TYPE Original Research PUBLISHED 17 January 2023 DOI 10.3389/fimmu.2022.1020227 Systemic CD4 cytotoxic T cells OPEN ACCESS improve protection against EDITED BY Nicolas Bertho, INRA Biologie, Épidémiologie et PRRSV-1 transplacental infection Analyse de Risque en santé animale (BIOEPAR), France Yanli Li 1*, Ivan Dı́az 2, Gerard Martı́n-Valls 1, Niklas Beyersdorf 3 REVIEWED BY Tobias Kaeser, and Enric Mateu 1* North Carolina State University, United States 1 Departament de Sanitat i Anatomia Animals, Facultat de Veterinària, Universitat Autònoma de Andrea Ladinig, Barcelona (UAB), Cerdanyola del Vallès, Spain, 2 Centre de Recerca en Sanitat Animal, Institut de University of Veterinary Medicine Recerca en Tecnologies Agroalimentáries (IRTA-CReSA), Universitat Autònoma de Barcelona (UAB), Vienna, Austria Cerdanyola del Vallès, Spain, 3 Institute for Virology and Immunobiology, University of Würzburg, Würzburg, Germany *CORRESPONDENCE Yanli Li yanli_li@med.unc.edu Enric Mateu enric.mateu@uab.cat Background: Porcine reproductive and respiratory syndrome virus (PRRSV) is one of the major swine pathogens causing reproductive failure in sows. SPECIALTY SECTION This article was submitted to Although modified-live virus (MLV) vaccines are available, only partial Comparative Immunology, protection against heterologous strains is produced, thus vaccinated sows a section of the journal can be infected and cause transplacental infection. The immune effector Frontiers in Immunology mechanisms involved are largely unknown. RECEIVED 16 August 2022 ACCEPTED 19 December 2022 PUBLISHED 17 January 2023 Methods: The present study investigated the role of cytotoxic lymphocytes, including cytotoxic T cells (CTL), NKT, and NK cells, from blood in preventing CITATION Li Y, Dı´az I, Martı´n-Valls G, PRRSV-1 transplacental infection in vaccinated primiparous sows (two doses Beyersdorf N and Mateu E (2023) vaccinated). Sows from a PRRSV-1 unstable farm were bled just before the last Systemic CD4 cytotoxic T cells improve protection against PRRSV-1 month of gestation (critical period for transplacental infection), then followed transplacental infection. to determine whether sows delivered PRRSV-1-infected (n=8) or healthy Front. Immunol. 13:1020227. (n=10) piglets. After that, functions of CTL, NKT, and NK cells in the two doi: 10.3389/fimmu.2022.1020227 groups of sows were compared. COPYRIGHT © 2023 Li, Dı´az, Martı´n-Valls, Beyersdorf and Mateu. This is an open-access Results: No difference was found through cell surface staining. But upon in article distributed under the terms of vitro re-stimulation with the circulating field virus, sows that delivered healthy the Creative Commons Attribution License (CC BY). The use, distribution piglets displayed a higher frequency of virus-specific CD107a + IFN-g- or reproduction in other forums is producing T cells, which accumulated in the CD4+ compartment including permitted, provided the original CD4 single-positive (CD4 SP) and CD4/CD8a double-positive (CD4/CD8a DP) author(s) and the copyright owner(s) are credited and that the original subsets. The same group of sows also harbored a higher proportion of publication in this journal is cited, in CD107a+ TNF-a-producing T cells that predominantly accumulated in CD4/ accordance with accepted academic practice. No use, distribution or CD8a double-negative (CD4/CD8a DN) subset. Consistently, CD4 SP and reproduction is permitted which does CD4/CD8a DN T cells from sows delivering healthy piglets had a higher not comply with these terms. virus-specific proliferative response. Additionally, in sows that delivered PRRSV-1-infected piglets, a positive correlation of virus-specific IFN-g response with average Ct values of umbilical cords of newborn piglets per litter was observed. Frontiers in Immunology 01 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 Conclusion: Our data strongly suggest that CTL responses correlate with protection against PRRSV-1 transplacental infection, being executed by CD4 T cells (IFN-g related) and/or CD4/CD8a DN T cells (TNF-a related). KEYWORDS CTL, NKT, NK, PRRSV-1, transplacental infection Introduction granzyme and perforin, or inducing death receptor-mediated apoptosis i.e. Fas ligand or TRAIL; 2) production of antiviral Porcine reproductive and respiratory syndrome virus cytokines, most notably IFN-g and TNF-a [reviewed by Barry (PRRSV) is a positive-sense single-stranded RNA virus causing and Bleackley (9)]. The classical cytotoxic cells are innate natural one of the costliest swine diseases worldwide. The virus causes killer (NK) cells and the adaptive CD8 cytotoxic T cells (10). reproductive failure in sows (i.e. late-term abortions, premature However, several studies have also uncovered the cytotoxic farrowing, stillbirths, and fetal mummification or weak-born function of CD4 T cells that kill infected cells through piglets) and increased mortality of suckling piglets in farrowing mechanisms similar to NK and CD8 T cells (11, 12). CD4 units (1, 2). When weaned piglets are infected, the infection cytotoxic lymphocytes (CD4 CTL) are MHC-II restricted, in manifests as a respiratory disease often complicated by contrast to classical CD8 CTL which are mediated by the MHC-I secondary agents (3). pathway (11, 12). Modified-live virus (MLV) vaccines are widely used for the The role of cytotoxic lymphocytes in protection against control of PRRSV infections in endemic farms. For naïve breeders, transplacental infection of PRRSV is largely unknown. In this vaccination usually starts before the first mating (1 or 2 doses), study, we followed a batch of primiparous sows and determined then recall vaccinations are administered periodically (often using two groups, sows transmitting the infection to fetuses and sows a blanket protocol, every 3-4 months) to maintain the immunity. that did not suffer transplacental infection. We then compared MLV vaccines are very effective to protect against homologous their CD4/CD8 CTL, NKT, and NK cells in blood through a strains; however, only partial efficacy is shown during range of assays, including cell surface staining, intracellular heterologous infections (4). Accordingly, the administration of staining of IFN-g and TNF-a, and proliferative responses to PRRSV MLV vaccines helps to reduce the abortion rate but does the field circulating virus. The results showed that upon in vitro not produce sterilizing immunity. re-stimulation, sows delivering healthy piglets had a higher Vilalta et al. (5) showed that, in PRRSV-2 affected farms, frequency of CD107a + IFN-g-producing T cells, which birth of viremic piglets could be detected up to 23 weeks after the accumulated within the CD4+ compartment including CD4 onset of an outbreak. It was reported that vertical transmission single-positive (CD4 SP) and CD4/CD8a double-positive occurred particularly in younger sows, although they had been (CD4/CD8a DP) subsets. Also, the same group of sows immunized and the inflow of new susceptible pigs was stopped. harbored a higher proportion of CD107a+ TNF-a-producing T Our group found that the percentage of sows delivering infected cells that mainly accumulated in the CD4/CD8a double- piglets may reach 5-10% in PRRSV-1-vaccinated and -endemic negative (CD4/CD8a DN) subset. The proliferative response farms in Spain, surprisingly, with no overt reproductive disease of CD4 SP and CD4/CD8a DN subsets to the virus was also (unpublished data). higher in sows that delivered healthy piglets. Our data strongly Previous studies suggested that homologous titers