Prostaglandin E₂-Induced Immune Suppression via Cytotoxic T-Lymphocyte Antigen 4 in Paratuberculosis

ABSTRACT

Paratuberculosis is a chronic enteritis of ruminants caused by the facultative intracellular pathogen Mycobacterium avium subsp. paratuberculosis. The Th1 response inhibits the proliferation of M. avium subsp. paratuberculosis during the early subclinical stage. However, we have previously shown that immune inhibitory molecules, such as prostaglandin E₂ (PGE₂), suppress M. avium subsp. paratuberculosis-specific Th1 responses as the disease progresses. To date, the mechanism underlying immunosuppression during M. avium subsp. paratuberculosis infection has not been elucidated. Therefore, in the present study, we investigated the function of cytotoxic T-lymphocyte antigen 4 (CTLA-4) expressed by peripheral blood mononuclear cells (PBMCs) from cattle with paratuberculosis because CTLA-4 expression is known to be elevated in T cells under an M. avium subsp. paratuberculosis experimental infection. M. avium subsp. paratuberculosis antigen induced CTLA-4 expression in T cells from cattle experimentally infected with M. avium subsp. paratuberculosis. Interestingly, both PGE₂ and an E prostanoid 4 agonist also induced CTLA-4 expression in T cells. In addition, a functional assay with a bovine CTLA-4-immunogobulin fusion protein (CTLA-4-Ig) indicated that CTLA-4 inhibited gamma interferon (IFN-γ) production in M. avium subsp. paratuberculosis-stimulated PBMCs, while blockade by anti-bovine CTLA-4 monoclonal antibody increased the secretion of IFN-γ and tumor necrosis factor alpha production in these PBMCs. These preliminary findings show that PGE₂ has immunosuppressive effects via CTLA-4 to M. avium subsp. paratuberculosis. Therefore, it is necessary to clarify in the future whether CTLA-4-mediated immunosuppression facilitates disease progression of paratuberculosis in cattle.

INTRODUCTION

Johne’s disease, which is also known as paratuberculosis, is a chronic enteric disease of ruminants caused by the facultative intracellular pathogen Mycobacterium avium subsp. paratuberculosis (1). Johne’s disease is common worldwide; indeed, no country or region is reported to be free of M. avium subsp. paratuberculosis (2). During the early stage of infection, the production of Th1 cytokines, such as gamma interferon (IFN-γ), is strongly induced by the pathogenic bacteria, which is killed by macrophages activated by IFN-γ (3–5). In contrast, M. avium subsp. paratuberculosis-specific Th1 responses are gradually suppressed during late subclinical and clinical stages (6–8). Our previous studies revealed one mechanism of this suppression, which was observed as the disease progresses (9–11). Programmed death 1 (PD-1) and PD-ligand 1 (PD-L1) are immunoinhibitory molecules that contribute to T-cell dysfunction, i.e., exhaustion, in humans and several animals, including cattle (12, 13). We showed previously that the expression levels of PD-1 and PD-L1 are upregulated in cattle infected with M. avium subsp. paratuberculosis (9). In addition, we found that blockade of the PD-1/PD-L1 pathway using anti-bovine PD-L1 antibody activates M. avium subsp. paratuberculosis-specific Th1 responses both in vitro and in vivo (10, 11). In addition, we demonstrated that prostaglandin E₂ (PGE₂) was associated with the suppression of Th1 responses in M. avium subsp. paratuberculosis-infected animals (10). In cattle, PGE₂ induces PD-L1 expression in bovine immune cells (14) and suppresses Th1 responses via its receptor, E prostanoid 4 (EP4) (10, 15). Serum PGE₂ concentrations are upregulated in cattle with Johne’s disease, and treatment with a COX-2 inhibitor, which blocks PGE₂ production, activates M. avium subsp. paratuberculosis-specific Th1 responses in vitro (10). Collectively, these results suggest that PGE₂ is involved in the progression of Johne’s disease. However, the mechanisms underlying the suppression of Th1 responses have yet to be fully elucidated.

Cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), also known as CD152, is an immunoinhibitory molecule that is expressed in T cells and an essential negative regulator of T-cell responses (16). CTLA-4 binds to CD80 (B7-1) and CD86 (B7-2), which are expressed in antigen-presenting cells, and suppresses host immunity (17, 18). In humans, CTLA-4 has been associated with the progression of chronic diseases caused by human immunodeficiency virus (HIV) and hepatitis C virus (19, 20). In the veterinary field, CTLA-4 has been associated with the progression of bovine leukemia virus (BLV) infection (a bovine chronic infection); specifically, CTLA-4 expression in T cells is increased in BLV-infected cattle, and the blockade of CTLA-4 using anti-bovine CTLA-4 monoclonal antibody (MAb) activates BLV-specific Th1 responses in vitro (21–23). In addition, the involvement of CTLA-4 in cattle infected with M. avium subsp. paratuberculosis has been shown using anti-human CTLA-4 MAb (24). However, the involvement of CTLA-4 in M. avium subsp. paratuberculosis-infected cattle has yet to be investigated using anti-bovine CTLA-4 MAb.

Therefore, in the present study, we analyzed CTLA-4 expression in T cells in M. avium subsp. paratuberculosis-infected cattle using anti-bovine CTLA-4 MAb, which was established in our previous study (23). In addition, we investigated PGE₂/EP4 signaling, finding that it was involved in CTLA-4 upregulation in T cells from cattle. Furthermore, we established a bovine CTLA-4-immunogobulin fusion protein (CTLA-4-Ig), using it and anti-bovine CTLA-4 MAb to examine whether CTLA-4 suppresses Th1 responses to antigens of M. avium subsp. paratuberculosis.

RESULTS

Expression analysis of CTLA-4 in M. avium subsp. paratuberculosis-infected cattle.

In our previous studies, we demonstrated that CTLA-4 is associated with disease progression in BLV-infected cattle (21, 22). Here, we performed expression analysis of CTLA-4 in cattle infected with M. avium subsp. paratuberculosis using anti-bovine CTLA-4 MAb (4C2-D9). The gating strategy and representative plots of CTLA-4 expression analysis are shown in Fig. S1 in the supplemental material. Compared to cattle not infected with M. avium subsp. paratuberculosis, CTLA-4 expression in CD8⁺ T cells but not CD4⁺ T cells was significantly increased in M. avium subsp. paratuberculosis-infected cattle (Fig. 1). Peripheral blood mononuclear cells (PBMCs) from M. avium subsp. paratuberculosis-infected cattle were cultivated with or without J-PPD, and flow cytometric analysis revealed that stimulation with J-PPD significantly upregulated CTLA-4 expression in both CD4⁺ and CD8⁺ T cells (Fig. 2). Thus, CTLA-4 expression is apparently upregulated by antigen stimulation in vitro in M. avium subsp. paratuberculosis-infected cattle.

FIG 1.

Expression analysis of CTLA-4 in T cells from M. avium subsp. paratuberculosis-infected cattle. The expression of CTLA-4 in CD4⁺ T cells (CD3⁺ IgM⁻ CD4⁺) (a) and CD8⁺ T cells (CD3⁺ IgM⁻ CD8⁺) (b) was analyzed using flow cytometry (uninfected [cattle not infected with M. avium subsp. paratuberculosis], n = 10; infected [M. avium subsp. paratuberculosis-infected cattle], n = 10; n.s., not significant [the P value was obtained using Mann-Whitney U test]).

FIG 2.

Upregulation of CTLA-4 expression through stimulation with J-PPD. PBMCs from cattle experimentally infected with M. avium subsp. paratuberculosis (n = 6) were cultured with J-PPD, and the expression of CTLA-4 in CD4⁺ T cells (a) and CD8⁺ T cells (b) was assayed using flow cytometry. N.S., no stimulation; J-PPD, Johnin-purified protein derivative. P values were obtained using Mann-Whitney U tests.

Effects of PGE₂ on CTLA-4 upregulation.

