U-937, THP-1, and HeLa human cell lines (ATCC), and human and murine primary cells, were grown in RPMI or DMEM (Sigma) supplemented with fetal calf serum (FCS, 10%), 4 mM l-glutamine, 1 mM sodium pyruvate, penicillin (100 U/mL), and streptomycin (0.1 mg/mL) (Biological Industries, Israel) at 37°C and 5% CO₂. U-937 and THP-1 monocytes were differentiated to macrophages with 5 ng/mL PMA (Cayman Chemical, Ann Arbor, MI, USA) for 3 days. Primary human CD4⁺ T cells and monocytes were purified from buffy coats by the RosetteSep Human CD4⁺ T or the RosetteSep human monocyte cell enrichment cocktail according to the manufacturer’s instructions (StemCell Technologies, USA). Primary monocytes were differentiated to macrophages (confirmed by morphology and adhesion) with 10 ng/mL GM-CSF (Peprotec, Rehovot, Israel) for 6 days.

Five different MyD88 shRNA lentiviral transduction particles (Sigma) were independently transduced, and stable clones (of THP-1, U-937, and HeLa cells) were obtained based on resistance to puromycin (Gold Biotechnology. St. Louis, MO, USA). The shRNA sequence that reduced human MyD88 to the greatest extent was 5′-CCGGCCTGTCTCTGTTCTTGAACGTCTCGAGACGTTCAAGAACAGAGACAGGTTTTTG-3′. Non-target shRNA control cells were generated using an irrelevant sequence (Sigma). To transduce, cells were plated at 75% confluency 3 h prior to transduction and then the corresponding lentiviral particles were added at a Multiplicity of Infection of 10 overnight. After 48 h, cells were selected by addition of medium containing 2 µg/mL puromycin for THP-1 cells and 0.5 µg/mL for U-937 and Hela cells.

Total cellular RNA of the transduced cell lines was extracted using tri-reagent (Sigma) and the mRNA was reverse transcribed to cDNA with Maxima H minus Enzyme mix, according to the manufacturer’s instructions (Thermo Scientific, Waltham, MA, USA). Amplification for qPCR was performed with primers to hMyD88 (forward 5′-ACAGGCACCAGCATACAC-3′; reverse 5′-TTGGGTCCTTTCCAGAGT-3′) and hβactin (forward 5′-CACGGCATCGTCACCAACT-3′; reverse 5′-TGATCTGGGTCATGTTCTCGC-3′). qPCR was performed using Bio-Rad CFX manager (Bio-Rad, Hercules, CA, USA) and the reactions with KAPA SYBR Fast ABI prism (KAPA systems, Salt River, Cape Town). The final results were calculated by dividing the relative transcript levels of the test genes by the relative amount of βactin RNA.

For MyD88 protein detection, 10⁷ cells were lysed in Ripa buffer in the presence of protease inhibitors and incubated on ice for 25 min. The lysates were then centrifuged at 10,000g for 15 min. Proteins were separated by gel electrophoresis and transferred to nitrocellulose membranes. The nitrocellulose membranes were blocked with 5% skim milk for 1 h and MyD88 was detected using 1 µg/mL polyclonal rabbit anti-MyD88 (eBioscience, San Diego, CA, USA) followed by 100 ng/mL goat anti-rabbit IgG-HRP (Abcam, Cambridge, UK). After washing extensively with Tris-buffered saline (TBS) containing 0.2% Tween 20, the membranes were exposed to chemiluminescent substrate in the presence of hydrogen peroxide, using the EZ-ECL-chemiluminescence detection kit (Biological Industries, Israel). A Gel-Doc imaging system (Bio-Rad, Hercules, CA, USA) was used to capture the image.

