key: cord-0937708-u61fc0sg
authors: Charley, B.; McCullough, K.; Martinod, S.
title: Antiviral and Antigenic Properties of Recombinant Porcine Interferon Gamma
date: 1988-09-30
journal: Veterinary Immunology and Immunopathology
DOI: 10.1016/0165-2427(88)90001-3
sha: d55f8c96c7d28ad32a7975c32d6644b8abd4f9ad
doc_id: 937708
cord_uid: u61fc0sg

Abstract Recombinant porcine interferon gamma (rPoIFNγ) induced a dose-dependent inhibition of the cytopathic effect produced by vesicular stomatitis virus (VSV) challenge of both homologous and heterologous (bovine) cell lines. In addition, an antiviral effect of rPoIFNγ was demonstrable against the coronavirus transmissible gastroenteritis virus (TGEV) infection of porcine epithelial cells and of pulmonary macrophages. A rabbit anti-PoIFNγ antiserum was prepared and shown to specifically neutralize the antiviral effects of natural and recombinant porcine IFNγ preparations. This antiserum could also neutralize recombinant bovine IFNγ but not recombinant human IFNγ. These results suggest antigenic homology of porcine and bovine IFNγ but antigenic differences between these molecules and human IFNγ.

Interferon 7 (IFN7), produced by activated T lymphocytes, can exert antiviral activity and a number of immunomodulatory effects such as enhancement of NK-and T cell-mediated cytotoxicity, B cell differentiation, surface antigens expression and macrophage activation (reviewed by Trinchieri and Perussia, 1985 ) .

Although extensively studied in rodents and man, IFN7 has only recently received increasing attention in domestic animals. Adequate evaluation of IFNy for its potential use against viral diseases or as an immunomodulator was difficult mainly because of the paucity of purified material. Cloning and expression of DNA encoding bovine and porcine IFNs (Capon et al., 1985; Ceretti et al., 1986; Lefevre and La Bonnardi~re, 1986) , however, has enabled evaluation of their biological activities, particularly in the bovine species Babiuk, 1985, 1986; Czarniecki et al., 1986) . Briefly, recombinant bovine IFNy (rBoIFNT) was shown to inhibit viral replication, to exert antiproliferative effects in vitro (Czarniecki et al., 1986) and to modulate neutrophil and lymphocyte functions both in vitro and in vivo Babiuk, 1985, 1986) .

In contrast, very little is presently known about porcine IFNT: supernatants of PHA-stimulated porcine blood mononuclear cells were found to afford antiviral protection to ovine cells challenged with vesicular stomatitis virus (VSV) (Yilma, 1983) . Sauvagnac (1987) described the production of IFNy and the synthesis of IFNT-specific messenger RNA by porcine lymphocytes pretreated by a phorbol ester prior to induction by PHA. There is indirect evidence that IFN7 may be responsible for the stimulatory effects of supernatants from PHAinduced porcine lymphocytes on newborn pig NK activity (Charley and Fradelizi, 1987 ) .

The gene coding for porcine IFN?' was cloned and sequenced by Genentech Inc. (U.S.A.). E. coli-derived rPoIFN7 was shown to contain 166 residues, for a molecular weight of 31.6 kD, as expected for a dimer (unpublished data from Genentech, U.S.A. ).

The present report describes the effects of recombinant porcine IFN7 (rPoIFNT) in vitro on multiplication of VSV in bovine and porcine cells as well as of coronavirus transmissible gastroenteritis virus (TGEV) in porcine kidney cells and pulmonary macrophages. A polyclonal anti-porcine IFN7 antiserum was also used to delineate the antigenic relationships between porcine, bovine and human IFN7 with respect to their biological activity.

Cell lines of bovine (Madin-Darby bovine kidney, MDBK), porcine (PD5 and RPTG pig kidney cells ) and human (Wish amnion cells ) origin were used. All cell lines were cultured in Eagle's Minimum Essential Medium (MEM) supplemented with 10% fetal calf serum (FCS).

Porcine alveolar macrophages were obtained by lung washings of exsanguinated animals (Charley et al., 1983) , kept frozen in liquid nitrogen and thawed before culture.