We previously showed that PGE₂ production from the PBMCs of M. avium subsp. paratuberculosis-infected cattle was increased by J-PPD stimulation (10). In the present study, we identified the major source of PGE₂ production under antigen stimulation. PBMCs and CD14⁺ cell-deleted PBMCs from M. avium subsp. paratuberculosis-infected cattle were cultured with or without J-PPD, and PGE₂ concentrations in culture supernatants were measured using enzyme-linked immunosorbent assay (ELISA). PGE₂ production was significantly induced from PBMCs following J-PPD treatment in vitro, whereas PGE₂ production was not induced from CD14⁺ cell-deleted PBMCs (see Fig. S2). Therefore, in M. avium subsp. paratuberculosis-infected cattle, CD14⁺ cells could be a major source of PGE₂ production under antigen stimulation. To determine whether CTLA-4 expression is induced by PGE₂ signaling, PBMCs from uninfected cattle were cultivated with PGE₂ or an EP4 agonist, and the expression of CTLA-4 was measured using flow cytometry. Treatment with PGE₂ tended to induce CTLA-4 expression in CD4⁺ and CD8⁺ T cells (Fig. 3a and b), and treatment with the EP4 agonist significantly promoted CTLA-4 expression in both CD4⁺ and CD8⁺ cells (Fig. 3a and b). In further tests, CD3⁺ cells were sorted from uninfected cattle and cultured with PGE₂ or the EP4 agonist. The gene expression of CTLA-4 in CD3⁺ cells was significantly increased by culture with PGE₂ and the EP4 agonist (Fig. 3c and d). Collectively, these data suggest that J-PPD-induced PGE₂ is involved in the upregulation of CTLA-4 expression, at least in part, via its receptor EP4.

FIG 3.

Upregulation of CTLA-4 expression via treatment with PGE₂ or an EP4 agonist. (a and b) First, PBMCs of cattle not infected with M. avium subsp. paratuberculosis were cultured with PGE₂ or the EP4 agonist, and dimethyl sulfoxide (DMSO) was used as a vehicle control. Subsequently, CTLA-4 expression in CD4⁺ T cells (a) and CD8⁺ T cells (b) was measured by using flow cytometry (n = 10; P values were obtained using Mann-Whitney U tests). (c and d) CD3⁺ cells were sorted from the PBMCs of cattle not infected with M. avium subsp. paratuberculosis cattle and cultured with PGE₂ (c) or the EP4 agonist (d), and DMSO was used as a vehicle control. The gene expression of CTLA-4 was quantitated using qPCR (n = 8; P values were obtained using Steel-Dwass tests).

The PGE₂/EP4 pathway is known to stimulate the production of intracellular cyclic AMP (cAMP) (25). In the present study, we examined whether the upregulation of intracellular cAMP is involved in upregulation of CTLA-4 expression in cattle. Treatment with dibutyryl cAMP (dbcAMP), a cell-permeable analog of cAMP that activates cAMP-dependent protein kinases, significantly induced the gene expression of CTLA-4 in T cells (Fig. 4a). In addition, flow cytometric analysis revealed that treatment with dbcAMP upregulated CTLA-4 expression in both CD4⁺ and CD8⁺ T cells (Fig. 4b and c), suggesting that CTLA-4 expression is induced by the PGE₂/EP4/cAMP pathway in cattle.

FIG 4.

Upregulation of CTLA-4 expression via treatment with dbcAMP. (a) CD3⁺ cells were sorted from the PBMCs of cattle not infected with M. avium subsp. paratuberculosis (n = 6) and cultured with dbcAMP. Double-distilled water (DDW) was used as a vehicle control. Gene expression of CTLA-4 was quantitated using qPCR. (b and c) PBMCs of cattle not infected with M. avium subsp. paratuberculosis (n = 7) were cultured with dbcAMP, DDW was used as a vehicle control, and CTLA-4 expression in CD4⁺ T cells (b) and CD8⁺ T cells (c) was assayed by using flow cytometry. All P values were obtained using Mann-Whitney U tests.

Establishment and functional analysis of CTLA-4-Ig.

To investigate whether CTLA-4 is involved in the suppression of M. avium subsp. paratuberculosis-specific immune responses, we generated an Fc-fusion protein of bovine CTLA-4, namely, CTLA-4-Ig, which was expressed using a pDC62c5-U533 expression vector and CHO-DG44 cell expression system in vitro. After purification, SDS-PAGE was performed under nonreducing (Fig. 5a) and reducing (Fig. 5b) conditions. Flow cytometric analysis revealed that CTLA-4-Ig bound to both bovine CD80 and CD86 (Fig. 5c and d). In addition, flow cytometric analysis showed that CTLA-4-Ig inhibited the binding of CD80-Ig or CD86-Ig to CD28-expressing cells on the cell surface (Fig. 5e and f). We also used CTLA-4-Ig to assess the involvement of CTLA-4 in the suppression of M. avium subsp. paratuberculosis-specific immune responses. PBMCs from M. avium subsp. paratuberculosis-infected cattle were cultured with CTLA-4-Ig in the presence of J-PPD, and CTLA-4-Ig treatment significantly downregulated the production of IFN-γ from PBMCs (Fig. 6). Therefore, CTLA-4 may be associated with the immune suppression in M. avium subsp. paratuberculosis-infected cattle.