of suggest that CTL responses are correlated with the protection neutralizing antibodies (NAb) ≥1:8-1:16 might protect against against PRRSV-1 transplacental infection, being executed by abortion or viremia in piglets (6, 7). Unfortunately, NAb against CD4 T cells or CD4/CD8a DN cells. PRRSV usually have a narrow breadth and limited and unpredictable cross-neutralization efficacy (8). As a result, and given the high genetic/antigenic diversity of PRRSV, most natural Materials and methods exposures can be considered heterologous challenges. Protection in the absence of cross-neutralizing antibodies has been suggested Farm and sampling of gilts to be the result of cell-mediated immune responses (4). Cytotoxic lymphocytes play a key role in eliminating virus- Gilts that were followed in this study belonged to a 500-sow infected cells through two mechanisms: 1) direct cytolytic farm with farrowing batches every two weeks. No reproductive activity mediated by releasing cytolytic proteins such as failure above normal values was observed when the farm was Frontiers in Immunology 02 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 selected. Figure 1 shows the schematic of gilt management in the Both were diagnosed positive for PRRSV-1. The isolated virus farm. Specifically, gilts were purchased at 6-6.5 months of age was a strain with increased virulence corresponding to a cluster from a certified PRRSV-free source. Then gilts were allocated to recently reported (Genbank accession number ON571708). an isolated quarantine where they stayed for at least 6 weeks. All Whole-genome sequencing revealed that the virus was derived gilts were vaccinated with a PRRS MLV vaccine (Porcilis® PRRS, from the highly virulent Italian strain PR40 (13) with at least MSD Animal Health, Salamanca, Spain) upon entry into the three additional recombination events with other PRRSV-1 quarantine unit and were revaccinated 4 weeks later with the isolates (14). In the affected farm, 10% of sows suffered from same vaccine. reproductive failures, i.e. late abortions, premature farrowing, After quarantine, gilts were mated (insemination with fetal death, and the birth of congenitally infected piglets, since certified PRRSV-free semen) and kept in the mating units introduction of the new virulent strain. The impact of this virus until gestation was confirmed (day 35 of gestation). resulted in a decrease in the global production of the farm, from Afterwards, gilts were re-located to a gestation area and stayed 32 to 27 piglets/sow/year. In addition, mortality reached 20% in until day 105 of gestation. Finally, gilts were moved to the nurseries as a consequence of the outbreak, mostly caused by farrowing units. The farm housed pregnant sows with static secondary bacterial infections. groups. In other words, animals in the same batch were allocated to the same pen, where they could have contact with each other but physically separated from other batches of sows. Therefore, Isolation of peripheral blood the involved gilts only contacted with gilts in the same batch mononuclear cells (PBMC) before being moved to the farrowing units (day 105), where they interacted with sows from other batches (until the expected PBMC were separated by density-gradient centrifugation delivery date, days ~114-115 of gestation). Gilts were bled on day within 4 h after sampling. Blood was diluted 1:2 with Hank’s 70 of gestation, which was 35 days after moving to the gestation balanced salt solution (Fisher Scientific, Spain) and layered onto units and 35 days before transferring to the farrowing units. Histopaque 1.077 (Sigma-Aldrich, Spain) in SepMate tubes No clinical signs were observed before gilts were moved to (STEMCELL Technologies, France) before centrifugation (400 the farrowing units. Upon delivery, umbilical cords from the × g, 30 min) at room temperature. The cellular interface was newborns were collected for PRRSV diagnosis by RT-qPCR. retrieved, and red blood cells were lysed in 0.15M NH4Cl. Weaners (6-7 weeks of age) from other batches were also tested. Between each step, washing 3× with Dulbecco’s PBS (DPBS, FIGURE 1 Schematic of gilt management, sampling, and determination of transplacental infection. Frontiers in Immunology 03 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 Fisher Scientific, Spain) was done. Cells were counted and the after thawing, and 85.8 ± 14.3% and 85.0 ± 12.8% after rest. Cell viability was determined by trypan blue staining in a Neubauer number mentioned below means viable cells. chamber. Finally, cells were frozen in Cryostor CS10 (Sigma- Five-color staining (live/dead and CD3/CD4/CD8a/CD16) Aldrich, Spain) at approximately 20 × 106/vial (1 ml CS10) and was performed to define CTL, NKT, and NK cells. PBMC stored in liquid nitrogen until used. (500,000) were initially stained with LIVE/DEAD Near-IR dead cell stain (1 ml/ml, Thermo Fisher Scientific, Spain) on ice for 30 min. Then, cells were labeled with an anti-CD16 antibody (clone Determination of vertical transmission: FcG7, Bio-Rad, Spain) followed by anti-mouse IgG1 conjugated to sows delivering PRRSV-1-infected or BV421 (BioLegend, Spain), and finally with a cocktail of CD3-PE PRRSV-1-free piglets (clone BB23-8E6-8C8, BD Biosciences), CD4-FITC (clone 74-12-4, BD Biosciences, Spain), and CD8a-AF647 (clone 76-2-11, BD Umbilical cords of newborn piglets were collected to assess Biosciences, Spain). To further characterize NK cells, CD4-FITC the incidence of transplacental infection. Briefly, 3-5 cm of was replaced with NKp46-AF488 (clone VIV-KM1, Bio-Rad). At umbilical cords were aseptically collected in sterile DPBS least 100,000 cells were acquired using a MACSQuant Analyzer 10 (Fisher Scientific, Spain) immediately after the birth of piglets. (Miltenyi Biotec, Bergisch Gladbach, Germany). Once received in the laboratory, umbilical cords were sliced in In all assays of the present study, cell surface staining was sterile PBS. Within each litter, suspensions of umbilical cords performed in DPBS containing 5% FBS and 5% horse serum were pooled every two piglets. RNA was then extracted using (Sigma-Aldrich, Spain) for 30 min on ice. Between each step, MagMAX Pathogen RNA/DNA Kit (Thermo Fisher Scientific, cells were washed twice with DPBS (2% FBS). Antibodies Spain). Transplacental infection was determined when at least involved were titrated before use. Their specificities were one pool was positive in PRRSV-1 RT-qPCR (VetMAX PRRSV affirmed by comparing with the matched isotype antibodies. EU & NA 2.0 Kit, Thermo Fisher Scientific, Spain). Live lymphocytes were defined by forward/side scatter, height/ area forward scatter, and live/dead stain. Single staining of each fluorochrome was prepared for compensation. Positive events Isolation and propagation of the virus were gated using fluorescence minus one (FMO) controls. Analysis was performed using FCS Express 7 (de novo Sera from infected sows and piglets were used to isolate the Software, Glendale, CA, United States). circulating virus in porcine alveolar macrophages (AM). AM were seeded in 24-well plates (1 × 106/well, Sigma-Aldrich, Spain) and allowed to attach overnight. Then, cells were inoculated with the Cell Trace Violet proliferation assay RT-qPCR positive sera (250 ml/well) that had been diluted 1:5 with MEM containing 100 units/ml penicillin, 100 mg/ml streptomycin, PBMC were labeled with 5 mM CellTrace Violet (Thermo and 50 mg/ml gentamicin (all from Fisher Scientific, Spain). At 1.5 h Fisher Scientific, Spain) at 37°C for 20 min in the dark. After