Male Holstein × Friesian cattle at 3 months of age were used, animals were tested for prior exposure to F. hepatica before experimental infections using both serological analysis and fecal egg counts (FECs). Animals were infected with 250 metacercariae from Ridgeway Scientific and maintained until 13 weeks of postinfection. Throughout the experiment, animals had ad libitum access to feed and water. After, the infection period animals were euthanized by stunning with a captive bolt and exsanguination. Blood was collected weekly and, at PM, lymph nodes were removed and collected into sterile PBS.

Parasite-specific LFH IgG1 was determined as previously described using LFH (10 µg/mL) as a capture antigen (9), plates were probed with serially diluted serum samples. Thereafter, bound IgG1 was detected using mouse-anti-bovine IgG1:HRP (Abdserotec clone IL-A60) and TMB. Additionally, entire livers were removed and examined to determine the final parasite burden, parasites were determined to be mature or immature on the basis of body shape and the development of the “shoulder” area of the fluke as previously described (10, 11). FECs were performed on 1 g of feces using the standard sedimentation technique.

Experiments were conducted under Home Office License PPL 30/3239 in accordance with Animal Scientific Procedures Act 1986 and ethical approval for the study was provided by the University of Nottingham SVMS and ADAS UK Animal Welfare and Ethics Review Committee. Animal involved in experiments had their suffering minimized and their welfare was as outlined in the guidance provided by EU Directive 2010/63/EU.

Single cell suspensions were prepared from the lymph nodes by forcing through a 70 µm cell strainer, thereafter, red blood cells were lysed with ammonium chloride lysis buffer. Cells were washed twice in 1× PBS and resuspended at 10⁷ cells/mL for cell separation. Miltenyi magnetic cell separation was used, as per the manufacturer’s instructions, following a primary incubation with the mouse anti-bovine CD4 monoclonal antibody (AbDserotec—clone CC30; mouse IgG1) at 1/50 dilution followed by incubation with secondary Miltenyi magnetic beads coated with anti-mouse IgG as per manufacturer’s instructions. Purified cells were cultured at a final density of 2 × 10⁶/mL and stimulated with endotoxin-free total adult antigen extract (LFH) at 10 µg/mL or anti-bovine CD3 (VMRD—clone MM1A) that had been coated onto tissue culture plates at 1 µg/mL overnight. Where indicated, anti-IL-10 (AbDserotec—clone CC320) (12, 13) or anti-TGF (R&D Systems—polyclonal AB-100-NA) (14) were used at 1 µg/mL, alternatively cells were cultured in the presence of recombinant bovine IL-2 (Kingfisher Biotech RP0026B at the indicated concentrations), all cultures were in RMPI-1640 with 10% heat-inactivated fetal calf serum (FCS), penicillin (100 U/mL), and streptomycin (100 µg/mL).

For coculture, experiments requiring PBMCs cells were purified as before (8); PBMCs were subsequently stored in liquid nitrogen in 80% FCS/20% DMSO. After thawing, cells were washed immediately twice in complete RMPI prior to separation of CD4 T-cells as above. Cells were then cocultured together in the indicated ratios prior to analysis for either IFN-γ or proliferation using the MTT assay (Sigma-Aldrich).

Single cell suspensions were stained with combinations of antibodies indicated below for 25 min at 4°C protected from the dark. Cells were acquired on a BD FACS Aria (BD Biosciences, San Jose, CA, USA) and gated on forward and side scatter to exclude damaged cells and debris. Cells were stained with combinations of the following; anti-bovine CD4 (Alexa Fluor^® 647 clone CC8, mouse IgG2a, AbDserotec), anti-bovine CD25 (RPE clone ILA111, mouse IgG AbDserotec), anti-bovine WC1 (FITC clone CC15, AbDserotec), anti-bovine Foxp3 (APC FJK-16s, rat IgG2a, eBioscience), anti-bovine CD62L (FITC clone CC32 AbDserotec), and anti-bovine CD45RO (RPE clone IL-A116, AbDserotec). Appropriate isotype and unstained controls were captured. Data were analyzed using FlowJo version 7.6.1 analysis software (TreeStar, San Carlos, CA, USA).

Supernatants were tested for the presence of cytokines using commercial ELISAs as follows; TGF-β (Promega-G7591), IL-4 (Thermofisher-ESS0031), IFN-γ (MABtech-3119-1H-6), IL-5 (LSBio LS-F6133), IL-13 (Kingfisher Biotech VS0148B), and IL-2 (Kingfisher Biotech DIY1100B). For the detection of IL-10, IL-10 capture antibody (AbDserotec, clone CC318) was used at 6 µg/mL and detection IL-10 antibody (AbDserotec, biotin-conjugated clone CC320) was used at 2 µg/mL as previously described (15), with the inclusion of recombinant IL-10 as a standard curve. All assays were carried out as per manufacturer’s instructions with optical densities captured on a LT-4000 microplate reader (Labtech, UK) with standard curves used for quantification throughout.