FIG 5.

Establishment and functional analyses of CTLA-4-Ig. (a and b) Purified CTLA-4-Ig was confirmed via SDS-PAGE under nonreducing (a) and reducing (b) conditions. (c and d) Binding of CTLA-4-Ig to bovine CD80 (c) and bovine CD86 (d) was determine using flow cytometry. (e and f) CTLA-4-Ig inhibited the binding of bovine CD80 (e) or bovine CD86 (f) to bovine CD28. 2ME, 2-mercaptoethanol; isotype: purified bovine IgG₁; control-Ig, purified rabbit IgG.

FIG 6.

Suppression of IFN-γ production via CTLA-4-Ig. PBMCs from M. avium subsp. paratuberculosis-infected cattle (n = 6) were cultured with CTLA-4-Ig or Control-Ig (39) in the presence of J-PPD. IFN-γ production was measured using ELISA. P values were obtained using Steel-Dwass tests.

Activation of J-PPD-stimulated immunity via CTLA-4 blockade.

As described previously, the blockade of CTLA-4 activates immune responses in BLV-infected cattle in vitro (22, 23). To examine the effects of CTLA-4 blockade on immune responses in M. avium subsp. paratuberculosis-infected cattle, PBMCs derived from the infected animals were cultured with anti-bovine CTLA-4 MAb (4C2-D9) (23) in the presence of J-PPD. The production of Th1 cytokines, i.e., IFN-γ and tumor necrosis factor alpha (TNF-α), was significantly increased in the group treated with anti-bovine CTLA-4 MAb (Fig. 7).

FIG 7.

Activation of Th1 responses via treatment with anti-CTLA-4 MAb. PBMCs from M. avium subsp. paratuberculosis-infected cattle (n = 8) were cultured with anti-CTLA-4 MAb in the presence of J-PPD, and the concentrations of IFN-γ (a) and TNF-α (b) in culture supernatants were measured using ELISA. P values were obtained using Mann-Whitney U tests. Control-Ig, mouse IgG1 isotype control.

DISCUSSION

In the present study, as well as a previous study (10), we have shown that antigen stimulation activates PGE₂ production from the PBMCs of M. avium subsp. paratuberculosis-infected cattle. Furthermore, we showed here that PGE₂ upregulated CTLA-4 expression in T cells via the EP4/cAMP pathway (see Fig. S3). cAMP is known to play a critical role in upregulating CTLA-4 expression in T cells (26) and has been associated with the progression of chronic infections, such as HIV (27). However, information was previously lacking on the relationship between antigen stimulation and cAMP-induced CTLA-4; thus, the present study provides new insights in this regard. CTLA-4 expression in T cells is upregulated in many other chronic infections, including HIV (19), hepatitis C virus (20), and BLV (21, 22) infections. Therefore, further studies will be needed to investigate whether antigen stimulation is involved in CTLA-4 expression via the PGE₂/EP4/cAMP pathway in bovine chronic infections. In the present study, M. avium subsp. paratuberculosis-experimentally infected cattle showed high expression of CTLA-4 but did not exhibit clear clinical signs because they might be in the latent period. The detailed involvement of CTLA-4 in the process leading to the development of granulomas in Johne’s disease should also continue to be investigated.

CTLA-4 is expressed in regulatory T cells (Tregs) and is known as a Treg-associated molecule (28, 29). In M. avium subsp. paratuberculosis-infected cattle, Tregs have been associated with immune responses to antigen stimulation (30). However, the mechanism underlying Treg proliferation in M. avium subsp. paratuberculosis-infected cattle is still unclear. Our previous studies showed that PGE₂ concentrations in sera are increased in infected cattle and that PGE₂ induces the expression of the Treg-associated molecules Foxp3 and TGF-β₁ in cattle (10, 31). Thus, in M. avium subsp. paratuberculosis-infected cattle, antigen stimulation-induced PGE₂ might contribute to increasing the Treg population via its receptor EP4.