post-infection, the inoculum was washed away and fresh medium washing 3×, 250,000 cells/well were added to the 96-well round- containing 7.5% fetal bovine serum (FBS, Sigma-Aldrich, Spain) bottom plates (Sigma-Aldrich, Spain). Then, cells were cultured for was added. The isolated virus, designated as JA2, was then titrated 3 days with PRRSV-1 JA2 (multiplicity of infection (MOI) 0.1) in on AM with titer determined by the immunofluorescence assay quadruplicates (final volume 200 ml) in complete RPMI using an anti-PRRSV-1 nucleocapsid (N) antibody (clone 1C5H, supplemented with 10% FBS, 100 units/ml penicillin, and 100 mg/ Ingenasa, Spain). Further propagation of JA2 was performed in cell ml streptomycin. Mock (complete RPMI) and PHA (10 µg/ml) culture flasks (25 cm2, Sigma-Aldrich, Spain) with 10 × 106 stimulation were used as negative and positive controls, respectively. macrophages seeded. The 4th passage with titer 106.5 TCID50/ml Proliferation for 3 days was determined according to a kinetic study was used in the present study. The full genome of the virus was (1, 3, and 5 days), where the viability of lymphocytes on day 3 was sequenced using RNAseq Illumina Technology as reported around 45% and 5-10% of CD3+ T cells had proliferated upon JA2 before (15). stimulation, whereas after 5 days, < 20% of viable lymphocytes were remained (data not shown). At the end of incubation, cells from four wells were mixed and stained with Near-IR dead cell stain Flow cytometry analysis of CTL NKT followed by biotinylated anti-CD16 antibody (clone FcG7, BD NK phenotype Biosciences) that were revealed by streptavidin PerCP-Cy5.5 (BD Biosciences, Spain). A mixture of CD3-PE, CD4-FITC, and CD8a- Cryopreserved PBMC were thawed, rested overnight at 37° AF647 was finally added. At least 250,000 events were acquired. The C, then counted for use in the following experiments. The proportion of cells that had proliferated in the culture was viability of PBMC from sows that delivered infected and calculated with the following formula: (sum of cells in all healthy piglets was respectively 91.8 ± 3.1% and 92.9 ± 3.5% generations)/(total number of cells of the subset) × 100%. Frontiers in Immunology 04 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 IFN-g and TNF-a intracellular cytokine Results staining assay Examination of umbilical cords from the delivered piglets by After overnight rest, PBMC (250,000) were stimulated with RT-qPCR divided sows into two groups: one (n = 8) delivering PRRSV-1 JA2 (MOI 5.0) or mock (complete RPMI) stimulated PRRSV-1-infected piglets and the other (n = 10) giving birth to in the presence of CD107a-FITC (2 mg/ml, clone 4E9/11, Bio- PRRSV-1-free piglets (also mentioned as healthy piglets). Rad, Spain) and co-stimulatory monoclonal antibody anti-CD28 [1 mg/ml, clone 3D11 (16)] at a final volume of 200 ml. Cultures were produced in octuplicates and incubated for 8 h with Surface profiling of potential CTL, NKT, brefeldin A and monensin added for the final 4 h. After and NK cells did not detect differences incubation, cells were harvested by washing once with DPBS, between sows that delivered PRRSV-1- and then every four replicas were mixed to reach 1 × 106 cells for infected and healthy piglets staining in 96-well V-bottomed plates. Live cells were identified by using the fixable Near-IR staining as described above. We firstly examined the proportion of potential CTL, NKT, and Afterwards, cells were stained with a mixture of CD3-Pacific NK cells in PBMC obtained from blood of sows at the 10th week of Blue (clone PPT3, Bio-Rad), CD4-PE-Cy7 (clone 74-12-4, BD gestation through cell surface staining. Potential CTLs were Biosciences, Spain), and CD8a-PE (clone 76-2-11, BD considered to belong to the CD3+CD16– subset with further Biosciences, Spain). Next, cells were fixed and permeabilized classification based on CD4 and CD8a expression (gating (BD Cyfix/Cytoperm, BD Biosciences, Spain), and finally hierarchy shown in Figure 2). Potential NKT cells were defined as intracellularly stained with an anti-IFN-g-AF647 antibody CD3+CD16+CD8a+ (Figure 2). Statistically, no difference in the (clone P2G10, BD Biosciences, Spain) at 4°C for 30 min. For proportion of potential CTL or NKT was shown between sows with TNF-a staining, CD8a-PE was substituted by CD8a-AF647 and and without transplacental infection (Table 1), but a slightly higher an antibody anti-TNF-a (clone 103302, R&D Systems, Spain) frequency of CD4 single-positive T cells (CD4 SP, CD4+CD8a–) conjugated to LYNX Rapid RPE (Bio-Rad, Spain) was used. A and a slightly lower frequency of CD4/CD8a double-negative T minimum of 50 positive events were acquired for either IFN-g+ cells (CD4/CD8a DN) were seen in sows delivering healthy piglets or TNF-a+ in CD3+ cells. in the whole PBMC and also in T cells (CD3+) (Table 1). For NK cells, three subsets were defined by the expression of CD8a and NKp46 within CD3–CD16+ cells (Figure 2). They were IgG ELISPOT assay two classical CD8a-positive subsets CD8a+NKp46– and CD8a+ NKp46+, and a third subset assumed to have a higher cytotoxic PRRSV-specific IgG secreting cells (IgG-SC) were measured capacity CD8adim/–NKp46hi (17). Sows delivering infected piglets utilizing a commercial ELISPOT kit (Porcine IgG ELISPOT showed a higher frequency of CD8a+NKp46+ in CD3– cells (on BASIC, Mabtech, Nacka Strand, Sweden) according to the average 2.9 vs 1.8%) although the difference was not significant and manufacturer’s instructions. Briefly, nitrocellulose-bottomed was narrowed in the whole PBMC (0.5 vs 0.4%) (Table 1). plates (MultiScreen-HA plates, Merck, Spain) were coated overnight at 4°C with PRRSV-1 JA2 at MOI 0.1 in DPBS. The MOI was chosen according to a preliminary dose-response test. Sows that did not suffer transplacental PBMC were added at 500,000 cells/well (in 200 µl complete infection harbored a higher frequency RPMI) and left stimulated overnight before plate development. of degranulated IFN-g- and TNF-a- All tests were run in triplicates. The number of PRRSV-specific producing T cells that accumulated IgG-SC were calculated by subtracting counts of spot-forming in CD4 T cells and CD4/CD8a DN units (SFU) in unstimulated wells from counts in virus- T cells respectively stimulated wells. PRRSV-specific IgG-SC were expressed as SFU/106 cells. To investigate whether there was a functional divergence of cytotoxic lymphocytes between sows delivering PRRSV-1-free and -infected piglets, we performed an intracellular cytokine Statistical analysis (IFN-g and TNF-a) staining combined with degranulation marker CD107a and cell surface CD3, CD4, and CD8a. For All data were analyzed using the GraphPad Prism 9.3 sows delivering healthy piglets, two were not included in this software package (GraphPad Software, La Jolla, CA, United analysis due to low accessible cell number. Antigen-specific T States). The method of statistical tests applied to each data set cells were defined as IFN-g- or TNF-a-producing T cells, while is indicated in the figure legend. potential CTL were defined as degranulated antigen-specific T Frontiers in Immunology 05 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 FIGURE 2 Gating hierarchy of surface staining to identify potential CTL, NKT, and NK cells and their proliferation. The gating hierarchy is shown as contour plots. For phenotyping, a five-color staining scheme was performed: live/dead Near-IR/CD3/CD4/CD8a/CD16. After removal of cell debris and gating on singlets, live cells were divided based on the expression of CD3. Within CD3-positive