ORIGINAL RESEARCH published: 02 June 2021 doi: 10.3389/fimmu.2021.674185 Interaction of Type 1 Porcine Reproductive and Respiratory Syndrome Virus With In Vitro Derived Conventional Dendritic Cells Yanli Li 1* and Enric Mateu 1,2 1Departament de Sanitat i Anatomia Animals, Facultat de Veterinària, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain, 2 IRTA, Centre de Recerca en Sanitat Animal (CReSA, IRTA-UAB), Bellaterra, Spain The present study delineates the interaction of a typical PRRSV1.1 isolate 3267 (moderate virulence) with in vitro derived pig conventional dendritic cells, cDC1, cDC2, and a CD14+ population (designated as CD14+ DCs). cDC1 and cDC2 were not susceptible to 3267 Edited by: infection, but a fraction of CD14+ DCs were infected. After exposure to the virus, all three Leon Grayfer, DC types remained immature as determined by no increase of maturation molecules George Washington University, (MHC-I, MHC-II, CD80/86, CCR7), no release of cytokines, no modification of antigen United States presentation abilities, and no alteration of endocytic/phagocytic capabilities. However, Reviewed by: Ayako Miyazaki, when infected MARC-145 cells were used as a source of viral antigens, cDC2 and CD14+ National Agriculture and Food DCs showed a significant increase in the expression of maturation molecules and Research Organization, Japan Viskam Wijewardana, substantial release of cytokines, notably IL-12/IL-23p40 (by both DC types) and IL-10 International Atomic Energy Agency, (by CD14+ DCs). To address the impact of PRRSV1 3267 on TLR3- and TLR7-mediated Austria activation, cDC1, cDC2, and CD14+ DCs were inoculated by the virus (live or UV- *Correspondence: inactivated) for 6 h prior to or simultaneously with the addition of poly I:C (TLR3 ligand) Yanli Li yanli.li@uab.cat or gardiquimod (TLR7 ligand; not used for cDC1). Compared with using TLR ligand alone, combination with the virus did not result in any alteration to the maturation markers on all Specialty section: DC types but changed the cytokine response to either TLR3 or TLR7 ligand. Pre-exposure This article was submitted to Comparative Immunology, of cDC2 or CD14+ DCs to the live virus resulted in an increased production of IFN-a upon a section of the journal poly I:C stimulation, while pre-exposure to UV-inactivated virus tended to enhance the Frontiers in Immunology release of IL-10 upon gardiquimod stimulation. Simultaneous addition of the live virus and Received: 28 February 2021 Accepted: 12 May 2021 the TLR ligand either had no effect (mainly in cDC2) or impaired most of the cytokine Published: 02 June 2021 release after gardiquimod stimulation (in CD14+ DCs). When used as antigen presenting Citation: cells, cDC2 pre-inoculated by the live virus before addition of gardiquimod impaired the Li Y and Mateu E (2021) proliferation of CD4–CD8– T cells. In the case of CD14+ DCs, pre-exposure to the live virus Interaction of Type 1 Porcine Reproductive and Respiratory or simultaneously added with TLR3 or TLR7 ligand largely decreased the proliferation of Syndrome Virus With In Vitro Derived CD4–CD8+ and CD4–CD8+ T-cell subsets. For cDC1, no significant changes were Conventional Dendritic Cells. Front. Immunol. 12:674185. observed in cytokine responses or T-cell proliferation after poly I:C stimulation. Of note, doi: 10.3389/fimmu.2021.674185 cDC1 had a short life during in vitro culturing, for which the results obtained might be biased. Frontiers in Immunology | www.frontiersin.org 1 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs Overall, exposure to PRRSV1 did not induce maturation of cDC1, cDC2, or CD14+ DCs, but modified TLR3 and TLR7-associated responses (except for cDC1), which may affect the development of adaptive immunity during PRRSV1 infection. Moreover, the sensing of infected cells was different from that of the free virus. Keywords: PRRSV1, conventional dendritic cells, TLR3, TLR7, innate immune response INTRODUCTION PRRSV1, and that the highly virulent PRRSV1.3 Lena induced a higher Th1 polarization compared to PRRSV1.1 strains. Porcine reproductive and respiratory syndrome viruses (PRRSV1 Although the ex vivo approach is highly representative of and PRRSV2) are positive-stranded RNA viruses, classified in the what happens in the infected animals, the scare number of DCs Family Arteriviridae, Order Nidovirales. Since the emergence in obtained from lymphoid tissues, blood, or lungs makes it 1980s, they spread rapidly to most pig-producing countries, difficult to proceed with in-depth functional studies. For that, causing significant economic losses to the swine industry. The a reliable alternative model for in vitro studies is demanded. The infection produces respiratory disease in piglets and reproductive Flt3 ligand (Flt3L)-stimulated in vitro derivation model was failure in pregnant sows; it could also produce asymptomatic recently validated for generating in vivo equivalent cDC1 and infections in animals with some level of previous immunity or cDC2 as well as an unclassified CD14+ population from bone when the infecting strain is of low virulence (1–4). For both marrow hematopoietic cells (26). In the present study, we used PRRSV1 and 2, infected animals are able to clear the virus and this model to explore the impact of a typical PRRSV1.1 strain on develop a solid homologous immunity, although this happens maturation, cytokine production, and antigen presentation after a long viremic period (weeks to months) and several weeks abilities of the three generated DC populations, called cDC1, more with the virus detected in lymphoid tissues (5). cDC2, and CD14+ DCs for the present study. The use of The immune response of pigs to PRRSV is unusual with a infected cells as a source of antigens and interference of delayed development of neutralizing antibodies (NAb) and an PRRSV1 with TLR3- and TLR7-mediated cDC activation irregular cell-mediated response during the first weeks of were also studied. infection (6–8). For more complexities, PRRSV viremia was found in the presence of neutralizing antibodies (9); moreover, it could be resolved before the development of neutralizing antibodies (10). Such features suggest that PRRSV may deceive the immune system. Interaction with dendritic cells (DCs) could MATERIALS AND METHODS be one of the mechanisms of deception. DCs are specialized in sampling antigens and launching Obtention of Bone Marrow Hematopoietic immune response by conveying differentiation signals to naive Cells and Alveolar Macrophages, and CD4+ T cells (11–13); they thus are presumed to be the central Isolation of Peripheral Blood players in orchestrating adaptive immune responses during viral Mononuclear Cells infection. Before identification and characterization of bona fide Bone marrow hematopoietic cells (BMHCs) were aseptically DCs in pigs, bone marrow- (BMDCs) and monocyte-derived obtained from the femora and humeri of 4-week-old pigs. dendritic cells (moDCs) were widely employed to study the Briefly, bones were cut into approximately 1 cm3 pieces and interaction of PRRSV with DCs. Although with controversies, agitated in PBS at room temperature (RT) for at least 1 h. The most of the studies showed that the virus may dysregulate DCs resulting cell suspension was filtered through a 70 mm strainer by preventing DC maturation and the production of cytokines before being depleted of erythrocytes by lysis with 0.15 M NH4Cl (14–20). Also, some reports indicated PRRSV-exposed DCs may solution. Finally, cells were filtered through a 40 mm strainer, induce regulatory T cells (Tregs) (21). But it must be noted that washed, and resuspended in a freezing medium (90% FBS, 10% BMDCs and moDCs represent models of inflammatory DCs dimethyl sulfoxide [DMSO]) (Sigma-Aldrich, Spain). Cells were (22); thereby, the results obtained would be biased from what frozen in liquid nitrogen until used. may happen in the in vivo environment. Alveolar macrophages (AMs) were collected by broncho- With the access to conventional (c) cDC1, cDC2, and alveolar lavage. In brief, lungs were removed from euthanized plasmacytoid DCs (pDCs), the interaction of