key: cord-260109-vgbrwegt
authors: Charley, B.; Lavenant, L.
title: Characterization of blood mononuclear cells producing IFNα following induction by coronavirus-infected cells (porcine transmissible gastroenteritis virus)
date: 1990-02-28
journal: Research in Immunology
DOI: 10.1016/0923-2494(90)90133-j
sha: 
doc_id: 260109
cord_uid: vgbrwegt

Abstract Porcine blood mononuclear cells (PBMC) were shown to produce interferon-α (IFNα) following incubation with cells infected by a coronavirus, transmissible gastroenteritis virus. Monoclonal antibodies (mAb) with specificities for leukocyte subsets and major histocompatibility complex (MHC) antigens were used to characterize IFNα producer cells. The production of IFNα was found to be a function of non-phagocytic, non-adherent, non-T, non-B, CD4+ (and to a lesser extent CD8+) MHC-class-II-positive cells. Furthermore, addition of anti-MHC (class II) mAb during PBMC incubation with virus-infected cells reduced IFN yields, suggesting that masking of these surface antigens alters PBMC responsiveness to IFN induction.

INTRODUCTION Peripheral blood leukocytes have been shown to produce interferon alpha (IFNa) upon induction by viruses, bacterial products or tumour cells (Dianzani and Capobianchi, 1987) . In contrast to IFN beta, for which the critical inducing factor appears to be viral RNA, a different mechanism seems to be involved in induction of IFNa. Thus, inactivated virus particles, virus-infected cells, mycoplasmas and cell membranes are capable of inducing IFNa (Lebon et ai., 1982; Hughes and Blalock, 1983; Goblet al., 1988; Capobianchi et al.., 1987) , which suggests that membrane interactions between the inducer structure and the leukocyte membrane is sufficient to trigger IFNa produc-tion. In the case of transmissible gastroenteritis virus (TGEV), a coronavirus which induces acute diarrhoea and intense IFN synthesis in newborn piglets (La Bonnardiere and Laude, 1981) , we have previously shown that early IFN~ production could result from exposure of non-immune pig lymphocytes to virus-infected cells. Moreover, experiments in which IFN0c production was inhibited by monoclonal antibodies (mAb) directed at some epitopes of the TGEV glycoprotein E1 suggested that a defined domain of this transmembrane viral protein played a key role in the IFN0t induction process (Charley and Laurie, 1988) .

The nature of IFN~ producer cells (IPC) is not clear, since several cell types appear to be involved depending upon the inducer or the induction protocol (Dianzani and Capobianchi, 1987) . Thus, among non-adherent human mononuclear cells, "nuli" (non-T, non-B) lymphocytes and large granular lymphocytes (LGL) were shown to produce IFN0c upon exposure to herpes, influenza or dengue virus (Peter et ai., 1980; Kirchner et al., 1979; Lebon et al., 1982; Djeu et al., 1982; Kurane et al., 1986) . More recent reports have indicated that a common phenotypic feature for murine and human IPC in response to different stimuli is their surface expression of MHC class II molecules (Abb et al., 1983 ; Reiss et al., 1984; Perussia et al., 1985 ; Kurane and Ennis, 1987; Hughes and Blalock, 1983 ; Capobianchi et al., 1987; Oh et al., 1987; Fitzgerald-Bocarsky et al., 1988; Sandberg et al., 1989a) .

In the present report, we analysed the nature of lymphocyte subpopulation(s) responsive to IFNa induction by TGEV-infected cell monolayers by using cell-depletion experiments with mAb plus complement. In addition, IFN induction-blocking experiments conducted with anti-MHC (SLA, swine leukocyte antigens) class II mAb without complement suggest a functional role for these membrane molecules in the IFN induction process.

Porcine peripheral blood mononuclear cells (PBMC) were obtained from heparinized blood collected from 2-to 4-month old animals. Phagocytic cells were depleted by carbonyl iron ingestion (Salmon, 1979) before PBMC isolation by Ficoll density centrifugation on MSL (density 1.077, Eurobio, Paris). PBMC were suspended in RPMI-1640 medium supplemented with 10 °7o foetal calf serum. IFN = interferon. IPC = IFNQt producel cell.