The cell-mediated immunity plays an essential role in controlling the progression of Johne’s disease. However, M. avium subsp. paratuberculosis-specific immune responses, especially IFN-γ production, are gradually suppressed in M. avium subsp. paratuberculosis-infected cattle, contributing to progression to the clinical stage (6–8, 32). In the present study, we found that CTLA-4-Ig treatment reduced IFN-γ production from the PBMCs of M. avium subsp. paratuberculosis-infected cattle. In addition, anti-bovine CTLA-4 MAb increased anti-bacterial cytokine productions in vitro. These data suggest that CTLA-4 is involved in the suppression of anti-bacterial responses in M. avium subsp. paratuberculosis-infected cattle.

On farms, calves become infected with M. avium subsp. paratuberculosis mainly via the fecal-oral route (1). Therefore, M. avium subsp. paratuberculosis-infected cattle that exhibit bacterial fecal shedding can be considered a major infection source. Our previous study demonstrated that treatment with anti-bovine PD-L1 antibody significantly activated Th1 responses in M. avium subsp. paratuberculosis-infected cattle in vivo, although the inhibitory effect of anti-PD-L1 antibody alone on bacterial shedding was limited (11). Thus, to reduce bacterial load, the development of a novel control method that induces potent antibacterial responses is desirable. The upregulation of PGE₂ is one possible mechanism underlying the inhibition of anti-bacterial responses during PD-1 blockade (15). Our previous study revealed that PGE₂ concentrations in sera were increased in BLV-infected cattle after treatment with anti-bovine PD-L1 antibody (15). In the present study, we revealed the role played by PGE₂ in the upregulation of CTLA-4 expression in cattle. Therefore, CTLA-4 levels might increase after PD-1 blockade, and CTLA-4 could be a target molecule for inducing potent immune responses. Indeed, in cancer research, the dual blockade of PD-1/PD-L1 and CTLA-4 using specific antibodies enhances therapeutic effects compared to the effects achieved using single blockade of PD-1/PD-L1 or CTLA-4 (33). Moreover, combined treatment with anti-PD-1 antibody and anti-CTLA-4 antibody leads to potent antitumor effects in patients with advanced melanoma (34, 35). In our previous study, we found that the dual blockade of PD-L1 and CTLA-4 enhances antiviral cytokine production from the PBMCs of BLV-infected cattle in vitro (23). Therefore, combination therapy involving anti-PD-1 and CTLA-4 antibodies could be a novel control strategy for Johne’s disease. Future studies should investigate the synergistic effects of dual blockade in cattle infected with M. avium subsp. paratuberculosis.

MATERIALS AND METHODS

Animals and blood samples.

Blood samples of cattle (adults of the Holstein breed) not infected with M. avium subsp. paratuberculosis, which were tested for specific antibodies against M. avium subsp. paratuberculosis using a commercial ELISA test kit (Kyoritsu Seiyaku Corporation, Tokyo, Japan), were obtained at the Field Science Center for Northern Biosphere, Hokkaido University and from dairy farmers in Hokkaido. Blood samples of cattle experimentally infected with M. avium subsp. paratuberculosis were obtained from the National Institute of Animal Health, Tsukuba, Japan. For the experimental infection, Holstein calves 1 to 4 weeks of age were inoculated orally with intestinal tissue homogenate (6.8 × 10⁶ CFU) from M. avium subsp. paratuberculosis-infected cattle once daily for 20 days (10) or once with a high concentration of homogenate (1.36 or 2.50 × 10⁸ CFU), as described previously (11). All infected cattle were kept in a biosafety level II animal facility at this institute and did not show any clinical symptoms. All experiments using bovine blood samples were approved by the National Institute of Animal Health Ethics Committee (approval numbers 15-001, 18-004, and 18-077) and/or the local committee for animal studies at Hokkaido University (approval number 17-0024).

Cell preparation and culture.

PBMCs from blood samples were purified using density gradient centrifugation on Percoll (GE Healthcare, Little Chalfont, UK) as described previously (36). CD3⁺ and CD14⁺ cells were isolated from PBMCs using an autoMACS Pro (Miltenyi Biotec, Bergisch Gladbach, Germany) as described previously (15, 31). All cell cultures were grown in 96-well plates (Corning, Inc., Corning, NY) using RPMI 1640 medium (Sigma-Aldrich, St. Louis, MO) including 10% heat-inactivated fetal calf serum (Thermo Fisher Scientific, Waltham, MA), 100 U/mL of penicillin (Thermo Fisher Scientific), 100 μg/mL of streptomycin (Thermo Fisher Scientific), and 2 mM l-glutamine (Thermo Fisher Scientific).