THP-1 cells were seeded at 1.5 × 10⁶ in six well plates with 5 ng/mL PMA (Cayman Chemical, Ann Arbor, MI, USA) for 3 days differentiation. When indicated, 100 ng/mL Pam3csk4 (Invitrogen, San Diego, CA, USA) + 20 ng/mL IFNγ (Peprotec, Rehovot, Israel) was added to the cells for the final 24 h of incubation to induce differentiation (concentrations of stimulants were determined in preliminary calibration experiments). Cells were then collected, washed, and stained using APC-conjugated human-specific antibodies (Biolegend, San Diego, CA, USA) to CD80, HLA-A2, and HLA-DR for 30 min on ice. The purity of CD4⁺ T cells isolated from buffy coats of healthy donors was confirmed by flow cytometry using FITC-anti-CD4 (Tonbo Biosciences, San Diego, CA, USA). Intracellular cytokine staining was analyzed on T cells incubated for 6 days with APCs. Cells were incubated with cell stimulation cocktail (eBioscience, San Diego, CA, USA) and Brefeldin (Biolegend, San Diego, CA, USA) for 6 h prior to collection, and then fixed 15 min at RT. Cells were then permeabilized and stained with Leucoperm permibilization reagent (AbD Serotec, Kidlington, UK) and APC-anti-IL-4 (Biolegend, San Diego, CA, USA) for 30 min on ice. Samples were analyzed using a BD LSRII flow cytometer or a BD ACCURI C6 cytometer (BD Biosciences, San Jose, CA, USA), and data were analyzed using FCS Express software (De Novo Software, Glendale, CA, USA).

Hela shCnt and HeLa shMyD88 cells were activated with 20 ng/mL IL-1β or 10 ng/mL TNFα (ProSpec, Rehovot, Israel) for 30 min or 1 h, respectively. Cells were then fixed (3.7% PFA in PBS for 10 min), permeabilized (0.25% Triton-X100) and blocked (2% BSA in TBS) at 4°C for 16 h. Cells were then stained with 0.6 µg/mL rabbit anti-p65 in 2% BSA in TBS (Santa Cruz Biotechnology, Dallas, TX, USA) followed by 0.5 µg/mL CY-3 goat anti-rabbit antibody (Jackson ImmunoResearch, Baltimore Pike, PA, USA). Cytoplasmic vs. nuclear localization was analyzed by fluorescent microscopy (15) (Nikon-Ti microscope).

Myeloid differentiation factor 88 inhibitor peptide (MyDI) (RDVLPGT), or the scrambled version of the same sequence, (MyDI-sc) (PTDLVRG), were synthesized in the Weizmann Institute of Science and purified by HPLC.

shMyD88 THP-1 vs. shCnt THP-1 cells were differentiated as above with PMA alone or PMA with Pam3csk4/IFNγ. Cells were treated with 10 µg/mL Mitomycin c (Sigma) to prevent replication, and added in various concentrations to CD4⁺ T cells in 96 well round bottom plates (Thermo Scientific). T-cell proliferation and cytokine production were analyzed after 6 days of coculture. Proliferation was assayed using a non-radioactive Tetrazolium-based kit (Promega, Madison, WI, USA) according to the manufacturer’s instructions.

Primary human macrophages were incubated for 24 h with or without 80 µM MyDI vs. MyDI-sc peptides. Macrophages were treated with 10 µg/mL mitomycin c (Sigma) to prevent replication and added in various concentrations to CD4⁺ T cells in 96-well round bottom plates (Thermo Scientific). T-cell proliferation and cytokine production were analyzed after 6 days of coculture. Proliferation was assayed using a non-radioactive Tetrazolium-based kit (Promega, Madison, WI, USA) according to the manufacturer’s instructions.