To measure PD-L1 and CTLA4 via qPCR, RNA was isolated from freshly prepared cells and converted to cDNA using the GoScript Reverse Transcription Kit (Promega). Primers for pdcd1 and ctla4 (Applied Biosystems #4351372 and #4351372) and conditions used are as previously reported (16). Assays were run in quadruplicate on a Roche Lightcycler II, data were analyzed using the ΔΔCT method to determine fold expression changes relative to rpl19 (Applied Biosystems #4331182), which has previously been shown to be a suitable housekeeping gene for immunological analysis (17).

Data were plotted within GraphPad Prism (La Jolla, CA, USA), a P-value cutoff of 0.05 was taken as significant, and details of specific tests used are given in the relevant figure legends.

Preinfection examination of the animals revealed neither specific anti-Fasciola IgG1 antibodies (Figure 1A) nor parasite eggs present within feces (Figure 1B). However, as infection progressed, the development of parasite-specific IgG1 became clear at week 2 postinfection with titers reaching a plateau by week 8 and remaining at these levels until completion of the trial (Figure 1A). Similarly, when animals were examined for the presence of parasite eggs in the feces, we found all animals to be egg positive, with a range of 2–13 eggs/g and a median count of 5 (Figure 1B). At postmortem, all livers were examined for the presence of infiltrating parasites in the bile ducts, gall bladder, and throughout the tissues; total parasite numbers and both immature and mature forms were counted (Figure 1C). All animals harbored parasites in their liver at examination; the range of total parasites recovered was 28–88, while all animals had greater numbers of mature parasites (range: 21–72) compared to immature parasites (range: 7–28; Figure 1C).

Previous studies of the bovine cytokine response have either focused on the systemic response or used a transcriptomics approach to whole tissues. Using a well-established model of experimental infection, we collected the MLN and HLN from infected animals after the development of patency. CD4 T-cells were isolated by magnetic bead selection and cultured for 72 h in the presence of anti-CD3 or LFH; thereafter, supernatants were collected for cytokine ELISA. In both the infected MLN (Figure 2A, right panel) and HLN (Figure 2B, right panel), there was no significant IFN-γ response either to CD3 or LFH; only CD3 stimulation of the MLN CD4 T-cells produced some detectable cytokine. This is in contrast to the findings in naïve tissues, where T-cells from both the MLN and HLN produced significant IFN-γ after anti-CD3, but not LFH stimulation (Figures 2A,B, left panels).

Of the type-2 cytokines examined, IL-5 was the most abundant at this point with CD3 stimulation but not LFH in the MLN and both CD3 and LFH stimulation within the HLN. In contrast, the naïve tissues produced IL-5 in response to anti-CD3 but not LFH; while IL-13 was non-detectable in naïve HLN T-cells low but variable levels were present in naïve MLN T-cells (Figures 2A,B, left panels). IL-13 was not significantly elevated except in response to LFH stimulation in the HLN. Indeed, only IL-10 and TGF-β were readily detectable in both the MLN and HLN. IL-10 was significantly upregulated within both the MLN and HLN in response to both CD3 and LFH, and for these responses in the HLN parasite specific IL-10 was more readily produced in comparison to CD3. TGF-β was spontaneously produced in PBS cultures in levels comparable to those produced by LFH. CD3-induced TGF-β was significantly raised, in comparison to both PBS and LFH, within the MLN but not the HLN (Figure 2B). IL-4 was not detectable in response to either CD3 or LFH at either site (data not shown). Overall, this would suggest that, even in the presence of a current infection, the draining HLN presents a strong immunoregulatory pattern with minimal type-2 cytokines.

We next examined CD4 T-cells from the HLN of infected animals, overall, there was a significant increase in the total number of cells from infected animals (95.2 ± 26.5 × 10⁶ cells; P < 0.05) compared with naïve controls (52.9 ± 20.4 × 10⁶ cells). The percentage of CD4 cells expressing CD25, as a marker of activation was 90 vs 30% within the controls (Figure 3A, left panel), in parallel, we saw a decrease in the percentage of naïve CD62L⁺ CD4 T-cells, an average of 44% in controls vs 6% from infected animals (Figure 3A, right panel).