LGL = large granular lymphocyte. mAb = monocional antibody.

= major histocompatibility complex. PBMC = peripheral blood mononuclear cell, SLA = swine leukocyte antigen. TGEV = transmissible gastroenteritis virus.

Anti-PBMC _.'nAb MSA4 (anti-CD2), 76-7-4 (anti-B), 74-12-4 (anti-CD4) were kindly provided by J. Lunney (USDA, Beltsville, MD, USA); 295/33 (anti-CD8) mAb was kindly provided by U. Koszinowski (Tiibingen, FRG). mAb 2-2-13 (anti-SLA class I) and MSA3 (anti-SLA class II) were provided by J. Lunney. mAb TH22A5 (anti-SLA class II) was purchased from VMRD (Pullman, WA, USA). Other anti-SLA class II mAb (TH14B, H42A and TH81A) were kindly provided by W. Davis (Pullman, WA, USA) (table I). These antibodies were used for cell-depletion experiments as ascitic fluids, excepts for MSA3 which was used as hybridoma cell supernatant.

PBMC were incubated for 30 min with mAb and complement at 37°C as described previously (Charley et al., 1987) : briefly, one-month old rabbit serum served as a source of complement at a final dilution of 1/9 and mAb were used at a final dilution of 1/300. The percentage of dead cells was determined by trypan blue dye exclusion and ceils were resuspended in the initial volume, i.e. without readjustment of the viable cell concentration, before being used in the assays.

PBMC were induced to produce IFN0t by overnight incubation on TGEV-induced, glutaraldehyde-fixed cell monolayers as described previously (Charley and Laude, 1988) : briefly, pig kidney cells were plated in 96-well microplates, infected by the coronavirus TGEV for 18 h, then fixed with 0.25 % glutaraldehyde (1 h at 4°C) and stored with 3 % glycine. Monolayers were washed before addition of PBMC (100 Izl/well at 5 x 106/ml). Supernatants were collected after 18 h of incubation at 37°C and assayed for IFN activity.

Various dilutions of dialysed mAb preparations, as indicated in "Results", were added during overnight incubation of PBMC with TGEV-infected monolayers or with virus particles. Alternatively, PBMC were pretreated with various amounts of dialysed mAb Hammerberg and Schurig, 1986 Pescovitz et aL, 1984 Jonjic and Koszinowski, 1984 Pescovitz et aL, 1984 Davis et al., 1984 Hammerberg and Schurig, 1986 VMRD, Inc., Pullman WA Davis et ai., 1984 Davis et aL, 1984 Davis et aL, 1984 mAb preparations for 60 min at 37°C, washed to remove unbound material, resuspended at 5 × 106/ml and incubated overnight on T~EV-infected monolayers. Supernatants were assayed for IFN.

Log 3 dilutions of PBMC supernatants were assayed for IFN on bovine MDBK cells using vesicular stomatitis virus as a challenge (La Bonnardi6re and Laude, 1981) . A standard porcine IFNa was included in each assay. This standard was calibrated on MDBK ce!ls with the human international reference IFN B69/19 (NIH, Bethesda, MD, USA). In our results, 1 U is equivalent to IlU of human IFN.

We have previously shown that phagocyte-depleted porcine PBMC could secrete IFN0t following incubation on TGEV-infected glutaraldehyde-fixed cell monolayers (Charley and Laude, 1988) . Antibody plus complement depletion experiments were conducted to characterize the IPC nature. When PBMC were pretreated with rabbit serum as a source of complement, they produced high amounts of IFN (4 800 +_ 3 000 U/ml in 10 different experiments). Table II shows the effect of treatment with various anti-lymphocyte mAb and complement on IFN production. Since IFN assays are performed on log 3 dilutions of PBMC supernatants, any reduction lower than 33 % was considered as negligible. In fact, pretreatment of PBMC with anti-T or anti-B cell mAb plus complement did not alter IFN0~ production" IFN titres obtained were ~c~p~t~vc~y equa~ to 45 and 71 ~o oI-"~ titres obtained With complement-treated PBMC (table II) . In contrast, pretreatment with 245/33 (anti-CD8) mAb and, No. = number of experiments. Viable cells = 100 °7o -(% dead cells after effect of mAb + complement-% dead cells with complement alone).