Anti-MOG35-55 resting T cells were treated in vitro with MyDI or MyDI-sc (20 µM) prior to stimulation with IL-18 (30 ng/mL) (R&D systems, Inc., Minneapolis, MN, USA). The effect of MyDI or MyDI-sc was also tested on anti-MOG35-55 resting T cells re-stimulated with MOG35-55 (5 µg/mL) in the presence of irradiated splenocytes with or without the addition of IL-18. Cytokines were measured in the cell supernatants after overnight incubation. Primary human CD4⁺ T cells that were isolated from buffy coats were incubated for 3 h in the absence or presence of MyDI, or MyDI-sc (20 µM) and stimulated with IL-18 + IL-12 (320 and 160 ng/mL, respectively) overnight (Peprotec, Rehovot, Israel). Cytokines were measured in the cell supernatants after overnight incubation.

C57BL/6 (B6) mice were purchased from Harlan (Jerusalem, Israel). Female, 8- to 14-week-old mice were used in the experiments. The mice were housed at the SPF units of our universities and all experiments were approved by the Hebrew University-Hadassah Institutional Animal Care and Use Committee (protocol number 16-14745).

Mice immunized s.c. with 100 µg MOG_35-55/CFA and treated with MyDI and control were sacrificed 11 days after immunization, and popliteal, inguinal, and axillary LNs were collected and single cell suspensions were prepared. Cells were cultured in 96-well plates (0.5 × 10⁶ per well) for 72 h with or without increasing concentrations of Myelin Oligodendrocyte Glycoprotein (MOG_35-55) or purified protein derivative (PPD) or ovalbumin (OVA) peptide in the culture.

For in vivo treatment, we synthesized the MyDI or MyDI-scr peptides with a basic peptide from the Drosophila Antennapedia homeodomain (DRQIKIWFQNRRMKWKK) at the C-terminus [12]. All peptides were synthesized at the Weizmann Institute of Science and purified by HPLC. Mice were immunized s.c. in the flank with 200 µg MOG_35-55 emulsified in CFA supplemented with 300 µg Mycobacterium tuberculosis (Mt) H37RA (Difco). Pertussis Toxin (PTX, List Biological Laboratories, CA, USA) was injected i.v. at the time of immunization and 48 h later. In some experiments, animals were sacrificed prior to onset of clinical symptoms in order to analyze the lymph node response. EAE was scored on a scale of 0–6: 0, no impairment; 1, limp tail; 2, limp tail and hind limb paresis; 3, ≥1 hind limb paralysis; 4, full hind limb and hind body paralysis; 5, hind body paralysis and front limb paresis; and 6, death. Mice were treated with MyDI i.p. (2 mg/kg) vs. control at day 0 and every 48 h.

Levels of hTNF-α, h INF-γ and hIL-17 and mIL-10, mIL-5, mIL-17, and mINF-γ were determined using human/mouse OptEIA sets (BD Biosciences, CA, USA) according to the manufacturer’s instructions.

The two-tailed t-test was used for statistical evaluation of all the results except the two-way analysis of variance test that was used for the EAE model assay. Values are shown for data that reached a significance of P ≤ 0.05 (*), P ≤ 0.01 (**), P ≤ 0.005, (***), and P ≤ 0.001 (****). Bars show mean and SD and in Figure 6D SEM (Prism v.5, GraphPad Software Inc., San Diego, CA, USA).

To determine the role of MyD88 in human APCs, we first established stable shRNA knock-down cells. We screened lentivirally expressed shRNA sequences in HeLa cells to find a sequence that reduced MyD88 efficiently and did not affect inflammatory pathways driven by other signals. Sequences that reduced MyD88 mRNA expression also blocked NFκB nuclear translocation in response to IL-1 stimulation, a MyD88-dependent event, but had no effect on NFκB translocation in response to TNFα stimulation that occurs independently of MyD88 (Figure 1B). Control shRNA transduction did not affect either signaling pathway (Figure 1A). We next transduced human THP-1 monocyte/macrophage cells with this shRNA and found efficient reduction of MyD88 mRNA and protein (Figure S1 in Supplementary Material), and a near complete reduction in the response to stimulation with PAM3Csk4, a synthetic triacylated lipopeptide ligand of TLR2/TLR1 (Figure 1C).