To determine if the CD4 pool in the lymph node of infected animals was committed to a memory phenotype, we examined the CD4⁺CD62L⁺ subpopulation for CD45RO expression. There was no difference in CD45RO expression between infected and naïve animals (Figure 3B, left panel), 5.4 ± 1.8 and 4 ± 1% of the CD4⁺CD62L⁺ cells, respectively, suggesting no change in the commitment of T-cells to the central memory phenotype. Similarly, gating on the CD62L⁻ pool demonstrated that 3.2 ± 1% of naïve CD4 T-cells were also CD45RO⁺ while the corresponding population in infected animals was 1.7 ± 0.5%; demonstrating the lack of an memory cell expansion (Figure 3B, right panel). We stained the CD4⁺CD25⁺ T-cell population for Foxp3, we found only 5.8 ± 1.2% of infected animals, and 5.1 ± 0.9% of naïve animals, harbored traditional Foxp3 Tregs (Figure 3C, left panel). Indeed, we found that almost 50% (±6%) of γδ (WC1⁺) T-cells were Foxp3⁺ in infected animals compared with approximately half that proportion in naïve animals (Figure 3C, right panel). Thus, it would appear that the lack of proliferation and cytokine production in infected CD4 T-cells is not due to the presence of Foxp3⁺ CD4 T-cells.

IL-2 is a limiting factor in the responsiveness of lymphocytes, to determine if IL-2 played a role in the observed pattern of responses, we repeated the above CD4⁺ T-cell isolation and culture in the presence of LFH or α-CD3 stimulation. Cultures were sampled over a 72 h period, and Figure 4A shows that neither CD3 nor LFH drive IL-2 secretion but that naïve animals respond to CD3 stimulation with a steady accumulation of IL-2. Next, we conducted qPCR to determine if markers of anergy, ctla4, and pdcd1 were upregulated. Both ctla4 and pdcd1 are upregulated in CD4 T-cells derived from animals 13 weeks of postinfection compared to those sampled at 2 weeks of postinfection or naïve controls (Figure 4B).

To test if the anergy could be reversed we repeated the above in the presence of IL-2, α-IL-10 (clone CC320), or α-TGF-β (AB-100-NA). While IFN-γ secretion was restored in response to LFH and CD3 with exogenous IL-2, neither α-IL-10 nor α-TGF-β had a significant effect on IFN-γ (Figure 4C). Conversely, IL-13 production was restored, compared to controls, in the presence of all three treatments during LFH and CD3 stimulation. Here, a graded effect was evident with an increasing capacity of IL-2, α-IL-10, and α-TGF-β treatments in restoring IL-13 production (Figure 4C). To determine if the mechanisms controlling cytokine production and proliferation were linked, we examined antigen-specific proliferation, by MTT assay, in the presence of the IL-2, α-IL-10, and α-TGF-β (Figure 4D). There was a high degree of variation in the responses of animals to these treatments. α-IL-10 treatment succeeded in significantly restoring the capacity of cells to respond to CD3 stimulation; while IL-2 or α-TGF-β had some effect this was not significant. Similarly, when stimulated with LFH, there was a small, but significant increase, in proliferation with α-IL-10 while neither IL-2 nor α-TGF-β had any effect on proliferation.

The presence of these anergic T-cells may impact upon host responses to reinfection, or upon future vaccination. To test if the anergic T-cells could impact on normally responsive T-cells, we isolated CD4 T-cells from circulation at week 2 postinfection, when F. hepatica-infected animals are normally responsive, and cocultured these with anergic cells in the presence of LFH. While week 2 CD4 T-cells could respond normally, in the presence of week 13 cells, this response was abrogated (Figure 5A).

Addition of α-IL-10 resulted in an enhanced response in week 2 cells in response to LFH, suggesting that there was inhibition of low-level paracrine IL-10, and an increase in the week 13 only cells LFH-specific response (Figure 5B), consistent with our findings above. When α-IL-10 was added to the coculture, there was a reversal of the inhibitory effect of week 13 CD4 T-cells, and the IL-10 they produce, on week 2 CD T-cells. Neutralization of IL-10 in the cocultures of these cells resulted in a threefold increase in proliferation. Figure 5B demonstrating that pre-existing anergic T-cell can be a major contributor of IL-10, altering the response of freshly responding non-anergic T-cells. This same trend is visible in terms of IL-2 production and utilization. Addition of exogenous IL-2 in the same system had effects on week 2 only CD4 cells while no effect was apparent on week 13 CD4 T-cells. Figure 5B in our coculture system, the effect of IL-2 addition was weaker compared with IL-10 neutralization as a small but not significant effect was seen on LFH-specific proliferation (Figure 5C); suggesting that proliferative hyporesponsiveness in this setting is mediated by IL-10. To test if the dissociation seen in the actions of IL-2 and IL-10 extended to cytokine production, we examined the secretion of IFN-γ. Figure 5C clearly shows that IL-2 has a more profound effect on IFN-γ secretion in cocultured week 2 and week 13 cells when compared with α-IL-10 treatment; demonstrating that the cell-intrinsic effects we found in chronically infected animals can be influential on normally responsive week 2 cells.