IFN production is expressed as % of IFN produced by complement-pretreated PBMC. Table III shows that pretreatment by anti-SLA class I mAb and complement, which destroyed almost all PBMC, completely abolished IFN production. Pretreatment by anti-SLA class II mAb TH22A5 plus complement markedly lowered IFN production, along with the lysis of 14 o70 PBMC. These experiments therefore indicated that porcine IPC were largely SLA-class-II-positive cells. at 37°C), washed and incubated overnight on TGEV-infected glutaraldehyde-fixed cell monolayers. IFN activity was assayed in cell supernatants.

During cell-depletion experiments, controls performed in the absence of complement indicated that most mAb used had no direct inhibitory effects on IFN production. However, two preparations of anti-SLA class II mAb (TH22A5 and MSA3) could directly block IFN induction when added during co-incubation of PBMC with i'GEV-infected fixed monolayers or TGE alone (5 independent experiments). Thus, rII22A5 mAb used as an ascitic fluid could hLhibit up to 99.9 07o IFN production at a final Ig concentration of 12 ,.6/ml ( fig. la) . MSA3 hybridoma culture supernata,~t blocked up to 99 070 IFN production at a final Ig concentration of 0.25 [~g/ml ( fig. Ib) . Three other ascitic fluids, directed at SLA class II, also showed dose-dependent iaS~hition of IFN production ( fig. 2 ). In addition, the latter blocking experiment was achieved by pretreatment of PBMC with mAb, followed by washing of cells before induction on TGEV-infected fixed-cell monolayers, which argues for a direct effect of mAb on PBMC as opposed to an antibody effect on infected cell monolayers. In addition, such inhibiting effects could not be related to toxic effects of mAb on PBMC (not shown). IFN induction by other viruses such as Newcastle disease virus (NDV), Sendai virus and types A and B myxoviruses was also blocked by co-incubation of PBMC with TH22A5 mAb (not shown).

We have previously shown that phagocyte-depleted PBMC incubated with TGEV-infected cells are rapidly induced to secrete IFNa (Charley and Laude, 1988) . This IFN~ induction seems to result from membrane interactions between PBMC and a defined domain of the viral glycoprotein E1 expressed on the surface of infected cells (Charley and Laude, 1988 ). In the present paper, cell-depletion experiments by mAb and complement-dependent lysis indicate that porcine IFNa producer cells (IPC) are mostly MSA4-(CD2 or pan-T)-and 76-7-4-(B) cells, whereas depletion of CD4 + cells, and to a lesser extent CD8 ÷ cells reduced IFN yields by 95 and 82 °70, respectively (table II) . In addition, IPC are also SLA class I ÷ and class II ÷ cells. Furthermore, the addition of anti-SLA class II mAb reduces IFN~t production.