bona fide DCs pigs and then were filled by pouring phosphate-buffered saline with PRRSV can be better studied. pDCs isolated from blood (PBS) into the trachea. After a gentle massage, the bronchoalveolar were used to determine type I IFN responses to PRRSV. Little or lavage was collected and centrifuged. AM pellets were washed with no inhibition of type I IFN responses were detected except for PBS and finally frozen as described for BMHCs. highly virulent PRRSV2 strains (23). Recently, the study of Nazki PBMCs were isolated from the blood of 12-week-old pigs by et al. (24) indicated that respiratory cDCs were recruited at the density gradient centrifugation using Histopaque 1.077. Cells peak of PRRSV viremia. Bordet et al. (25) using ex vivo lung were frozen using CryoStor cell cryopreservation media (Sigma- cDCs demonstrated cDC1, cDC2 were not susceptible to Aldrich, Spain) in liquid nitrogen. Frontiers in Immunology | www.frontiersin.org 2 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs Donor animals were euthanized using approved methods Virus inactivation was performed by exposing the viral according to Spanish (Royal Decree 53/2013) and European Union suspensions to a UV-254 source at a dose of 200 mJ/cm2. regulations (Directive 2010/63). Animals were sedated before Inactivation was validated by titration of the UV-treated virus euthanasia with a pentobarbital overdose. All donor pigs were on AMs or MARC-145 cells according to the system where the negative of PRRSV, porcine circovirus type 2 (PCV2), Mycoplasma virus was initially propagated. hyopneumoniae, torque teno sus virus (TTSuV) 1 and 2, and influenza A virus as determined by Real-Time quantitative PCR (RT-qPCR). Generation and Sorting of DCs Batches of BMHCs, AMs, and PBMCs used had viabilities > 90% DCs were produced using Flt3L as described previously (26). after thawing as assessed by trypan blue (Sigma-Aldrich, Spain) Briefly, BMHCs were seeded in the 24-well plates (non-treated; staining. For PBMCs, the capability of T-cell proliferation was also Corning, Spain) at a density of 1 × 106 cells/600 µl in RPMI 1640 verified. Cells from four animals were used. medium (Sigma-Aldrich, Spain) containing 20 ng/ml recombinant human Flt3L (rhuFlt3L, Fisher Scientific, Spain), 10% of fetal calf serum (FCS), 2mM glutamine, 20mM HEPES, Virus Production and Titration 100 units/ml penicillin, and 100 mg/ml streptomycin. Cells were PRRSV1.1 isolate 3267 (Genbank accession n° JF276435) was cultured for 14 days, with half of the medium replaced every used in the present study. This strain has been extensively used three days. previously (23, 27–30). Strain 3267 was propagated in AMs and DCs were stained for CADM1/MHC-II/CD172a/CD14 as used as a sixth passage (used in all experiments except reported before (26). Fluorescence minus one (FMO) controls, otherwise indicated). There were not cytokines (IFN-a, IFN-g, were used for gating analysis. Primary antibodies (conjugated or IL-4, IL-6, IL-12/IL-23p40, IL-10, IL-1b, TNF-a) detected in not) and their working dilutions are listed in Table 1. cDC1 3267 produced in AMs. The titre reached 7.2 log10 TCID50/ml, (CADM1+CD14–MHC-IIhiCD172a–/lo), cDC2 (CADM1+CD14– whereby only a low volume was needed for a specific MHC-IIhiCD172a+), and CD14+ DCs (CADM1+CD14+MHC multiplicity of infection (MOI) in the following experiments. IIhiCD172a+) were sorted on the BD FACSJazz sorter (BD In parallel, strain 3267 was adapted to MARC-145 cells and Biosciences, Oxford, UK). The other flow cytometric analyses used as a third passage (named MARC3267). Titration of the were acquired on the MACSQuant Analyzer 10 (Miltenyi Biotec, produced viruses was performed on AMs or MARC-145 Bergisch Gladbach, Germany). Data were examined using the according to the system where the virus was initially FCS Express 7 software (de novo Software, Glendale, CA, propagated. Titres were calculated using the Reed-Muench United States). method (31). The endpoint of infection in cell cultures was identified by immunofluorescent staining of PRRSV1 PRRSV Infection nucleocapsid (N) with a specific antibody (clone 1C5H; In the first experiment, unsorted DCs (500,000 cells) were Ingenasa, Spain). inoculated by PRRSV1 3267 at MOI 0.1 for 1.5 h (5% CO2, 37°C) TABLE 1 | Antibodies used for flow cytometry analysis. Antibody Clone Isotype Species produced Working dilution Supplier Anti-human CADM1 3E1 IgY Chicken 1/1000 MBL Anti-CD14 MIL2 IgG2b Mouse 1/100 Bio-Rad FITC conjugated Anti-CD172a BL1H7 IgG1 Mouse 1/500 Bio-Rad Anti-SLA II DR 2E9/13 IgG2b Mouse 1/500 Bio-Rad Anti-CD11R1 MIL4 IgG1 Mouse 1/500 Bio-Rad Anti-CD11R3 2F4/11 IgG1 Mouse 1/500 Bio-Rad Anti-CD1 76-7-4 IgG2a Mouse 1/50 Bio-Rad Anti-CD163 2A10/11 IgG1 Mouse 1/250 Bio-Rad Anti-CD3 PPT3 IgG1 Mouse 1/200 Bio-Rad Anti-CD25 K231.3B2 IgG1 Mouse 1/80 Bio-Rad Anti-CD4 74-12-4 IgG2b Mouse 1/200 BD Pharmingen FITC conjugated Anti-CD8a 76-2-11 IgG2a Mouse 1/100 BD Pharmingen Alexa Fluor 647 conjugated Anti-human CCR7 (CD197) PE-Cy7 conjugated 3D12 IgG2a Rat 1/20 BD Pharmingen Anti-DEC-205 9HZF7 IgG1 Mouse 1/1 Provided by Lab. de Inmunologıá , (hybridoma supernatant) CIAD, A.C. Anti-MHC-I 4B7/8 IgG2a Mouse 1/10 Gift from Dr. J. Domıń guez* (hybridoma supernatant) human CD152 (CTLA-4)-muIg fusion protein – IgG2a Mouse 1/50 Ancell Anti-mouse/rat FoxP3 FJK-16s IgG2a Rat 1/40 eBioscience Antibodies without specific indications are anti-pig antibodies. *Department of Biotechnology, INIA, Madrid, Spain. Frontiers in Immunology | www.frontiersin.org 3 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs in the 15 ml tubes (cap loosed). Unbound virus was washed CD80/86, and CCR7 as described above. Cell culture away, and a fresh RPMI 1640 medium was added (containing supernatants were frozen at –80°C until cytokine assessment. 10% FCS, now onwards referred to as complete RPMI). Cells To establish the dose-response curve, an increasing proportion of were dispensed in the flat-bottomed 96-well plates with a final infected cells was generated by infecting MARC-145 cells with volume of 200 ml. At 48 h post-inoculation (hpi), cell cultures different doses of MARC3267. Infection was quantified by flow were collected and subject to a three-color staining for PRRSV1, cytometry examination of PRRSV1 N. CD14, and a third molecule (MHC-II, DEC205, CD163, To examine the role of apoptotic cells in DC activation, MARC- CD172a, CD11R1, or CD11R3). Briefly, cells were incubated 145 cells were exposed to UV radiation (312 wavelength) and then with primary antibodies of the third molecules, followed by cultured for 4 h at 37°C 5% CO2 to develop apoptosis. A radiation Alexa Fluor 647-conjugated secondary antibodies (anti-mouse dose (dose 300 mJ/cm2) that generated an apoptosis profile IgG1 or IgG2a). Then, the anti-CD14-FITC antibody was added. resembling 3267-infected MARC-145 cells (MOI 0.01, 48 hpi) PRRSV1 was detected by using an RPE conjugated (Bio-Rad, were used to co-culture with cDC2 or CD14+ DCs. The apoptosis Spain) PRRSV1 N specific antibody (clone 1C5H; Ingenasa, was assessed by Annexin V and propidium iodide (PI) staining. Spain) after cells were fixed (4% paraformaldehyde) and DCs exposed to plain medium, mock-infected MARC-145 cells, or permeabilized (0.3% saponin). supernatants from infected or mock-infected MARC-145 cells were In the second experiment, sorted cDC1, cDC2, or CD14+ DCs used as negative controls. DCs exposed to poly I:C (10 mg/ml) were (50,000 cells) were infected as described above. At 48 hpi, used as the positive control. At 24 h, the expression of MHC-I, supernatants were collected and titrated in AMs (as described MHC-II, CD80/86, and CCR7, and the antigen presentation ability above). Cells were stained for PRRSV1 N protein as described (as described below in “Mixed Lymphocyte Reaction (MLR) Assay”) above and examined on a flow cytometer. were determined. Stimulation of cDC1, cDC2, and Mixed Lymphocyte Reaction Assay CD14+ DCs cDC1, cDC2, and CD14+ DCs that have been exposed to + Sorted cDC1, cDC2, and CD14 DCs were dispensed in the different stimuli were washed to remove the remaining virus, round-bottomed 