cells, NKT cells were defined as CD16+CD8a+ and potential CTL as CD4/CD8a subsets within CD3+CD16– cells. In the CD3-negative population and after gating CD16+ cells, three subsets of NK cells were shown: CD8a+NKp46–, CD8a+NKp46+, and CD8adim/–NKp46hi. For proliferation assays, CellTrace Violet-labeled PBMC were stimulated with PRRSV-1 JA2 (MOI 0.1) for 3 days, then collected and stained for live/dead Near-IR/CD3/CD4/CD8a/CD16. The proliferated CTL and NKT were defined as CellTrace Violetdim within each subset. Stimulation with medium only or PHA was used as negative and positive controls, respectively. cells (18), namely cells with positive labeling for CD107a and (0.15 ± 0.07%) compared to their counterparts suffering IFN-g or TNF-a (Figure 3A). NKT cells were not included in transplacental infection (0.07 ± 0.04%) (p < 0.05) (Figure 3B). this analysis due to the limitation of fluorochrome combinations. This is by contrast with in CD3– cells where no difference was For total IFN-g-producing T cells upon PRRSV-1 JA2 shown between the two groups (Supplementary Figure 2A). stimulation, sows delivering healthy piglets showed a significantly Further phenotyping revealed that CD107a+ IFN-g-producing higher frequency than their counterparts delivering infected piglets T cells mostly accumulated in the CD4+ T compartment, (p < 0.05, Figure 3B). A similar tendency was also seen for TNF-a- including CD4 SP and CD4/CD8a DP (Figure 3D). The same producing T cells (p = 0.08, Figure 3C). Notably, in sows that pattern was seen in both groups of sows (without transplacental delivered infected piglets, 7/8 and 5/8 of them (marked as dashed infection: 40.8 ± 13.3% in CD4 SP and 44.3 ± 20.0% in CD4/ red circles) did not reach the positivity criteria (> 2-fold of mock- CD8a DP; with transplacental infection 41.5 ± 15.9% and 47.0 ± stimulated cells) for frequencies of IFN-g+ (Figure 3B) and TNF-a+ 11.8% respectively; non-significant) (Figure 3D). (Figure 3C) T cells, respectively. By contrast in the group that For CTL response with regards to TNF-a production did not suffer transplacental infection, only 2/8 of sows were off (CD107a+ TNF-a-producing T cells), sows that delivered healthy the positivity criteria. This suggests that sows suffering piglets also showed a higher frequency, on average 0.63 ± 0.24%, transplacental infection harbored much lower or even negligible contrasting with 0.38 ± 0.16% in sows suffering transplacental virus-specific T cell. Further combination with degranulation infection (p < 0.05) (Figure 3C). Responsive cells were marker CD107a improved the selection of virus-specific T cells, predominantly accumulated in the CD4/CD8a DN subset (48.6 where only 3/8 and 1/8 of sows delivering infected piglets were ± 12.9 vs 48.8 ± 7.2% of sows delivering healthy and infected piglets, off the positivity criteria for CD107a+IFN-g+ and CD107a+TNF-a+ respectively) (Figure 3E). Within CD3– population, a higher T cells, respectively. No difference was seen in the total proportion of CD107a+ TNF-a-producing cells was also detected proportion of CD107a+ T cells between two groups of sows in sows that gave birth to healthy piglets (Supplementary (Supplementary Figure 1). Figure 2B). But more than 90% of those cells were CD4/CD8a CTL response with regards to IFN-g production (CD107a+ DN, indicating they were not NK cells. IFN-g-producing T cells) to in vitro PRRSV-1 JA2 (MOI 5.0) On the individual level (considering all examined sows), the stimulation is shown in Figure 3B. Although CD107a+ IFN-g- frequency of virus-specific CTL responses showed a significant producing T cells were detected in both groups, sows that correlation between IFN-g- and TNF-a production (IFN-g = delivered healthy piglets showed a significantly higher level 0.19 + 0.015 TNF-a; R2 = 0.34, p = 0.02) (Figure 4). Comparing Frontiers in Immunology 06 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 Table 1 Percentage of T cell subsets, NKT, NK cells, and B cells. Sows Sows Sows Sows Sows Sows _infected piglets _healthy piglets _infected piglets _healthy piglets _infected piglets _healthy piglets In whole PBMC In CD3+ In CD3– T cells 64.0 ± 6.5 63.3 ± 6.3 – – – – CD4 SP 7.2 ± 3.5 9.8 ± 4.0 12.9 ± 6.8 16.7 ± 5.5 – – CD8a SP 21.0 ± 5.1 20.5 ± 3.0 40.8 ± 6.2 41.4 ± 3.3 – – CD4/CD8a DP 17.0 ± 8.2 16.5 ± 4.2 29.1 ± 12.2 28.5 ± 6.4 – – CD4/CD8a DN 9.6 ± 3.6 7.4 ± 3.3 17.2 ± 6.9 13.4 ± 6.1 – – NKT 3.8 ± 1.4 4.0 ± 1.3 5.9 ± 2.0 6.4 ± 1.7 – – NK 1.4 ± 0.7 1.2 ± 0.3 – – 7.7 ± 4.2 6.1 ± 1.8 – + CD8a NKp46 0.5 ± 0.3 0.5 ± 0.2 – – 2.8 ± 1.3 2.4 ± 1.4 + CD8a NKp46 + 0.5 ± 0.4 0.4 ± 0.2 – – 2.9 ± 2.1 1.8 ± 0.8 CD8adim/-NKp46hi 0.4 ± 0.2 0.4 ± 0.2 – – 2.0 ± 1.3 1.8 ± 1.0 B cells 19.0 ± 6.5 17.5 ± 5.4 – – – – No significant difference was found between the two groups of sows for all subsets. Whole PBMC: Acquired events with debris and dead cells excluded; T cells: CD3+; CD4/CD8a subsets: All subsets were gated within CD3+CD16–. CD4+CD8a– (CD4 single-positive, CD4 SP), CD4–CD8a+ (CD8a single-positive, CD8 SP), CD4+CD8a+ (CD4/CD8a double-positive, CD4/CD8a DP), CD4–CD8a– (CD4/CD8a double-negative, CD4/CD8a DN) subsets within CD3+CD16– population. NKT: CD3+CD16+CD8a+; NK: CD8a+NKp46–, CD8a+NKp46+, and CD8adim/–NKp46hi subsets within CD3–CD16+ population; B cells: CD21+. the two groups, there was a tendency towards higher responses No correlation was observed with virus-specific TNF-a for both cytokines in sows giving birth to healthy piglets (slope response (Figures 5C, D) or overall T cell degranulation (Figure 5E). regression 0.13 vs 0.10, Figure 4), although the difference was not statistically significant. This means the relationship between IFN-g and TNF-a responses was similar in both groups, but CD4 SP and CD4/CD8a DN T cells sows that did not suffer transplacental infection harbored a proliferated to a higher extent in sows higher strength of response. that delivered healthy piglets We then measured the proliferative response of Infection status of newborn piglets lymphocytes from each sow (gating hierarchy shown in correlated with the frequency of IFN-g- Figure 1). Upon in vitro re-stimulation with PRRSV-1 JA2 producing T cells in sows (MOI 0.1), CD4 SP and CD4/CD8a DN T cells proliferated to a higher extent in sows that delivered healthy piglets than their In the group of sows delivering JA2-infected piglets, we counterparts delivering JA2-infected fetuses (5.2% vs 1.6% for observed that average Ct values of newborn piglets from the CD4 SP, and 4.3% vs 2.3% for CD4/CD8a DN, p < 0.05) same littler showed a significant correlation with the frequency (Figure 6). Of note, the proliferated cells (CellTrace Violetdim) of total IFN-g-producing T cells (R2 = 0.55, p = 0.04) (Figure 5A) were mostly Near-IR+, namely dead cells (data not shown). but not with CD107a+ IFN-g-producing T cells (R2 = 0.30, p = This indicates a short half-life of the virus-specific cells in in 0.16) (Figure 5B) in the blood of sows. This suggests a higher vitro re-stimulated cultures, possibly caused by exhaustion or level of virus-specific IFN-g-producing T cells obtained from activation-induced cell death (AICD) (19). By contrast, CD8a MLV vaccination may contribute to alleviating infection severity SP and CD4/CD8a DP as well as NKT and NK cells rarely of fetuses when transplacental infection happens. proliferated in either group of sows (data not shown). Frontiers in Immunology 07 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 A B D C E FIGURE 3 Virus-specific IFN-g and TNF-a responses as measured by intracellular cytokine