Our present data on the nature of porcine IPC are in agreement with several reports about the characterization of human IPC. Thus, following induction by different viruses including herpes virus, influenza virus, dengue virus or cytomegalovirus, as well as by mycoplasma membranes, human mononuclear leukocytes producing IFN~ were described as non-adherent, non-phagocytic, non-T,non-B cells (Trinchieri et aL, 1978; Kirchner et aL, 1979; Peter et aL, 1980; Lebon et aL, 1982; Abb et aL, 1983; Djeu et aL, 1982; Perussia et aL, 1985; Kurane et aL. 1986 ). Human IPC were generally shown to lack natural killer function (Lebon et aL, 1982; Abb et al., 1983; Fitzgerald-Bocarsly et al., 1988) , but to express MHC class II antigens (Abb et al., 1983 ; Perussia et al., 1985; Capobianchi et al., 1987; Oh et al., 1987; Fitzgerald-Bocarsly et aL, 1988) . A recent report using combined immunocytochemistry ,~o.,u,,a,~t ,,t on numan rt~Mt; stimulated by HSVinfected cells clearly showed that IPC lacked antigens typical of T and B lymphocytes, but expressed HLA-class II antigens (Sandberg et al., 1989a) . The same laboratory observed that IPC also expressed CD4 antigens (Sandberg et aL, 1989b) . The expression of MHC class II antigens led to the hypothesis that human IPC could be dendritic cells (Fitzgerald-Bocarsly et aL, 1988) . However, HLA-DR + cells could recently be divided into two separate subsets: a loosely adherent population meeting the functional criteria of dendritic cells but distinct from IPC which were non-adherent (Chehimi et ai., 1989) . Our data therefore indicate that porcine IFNa-producing cells in response to TGEV-infected cells have the same properties as human IPC: porcine IPC are non-phagocytic, non-adherent (Salmon et al., 1989) , non-T,non-B, MltC-clas~-II-positive and CD4 + cells. Preliminary experiments using DNA-RNA in situ hybridization suggested that porcine IFNa-mRNAcontaining cells were infrequent (around 1/104 PBMC; Buseyne and Charley, unpublished observations) as already described for human IPC (Goblet al., 1988) . Interestingly, one might hypothesize that a defined, albeit infrequent cell population exhibiting unusual phenotypic features could produce IFNa in response to a wide range of IFN inducers. In order to further analyse these cells and their mode of activation, it will be necessary to use positive selection procedures such as cell sorting (Sandberg et al., 1989b) .

Blocking experiments conducted with anti-SLA-class II MSA3 mAb used as hybridoma culture supernatant revealed a reduction in IFN yield of more than 95 °70 ( fig. 1 ). Comparable inhibition was obtained with 4 other anti-SLA class II mAb used as ascitic fluids. This inhibition is not observed when virus-infected cells are pretreated with mAb, then washed before PBMC are added. However, when PBMC are pretreated with mAb, then washed before induction, IFN yield is still reduced, which argues for a direct effect of anti-SLA class II antibodies on PBMC. In addition, PBMC are not lysed by such treatments. Therefore, these results suggest that masking of MHC class II antigens on the PBMC membrane reduces their responsiveness to IFN induction. This anti-SLA class-II-Ab-mediated blockage is observed when PBMC are induced by TGEV, Sendai, NDV or influenza virus. A similar observation was reported by Capobianchi et al. (1987) : pretreatment of human PBMC by anti-HLA-DR antibody reduced IFN0t yield after induction with mycoplasma membranes. Similarily, Ia antigens were shown to bind cell surface glycoproteins responsible for IFN induction (Hughes et al., 1986) . Therefore, our findings suggest that, in addition to mycoplasmas and cell surface glycoproteins, viruses could also interact with MHC class II antigens to induce IFN0t synthesis.

The fact that masking of MHC class II could reduce PBMC responsiveness to several IFN inducers suggests that these surface antigens are not virusspecific receptors on lymphoid cells. In fact, anti-SLA class II mAb did not block TGEV replication in susceptible pig kidney cells (data not shown). The actual functional role of MHC class II molecules in the IFN0t induction process remains to be clarified" they could represent broadly reactive recogni-t~u. ~t, m.tu~e~ tto~c to omd utllcrcnt IFN-inducing components (as suggested by Hughes et al., 1986) . Alternatively, MHC class II molecules could be involved in post-recognition events, such as internalization or recycling of membrane structures, which seem to precede activation of IFN0c synthesis (Lebon, 1985) . Experiments are in progress to further define the nature of virus lymphocyte interactions leading to IFN0t production. Des cellules sanguines mononucl6es du porc produisent de l'interf6ron cx (IFN~x) h la suite de leur incubation avec des cellules infect6es par le coronavirus GET (gas-troent6rite transmissible). Les cellules productrices d'interf6ron ont 6t6 caract~ris6es l'aide d'anticnrps monoclonaux (AcM) sp~cifiques des sous-populations leucocytaires et des antigbnes du complexe majeur d'histocompatibilit6 (CMH). Les cellules productrices d'lFN0t sont des cellules non phagocytaires, non adh&entes, ni T, ni B, CD4÷ (pour une moindre part CD8+) et CMH-classe-II÷. De plus, l'addition d'AcM dirig6s contre les antig6nes de classe II du CMH, au m61ange d'incubation cellules mononucl6es plus cellules infect6es par le virus GET, r6duit la production d'IFN~, ce qui sugg6re que le masquage de ces antig6nes de surface modifie la capa-cit6 des cellules mononucl6es/l r6pondre aux signaux inducteurs d'IFN0t.