96-well plates at a density of 15,000 cells/well poly I:C, or gardiquimod. Then they were 1: 5 mixed with for cDC1 and 30,000 cells/well for cDC2 and CD14+ DCs. Then, allogeneic PBMCs that were labeled with CellTrace Violet. cells were inoculated with PRRSV1 3267 at an MOI of 10 or After 5 days of incubation, cells were harvested and stained for incubated with complete RPMI for 6 h. After that, poly I:C (10 CD3, CD4, and CD8a. mg/ml, to all DC types) or gardiquimod (10 mg/ml, to only cDC2 For determining the frequency and proliferation of Tregs, and CD14+ DCs) (both from InvivoGen, Spain) was added to only DCs exposed to viral suspensions were examined. In this the cultures. With additional 18 h, DCs were harvested, and the case, half of the supernatant of the original DC culture was kept viability was determined by Near-IR dead cell staining kit with cells for mixing with PBMCs. After 5 days, cells were (Fisher Scientific, Spain). Expression of MHC-I, MHC-II, collected and incubated with an anti-CD25 antibody (Bio-Rad, CD80/CD86, and CCR7 was examined by flow cytometry. For Spain) biotinylated by EZ-Link Sulfo-NHS-SS-Biotin (Fisher CD80/86, a dimeric protein with the extracellular domain of Scientific, Spain); then, streptavidin PerCP-Cy5.5 was added human CD152 fused to murine IgG2a Fc was used. CCR7 was followed by a CD4a-FITC antibody. Foxp3 intracellular labeled by an antibody anti-human CCR7 (CD197), which has staining was performed with an anti-Foxp3 PE antibody using been verified to cross-react with porcine lymphocytes (32). the Foxp3/Transcription Factor Staining Buffer Set (eBioscience, Alexa Fluor 647-conjugated anti-mouse IgG2a (Fisher United States). Proliferation was determined by examining the Scientific, Spain) was used as the secondary antibody for intensity of CellTrace Violet fluorescence. MHC-I and CD80/86 staining. The referred antibodies are listed in Table 1. DCs were also used to perform mixed lymphocyte reaction as described below in the “Mixed Endocytosis/Phagocytosis Assay Lymphocyte Reaction Assay” subsection. cDC2 and CD14+ DCs were inoculated with PRRSV1 3267 at an MOI of 10 and incubated for 24 h at 37°C, 5% CO2. Then, cells Co-Culture of cDC1, cDC2, and CD14+ were centrifuged, seeded at a density of 10,000 cells/well in DCs With Infected MARC-145 Cells round-bottomed 96-well plates, and pulsed with dextran-FITC Two days prior to co-culture, MARC-145 cells were infected with (molecular weight 40,000; Sigma-Aldrich, Spain) at 1 mg/ml or MARC3267 (MOI 0.1), then detached and 10:1 mixed with Alexa 488-labeled S. aureus (Fisher Scientific, Spain) at 20 mg/ml cDC1, cDC2, or CD14+ DCs at a final concentration of 1.1 × (equivalent to 60 particles/cell) for 2.5 h and 1 h, respectively. 106/ml for cDC2 and CD14+ DC, and 0.55 × 106/ml for cDC1 (a Cells were incubated either at 37°C or on ice (negative control). final volume of 200 ml in the 96-well round-bottomed plates). Free Dextran-FITC was washed away. Endocytosis of dextran DCs co-cultured with mock-infected MARC-145 cells or with was determined by comparing the median fluorescence intensity only complete RPMI were used as negative controls; DCs (MFI) of cells incubated at 37°C with the MFI of cells incubated stimulated with poly I:C was used as the positive control. After on ice. Extracellular bound S. aureus was quenched by the 24 h of incubation, DCs were labeled for MHC-I, MHC-II, addition of 1 mg/ml trypan blue. The phagocytic capability Frontiers in Immunology | www.frontiersin.org 4 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs was determined by examining the proportion of AF488-S. aureus positive cells as published before (26). PRRSV1 nucleocapsid (N) protein and titrating supernatants in alveolar macrophages (AMs). Mock-infected cells were used as the negative control (mock-infected). Results were obtained with cells from two pigs. FIGURE 1 | Susceptibility of cDC1, cDC2, and CD14+ DCs to PRRSV1 3267 infection. Cells were infected by PRRSV1 3267 at MOI of 0.1 for 48 h. The infection was determined by flow cytometry staining of For examining the phagocytosis of infected cells, cDC2 or CD14+ DCs were co-cultured (1 h, 37°C) with PRRSV1 3267- infected MARC-145 cells (ratio of 1:5) that were labeled with CellTrace Violet (Fisher Scientific, Spain). Alveolar macrophages, the natural target of PRRSV, were not used because they are also phagocytes and secret substantial amount of pro-inflammatory cytokines (TNF-a, IL-1b). Cells incubated on ice or co-cultivated with mock-infected MARC-145 cells were used as controls. DCs (MHC-II+) harboring Violet fluorescence were considered as positive of phagocytosis. Detection of Cytokines/Chemokines The presence of cytokines/chemokines IFN-a, IFN-g, IL-4, IL-6, IL-12/IL-23p40, IL-10, IL-1b, TNF-a, and IL-8 (CXCL8) in cell culture supernatants was determined using Porcine ProcartaPlex Multiplex Immunoassay (Invitrogen, Spain) according to the manufacturer’s instructions. Cytokines/chemokines were captured by fluorescent antibody-coated beads, thereafter identified by adding 25 ml of biotinylated detection antibodies followed by 50 ml of streptavidin-PE antibodies. TGF-b1 was measured by Simplex ProcartaPlex Immunoassay (Invitrogen, Spain). Cell culture supernatants were acid treated to activate TGF-b1 before adding to the antibody-coated beads. Results were quantified on a MAGPIX system using a four-parameter logistic regression curve. All samples were run in duplicate. Statistical Analysis All data were analyzed using the GraphPad Prism 9.0 software package (GraphPad Software, La Jolla, CA, United States). Statistical tests applied to each data set are indicated in figure legends. RESULTS cDC1 and cDC2 Were Not Susceptible to PRRSV1 Infection, a Fraction of CD14+ DCs Were Infected The contour plots from one pig are shown as the representative. cDC1, cDC2, and CD14+ DCs were sorted and infected with PRRSV1 3267 at an MOI of 0.1 for 48 h. As shown in Figure 1, PRRSV1 N protein labeling was only detected in a small but consistent fraction of CD14+ DCs (6.2 ± 1.3%). The supernatants of the infected CD14+ DCs contained high viral titers, on average 6.2 ± 0.8 log10 TCID50/ml. No virus was detected in cDC1 or cDC2 (Figure 1). Parallel infection of the unsorted DCs showed that all Npos cells were CD14– (Supplementary Figure S1A). The proportion of CD14+ cells decreased significantly in the inoculated cultures compared with the mock-inoculated ones (30.7 ± 5.0% versus 18.2 ± 3.7%, respectively, p<0.05; 33.2 ± 2.8% before infection) (Supplementary Figure S1B). The proportion of cells harboring other cell surface markers were not significantly changed. Infected cells were further characterized as MHC-II+ CD172a+ CD163+/– DEC205 – CD11R1 +/– CD11R3 +/– CD1 + (Supplementary Figure S1C). Frontiers in Immunology | www.frontiersin.org 5 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs Exposure of CDC1, CDC2, and CD14+ DCs IL-4 (Figure 3). When UV-inactivated PRRSV1 3267 (referred to PRRSV1 3267 Suspensions Did Not to as UV3267 below) was used, the result was similar, suggesting Change the Expression of MHC-I, MHC-II, that the viability of the virus was not relevant to these variables. As the positive controls, exposure of DCs to poly I:C or CD80/86, or CCR7, Did Not Induce any gardiquimod induced an evident increase in the expression of Significant Cytokine Release, Did Not MHC-I, MHC-II, CD80/86, and CCR7, as well as the release of Modify T-Cell Proliferation in the MLR, and cytokines for all DC types (Figures 2 and 3). Of note, the viability Did Not Affect the Endocytic/Phagocytic of cDC2 and CD14+ DCs did not drop after exposure to the virus Capabilities for 24 h compared with exposure to plain medium. However, the Figure 2 shows the result of the flow cytometry assessment of viability of cDC1 dropped significantly if they were not MHC-I, MHC-II, CD80/86, and CCR7 expression on cDC1, stimulated by poly I:C, which maintained the viability of cDC1 cDC2, and CD14+ DCs after incubation with PRRSV1 3267 up to 70.8 ± 3.2% after 24 h of incubation. When exposed to the (MOI 10, 24 h). As shown, exposure