staining assays. PBMC from sows with or without the occurrence of transplacental infection were stimulated with PRRSV-1 JA2 strain in vitro for 8 h in the presence of anti-CD107a and anti-CD28 antibodies and with brefeldin A and monensin added for the final 4 h. After that, cells were stained for live/dead Near-IR, CD3/CD4/CD8a, and intracellular IFN-g or TNF-a. (A) Gating hierarchy to identify virus-specific IFN-g- and TNF-a-producing T cells with cell degranulation: exclusion of debris ! singlets ! live cells ! CD3+/CD3– ! total IFN-g+ or CD107a+IFN-g+, total TNF-a+ or CD107a+TNF-a+ ! CD4/CD8a; (B) Virus- specific IFN-g responses and (C) virus-specific TNF-a responses in T cells (CD3+). The graph shows the proportion of total IFN-g-producing and CD107a+ IFN-g-producing T cells, total TNF-a-producing and CD107a+ TNF-a-producing T cells in PRRSV-1 JA2-stimulated cultures. Sixteen sows (8 per group) were included. Each symbol, representing one sow, was the mean percentage of two replicas with background (cultures in the absence of PRRSV-1 JA2 stimulation) subtracted. Empty boxes (empty triangles) represent sows delivering infected fetuses; solid boxes (solid triangles) represent sows delivering healthy piglets. Symbols marked by dashed circles (in red) mean the frequency of IFN-g- or CD107a+ IFN-g- producing, or TNF-a- or CD107a+ TNF-a-producing T cells in JA2-stimulated cells was < 2-fold that of mock-stimulated cells; Allocation of virus-specific (D) CD107a+ IFN-g-producing T cells and (E) CD107a+ TNF-a-producing T cells to CD4/CD8a subsets. Statistical significance was measured by the Mann–Whitney nonparametric test. *p < 0.05. Frontiers in Immunology 08 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 FIGURE 4 Scatter plots showing the relationship between CD107a+ IFN-g- and CD107a+ TNF-a-producing T cells in sows delivering PRRSV-1 JA2- infected and healthy piglets. Dashed line (in red), linear regression line of the relationship between CD107a+ IFN-g-producing T cells and CD107a+ TNF-a-producing T cells considering all animals., Empty triangle, sows delivering infected piglets; solid triangle, sows delivering healthy piglets. Dashed line (in black), linear regression of sows delivering JA2-infected piglets; solid line (in black), linear regression of sows delivering healthy piglets. Slopes for the regression lines of healthy and infected animals were not significantly different, ns: not significant. A B C D E FIGURE 5 Correlation of virus-specific IFN-g, TNF-a, or CD107a T cells in sows with the average Ct values of umbilical cords of newborn piglets per litter. Average Ct values of umbilical cords from the same litter of newborn piglets were used to indicate the infection status of newborn piglets. Solid lines, linear regression of the correlation between Ct values of newborn piglets and PRRSV-1 JA2-specific (A) total IFN-g-producing or (B) CD107a+ IFN-g-producing T cells, (C) total TNF-a-producing or (D) CD107a+ TNF-a-producing T cells, and (E) total degranulated (CD107a+) T cells detected in each sow. Each solid circle represents one sow that delivered infected piglets. The value was the mean percentage of two replicas with background (cultures in the absence of PRRSV-1 JA2 stimulation) subtracted. Dashed lines, 95% confidence intervals; R2, fit goodness; p values, slope significance. Frontiers in Immunology 09 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 This, together with changes in the endometrial vascularization, may explain why transplacental infection happens only at the end of gestation (22). Although MLV vaccines help reduce the impact of PRRSV- associated reproductive problems, it is not uncommon to have vaccinated sows transmitting the virus to fetuses after exposure to a heterologous strain during gestation. The underlying mechanisms by which vaccinated sows remain susceptible or become resistant to transplacental heterologous infection are largely unknown. NAb were proposed to afford protection against abortion in the homologous challenge model (6). Unfortunately, NAb usually have a narrow neutralization breadth and lack broadly neutralizing capacity (8). Hence in our study, protection from transplacental infection in some sows FIGURE 6 Proliferative responses of T cells from sows that delivered was unlikely to arise from NAb, especially when the circulating infected or healthy piglets to PRRSV-1 JA2 stimulation. PBMC strain was highly virulent and genetically distant from the labeled with CellTrace Violet were stimulated in vitro with the circulating field virus at MOI 0.1. After 3 days, a five-color vaccine strain (>18% difference, at the whole genome level, in staining, live/dead Near-IR/CD3/CD4/CD8a/CD16, was our study). In an unpublished study from our group, sera from performed. In each T cell subset (CD4 SP, CD8a SP, CD4/CD8a gilts vaccinated with the same protocol were unable to neutralize DP, and CD4/CD8a DN), the proportion of CellTrace Violetdim cells was defined as the ratio of proliferation. Mock-stimulated any of the heterologous PRRSV-1 strains tested (n = 5). The cells were subtracted from the virus-stimulated cultures. Empty scarce capacity of vaccine-induced NAb for heterologous boxes, sows delivering JA2-infected piglets; grey-filled boxes, sows delivering healthy piglets. Statistical significance was infection suggests an essential role of cell-mediated immunity. measured by the Mann–Whitney nonparametric test, *p < 0.05, The primary goal of this study was to explore the role of ****p < 0.0001. virus-specific cytotoxic lymphocytes including CTL, NKT, NK cells in preventing PRRSV-1 transplacental infection. To this end, we followed a cohort of gilts during gestation and determined two groups of sows: the one delivering healthy piglets and the other delivering infected piglets. Using PBMC A higher frequency of virus-specific collected just before the last month of gestation, we compared IgG-SC was detected in sows delivering the frequency of virus-specific cytotoxic responses between two healthy piglets in ex vivo groups of sows. Gilts were kept in isolation during the IgG ELISPOT immunization phase with no contact with older sows before late gestation. This guaranteed the immunity against PRRSV-1 We also examined the frequency of memory B cells through at the moment of sampling was gained only from vaccination. A ex vivo IgG ELISPOT. The results showed that only sows lateral introduction of a new strain at the end of the gestation delivering healthy piglets showed apparent counts in the IgG allowed us to investigate the efficacy of immunity developed in ELISPOT (14.8 ± 9.3 vs 0.1 ± 0.4 SFU/106 cells) (Figure 7). Since sows against a heterologous virus. Considering fast transmission the ELISPOT was performed without expansion of Ag-specific to sows in the farrowing units, high rate of infection, and the fact B-cell clones, the SFU enumeration would correspond mostly to that mortalities surpassed 20% in both maternity and nursery plasma cells or long-lived plasma cells (20). phases, the new strain presented a worst-case scenario, a virulent easy-to-transmit PRRSV-1 strain. With all gilts sharing the same air space in the farrowing units, it is unlikely that gilts not deliver Discussion infected piglets were not exposed to the virus. But it is not assured whether the exposed gilts did not develop viremia or Transplacental infection of PRRSV has profound antigen-specific T cells blocked the transplacental infection. An implications in the field: the crucial mechanism for experimental model will be needed