with the exception of T ceils. This agrees with the results of Trinchieri et al. (1978). Non-T, non-b cells may thus react preferentially with virus-carrier cells, and this may reflect host defence mechanisms against virus infection in vivo. The T cells did not produce IFN even in the presence of macrophages (M~). Although IFN was induced in the mixed cultures with T cells and M~ in response to mumps virus and Pc-10/MpV cells, all the acid-stable IFN-a could be accounted for by the M~ only. In addition, T cells activated non-specifically with PHA-P or Con A and specifically with Pc-10/MpV cells pretreated with mitomycin C or u.v.-irradiated mumps virus, did not produce IFN in response to mumps virus or Pc-10/MpV cells. Therefore, we conclude that mumps virus or Pc-10/MpV cells stimulate lymphocyte transformation, but do not induce IFN in T cells obtained from seropositive individuals. Our results agree with those of Lazar &

7 IFN induced in human PBML by mumps virus 299 Wright (1980) who reported that influenza A virus induced IFN<t in human PBML obtained from seropositive as well as seronegative individuals. However, Valle et al. (1975) reported that a mixture of T cells and M~ produced acid-labile IFN-7 in response to herpes simplex virus, but that whole PBML obtained from the same donor produced acid, stable IFN-a. Therefore, it is possible that IFN-a is produced generally in blood in a virus infection, while IFN-y may be produced locally in the infected foci. The production and significance of IFN-y in virus infection still remains to be investigated. We wish to thank Dr H. Iida of this Department for his advice and encouragement and Ms T. Abe and M. Shiozaki for their technical assistance. This study was supported in part by a grant from the Ministry of Education, Science and Culture, Japan. REFERENCES ARMSTRONG, J. A. (1971). Semi-micro, dye-binding assay for rabbit interferon. Applied Microbiology 21, DUC-NGUVEN, a. & HENLE, W. (1966). Replication of mumps virus in human leukocyte cultures. Journal of Bacteriology 92, EPSTEIN, L. B., STEVENS, D. A. & MERICAN, r. C. (1972). Selective increase in lymphocyte interferon response to vaccinia antigen after reactivation. Proceedings of the National Academy of Sciences of the United States of America 69, I~AVELL, E. A., WP, V. K. ~ VmCEK, J. (1977). Characteristics of human lymphoblastoid (Namalva) interferon. Journal of General Virology 38, INTERFERON NOMENCLATURE. (1980). Nature, London 286, 110. IVO, Y., gimxma, v., NAGATA, L ~ glrmi, A. (1974). Production of interferon-like substance by mouse spleen cells through contact with BHK cells persistently infected with HVJ. Virology 60, JULIUS, u. H., SIMPSON, E. & HERZENBEaG, L. A. (1973). A rapid method for isolation of functional thymus-derived murine leukocytes. European Journal of Immunology 3, KADISH, A. S., TANSEY, F. A., YU, G. S. M., DOYLE, A. T. & BLOOM, B. R. (1980). Interferon as a mediator of human lymphocyte suppression. Journal of Experimental Medicine 151, KIRCHNER, H., PETER, H. H., HIRT, H. M., ZAWATZKY, R., DALUGGE, H. & BRADSTREET, P. (1979). Studies of the producer cell of interferon in human lymphocyte cultures. Immunobiology 156, LAZAR, A. ~ WRICaHV, P. (1980). Cell-mediated immune response of human lymphocytes to influenza A/USSR (H 1N 1) virus infection. Infection and Immunity 27, MCFARLAND, n. F., PEDONE, C. A., MINGIOLI, E. S. & McFARL~, D. E. (1980). The response of human lymphocyte subpopulations to measles, mumps, and vaccinia viral antigens. Journal of Immunology 125, SANTOLI, D., rmnci-iieri, Ca. ~ gopgowsra, I~. (1978). Ceil-mediated cytotoxicity against virus-infected target cells in humans. II. Interferon induction and activation of natural killer cells. Journal of Immunology 121, TR1NCHIERI, Ca., SANTOLI, D. & KNOWLES, B. B. (1977). Tumor cell lines induce interferon in human lymphocytes. Nature, London 270, TRINCHIERI, G., SANTOLI, D., DEE, R. R. & KNOWLES, B. B. (1978). Anti-viral activity induced by culturing lymphocytes with tumor-derived or virus-transformed cells. Identification of the anti-viral activity as interferon and characterization of the human effeetor lymphocyte subpopulation. Journal of Experimental Medicine 147, VALLE, M. J., JORDAN, G. W., HAAHR, S. & MERI~AN, X. C. (1975). Characteristics of immune interferon produced by human lymphocyte cultures compared to other human interferons. Journal of Immunology 115, WVBRAN, J., CI-XANTLER, S. & FUDENBERG, a. (1973). Human blood T cells: response to phytohemagglutinin. Journal of Immunology 110, YAMAGUCHI, T., HANDA, K., SHIMIZU, Y., ABO, T. & KUMAGAI, K. (1977). Target cells for interferon production in human leukocytes stimulated by Sendal virus. Journal of Immunology 118, (Received 26 November 1980)