to the virus did not induce a virus or plain medium, less than 50% of cDC1 were viable at 6 h significant increase in the expression of the mentioned molecules and much less (< 10%) at 24 h. compared to DCs cultured with plain medium. Results were Virus-stimulated cDC1, cDC2, and CD14+ DCs were further similar when using MOI 1.0 (data not shown). used as antigen-presenting cells in an MLR. cDC1 pre-incubated Exposure to the virus did not induce any significant release with either live or inactivated virus were not efficient in sustaining of IL-12/IL-23p40, IL-10, IFN-a, IL-6, IL-1b, TNF-a, IFN-g, or T-cell proliferation as expected by the damaged cell viability. A B C FIGURE 2 | Maturation of cDC1, cDC2, and CD14+ DCs. DCs were inoculated with PRRSV1 3267 (MOI 10), live or UV-inactivated (UV3267), for 6 h prior to (for example, 3267+Poly I:C) or simultaneously with (for example, 3267/Poly I:C) poly I:C (10 mg/ml, to cDC1, cDC2, and CD14+ DCs) or gardiquimod (10 mg/ml, to only cDC2 and CD14+ DCs). Plain medium (complete RPMI 1640) was used as the negative control (mock) in each step. After incubation for extra 18 h, cells were harvested and stained for MHC-I, MHC-II, CD80/CD86, and CCR7 on cDC1 (A), cDC2 (B), and CD14+ DCs (C). Data of cDC1 were obtained from two pigs, while cDC2 and CD14+ DCs were from four pigs. The histograms shown are from one pig as the representative. Frontiers in Immunology | www.frontiersin.org 6 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs A B C FIGURE 3 | Cytokine production by cDC1, cDC2, and CD14+ DCs. DCs were inoculated with PRRSV1 3267 (MOI 10), live or UV-inactivated (UV3267), for 6 h prior to (for example, 3267+Poly I:C) or simultaneously with (for example, 3267/Poly I:C) poly I:C (10 mg/ml, to cDC1, cDC2, and CD14+ DCs) or gardiquimod (10 mg/ml, to only cDC2 and CD14+ DCs). In each step, plain medium (complete RPMI 1640) was used as the negative control (mock). After additional 18 h, cytokines (IFN-a, IFN-g, IL-4, IL-6, IL-12p40, IL-10, IL-1b, TNF-a, and IL-8 (CXCL8)) in the supernatants were assessed by a multiplex immunoassay, and TGF-b was assessed by a simplex immunoassay. Production of IL-12/IL-23p40 and TGF-b (the only two cytokines detected) in cDC1 (A), and IL-12/IL-23p40, IL-1b, IL-6, TNF-a, IFN- a, IL- 10, and TGF-b in cDC2 (B) and CD14+ DCs (C) are shown. IFN-g and IL-4 are not shown as they were rarely detected; IL-8 (CXCL8) is not shown as its standard was not good enough to use. The values are displayed as bars with mean and standard deviation indicated. Symbols in different colors represent different animals; symbols in red and orange are DCs from the same pig but derived in different days, the same as blue and bright green symbols. Symbols in the same color throughout (A–C) represent the same animal. Results of cDC1 were obtained from three animals, while cDC2 and CD14+ DCs were from four animals. Statistical significance was calculated by the Kruskal Wallis test with Tukey test for multiple comparisons. ****p < 0.0001, ***p < 0.001, **p < 0.01, and *p < 0.05. For cDC2 and CD14+ DCs, exposure to the live virus did not CD4+CD8– cells for both cDC2 and CD14+ DCs, and CD4–CD8– cause any significant effect on global T-cell proliferation, for CD14+ DCs. although some impairment was observed in the proliferation of The endocytic and phagocytic capabilities of cDC2 and the CD4–CD8– subset (Figure 4). By contrast, the use of UV3267 CD14+ DCs was also examined. Exposure to the virus did not resulted in a significantly higher (p < 0.05) proliferation rate of modify the capacity of cDC2 or CD14+ DCs to capture soluble Frontiers in Immunology | www.frontiersin.org 7 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs A B C FIGURE 4 | Proliferation of T cells induced by cDC1, cDC2, and CD14+ DCs. cDC1 (A), cDC2 (B), and CD14+ DCs (C) were inoculated with PRRSV1 3267 (MOI 10), live or UV-inactivated (UV3267), for 6 h prior to (for example, 3267+Poly I:C) or simultaneously with (for example, 3267/Poly I:C) poly I:C (10 mg/ml, to cDC1, cDC2, and CD14+ DCs) or gardiquimod (10 mg/ml, to only cDC2 and CD14+ DCs). The cultures were incubated for additional 18h. Plain medium (complete RPMI 1640) was used as negative control (mock) in each step. Such generated DCs were mixed with allogeneic pig PBMCs (labeled with CellTrace Violet dye) at a ratio of 1:5. After 5 days, cells were harvested and stained for CD3, CD4, and CD8a. Proliferation of CD3+, CD4+CD8a–, CD4– CD8a+, CD4+CD8a+, and CD4–CD8a– T cells was determined by CellTrace Violet dilution by flow cytometry. Gating strategy was CD3+ → CD4+CD8a–, CD4– CD8a+, CD4+CD8a+, CD4–CD8a– → proportion of diluted Violet (Supplementary Figure S3). In each group, one symbol represents DCs from one pig; symbols with the same color in the same box mean DCs were from the same pig but derived in different days; symbols with the same color throughout (A–C) represent the same animal. Data are shown as boxplots (25th–75th interquartile range), with median and whiskers showing minimum to maximum datapoints. For each type of DCs, proliferation induced by DCs cultured with plain medium (mock) was considered as 1.0, and the other populations were normalized to it. Data was normalization based on TLR3- or TLR7-related groups without cross-normalization. Data are from three independent experiments with two animals for cDC1 and four animals for cDC2 and CD14+ DCs. Statistical significance was calculated by the Kruskal Wallis test with Tukey test for multiple comparisons. ****p < 0.0001, ***p < 0.001, **p < 0.01, and *p < 0.05. dextran-FITC or particulate Alexa 488-labeled S. aureus with virus propagated in both AMs and MARC-145 cells, but not (Supplementary Figure S2). cDC1 were not included here with UV-treated virus. Of note, exposure of cDC2 to UV3267 because of the low number of available cells. resulted in a higher proportion of Tregs (Figure 5). Examination of IL-10 and TGF-b in the supernatants of MLR cultures did not find Pre-Exposure of cDC2 or CD14+ DCs, but any difference between cDC2 or CD14+ DCs exposed to the virus Not cDC1 to PRRSV1 3267 Suspensions and to the plain medium (data not shown). Enhanced Proliferation of CD4+CD25+Foxp3+ Regulatory T Cells (Tregs) PRRSV1 Modified the Response of cDC2 Next, we examined the capacity of DCs to induce Tregs after and CD14+ DCs to Poly I:C and exposure to PRRSV1. To this end, cDC1, cDC2, and CD14+ DCs Gardiquimod Stimulation were exposed to PRRSV1 3267, then mixed with allogeneic PBMCs To further investigate the impact of PRRSV1 3267 on cDC1, in an MLR assay. The proportion of CD4+CD25+Foxp3+ Tregs and cDC2, and CD14+ DCs, we examined whether exposure to their proliferation rate were determined. As shown in Figure 5, PRRSV1 could modify TLR3- and TLR7-mediated activation. exposure of DCs to the virus did not increase the frequency of To this end, cDC1, cDC2, and CD14+ DCs were exposed to Tregs but enhanced their proliferation. The enhancement occurred viable or UV-inactivated virus for 6 h prior to or simultaneously Frontiers in Immunology | www.frontiersin.org 8 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs A B FIGURE 5 | Proportion and proliferation of Tregs induced by PRRSV1-stimulated cDC1, cDC2, and CD14+ DCs. DCs were inoculated by live and UV-inactivated PRRSV1 3267 (MOI 10) that was propagated in either AMs or MARC-145 cells for 24 h, then mixed with allogeneic pig PBMCs (labeled with CellTrace Violet) at a ratio of 1:5. After 5 days, cells were harvested and stained for CD4, CD25, and Foxp3. DCs incubated with plain medium (mock) were used as the negative control. (A) Cells labeled as CD4+CD25+Foxp3+ were defined as Tregs, among which Violet-diluted cells were considered as proliferated Tregs. (B) The proportion of Tregs among CD4+ cells (black bars) and the proliferation of Tregs (red symbols) were shown with a dual Y-axis graph. Data are from three animals for each DC type. with poly I:C (TLR3 agonist) or gardiquimod (TLR7 For CD14+ DCs, pre-exposure to live 3267, but not UV3267, agonist) stimulation. enhanced IFN-a release upon addition of poly I:C (p<0.05), while Since