to clarify this point. maintaining the infection in herds and the leading cause of Initially, we did cell surface staining to examine overall increased mortality of suckling piglets [reviewed by Pileri and immune profile of different cell subsets, including potential Mateu (21)]. Transplacental infection takes place in late CTL, NKT, and NK cells, at 70 days of gestation. No gestation, after 80 days of pregnancy (22). Although the differences were found between the two groups of sows, precise causes are not fully known, it has been suggested that suggesting the outcome of exposure did not affect the overall CD163+ and sialoadhesin+ macrophages (PRRSV susceptible frequencies of immune cells in peripheral blood. But the use of cells) are enriched in the placenta at 80 days after gestation. distribution measurements in PBMC may differ from the result Frontiers in Immunology 10 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 kill the target cells in an MHC class II-restricted fashion. CD4 CTL have been mostly reported in chronic viral infections such as human cytomegalovirus and human immunodeficiency virus (26–28), but also in acute infections such as influenza when CD8 responses were impaired (29). A cytotoxic-Th1 cluster with high avidity and clonally expanded capability was also detected in COVID-19 patients who developed milder symptoms (30). The role of CD4 CTL in swine infectious diseases is largely unknown, although some clues could be traced from literatures. Lohse et al. (31) showed that depletion of CD8 cells did not exacerbate PRRSV infection. In another study (32), cytotoxic CD8+ T cells could not be detected in the first 56 days after inoculation and thereafter only in some animals at low intensity. These evidences suggest that classical CD8 CTL may not be the major actor in host defense against PRRSV. Rather, a higher level FIGURE 7 of CD4 CTL-skewed memory response gained from vaccination Frequencies of virus-specific IgG-secreting cells (IgG-SC) in would be crucial. In our study, a pivotal role of CD4 CTL in sows that delivered infected and healthy piglets. PBMCs (500,000 cells) were stimulated overnight in vitro with the protecting against transplacental infection was distinguished. circulating field virus JA2 at MOI 0.1 followed by IgG ELISPOT Whereas both groups of sows produced CD4 CTL after PRRSV- development. Statistical significance was measured by the 1 MLV prime-boost, sows harboring a higher frequency of CD4 Mann–Whitney nonparametric test, ***p < 0.001. CTL did not suffer transplacental infection. Therefore, we hypothesize that higher levels of PRRSV-specific CD4 CTL are of absolute counts in the whole blood. That should be considered crucial for sows to clear heterologous infection and to protect for future study. from transplacental infection. In the study of Kick et al. (33), We then performed the intracellular staining of IFN-g and CD4 T cells were also proved to be the major responder to TNF-a combined with the degranulation marker CD107a and PRRSV-2 in blood, but CD8 T cells responses were assumed to cell surface CD3, CD4, and CD8a. CD107a+ IFN-g-producing prevail in the lung. Therefore, the predominant T cell type T cells were detected in both groups of sows as expected after combating PRRSV infection is likely to be different between administration of the vaccine. Not surprisingly, sows that blood and tissues, indicating different roles of circulating and delivered PRRSV-1-free piglets showed a higher frequency. tissue-resident T cells. Also, more studies are required to resolve However, instead of the classical CD8 CTL, CD107a+ IFN-g- whether discrepancies exist in response to PRRSV-1 and producing T cells were mainly restricted to CD4+ T cells, PRRSV-2 or to strains with different virulence. Kick et al. (33, including both CD4/CD8a DP and CD4 SP subsets. 34) showed that PRRSV-2-vaccinated animals only responded to Correspondingly, CD4 SP displayed a higher proliferative the highly pathogenic strain. Moreover, the role of memory CTL response to the field virus in sows delivering healthy piglets, response (central/effector/stem-like memory) would be crucial whereas proliferation of CD8a SP was negligible. This agreed to uncover the development of host immunity against PRRSV. with a previous study, where the killing of autologous PRRSV- Two factors are assumed to drive the development of CD4 CTL, infected macrophages was carried out by CD4 but not CD8 T long-term exposure to viral antigens (35) and deficient or impaired cells (23). CD4/CD8 DP with CTL features were also detected CD8 CTL response (29). Coincidentally, a common feature of by Meier et al. (24) and Cao et al. (25), who additionally PRRSV infection is viral persistence in lymph nodes for an extended proposed the contribution by CD8 SP. Nevertheless, in Cao’s period after clearance of viremia. As demonstrated by Bierk et al. study, CD107a was rarely detected on CD8 SP (25), suggesting (36), transmission from infected sows may happen 86 days after the they were non-cytotoxic effectors. Also, Cao’s and Meier’s onset of infection. It is still unknown whether CD8 CTL could be studies used pigs infected by homologous strains, in contrast transiently induced and execute cross-cytotoxicity. In the future, it to sows exposed to a heterologous highly virulent strain in would be valuable to characterize the dynamic development of our study. CD4/CD8 effector and memory CTL elicited by PRRSV MLV In the past decade, CD4 CTL have been identified in vaccines, and the efficacy to protect homologous and heterologous human and mouse models, not only during viral responses infections. It is also critical to evaluate the presence of CD4 CTL in but also in anti-tumor and autoimmune responses [reviewed tissues at different phases as CD4 CTL are thought to reside mainly by Takeuchi and Saito (12)]. Like CD8 CTL, CD4 CTL in tissues (12). contribute to viral clearance via secreting antiviral cytokines Besides CD107a+ IFN-g-producing T cells, the second correlate and, in some conditions, secreting granzyme B and perforin to of protection observed in our study was CD107a+ TNF-a- Frontiers in Immunology 11 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 producing T cells. More than half of these cells accumulated in the further assessment of the role of nNAb in PRRSV. The results of CD4/CD8a DN subset, which contains mainly TCRgd+ T cells. As the present study suggest that protected sows were high responders has been shown in several species, CD4/CD8 DN T cells could compared to sows that suffered transplacental infection. Our group engage in innate and adaptive immune responses against different previously proved that increased IFN-g responses and higher levels pathogens, acting as CTL secreting TNF-a under certain conditions of NA in high responders to PRRSV-1 vaccination were related to [reviewed by (37)]. In the case of PRRSV, Ladinig et al. (38) reported the host genetic background (42). that higher levels of gd T cells were correlated with lower viral loads Taken together, our study demonstrated that differences in an experimental infection of gilts, although in that study fetal between sows delivering healthy and PRRSV-1-infected piglets mortality was not correlated. The role of gd T cells in the were mainly quantitative, not qualitative based on the fact that development of immunity against PRRSV was also proposed by PRRSV-1-specific T cell responses were detectable in both Olin et al. and Kick et al. (33, 34). groups of sows. The results highlighted the potential role