Flow cytometry.

To analyze CTLA-4 expression, flow cytometric analysis was performed using PBMCs from cattle infected or not infected with M. avium subsp. paratuberculosis. To prevent nonspecific reactions, Fc blocking was performed by incubating PBMCs with phosphate-buffered saline (PBS) including 10% goat serum (Thermo Fisher Scientific) for 15 min at 25°C. The cells were then stained with anti-bovine CTLA-4 MAb (4C2-D9) (23) or mouse IgG₁ isotype control (15H6; Southern Biotech, Birmingham, AL) for 20 min at 37°C. After being washed twice with PBS containing 1% bovine serum albumin (BSA; Sigma-Aldrich), the cells were stained with Alexa Fluor 647-conjugated anti-mouse IgG(H+L) F(ab′)₂ (Invitrogen, Carlsbad, CA) for 20 min at 25°C. They were then washed twice with PBS containing 1% BSA and stained with the following antibodies for 20 min at 25°C: PerCP/Cy 5.5-conjugated anti-bovine CD3 MAb (MM1A; Washington State University Monoclonal Antibody Center, Pullman, WA), FITC-conjugated anti-bovine CD4 MAb (CC8; Bio-Rad, Hercules, CA), PE-conjugated anti-bovine CD8 MAb (CC63; Bio-Rad), and PE/Cy7-conjugated anti-bovine IgM MAb (IL-A30; Bio-Rad). MM1A and IL-A30 were conjugated using Lightning-Link antibody labeling kits (Innova Biosciences, Cambridge, England). After the final staining, the cells were washed twice with PBS containing 1% BSA and analyzed immediately using FACSVerse (BD Biosciences, San Jose, CA, USA).

Enzyme-linked immunosorbent assay.

To measure the concentrations of IFN-γ, TNF-α, and PGE₂ in culture supernatants, ELISAs were performed using a bovine IFN-γ ELISA development kit (Mabtech, Nacka Strand, Sweden), a bovine TNF-α Do-It-Yourself ELISA (Kingfisher Biotech), and a PGE₂ Express ELISA kit (Cayman Chemical, Ann Arbor, MI), respectively.

Quantitative PCR.

RNA extraction and cDNA synthesis were performed in accordance with methods reported in a previous study (31). The gene expression of CTLA-4 was quantitated using a thermal cycler (LightCycler 480 System 2; Roche Diagnostic, Mannheim, Germany) and SYBR Premix DimerEraser (TaKaRa Bio, Otsu, Japan) according to the manufacturer’s instructions. β-actin (ACTB) was used as a reference gene. Relative expression levels were calculated using the ΔΔCt method, and the results were represented as changes relative to expression levels in the untreated group. The primers used were as follows: 5′-CCA GAG TCA TGG GAC TTG GT-3′ and 5′-TCA CAT GAG AAG CTG GCA AC-3′ for CTLA-4 and 5′-TCT TCC AGC CTT CCT TCC TG-3′ and 5′-ACC GTG TTG GCG TAG AGG TC-3′ for ACTB.

Establishment of CTLA-4-Ig.

To examine the effects of CTLA-4 on M. avium subsp. paratuberculosis-specific Th1 responses, we generated bovine CTLA-4-Ig using Chinese hamster ovary (CHO)-DG44 cells. The amino acid sequence of CTLA-4-Ig consisted of an extracellular domain fragment of bovine CTLA-4 and bovine IgG₁, triggering reduced Fc-mediated effector functions (ADCC−) (37). A plasmid vector encoding CTLA-4-Ig (pUCIDT-Kan-CTLA-4-Ig) was digested using AscI and AsiSI (both New England Biolabs, Ipswich, MA), after which the fragment was inserted into pDC62c5-U533 (38) using a DNA ligation kit (Mighty Mix; TaKaRa Bio). The vector (pDC62c5-U533-CTLA-4-Ig) was then transfected into CHO-DG44 cells using Lipofectamine LTX (Life Technologies, Carlsbad, CA). Transfectants (5 × 10⁵ cells/mL) were shaken for 14 days on a rotary shaker at 125 rpm and 37°C under 8% CO₂ using CD OptiCHO Medium (Thermo Fisher Scientific). Subsequently, the supernatants were collected and filtered using Centricon Plus-70 (Merck Millipore, Billerica, MA), and CTLA-4-Ig was purified using Ab-Capture ExTra (ProteNova, Tokushima, Japan) according to the manufacturer’s protocol. After purification, SDS-PAGE was performed using a 12% SDS-polyacrylamide gel under reducing and nonreducing conditions. The gel was stained using a Quick-CBB kit (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan), and Precision Plus Protein All Blue prestained protein standards (Bio-Rad) were used as a marker.