cDC1 are not responsive to gardiquimod stimulation (26), simultaneous addition of the live virus and poly I:C strongly only TLR3-mediated activation was examined. As a result, PRRSV1 decreased the production of IFN-a (p<0.05). This 3267 did not affect the maturation of cDC1 (Figure 2A) or the downregulation was also observed for IL-12, IL-6, TNF-a, and production of IL-12/IL-23 p40 (Figure 3A), the only responsive IL-10 when CD14 + DCs were exposed to the virus and cytokine of cDC1 upon poly I:C stimulation, regardless of 3267 was gardiquimod at the same time (Figure 3C). Of note, the highest added before or simultaneously with poly I:C. In the following induction of IL-10 was obtained by pre-exposing CD14+ DCs to MLR assay, all combinations of virus and poly I:C resulted in UV-3267 followed by TLR7 activation (p<0.05). In the MLR assay, similar T-cell proliferation values (Figure 4A). significant inhibition of T-cell proliferation, particularly in CD4– For cDC2, pre-exposure to the virus did not modify the CD8+ and CD4–CD8– subsets, was induced by CD14+ DCs that expression of cell surface maturation markers compared with were exposed to the virus prior to or simultaneously with TLR7 the use of poly I:C or gardiquimod alone (Figure 2B). Regarding activation (Figure 4C). A similar inhibition effect was also found in cytokine release, pre-exposure to the live virus resulted in an combination with TLR3 activation. The inhibition was not evident release of IFN-a upon poly I:C stimulation, although the observed when the UV-treated virus was used. amount was not significant compared with poly I:C was added TGF-b was not induced by the virus or the combination with alone (p=0.1) (Figure 3B). Gardiquimod stimulation produced a any TLR ligand above levels in the control samples (Figure 3). wide panel of cytokines. The production of IL-1b and IL-10 was enhanced (p<0.05) when cDC2 were pre-incubated with UV- cDC2 and CD14+ DCs, but Not cDC1 treated virus. This was not observed when live virus was used. Efficiently Sensed 3267-Infected Cells The exposed cDC2 were then used as antigen-presenting cells in We then investigated the response of different DC types to an allogeneic MLR assay. As shown in Figure 4B, pre-exposure PRRSV1-infected cells. For that purpose, we exposed cDC1, of cDC2 to the live virus before TLR7 activation resulted in an cDC2, and CD14+ DCs to 3267-infected MARC-145 cells impaired proliferation of CD4–CD8– T cells. This effect was not (62.6 ± 1.2% of PRRSV1 N+ cells). As shown in Figure 6A, the observed when UV3267 was used. No changes were observed expression of MHC-I, MHC-II, CD80/86, and CCR7 on cDC2 when the virus was combined with poly I:C (TLR3- was significantly increased (p < 0.05), in contrast to co-culture mediated activation). with mock-infected cells or plain medium. For CD14+ DCs, the Frontiers in Immunology | www.frontiersin.org 9 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs expression of MHC-I, MHC-II, and CD80/86 was also increased (2231.3 ± 762.0 pg/ml) along with lower quantities of IFN-a, IL- but not that of CCR7 (Figure 6A). MARC-145 cells inoculated 6, and TNF-a (Figure 6B). CD14+ DCs exposed to infected with UV-treated virus were unable to activate neither cDC2 nor MARC-145 cells had a similar IL-12/IL-23p40 response but also CD14+ DCs. produced IL-10 (147.0 ± 14.4 pg/ml) (Figure 6B). In contrast, Responsiveness of cDC2 and CD14+ DCs to infected MARC- cDC1 were not responsive to infected MARC-145 cells in any of 145 cells was also reflected in cytokine/chemokine production. the performed tests (data not shown). While co-culture of cDC2 with mock-infected MARC-145 cells To further confirm the observed effect was attributable to the did not induce any significant secretion of cytokines, the use of infection of MARC-145 or its consequences, a dose-response infected cells resulted in a substantial release of IL-12/IL-23p40 experiment was performed. cDC2 and CD14+ DCs were co- A B FIGURE 6 | Sensing of infected MARC-145 cells by cDC1, cDC2, and CD14+ DCs. cDC1, cDC2, and CD14+ DCs were co-cultured with MARC-145 cells that were infected with live or UV-inactivated PRRSV1 3267 that was propagated in MARC-145 (3267-MARC and UV3267-MARC, respectively; UV3267-MARC were not used for cDC1). DCs stimulated by poly I:C or cultured with plain medium were used as the positive and negative controls, respectively. After 24 hours, (A) DC maturation was determined by flow cytometry staining of MHC-I, MHC-II, CD80/86, and CCR7. (B) Cytokine production was assessed by a multiplex immunoassay for IFN-a, IFN-g, IL-4, IL-6, IL-12p40, IL-10, IL-1b, TNF-a, and IL-8 (CXCL8). Cytokines in the co-culture of DCs and UV3267-MARC were not examined and are not shown; rarely detected IFN-g and IL-4 and unworkable IL-8 (CXCL8) are also not shown. Data are obtained from three animals for cDC1 and four animals for cDC2 and CD14+ DCs. Statistical significance was calculated by the Kruskal Wallis test with Tukey test for multiple comparisons. ****p < 0.0001, ***p < 0.001, **p < 0.01, and *p < 0.05. Frontiers in Immunology | www.frontiersin.org 10 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs cultured with MARC-145 cells that displayed increasing whether the activation was triggered by the phagocytosis of proportion of infected cells (PRRSV1 N+), on average 0%, infected cells. As shown in Figure 8A, cDC2 and CD14+ DCs 4.1 ± 1.1%, 18.2 ± 0.7%, 39.6 ± 0.8%, 57.8 ± 0.2%, 63.4 ± 3.1%, were active in phagocytosing infected MARC-145 cells or respectively (Figure 7A). The results showed that phenotypical components derived from the infected cells, indicating the maturation (Figure 7B) and cytokine production (Figure 7C) of potential role of the phagocytic activity in DC activation. We cDC2 or CD14+ DCs were evident only when the infection also tested the role of apoptosis. In this case, cDC2 were co- reached at a higher level, > 39.6 ± 0.8%, where the cell viability cultivated with UV-treated MARC-145 cells (without infection) dropped significantly (Figure 7A). that had a proportion of apoptotic and necrotic cells (13.5 ± 2.6% Next, we examined some factors whereby infected cells may of Annexin V + PI – and 39.4 ± 1.4% of Annexin V + PI + ) trigger cDC2 or CD14+ DCs maturation. Firstly, cDC2 and comparable to that of infected MARC-145 cells (63.4 ± 3.1% of CD14+ DCs were exposed to the supernatants of infected PRRSV1 N+; 14.9 ± 1.5% of Annexin V+PI– and 33.8 ± 2.6% of MARC-145 cells (63.4 ± 3.1% of PRRSV1 N+ cells), but the Annexin V+PI+). The results showed that exposure of cDC2 to expression of maturation molecules was not increased in either infected or apoptotic MARC-145 cells resulted in a similar cDC2 or CD14+ DCs (data not shown). Then, we examined phenotypic maturation (data not shown) but differed in their A B C FIGURE 7 | Response of cDC2 and CD14+ DCs to different doses of PRRSV1 3267 infected MARC-145 cells. (A) An increasing % of infected (PRRSV1 N+) MARC-145 cells (black bars) was generated by infecting MARC-145 cells with different doses of MARC3267. The viability was indicated upper (orange line); (B) Expression of MHC-I, MHC-II, CD80/86, and CCR7 in cultures of cDC2 (upper) and CD14+ DCs (lower) exposed to MARC-145 cells containing different % of infected cells (indicated at the left of Y-axis); (C) Cytokine production in cultures of cDC2 (upper) and CD14+ DCs (lower) exposed to MARC-145 cells containing an increasing % of infected cells (indicated below IL-12/IL-23p40 of CD14+ DCs). IL-10 was not detected in cDC2, thus not shown. Frontiers in Immunology | www.frontiersin.org 11 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs ability to induce T-cell proliferation in the MLR. cDC2 co- COVID-19, the switch of CD14 expression was also observed cultured with infected MARC-145 promoted the proliferation (35–38). In those cases, the loss of CD14 was assumed to be of CD4–CD8+ and CD4–CD8– T-cell subsets which, however, caused by infection-induced apoptosis but not virus-specific was suppressed by the use of UV-treated cells (Figure 8B). regulation (35). This in vitro derived CD14 + DCs were supposed