of Notably, IFN-g and TNF-a CTL responses were significantly CD4 CTL and CD4/CD8a DN cells in protection against correlated in our study, suggesting both types of cells may transplacental infection. Because of the high genetic distance coordinate to protect sows from transplacental infection. Since between the vaccine and field-circulating strain (>18%), cross- the regression slopes were similar between two groups of sows, reactive T cell epitopes are assumed to be involved. It would be a sows with higher susceptibility to transplacental infection likely breakthrough if cross-reactive and promiscuous T-cell epitopes, represent a subpopulation of low responders with regards to the eliciting the IFN-g producing CD4 CTL and TNFa-secreting gd examined parameters. T cells, could be determined in the future. Exploration of newer Besides protecting sows from transplacental infection, virus- adjuvants promoting such responses is also demanded, as specific IFN-g-producing T cells were likely to reduce the viral differences in sows delivering PRRSV-1-free and -infected load reaching fetuses when transplacental infection happened. In piglets are mainly quantitative. Additionally, a promising the group of sows delivering infected piglets, the frequency of vaccine should consider the host’s genetic background. virus-specific IFN-g- but not TNF-a-producing T cells positively correlated with the Ct values of newborn piglets. The mechanism Data availability statement may relate to the reduction of virus produced in placenta, resulting then in less virus crossing placental barrier or largely The original contributions presented in the study are delay of the vertical transmission. In this case, no correlation included in the article/Supplementary Material. Further with virus-specific TNF-a-producing T cells was detected. But inquiries can be directed to the corresponding authors. given that only eight animals per group were examined, caution is advised in the interpretation of the results due to only factors with high impact could be detected in the present design. Ethics statement Although protection was very unlikely to arise from NA as The animal study was reviewed and approved by Committee mentioned above, it is worth mentioning that sows that did not for Human and Animal Experimentation of the Universitat transmit the virus to their offspring had a higher level of memory B Autònoma de Barcelona n° 5709. cells as detected by IgG ELISPOT. Since IgG ELISPOT was performed using virus-coated plates without expansion of pre- formed B cell clones, the result suggested an enhanced B cell Author contributions memory response to viral envelope proteins in the protected sows. Non-neutralizing antibodies (nNAb), induced by MLV YL and EM conceived the study, designed experiments, vaccination or natural infection, may not prevent PRRSV analyzed and interpreted data, wrote and edited the manuscript; infection or even enhance the severity of the disease. However, YL produced the virus, performed RT-qPCR and all flow cytometry their functional capabilities with regards to other mechanisms in assays; ID isolated PBMC and performed IgG ELISPOT; GM-V clearing PRRSV-infected cells are still largely unknown. For human contacted farms, sampled gilts, collected clinical data, performed viral pathogens, multifunctional nNAb have long been recognized RT-qPCR, and isolated the virus; NB produced anti-CD28 and studied in terms of antibody-dependent cell-mediated monoclonal antibody. EM supervised the work. All authors have cytotoxicity (ADCC), antibody-dependent cell-mediated approved the final version of the manuscript. phagocytosis (ADCP), complement-mediated cytotoxicity, and spatially blocking proteins for viral replication (39). Examples of Funding nNAb acting synergistically with T cells have also been reported (40, 41). This probably coincides with the observation in our study This work was supported by the Boehringer Ingelheim that enhanced virus-specific T cells along with a higher frequency of European PRRS Research Award 2019 “Role of cytotoxic T IgG-SC conferred heterosubtypic immunity to sows to protect lymphocytes in gilts after MLV vaccination in protection against transplacental infection. It would be interesting to perform a vertical transmission of PRRSV” and the Veterinary Diagnostic Frontiers in Immunology 12 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 Laboratory Service of Universitat Autònoma de Barcelona (UAB). Publisher’s note Anti-CD28 antibody production was supported by SFB1525 Cardio-immune Interfaces (DFG 453989101)-project A2. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated Acknowledgments organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or Our thanks to the farm owners and workers and Dr. Joan claim that may be made by its manufacturer, is not guaranteed Aparicio for assisting in the selection of farms and collection or endorsed by the publisher. of samples. Conflict of interest Supplementary material The authors declare that the research was conducted in the The Supplementary Material for this article can be found absence of any commercial or financial relationships that could online at: https://www.frontiersin.org/articles/10.3389/ be construed as a potential conflict of interest. fimmu.2022.1020227/full#supplementary-material References 1. Hopper SA, White ME, Twiddy N. An outbreak of blue-eared pig disease 14. Martı́n-Valls G, Cortey M, Allepuz A, Illas F, Tello M, Mateu E. Description (porcine reproductive and respiratory syndrome) in four pig herds in great Britain. of a new clade within subtype 1 of betaarterivirus suid 1 causing severe outbreaks in Vet Rec (1992) 131:140–4. doi: 10.1136/vr.131.7.140 Spain. Microbiol Resour Announc (2022) 11:e0030422. doi: 10.1128/mra.00304-22 2. Done SH, Paton DJ. Porcine reproductive and respiratory syndrome: clinical 15. Cortey M, Arocena G, Pileri E, Martı́n-Valls G, Mateu E. Bottlenecks in the disease, pathology and immunosuppression. Vet Rec (1995) 136:32–5. doi: 10.1136/ transmission of porcine reproductive and respiratory syndrome virus (PRRSV1) to vr.136.2.32 naïve pigs and the quasi-species variation of the virus during infection in 3. Saade G, Deblanc C, Bougon J, Marois-Cré han C, Fablet C, Auray G, et al. vaccinated pigs. Vet Res (2018) 49:107. doi: 10.1186/s13567-018-0603-1 Coinfections and their molecular consequences in the porcine respiratory tract. Vet 16. Uehlein S, Ding X, Flößer J, Schmidt S, Steitz J, Bille M, et al. Human-like Res (2020) 51:80. doi: 10.1186/s13567-020-00807-8 response of pig T cells to superagonistic anti-CD28 monoclonal antibodies. J 4. Dı́az I, Darwich L, Pappaterra G, Pujols J, Mateu E. Different European-type Immunol (2021) 207:2473–88. doi: 10.4049/jimmunol.2100174 vaccines against porcine reproductive and respiratory syndrome virus have 17. Mair KH, Müllebner A, Essler SE, Duvigneau JC, Storset AK, Saalmüller A, different immunological properties and confer different protection to pigs. et al. Porcine CD8adim/-NKp46high NK cells are in a highly activated state. Vet Virology (2006) 351:249–59. doi: 10.1016/j.virol.2006.03.046 Res (2013) 44:13. doi: 10.1186/1297-9716-44-13 5. Vilalta C, Sanhueza J, Alvarez J, Murray D, Torremorell M, Corzo C, et al. 18. Betts MR, Brenchley JM, Price DA, De Rosa SC, Douek DC, Roederer M, Use of processing fluids and serum samples to characterize porcine reproductive et al. Sensitive and viable identification of antigen-specific CD8+ T cells by a flow and respiratory syndrome virus dynamics in 3 day-old pigs. Vet Microbiol (2018) cytometric assay for degranulation. J Immunol Methods (2003) 281:65–78. 