Functional analysis of CTLA-4 Ig.

To confirm the binding of CTLA-4-Ig to bovine CD80 and CD86, Cos-7 cells expressing bovine CD80 or CD86 (22) were incubated with 10 μg/mL of CTLA-4-Ig for 30 min at 25°C. Purified bovine IgG₁ (10 μg/mL; Bethyl Laboratories, Montgomery, TX) was used as a negative control. After being washed, the cells were stained with Alexa Fluor 647-conjugated AffiniPure goat anti-bovine IgG (H+L) (Jackson ImmunoResearch Laboratories, West Grove, PA) for 20 min at 25°C and analyzed immediately using FACSVerse. Competitive and blocking assays were also performed to confirm whether CTLA-4-Ig inhibits interactions with CD28 and CD80 or CD86. CTLA-4-Ig was incubated with CD80-Ig or CD86-Ig (22) for 30 min at 25°C. Purified rabbit IgG (Southern Biotech) was used as a negative control of CTLA-4-Ig. Cos-7 cells expressing bovine CD28 were then incubated with a reaction mixture of Fc-fusion proteins for 30 min at 25°C. After incubation, the cells were stained with Alexa Fluor 647-conjugated anti-rabbit IgG(H+L) (Thermo Fisher Scientific) and analyzed immediately using FACSVerse.

Cell cultures.

PBMCs or CD14⁺ cell-depleted PBMCs from cattle experimentally infected with M. avium subsp. paratuberculosis were incubated with 1 μg/mL of Johnin-purified protein derivative (J-PPD). After 24 h, CTLA-4 expression in T cells was analyzed using flow cytometry as described above. After 72 h, PGE₂ concentrations in culture supernatants were measured using ELISA.

To determine whether the PGE₂/EP4/cAMP pathway upregulates CTLA-4 expression, PBMCs from cattle not infected with M. avium subsp. paratuberculosis were cultured with 2.5 μM PGE₂ (Cayman Chemical), 1 μg/mL of Rivenprost (an EP4 agonist; Cayman Chemical), or 400 μM dbcAMP (an analog of cAMP; Sigma-Aldrich). After 24 h, the cells were collected, and CTLA-4 expression was analyzed as described above.

Purified CD3⁺ cells of cattle not infected with M. avium subsp. paratuberculosis were cultured with 250 nM PGE₂, 1 μg/mL of Rivenprost, or 400 μM dbcAMP. After 24 h, the gene expression of CTLA-4 was measured using quantitative PCR (qPCR).

To examine the effects of CTLA-4-Ig on Th1 responses, PBMCs of M. avium subsp. paratuberculosis-infected cattle were cultured with 10 or 1 μM CTLA-4-Ig or with Control-Ig (39) in the presence of 1 μg/mL of J-PPD. After 24 h, IFN-γ concentrations in culture supernatants were measured using ELISA.

To determine whether anti-bovine CTLA-4 MAb activates M. avium subsp. paratuberculosis-specific Th1 responses, PBMCs from M. avium subsp. paratuberculosis-infected cattle were cultivated with 10 μg/mL of anti-bovine CTLA-4 MAb (4C2-D9) in the presence of 1 μg/mL of J-PPD. After 120 h, the concentrations of IFN-γ and TNF-α in culture supernatants were measured using ELISA.

Statistical analysis.

Statistical significance was determined using Mann-Whitney U tests and Steel-Dwass tests. MEPHAS (http://www.gen-info.osaka-u.ac.jp/MEPHAS/) was used to conduct statistical analyses. Statistical significance was set at P < 0.05.