to resemble moDCs. But whether this population exists in vivo and its roles during PRRSV infection is not DISCUSSION known yet. Regardless of the susceptibility, inoculation of cDC1, cDC2, The present study delineates the interaction of a typical and CD14+ DCs by PRRSV1 3267 did not induce any significant PRRSV1.1 isolate (3267) with in vitro derived cDC1, cDC2, expression of the maturation-associated molecules MHC-I, and CD14+ DCs. We show here that cDC1 and cDC2 were MHC-II, CD80/86, or CCR7. Also, no cytokines (of the chosen refractory to 3267 infection, in agreement with previous studies panel) were produced in response to the virus, and no alterations using lung- or tonsil-isolated cDCs (25, 33). A proportion of in the endocytic and phagocytic capabilities were observed. cDC1 CD14+ DCs was susceptible, likely due to the expression of was a particular case because their survival time was short in the CD163, which is the essential receptor for PRRSV infection (34). absence of poly I:C stimulation. The factors required to maintain Of note, the infection resulted in the loss of CD14 expression, as cDC1 survival in vitro remain to be determined. Taken together, deduced from the decrease in the proportion of CD14+ cells in DCs remained immature after exposure to PRRSV1. This is infected cultures and the fact that labeling of PRRSV N protein compatible with the results using ex vivo lung DCs (25), but was mostly found in CD14– cells, while, when sorted, CD14– cells opposed to what is expected from the functions of DCs (12). The were refractory to the infection. In infection models by other use of a high MOI 10 in our work makes it unlikely that DCs RNA viruses such as influenza virus, HIV, Zika virus, and were not sufficiently exposed to the virus. A previous study in our A B FIGURE 8 | (A) Phagocytosis of infected MARC-145 cells by cDC2 and CD14+ DCs. cDC2 and CD14+ DCs (with MHC-II revealed by Alexa Fluor 647) were co- cultured with PRRSV1 3267 infected or mock-infected MARC-145 cells (labeled with CellTrace Violet dye) at a ratio of 1:5 at 37°C or on ice for 1 h. cDC2 and CD14+ DCs shown as MHC-II+Violet+ were defined as phagocytotic cells. cDC2 and CD14+ DCs were derived from two animals with one shown as the representative; (B) Comparison of T-cell proliferation by cDC2 co-cultured with infected and UV-treated MARC-145 cells. cDC2 were co-cultured for 24 h with PRRSV1 3267 infected, UV-treated, or mock-infected MARC-145 cells (3267-MARC, UV-MARC, Mock-MARC), or with plain medium or supernatants collected from infected and mock-infected MARC-145 cells (S. mock-MARC, S. 3267-MARC). Then cells were mixed with allogeneic pig PBMCs (labeled with CellTrace Violet dye) at a ratio of 1:5 and cultured for five days. Cultures in the absence of cDC2 were also included. Proliferation of CD3+, CD4+CD8a–, CD4–CD8a+, CD4+CD8a+, and CD4–CD8a– T cells was determined by CellTrace Violet dilution (gating strategy is shown in Supplementary Figure S3). cDC2 were derived from two animals, as shown by the red and blue symbols. The same color means DCs were from the same pig but derived in different days. Frontiers in Immunology | www.frontiersin.org 12 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs group observed that the recall response against PRRSV1 3267 RNA could be processed and modified in a different way (IFN-g ELISPOT) was lower than against a heterologous, less compared to untreated RNA, and as a result, it could be virulent strain (designated as 3262) (29) that was able to induce distinctively recognized by cytosolic receptors. In any case, the several cytokines in AMs, PBMCs, and GM-CSF-derived activation of TLRs or other receptors such as Retinoic acid- BMDCs (18, 28). It would be coherent with a lower antigen inducible gene (RIG)-I-like receptors (RLRs) and NOD-like presentation ability of 3267-exposed DCs (or antigen-presenting receptors (NLRs) would be expected to end in cytokine cells in general). In the work of Bordet et al. (25), ex vivo DCs also production, which however was not observed with our live or remained immature after inoculation with PRRSV1 strains UV-treated PRRSV virus. Indeed, pre-exposure to the UV- (moderate or high virulence). These suggest that the treated virus enhanced IL-10 secretion upon gardiquimod impairment of DC maturation can be a general feature of stimulation, which was consistent with a higher frequency of PRRSV1 infection; however, some other strains, such as 3262, Tregs, at least for cDC2. may induce more effective responses based on the experimental The experiments combining PRRSV1 with TLR agonists data (29). Further studies are required to examine whether low- indicated that the virus may interact with TLR3 as indicated virulent strains modulate DCs distinctly. Impairment of DC by the enhanced IFN-a secretion when cDC2 or CD14+ DCs maturation has been reported for other viruses such as were pre-exposed to the live virus. Since the virus alone (either lymphocytic choriomeningitis virus (LCMV) that evades host live or inactivated) was unable to induce IFN-a, the boosting is immune response by preventing DC maturation (39). supposed not to be a summative effect but a synergistic Although unresponsive to viral suspensions, cDC2 and interaction. It is attempting to speculate that replication of CD14 + DCs were fully activated when co-cultivated with the virus could start in cDC2 but was aborted before the highly infected MARC-145 cells. Our preliminary analysis synthesis of the viral structural proteins (as viral N protein indicates phagocytosis of apoptotic or necrotic infected cells or was not detected in cDC2). Within the framework of an components derived from infected cells might be the mechanism. abortive replication, the replication activity and non- PRRSV-infected cells were also more potent inducers of IFN-a structural proteins (nsp) could be present in the cytoplasm, by pDCs than cell-free virions (40). A recent study showed that which may activate TLR3 signaling-associated modulators. cDCs and moDCs were only recruited at peak PRRSV viremia, Several proteins are known to regulate TLR3 folding, coinciding with the induction of protective T-cell responses and trafficking, and cleavage [see (45) for a review]. They might the resolution of infection in pigs (24). However, other reports be candidates for interaction with the virus. But this hypothesis were unable to show such an increase (25, 33). conflicts with the antagonizing effect of nsp4 and nsp11 in poly Based on all of these findings, it is tempting to speculate that I:C-induced type I IFN production (46, 47). the effective sensing of PRRSV infection by DCs is not reached For CD14+ DCs, simultaneous addition of the virus and TLR7 until the advanced infection stage when a massive number of agonist resulted in a pan-inhibition of cytokine production. This macrophages are infected and develop apoptosis or necrosis. was not associated with the susceptibility of CD14+ DCs because This scenario would partially explain why clearance of PRRSV is pre-exposure to the virus did not result in such effect. The pan- delayed during infection. Current commercial PRRSV vaccines, inhibition effect suggests that PRRSV1 may competitively inactivated or modified-live vaccines, are lack of satisfactory suppress the response to a coinfecting pathogen that is efficacy [reviewed by Hu and Zhang (41)]. The observations in recognized and signaled through TLR7. Swine influenza A our study would have implications for the development of DC- virus (swIVA), a virus recognized by pDCs through TLR7/ targeting vaccines, for instance, conjugating antigens to TLR MyD88 signalling (48), was demonstrated to be inhibited in ligands or targeting DC-specific endocytic receptors, i.e., C-type replication and interferon response by PRRSV that was added lectin receptors DEC205, DEC209, or mannose receptor (42), to simultaneously (49). But why is the inhibition restricted to when boost both the magnitude and quality of T- and B-cell responses. the TLR7 ligand was added simultaneously with PRRSV1 but not In the last decade, DC-targeting approaches have attracted