MOTS-CLI~S" lnterf6ron alpha, Lymphocyte, CMH, Coronavirus; Anticorps monoclonaux, Virus de la Gastroent6rite transmissible du pore.

Phenotype of human alpha interferonproducing leucocytes identified by monoclonal antibodies

Membrane interactions involved in the induction of interferon-alpha by Mycoplasma pneumoniae

Inhibition of transmissible gastroenteritis coronavirus (TGEV) multiplication in vitro by non-immune lymphocytes

Induction of alpha interferon by transmissible gastroenteritis

Dendritic cells and IFN-g-producing cells are two functionally distinct non-B,non-monocytic HLA-DR + cell subsets in human peripheral blood

The development and analysis of species-specific and cross-reactive monoclonal antibodies to leukocyte differentiation antigens and antigens of the major histocompatibility complex for use in the study of the immune system in cattle and other species

Mechanism of induction of alpha interferon

Positive self regulation of cytotoxicity in human natural killer cells by production of interferon upon exposure to influenza and herpes viruses

Human mononuclear cells which produce interferon-alpha during NK (HSV-FS) assays are HLA-DR-positive cells distinct from cytolytic naturel killer effectors

Different induction patterns of mRNA for IFN-a and -B in human mononuclear leukocytes after in vitro stimulation with herpes simplex virus-infected fibroblasts and Sendai virus

Characterization of monoclonal antibodies directed against swine leukocytes

Ia antigen: a murine B lymphocyte receptor for transformed cell induction of interferon-0t/B

Interferon-inducing transformed cell surface glycoproteins: purification by Ia antigen affinity chromatography

Monoclonal antibodies reactive with swine lymphocytes.-l. Antibodies to membrane structures that define the cytolytic T lymphocyte subset in the swine

Studies of the producer cell of interferon in human lymphocyte cultures

Induction of interferon-a from human lymphocytes by autologous, Dengue virus-infected monocytes

Induction of interferon alpha and gamma from human lymphocytes by Dengue virus-infected cells

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

Inhibition of Herpes Simplex virus type-l-induced interferon synthesis by monoclonal antibodies against viral glycoprotein D and by lysosomotropic drugs

Human lymphocytes involved in ~-interferon production can be identified by monoclonal antibodies directed ,o

Cooperation between CD16 (Leu-1 lb) + NK ceils and HLA-DR + cells in natural killing of herpesvirus-infected fibroblasts

A leukocyte subset bearing HLA-DR antigens is responsible for in vitro interf,~ron production to viruses

Preparation and characterization of monoclonal antibodies reactive with porcine PBL

Human peripheral null lymphocytes.-II. Producers of type 1 interferon upon stimulation with tumor cells, herpes simplex virus and Corynebacterium parvum

Interferon production by cultured murine splenocytes in response to influenza virus-infected cells

Surface markers of porcine lymphocytes and distribution in various lymphoi'd organs

atural killer (Nk) activity and interferon (IFN) production by a fraction of spleen and blood lymphocytes in swine

PORCINE CELLS i5i

Characterization of the blood mononuclear leucocytes producing alpha interferon after stimulation with herpes simplex virus in vitro, by means of combined immunohistochemical staining and in situ RNA-RNA hybridization

CD4 antigen are expressed on the interferon-0t-producing cells induced by herpes simplex-infected cells

Antiviral activity induced by culturing lymphocytes with tumor-derived or virus-transformed cells. Identification of the antiviral activity as interferon and characterization of the human effector lymphocyte subpopulation

We are grateful to Dr. J. Lunney (Beltsville, MD, USA), Dr. U. Koszinowski (Tiibingen, FRG) and Dr. W. Davis (Pullman, WA, USA), who kindly provided antiporcine lymphocyte subsets and anti-SLA monoclonal antibodies. We also thank Dr. F. Blecha (Manhattan, KS, USA) for revising the manuscript.