225:149–56. doi: 10.1016/j.vetmic.2018.09.006 doi: 10.1016/S0022-1759(03)00265-5 6. Osorio FA, Galeota JA, Nelson E, Brodersen B, Doster A, Wills R, et al. 19. Green DR, Droin N, Pinkoski M. Activation-induced cell death in T cells. Passive transfer of virus-specific antibodies confers protection against reproductive Immunol Rev (2003) 193:70–81. doi: 10.1034/j.1600-065X.2003.00051.x failure induced by a virulent strain of porcine reproductive and respiratory 20. Ionescu L, Urschel S. Memory b cells and long-lived plasma cells. syndrome virus and establishes sterilizing immunity. Virology (2002) 302:9–20. Transplantation (2019) 103:890–8. doi: 10.1097/TP.0000000000002594 doi: 10.1006/viro.2002.1612 21. Pileri E, Mateu E. Review on the transmission porcine reproductive and 7. Lopez OJ, Oliveira MF, Alvarez Garcia E, Kwon BJ, Doster A, Osorio FA. respiratory syndrome virus between pigs and farms and impact on vaccination. Vet Protection against porcine reproductive and respiratory syndrome virus (PRRSV) Res (2016) 47:108. doi: 10.1186/s13567-016-0391-4 infection through passive transfer of PRRSV-neutralizing antibodies is dose dependent. Clin Vaccine Immunol (2007) 14:269–75. doi: 10.1128/CVI.00304-06 22. Karniychuk UU, Nauwynck HJ. Pathogenesis and prevention of placental and transplacental porcine reproductive and respiratory syndrome virus infection. 8. Martı́nez-Lobo FJ, Dı́ez-Fuertes F, Simarro I, Castro JM, Prieto C. Porcine Vet Res (2013) 44:95. doi: 10.1186/1297-9716-44-95 reproductive and respiratory syndrome virus isolates differ in their susceptibility to neutralization. Vaccine (2011) 29:6928–40. doi: 10.1016/j.vaccine.2011.07.076 23. Chung CJ, Cha SH, Grimm AL, Ajithdoss D, Rzepka J, Chung G, et al. Pigs that recover from porcine reproduction and respiratory syndrome virus infection 9. Barry M, Bleackley RC. Cytotoxic T lymphocytes: All roads lead to death. Nat develop cytotoxic CD4 + CD8 + and CD4 + CD8 - T-cells that kill virus infected Rev Immunol (2002) 2:401–9. doi: 10.1038/nri819 cells. PloS One (2018) 13:e0203482. doi: 10.1371/journal.pone.0203482 10. Uzhachenko RV, Shanker A. CD8+ T lymphocyte and NK cell network: 24. Meier WA, Galeota J, Osorio FA, Husmann RJ, Schnitzlein WM, Circuitry in the cytotoxic domain of immunity. Front Immunol (2019) 10:1906. Zuckermann FA. Gradual development of the interferon-g response of swine to doi: 10.3389/fimmu.2019.01906 porcine reproductive and respiratory syndrome virus infection or vaccination. 11. Swain SL, McKinstry KK, Strutt TM. Expanding roles for CD4 + T cells in Virology (2003) 309:18–31. doi: 10.1016/S0042-6822(03)00009-6 immunity to viruses. Nat Rev Immunol (2012) 12:136–48. doi: 10.1038/nri3152 25. Cao QM, Tian D, Heffron CL, Subramaniam S, Opriessnig T, Foss DL, et al. 12. Takeuchi A, Saito T. CD4 CTL, a cytotoxic subset of CD4+ T cells, their Cytotoxic T lymphocyte epitopes identified from a contemporary strain of porcine differentiation and function. Front Immunol (2017) 8:194. doi: 10.3389/ reproductive and respiratory syndrome virus enhance CD4+CD8+ T, CD8+ T, and fimmu.2017.00194 gd T cell responses. Virology (2019) 538:35–44. doi: 10.1016/j.virol.2019.09.006 13. Canelli E, Catella A, Borghetti P, Ferrari L, Ogno G, De Angelis E, et al. 26. Van Leeuwen EMM, Remmerswaal EBM, Heemskerk MHM, Ten Berge Phenotypic characterization of a highly pathogenic Italian porcine reproductive IJM, Van Lier RAW. Strong selection of virus-specific cytotoxic CD4+ T-cell clones and respiratory syndrome virus (PRRSV) type 1 subtype 1 isolate in experimentally during primary human cytomegalovirus infection. Blood (2006) 108:3121–7. infected pigs. Vet Microbiol (2017) 210:124–33. doi: 10.1016/j.vetmic.2017.09.002 doi: 10.1182/blood-2006-03-006809 Frontiers in Immunology 13 frontiersin.org Li et al. 10.3389/fimmu.2022.1020227 27. Appay V, Zaunders JJ, Papagno L, Sutton J, Jaramillo A, Waters A, et al. cytomegalovirus infection. J Immunol (2004) 173:1834–41. doi: 10.4049/ Characterization of CD4 + CTLs ex vivo. J Immunol (2002) 168:5954–8. jimmunol.173.3.1834 doi: 10.4049/jimmunol.168.11.5954 36. Bierk MD, Dee SA, Rossow KD, Otake S, Collins JE, Molitor TW. 28. Aslan N, Yurdaydin C, Wiegand J, Greten T, Ciner A, Meyer MF, et al. Transmission of porcine reproductive and respiratory syndrome virus from Cytotoxic CD4+ T cells in viral hepatitis. J Viral Hepat (2006) 13:505–14. persistently infected sows to contact controls. Can J Vet Res (2001) 65:261–6. doi: 10.1111/j.1365-2893.2006.00723.x 37. Wu Z, Zheng Y, Sheng J, Han Y, Yang Y, Pan H, et al. CD3+CD4-CD8- 29. Zhou X, McElhaney JE. Age-related changes in memory and effector T cells (Double-negative) T cells in inflammation, immune disorders and cancer. Front responding to influenza A/H3N2 and pandemic A/H1N1 strains in humans. Immunol (2022) 13:816005. doi: 10.3389/fimmu.2022.816005 Vaccine (2011) 29:2169–77. doi: 10.1016/j.vaccine.2010.12.029 38. Ladinig A, Gerner W, Saalmüller A, Lunney JK, Ashley C, Harding JCS. 30. Lipsitch M, Grad YH, Sette A, Crotty S. Cross-reactive memory T cells and Changes in leukocyte subsets of pregnant gilts experimentally infected with porcine herd immunity to SARS-CoV-2. Nat Rev Immunol (2020) 20:709–13. doi: 10.1038/ reproductive and respiratory syndrome virus and relationships with viral load and s41577-020-00460-4 fetal outcome. Vet Res (2014) 45:128. doi: 10.1186/s13567-014-0128-1 31. Lohse L, Nielsen J, Eriksen L. Temporary CD8+ T-cell depletion in pigs does 39. Excler JL, Ake J, Robb ML, Kim JH, Plotkin SA. Nonneutralizing functional not exacerbate infection with porcine reproductive and respiratory syndrome virus antibodies: A new “old” paradigm for HIV vaccines. Clin Vaccine Immunol (2014) (PRRSV). Viral Immunol (2004) 17:594–603. doi: 10.1089/vim.2004.17.594 21:1023–36. doi: 10.1128/CVI.00230-14 32. Costers S, Lefebvre DJ, Goddeeris B, Delputte PL, Nauwynck HJ. Functional 40. Richter K, Oxenius A. Non-neutralizing antibodies protect from chronic impairment of PRRSV-specific peripheral CD3+CD8 high cells. Vet Res (2009) LCMV infection independently of activating FcgR or complement. Eur J Immunol 40:46. doi: 10.1051/vetres/2009029 (2013) 43:2349–60. doi: 10.1002/eji.201343566 33. Kick AR, Amaral AF, Cortes LM, Fogle JE, Crisci E, Almond GW, et al. The 41. Laidlaw BJ, Decman V, Ali MAA, Abt MC, Wolf AI, Monticelli LA, et al. T-cell response to type 2 porcine reproductive and respiratory syndrome virus Cooperativity between CD8+ T cells, non-neutralizing antibodies, and alveolar (PRRSV). Viruses (2019) 11:796. doi: 10.3390/v11090796 macrophages is important for heterosubtypic influenza virus immunity. PloS 34. Olin MR, Batista L, Xiao Z, Dee SA, Murtaugh MP, Pijoan CC, et al. gd Pathog (2013) 9:e1003207. doi: 10.1371/journal.ppat.1003207 lymphocyte response to porcine reproductive and respiratory syndrome virus. 42. Ait-Ali T, Dı́az I, Soldevila F, Cano E, Li Y, Wilson AD, et al. Distinct Viral Immunol (2005) 18:490–9. doi: 10.1089/vim.2005.18.490 functional enrichment of transcriptional signatures in pigs with high and low 35. van Leeuwen EMM, Remmerswaal EBM, Vossen MTM, Rowshani AT, IFN-gamma responses after vaccination with a porcine reproductive and Wertheim-van Dillen PME, van Lier RAW, et al. Emergence of a CD4 + CD28 – respiratory syndrome virus (PRRSV). Vet Res (2016) 47:104. doi: 10.1186/ granzyme b +, cytomegalovirus-specific T cell subset after recovery of primary s13567-016-0392-3 Frontiers in Immunology 14 frontiersin.org