Recombinant porcine (lot no. 4648-58) and bovine (lot no. 3229-38) IFNy were provided by Ciba Geigy Ltd. (Basel, Switzerland). Specific activities were 5-10 X 106 antiviral units/mg for rPoIFN7 and 2 X 106 units/mg for rBoIFNT.

Recombinant human ( Hu ) IFN7 with a titre of 107 units/mg (code ARN 3010 ) was purchased from Amersham (U.K.).

Porcine IFNa, obtained by in vitro infection of porcine lymphocytes with influenza virus (La , was kindly provided by C. la Bonnardi~re (Thiverval-Grignon, France) . The source of porcine IFNfl was supernatants of TGEV-infected PD5 cells. Natural porcine IFNy was prepared by the method developed by Sauvagnac (1987) : briefly, porcine blood mononuclear cells isolated by Ficoll density centrifugation (Charley et al., 1985) were cultured at a concentration of 5 × 106 cells/ml in RPMI 1640 plus 10% v~ v FCS for 3 h at 37 °C in the presence of 10 ng/ml phorbol-myristate-acetate (no. P8139 from Sigma, St Louis, U.S.A.); PHA-P (no. 3.110.56, Difco, Michigan, U.S.A. ) was then added to a final dilution of 1/700 and the cultures incubated for 20 h at 37°C.

Titrations were performed in microtitre plates (Falcon 3072, Becton Dickinson, Oxnard, U.S.A.). Serial dilutions of IFNs were assayed on monolayers of MDBK, RPTG or Wish cells challenged, 24 h after IFN treatment, with 50 plaque-forming units (PFU) of VSV per well (La Bonnardi~re and Laude, 1981 ) and on PD5 cells challenged with cell-adapted Purdue 115 strain of TGEV (Laude, 1981) at multiplicities of infection shown in the Results. IFN titres were expressed as the maximal dilution giving a total protection of cell monolayers against the viral challenge.

PoIFNy was also assayed on porcine alveolar macrophages infected with TGEV; 3 × 105 cells per well were cultured in microtitre plates in RPMI 1640 plus 20% v/v FCS for 24 h at 37 ° C, treated at 37 ° C for 24 h with fresh medium containing various concentrations of rPoIFNy, and then challenged with TGEV (103 PFU per well in RPMI 1640 plus 5% v/v normal calf serum) for 48 h at 38 ° C. The macrophage monolayers were then stained with neutral red as described before (Laude et al., 1984) : a 10 -4 dilution of neutral red was added for 30 min at 37 ° C, monolayers were then rinsed twice and treated with 100 pl/well of 90% v/v ethanol. Plates were read spectrophotometrically at 449 nm on a Titertek Multiskan ELISA reader (Flow Laboratories). Antiviral activity was expressed as percent protection by IFN, calculated from the ratio: (OD449 of infected wells/OD449 of control wells) × 100.

Sheep anti-human IFNa, goat anti-human IFNy and rabbit anti-bovine IFNy antisera were kindly provided by P. Adamovicz, S. Stefanos (Paris, France) and Ciba Geigy Ltd. (Basel, Switzerland). Rabbit anti-rPoIFN7 antiserum (no. 652) was prepared as follows: the primary immunization was an injection of 16 llg rPoIFN7 in complete Freund's adjuvant into a popliteal lymph node (Sigel et al., 1983) and monthly subcutaneous booster injections of 32/~g rPoIFN7 in incomplete Freund's adjuvant. IFN seroneutralization was assayed as described by La ; dilutions of antiserum were added to the cell monolayers in microtitre plates prior to serial dilutions of IFN, and for each serum dilution a neutralization index (NI) was determined as follows: NI = log3 (IFN titre with antiserum) -log3 (control IFN titre).