a lot 6 hours after PRRSV1 inoculation? Future studies aimed at of attention to developing vaccines against HIV, cancer, and delineating the precise mechanism whereby PRRSV in the diseases that currently lack effective vaccines [as reviewed by coinfection or superinfection modulates a heterologous virus Kastenmüller et al. (43)]. will be valuable for understanding the complex field cases. Exposure of cDC1, cDC2, or CD14+ DCs to the live virus did For both cDC2 and CD14+ DCs, pre-exposure to the virus not change their antigen presentation abilities. But when cDC2 affect the proliferation of CD4–CD8– T cells. This T-cell subset or CD14+ DCs were exposed to the UV-inactivated virus, the usually harbours TCR-gd and can be divided into two subsets proliferation of CD4+CD8– T cells was boosted. It is difficult to based on the expression of CD2 (50). Obviously, the reduction of see the reasons behind it as the use of UV- inactivated virus alone proliferation pointed towards the CD2+ memory/effector subset. did not enhance phenotypic maturation or induce secretion of The role of these cells in PRRSV infection is still unknown. the examined cytokines. Since the UV inactivation method In summary, our observations suggest that exposure to preserves the integrity of the viral envelop glycoproteins (44), PRRSV1 does not induce maturation of cDC1, cDC2, or the discrepancy of results between the use of live and UV3267 is CD14+ DCs, but it may modify TLR-associated responses unlikely caused by virus binding or internalization but by an (except for cDC1). Moreover, sensing of infected cells is event later in the replication process. Alternatively, UV-damaged different from that of the free viral suspensions. But how do Frontiers in Immunology | www.frontiersin.org 13 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs TLR sensor pathways coordinate the host response and the authors contributed to the article and approved the mechanisms resulting in a poor and delayed immune response submitted version. to PRRSV remain to be determined. FUNDING DATA AVAILABILITY STATEMENT This work was funded by the Veterinary Infectious Diseases The original contributions presented in the study are included in Diagnostic Laboratory of the UAB. the article/Supplementary Material. Further inquiries can be directed to the corresponding author. ACKNOWLEDGMENTS ETHICS STATEMENT We gratefully thank Ivan Dı́az, Esmeralda Cano, and Nú ria Navarro (CReSA, IRTA-UAB, Spain) who provided technical All the cells used in this work were obtained from healthy support in obtention of bone marrow cells, isolation of alveolar animals euthanized according to approved procedures by the macrophages and PBMCs. Ethics Committees of IRTA and UAB. SUPPLEMENTARY MATERIAL AUTHOR CONTRIBUTIONS The Supplementary Material for this article can be found online YL performed the experiments. YL and EM designed the at: https://www.frontiersin.org/articles/10.3389/fimmu.2021. experiments, analyzed the data, and wrote the manuscript. All 674185/full#supplementary-material 10. Dı́az I, Darwich L, Pappaterra G, Pujols J, Mateu E. Different European-Type REFERENCES Vaccines Against Porcine Reproductive and Respiratory Syndrome Virus 1. Rowland RRR, Lawson S, Rossow K, Benfield DA. Lymphoid Tissue Tropism Have Different Immunological Properties and Confer Different Protection to of Porcine Reproductive and Respiratory Syndrome Virus Replication During Pigs. Virology (2006) 351:249–59. doi: 10.1016/j.virol.2006.03.046 Persistent Infection of Pigs Originally Exposed to Virus in Utero. Vet 11. Shortman K, Naik SH. Steady-State and Inflammatory Dendritic-Cell Microbiol (2003) 96:219–35. doi: 10.1016/j.vetmic.2003.07.006 Development. Nat Rev Immunol (2007) 7:19–30. doi: 10.1038/nri1996 2. Chung W B, MW L, WF C, Hsu M, Yang PC. Persistence of Porcine 12. Guermonprez P, Valladeau J, Zitvogel L, Thé ry C, Amigorena S. Antigen Reproductive and Respiratory Syndrome Virus in Intensive Farrow-to- Presentation and T Cell Stimulation by Dendritic Cells. Annu Rev Immunol Finish Pig Herds. Can J Vet Res (1997) 61:292–8. (2002) 20:621–67. doi: 10.1146/annurev.immunol.20.100301.064828 3. Wills RW, Zimmerman JJ, Yoon KJ, Swenson SL, McGinley MJ, Hill HT, et al. 13. Banchereau J, Steinman RM. Dendritic Cells and the Control of Immunity. Porcine Reproductive and Respiratory Syndrome Virus: A Persistent Nature (1998) 392:245–52. doi: 10.1038/32588 Infection. Vet Microbiol (1997) 231–40. doi: 10.1016/S0378-1135(96)01337-5. 14. Wang X, Eaton M, Mayer M, Li H, He D, Nelson E, et al. Porcine Reproductive 4. Allende R, Laegreid WW, Kutish GF, Galeota JA, Wills RW, Osorio FA. and Respiratory Syndrome Virus Productively Infects Monocyte-Derived Porcine Reproductive and Respiratory Syndrome Virus: Description of Dendritic Cells and Compromises Their Antigen-Presenting Ability. Arch Persistence in Individual Pigs Upon Experimental Infection. J Virol (2000) Virol (2007) 152:289–303. doi: 10.1007/s00705-006-0857-1 74:10834–7. doi: 10.1128/jvi.74.22.10834-10837.2000 15. Rodrı́guez-Gó mez IM, Käser T, Gó mez-Laguna J, Lamp B, Sinn L, Rümenapf T, 5. Wills RW, Doster AR, Galeota JA, Sur JH, Osorio FA. Duration of Infection and et al. PRRSV-Infected Monocyte-Derived Dendritic Cells Express High Levels of Proportion of Pigs Persistently Infected With Porcine Reproductive and SLA-DR and CD80/86 But do Not Stimulate PRRSV-Naïve Regulatory T Cells Respiratory Syndrome Virus. J Clin Microbiol (2003) 41:58–62. doi: 10.1128/ to Proliferate. Vet Res (2015) 46:54. doi: 10.1186/s13567-015-0186-z JCM.41.1.58-62.2003 16. Weesendorp E, Stockhofe-Zurwieden N, Popma-De Graaf DJ, Fijten H, Rebel 6. Meier WA, Galeota J, Osorio FA, Husmann RJ, Schnitzlein WM, Zuckermann JMJ. Phenotypic Modulation and Cytokine Profiles of Antigen Presenting FA. Gradual Development of the Interferon-G Response of Swine to Porcine Cells by European Subtype 1 and 3 Porcine Reproductive and Respiratory Reproductive and Respiratory Syndrome Virus Infection or Vaccination. Syndrome Virus Strains. Vitro Vivo Vet Microbiol (2013) 167:638–50. Virology (2003) 309:18–31. doi: 10.1016/S0042-6822(03)00009-6 doi: 10.1016/j.vetmic.2013.09.021 7. Dı́az I, Darwich L, Pappaterra G, Pujols J, Mateu E. Immune Responses of 17. Peng YT, Chaung HC, luan CH, HC C, Chung W B. Modulations of Pigs After Experimental Infection With a European Strain of Porcine Phenotype and Cytokine Expression of Porcine Bone Marrow-Derived Reproductive and Respiratory Syndrome Virus. J Gen Virol (2005) Dendritic Cells by Porcine Reproductive and Respiratory Syndrome Virus. 86:1943–51. doi: 10.1099/vir.0.80959-0 Vet Microbiol (2009) 136:359–65. doi: 10.1016/j.vetmic.2008.11.013 8. Zuckermann FA, Garcia EA, Luque ID, Christopher-Hennings J, Doster A, 18. Gimeno M, Darwich L, Diaz I, De La Torre E, Pujols J, Martı́n M, et al. Brito M, et al. Assessment of the Efficacy of Commercial Porcine Cytokine Profiles and Phenotype Regulation of Antigen Presenting Cells by Reproductive and Respiratory Syndrome Virus (PRRSV) Vaccines Based on Genotype-I Porcine Reproductive and Respiratory Syndrome Virus Isolates. Measurement of Serologic Response, Frequency of Gamma-IFN-Producing Vet Res (2011) 42:9. doi: 10.1186/1297-9716-42-9 Cells and Virological Parameters of Protection Upon Challenge. Vet 19. Piñeyro PE, Subramaniam S, Kenney SP, Heffron CL, Gimé nez-Lirola LG, Microbiol (2007) 123:69–85. doi: 10.1016/j.vetmic.2007.02.009 Meng XJ. Modulation of Proinflammatory Cytokines in Monocyte-Derived 9. Vé zina SA, Loemba H, Fournier M, Dea S, Archambault D. Antibody Dendritic Cells by Porcine Reproductive and Respiratory Syndrome Virus Production and Blastogenic Response in Pigs Experimentally Infected With Through Interaction With the Porcine Intercellular-Adhesion-Molecule-3- Porcine Reproductive and Respiratory Syndrome Virus. Can J Vet Res (1996) Grabbing Nonintegrin. Viral Immunol (2016) 29:546–56. doi: 10.1089/ 60:94–99. vim.2016.0104 Frontiers in Immunology | www.frontiersin.org 14 June 2021 | Volume 12 | Article 674185 Li and Mateu Interaction of PRRSV1 With cDCs 20. Liu J, Wei S, Liu L, Shan F, Zhao Y, Shen G. The Role of Porcine Reproductive Virologically Suppressive Anti-Retroviral Therapy. Atherosclerosis (2014) and Respiratory Syndrome Virus Infection in Immune Phenotype and Th1/ 232:52–8. doi: 10.1016/j.atherosclerosis.2013.10.021 Th2 Balance of Dendritic Cells. Dev Comp Immunol (2016) 65:245–52. 