The antiviral activity of rPoIFN7 was evaluated in homologous and heterologous (bovine) cell systems infected by either VSV, the model pathogen for IFN assays, or coronavirus TGEV, a major pathogen for pigs. PoIFN7 proved not to be strictly species-specific since it caused a dose-dependent inhibition of VSV-induced cytopathic effects in bovine MDBK cells (Table 1 ). In the same assay system rBoIFN7 showed high antiviral activity (higher than rPoIFNT) whereas rHuIFN7 was inactive (Table 1) . Antiviral activity of rPoIFN7 was also observed on homologous (porcine) RPTG cells challenged with VSV (Table 1) . rPoIFNT, assayed on porcine (PD5) cells challenged by TGEV, was found to have a protective effect against up to 103 PFU/well (Fig.  1) . With TGEV titres greater than 104 PFU/well, no effect of the rPoIFN7 was observed. In contrast, natural porcine IFN~ could protect cells against as high a titre as 105 PFU/well of TGEV (Fig. 1 Since TGEV can also replicate in pulmonary macrophages (Laude et al., 1984 ) , the antiviral effect of rPoIFNy was also assayed in this cell system. Fig.  2 shows that a dose-dependent protective effect was obtained when porcine pulmonary macrophages were incubated with rPoIFN7 prior to TGEV challenge. Antiviral activity was obtained with rPoIFN7 doses as low as 3 ng/ml.

Production of rabbit polyclonal anti-rPoIFNy antisera was achieved by inoculation of purified rPoIFNy preparations into popliteal lymph nodes followed by subcutaneous booster injections. The sera obtained 3 weeks after the booster injection had a neutralizing titre which could not be enhanced by subsequent injections, and remained constant for at least 6 months without reimmunization. Preimmune sera were devoid of significant neutralizing activity.

The antiserum neutralized 98% of rPoIFN7 activity when diluted 100-fold, and more than 99% at a 10-fold dilution. This antiserum was specific for PoIFN7 since it showed little or no neutralizing activity against natural porcine IFN~ and fl (Fig. 3a) . The antiserum neutralized the antiviral activity of both rPoIFNy and that present in supernatants of porcine lymphocytes stimulated by PMA and PHA (Fig. 3a) . This demonstrated that the antiviral activity in the latter was probably due to IFNy.

The antigenic relationship with respect to biological activity, between por- cine, bovine and human rIFNy was investigated by comparing the neutralizing activity of three antisera tested against each IFN preparation. The rabbit anti-rPoIFNy antiserum neutralized both porcine and bovine IFNy with equal efficacy but did not have any such activity against rHuIFN7 (Fig. 3b ). In contrast, anti-human and anti-bovine IFN7 antisera could only neutralize homologous preparations (Table 2) .

Recombinant porcine IFN7 was shown to protect homologous as well as heterologous (bovine) cells against VSV-induced cytopathic effects. Although IFNy is generally considered species-specific, the fact that rPoIFNy was active on MDBK cells clearly indicates that certain cross-reactivities can exist. This would confirm in part, the report of Yilma (1983) , who demonstrated that crude preparations of bovine, caprine, equine and porcine IFNy had antiviral activity on heterologous (ovine choroid plexus) cells. Furthermore, rHuIFNy can act upon porcine lymphocytes by stimulating their natural killing (NK) activity (Charley, unpublished data, 1988) . Cross-species reactivities have also been described for IFNa (La Bonnardi~re and Laude, 1981; Yilma, 1983) and for IFNfl (Czarniecki et al., 1986) . We have also shown that rHuIFNfl, although devoid of antiviral activity on porcine cells, could activate porcine NK (Charley, unpublished results, 1988) .

Nevertheless, it still remained necessary to relate the antiviral activity of rPoIFN7 to the porcine situation. With this in mind, we were able to demonstrate that rPoIFN~ affords significant protection against TGEV challenge in vitro. Coronavirus TGEV induces acute and fatal diarrheas in young piglets (Haelterman, 1972) and replicates in two different cell populations: enterocytes (Haelterman, 1972) and pulmonary macrophages (Laude et al., 1984) . We show that rPoIFN7 is able to protect both epithelial cells and alveolar macrophages from destruction by the virus (Figs. 1 and 2) . This antiviral activity of rPoIFN7 in TGEV-infected porcine epithelial cells is highly dependent upon the multiplicity of infection of the virus, in contrast to the activity of the porcine IFNc~ preparation. The differences observed between IFNa and IFN7 preparations may reflect different modes of action on porcine epithelial cells. In addition, it is probable that the non-recombinant IFNa preparation contained a number of different forms of interferons and it is conceivable that the differences in antiviral activity observed between the IFN7 and IFNc~ preparations may be due to possible synergies between different interferon subtypes within the non-recombinant preparations. Certainly, different interferons appear to have different receptors on cells (for example, see Branca and Baglioni, 1981) and synergies have been seen between human IFN~ and IFNy (Weigent et al., 1983; Seow and Thong, 1986) .

The antigenic relationship of the active sites of porcine IFN~, IFNfl and IFNy, and of rPoIFNy, rBoIFN7 and rHuIFN7 was studied using antisera raised against IFNy of the porcine, bovine and human species. Anti-rPoIFN7 could neutralize the antiviral activity of rPoIFN7 and that present in the supernatants of PHA-activated porcine mononuclear cells, but had no such activity against non-recombinant porcine IFNa or IFNfl preparations. This supports the work done with other animal systems (for example, Branca and Baglioni, 1981 ) in demonstrating that in the porcine species the active site of, and probably the receptor for, IFN7 is different from IFNc~ and IFNfl.

Seroneutralization experiments conducted with anti-rPoIFN7 antiserum indicated that a high degree of antigenic homology exists between the active sites of porcine and bovine IFN7 whereas human IFN7 was immunologically different. Experiments with anti-HuIFN7 antiserum confirmed the absence of antigenic cross-reactivity between human and porcine IFN7. Conversely, antiserum to BoIFNy did not neutralize either HuIFNy or PoIFNT: this suggests the existence of immunodominant cross-reactive epitopes on rPoIFN7 which would induce the production of antibodies, whereas these determinants would not be immunogenic on rBoIFNy. Closer similarity between PoIFN7 and BoIFN7 than with HuIFN7 can also be seen in the degree of the sequence homologies between these three IFN molecules: rPoIFNT/rBoIFNT, 76%; rPoIFNy/rHuIFNy, 59%; rBoIFNy/rHuIFN7, 61% (McCracken, Genentech, U.S.A., personal communication, 1987) . In addition, antigenic cross reactivity of lymphokines from different species has been found with other molecules. PoIFNa (La and porcine ILl (Saktvala et al., 1985) were found to have antigenic homologies with their human counterparts whereas no antigenic relationship was observed between human and porcine IL2 (Charley et al., 1985) .

Following this initial description of several antiviral properties and of the antigenic characterization of rPoIFNT, a better understanding and appreciation of this compound as an immunomodulator in vitro and in immature or immunocompromised animals may now be obtained.

Effect of bovine ~1 interferon on bovine herpes virus type 1 induced respiratory disease

In vitro and systemic effects of recombinant bovine interferons on natural cell-mediated cytotoxicity in healthy and bovine herpes virus 1 infected cattle

Alteration of some leukocyte functions following in vivo and in vitro exposure to recombinant bovine alpha and gamma interferon

Evidence that type I and II interferons have different receptors

Two distinct families of human and bovine interferon a genes are coordinately expressed and encode functional polypeptides

Cloning, sequence and expression of bovine interferon 7

Differential effects of human and porcine interleukin 2 on natural killing (NK) activity of newborn piglets and adult pig lymphocytes

Effects of intravenous injection of BCG or Freund adjuvant on swine alveolar macrophages

Production of porcine interleukin 2 and its biological and antigenic relationships with human interleukin 2

In vitro biological activities of Escherichia coli derived bovine interferon c~, fl and

On the pathogenesis of transmissible gastroenteritis of swine

High interferon titer in newborn pig intestine during experimentally induced viral enteritis

Biological and antigenic relationships between virus induced porcine and human interferons

Thermal inactivation studies of a coronavirus, transmissible gastroenteritis virus

Replication of enteropathogenic coronavirus TGEV in swine alveolar macrophages

Molecular cloning and sequencing of a gene encoding biologically active porcine a interferon

An antiserum to pig ILl (Ca~abolin) reacts with the acidic but not the neutral form of human ILl

Contribution ~ l'~tude de l'interf~ron porcin. Caract~risation de rARN messager de l'interf~ron gamma

Augmentation of human polymorphonuclear leukocytes adherence by interferon

Production of antibodies by inoculation into lymph nodes

Immune IFN: a pleiotropic lymphokine with multiple effect

Potentiation of lymphocyte natural killing by mixture of alpha or beta interferon with recombinant gamma interferon

Sensitivity of ovine choroid plexus cells to human and other animals IFNs

We gratefully acknowledge the expert assistance of L. Lavenant and C. de Vaureix. We also thank C. la Bonnardi~re for helpful discussions and for providing IFNa.