37. Michlmayr D, Andrade P, Gonzalez K, Balmaseda A, Harris E. CD14+CD16+ doi: 10.1016/j.dci.2016.07.012 Monocytes are the Main Target of Zika Virus Infection in Peripheral Blood 21. Silva-Campa E, Flores-Mendoza L, Resé ndiz M, Pinelli-Saavedra A, Mata- Mononuclear Cells in a Paediatric Study in Nicaragua. Nat Microbiol (2017) Haro V, Mwangi W, et al. Induction of T Helper 3 Regulatory Cells by 2:1462–70. doi: 10.1038/s41564-017-0035-0 Dendritic Cells Infected With Porcine Reproductive and Respiratory 38. Alzaid F, Julla J, Diedisheim M, Potier C, Potier L, Velho G, et al. Monocytopenia, Syndrome Virus. Virology (2009) 387:373–9. doi: 10.1016/j.virol.2009.02.033 Monocyte Morphological Anomalies and Hyperinflammation Characterise 22. Summerfield A, Auray G, Ricklin M. Comparative Dendritic Cell Biology of Severe COVID -19 in Type 2 Diabetes. EMBO Mol Med (2020) 12:e13038. Veterinary Mammals. Annu Rev Anim Biosci (2015) 3:533–57. doi: 10.1146/ doi: 10.15252/emmm.202013038 annurev-animal-022114-111009 39. Sevilla N, McGavern DB, Teng C, Kunz S, Oldstone MBA. Viral Targeting of 23. Baumann A, Mateu E, Murtaugh MP, Summerfield A. Impact of Genotype 1 Hematopoietic Progenitors and Inhibition of DC Maturation as a Dual and 2 of Porcine Reproductive and Respiratory Syndrome Viruses on Strategy for Immune Subversion. J Clin Invest (2004) 113:737–45. Interferon-a Responses by Plasmacytoid Dendritic Cells. Vet Res (2013) doi: 10.1172/JCI20243 44:33. doi: 10.1186/1297-9716-44-33 40. Garcı́a-Nicolá s O, Auray G, Sautter CA, Rappe JCF, McCullough KC, Ruggli 24. Nazki S, Khatun A, Jeong CG, Mattoo SUS, Gu S, Lee SI, et al. Evaluation of N, et al. Sensing of Porcine Reproductive and Respiratory Syndrome Virus- Local and Systemic Immune Responses in Pigs Experimentally Challenged Infected Macrophages by Plasmacytoid Dendritic Cells. Front Microbiol With Porcine Reproductive and Respiratory Syndrome Virus. Vet Res (2020) (2016) 7:771. doi: 10.3389/fmicb.2016.00771 51:66. doi: 10.1186/s13567-020-00789-7 41. Hu J, Zhang C. Porcine Reproductive and Respiratory Syndrome Virus 25. Bordet E, Blanc F, Tiret M, Crisci E, Bouguyon E, Renson P, et al. Porcine Vaccines: Current Status and Strategies to a Universal Vaccine. Transbound Reproductive and Respiratory Syndrome Virus Type 1.3 Lena Triggers Emerg Dis (2014) 61:109–20. doi: 10.1111/tbed.12016 Conventional Dendritic Cells 1 Activation and T Helper 1 Immune 42. Geijtenbeek TBH, Gringhuis SI. Signalling Through C-Type Lectin Receptors: Response Without Infecting Dendritic Cells. Front Immunol (2018) 9:2299. Shaping Immune Responses. Nat Rev Immunol (2009) 9:465–79. doi: 10.1038/ doi: 10.3389/fimmu.2018.02299 nri2569 26. Li Y, Puebla-Clark L, Herná ndez J, Dı́az I, Mateu E. Development of Pig 43. Kastenmüller W, Kastenmüller K, Kurts C, Seder RA. Dendritic Cell-Targeted Conventional Dendritic Cells From Bone Marrow Hematopoietic Cells In Vaccines-Hope or Hype? Nat Rev Immunol (2014) 14:705–11. doi: 10.1038/ Vitro. Front Immunol (2020) 11:553859. doi: 10.3389/fimmu.2020.553859 nri3727 27. Gimeno M, Darwich L, Diaz I, De La Torre E, Pujols J, Martín M, Inumaru S, 44. Rauth AM. The Physical State of Viral Nucleic Acid and the Sensitivity of Cano E, Domingo M, Montoya M, et al. Cytokine Profiles and Phenotype Viruses to Ultraviolet Light. Biophys J (1965) 5:257–73. doi: 10.1016/S0006- Regulation of Antigen Presenting Cells by Genotype-I Porcine Reproductive 3495(65)86715-7 and Respiratory Syndrome Virus Isolates. Vet Res (2011) 42:9. doi: 10.1186/ 45. Leifer CA, Medvedev AE. Molecular Mechanisms of Regulation of Toll-Like 1297-9716-42-9 Receptor Signaling. J Leukoc Biol (2016) 100:927–41. doi: 10.1189/ 28. Darwich L, Gimeno M, Sibila M, Diaz I, de la Torre E, Dotti S, et al. Genetic jlb.2mr0316-117rr and Immunobiological Diversities of Porcine Reproductive and Respiratory 46. Huang C, Zhang Q, Guo X-k, Yu Z-b, Xu A-T, Tang J, et al. Syndrome Genotype I Strains. Vet Microbiol (2011) 150:49–62. doi: 10.1016/ Porcine Reproductive and Respiratory Syndrome Virus Nonstructural j.vetmic.2011.01.008 Protein 4 Antagonizes Beta Interferon Expression by Targeting the NF- 29. Dı́az I, Gimeno M, Darwich L, Navarro N, Kuzemtseva L, Ló pez S, et al. B Essential Modulator. J Virol (2014) 88:10934–45. doi: 10.1128/ Characterization of Homologous and Heterologous Adaptive Immune jvi.01396-14 Responses in Porcine Reproductive and Respiratory Syndrome Virus 47. Shi X, Wang L, Li X, Zhang G, Guo J, Zhao D, et al. Endoribonuclease Infection. Vet Res (2012) 43:30. doi: 10.1186/1297-9716-43-30 Activities of Porcine Reproductive and Respiratory Syndrome Virus Nsp11 30. Li Y li, Darwich L, Mateu E. Characterization of the Attachment and Infection was Essential for Nsp11 to Inhibit IFN-b Induction. Mol Immunol (2011) by Porcine Reproductive and Respiratory Syndrome Virus 1 Isolates in Bone 48:1568–72. doi: 10.1016/j.molimm.2011.03.004 Marrow-Derived Dendritic Cells. Vet Microbiol (2018) 223:181–188. 48. Diebold SS, Kaisho T, Hemmi H, Akira S, Reis E Sousa C. Innate Antiviral doi: 10.1016/j.vetmic.2018.08.013 Responses by Means of TLR7-Mediated Recognition of Single-Stranded RNA. 31. Reed LJ, Muench H. A Simple Method of Estimating Fifty Per Cent Endpoints. Science (2004) 303:1529–31. doi: 10.1126/science.1093616 Am J Epidemiol (1938) 27:493–7. doi: 10.1093/oxfordjournals.aje.a118408 49. Saade G, Mé nard D, Hervet C, Renson P, Hue E, Zhu J, et al. Porcine 32. Reutner K, Leitner J, Müllebner A, Ladinig A, Essler SE, Duvigneau JC, et al. Reproductive and Respiratory Syndrome Virus Interferes With Swine CD27 Expression Discriminates Porcine T Helper Cells With Functionally Influenza a Virus Infection of Epithelial Cells. Vaccines (2020) 8:1–22. Distinct Properties. Vet Res (2013) 44:18. doi: 10.1186/1297-9716-44-18 doi: 10.3390/vaccines8030508 33. Puebla-Clark L, Parra-Sá nchez H, Resé ndiz M, Valenzuela O, Herná ndez J. 50. Stepanova K, Sinkora M. Porcine gd T Lymphocytes can be Categorized Into Tonsil Conventional Dendritic Cells are Not Infected by Porcine Reproductive Two Functionally and Developmentally Distinct Subsets According to and Respiratory Syndrome Virus. Virology (2019) 529:65–72. doi: 10.1016/ Expression of CD2 and Level of TCR. J Immunol (2013) 190:2111–20. j.virol.2019.01.012 doi: 10.4049/jimmunol.1202890 34. Calvert JG, Slade DE, Shields SL, Jolie R, Mannan RM, Ankenbauer RG, et al. CD163 Expression Confers Susceptibility to Porcine Reproductive and Respiratory Conflict of Interest: The authors declare that the research was conducted in the Syndrome Viruses. J Virol (2007) 81:7371–9. doi: 10.1128/JVI.00513-07 absence of any commercial or financial relationships that could be construed as a 35. Hoeve MA, Nash AA, Jackson D, Randall RE, Dransfield I. Influenza Virus a potential conflict of interest. Infection of Human Monocyte and Macrophage Subpopulations Reveals Increased Susceptibility Associated With Cell Differentiation. PloS One Copyright © 2021 Li and Mateu. This is an open-access article distributed under the (2012) 7:e29443. doi: 10.1371/journal.pone.0029443 terms of the Creative Commons Attribution License (CC BY). The use, distribution or 36. Barbour JD, Jalbert EC, Chow DC, Gangcuangco LMA, Norris PJ, Keating reproduction in other forums is permitted, provided the original author(s) and the SM, et al. Reduced CD14 Expression on Classical Monocytes and Vascular copyright owner(s) are credited and that the original publication in this journal is Endothelial Adhesion Markers Independently Associate With Carotid Artery cited, in accordance with accepted academic practice. No use, distribution or Intima Media Thickness in Chronically HIV-1 Infected Adults on reproduction is permitted which does not comply with these terms. Frontiers in Immunology | www.frontiersin.org 15 June 2021 | Volume 12 | Article 674185