To probe the effect on T-cell phenotype of silencing MyD88 in APCs, we first tested costimulatory molecule (CD80) and HLA expression in shMyD88 compared to shCnt THP-1 cells differentiated with PMA and further stimulated with Pam3csk4/IFNγ. We focused on CD80 since CD86 expression was not strongly upregulated in response to stimulation with Pam3csk4/IFNγ. In shCnt cells, Pam3csk4 alone induces CD80 expression which is further enhanced by addition of IFNγ (Figures 2A,B). Silencing MyD88 prevented the induction of CD80 expression in response to Pam3csk4/IFNγ (Figures 2A,C), since enhanced CD80 expression is mostly due to TLR2 stimulation by PAM3csk4 (Figures 2B,C). However, the induction of HLA expression (both class I and class II), an effect attributable to IFNγ stimulation (Figures 2E,F), remained intact in the absence of MyD88 (Figures 2D–F; Figure S2 in Supplementary Material). Interestingly, Pam3csk4 reduced the basal level of MHC-II expression in differentiated shCnt THP1 cells (Figure 2E); however, this effect was not seen in the WT THP1 cells (not shown). Therefore, MyD88 affects the regulation of costimulatory signals without inhibiting the regulation of HLA.

We next performed a MLR using the shMyD88 compared to shCnt THP-1 cells. The THP-1- transfected cells were differentiated for 3 days, and during the final 24 h the cells were stimulated with Pam3csk4/IFNγ. The THP-1 cells were treated with Mitomycin c to prevent their proliferation, and then seeded together with CD4⁺ cells isolated from buffy coats. After 6 days, we measured T-cell proliferation and cytokine production. ShCnt and shMyD88 APCs induced T-cell proliferation to a similar extent (Figures 3E,F, no significant differences were detected between the proliferation induced by each type of APC). Nevertheless, there was a marked difference in the cytokine profile of the proliferating T cells. Whereas shCnt APCs induced IFNγ (Figure 3A) and IL-17 (Figure 3C) in a concentration-dependent manner, there was no production of either of these cytokines by T cells stimulated with shMyD88 APCs (Figures 3B,D). In contrast, the shMyD88 APCs induced IL-4-producing T cells, which were not present in the cultures of cells stimulated with shCnt APCs (Figures 3G,H).

To strengthen the results, we next examined the effect of MyD88 inhibition using primary human APCs and an additional means of MyD88 inhibition. To block MyD88, we incubated differentiated primary macrophages with the MyD88 inhibitory BB loop peptide (MyDI) (12), and as a control we used a scrambled control peptide (MyDI-sc). MyDI treatment specifically blocks MyD88-dependent signaling, as demonstrated by reduced NFκB activation in response to IL-1, but not in response to TNFα stimulation. In contrast, MyDI-sc does not inhibit NFκB activation in response to either stimulant (Figure S4 in Supplementary Material). MyDI treatment also blocks inflammatory cytokine production by primary cells or THP-1 cells activated with the TLR2 ligand Pam3csk4, although to a lesser extent than shRNA inhibition (Figures S5A,B in Supplementary Material). Primary APCs were treated with Mitomycin c to prevent their proliferation, and then seeded together with donor-mismatched CD4⁺ T cells isolated from buffy coats. After 6 days, we measured T-cell proliferation and cytokine production. Inhibition of MyD88 did not affect the ability of primary APCs to induce T cell proliferation (Figure 4C), however inhibition shifted the cytokine profile of the proliferating T cells. Similar to our findings using shMyD88 THP-1 cells, peptide inhibition of MyD88 in primary APCs led to significantly diminished IFNγ and IL-17 by responding T cells, and a shift to IL-4 production (Figures 4A,B,D). Importantly, peptide blocking of MyD88, a less efficient method compared to stable shRNA inhibition, produced a similar effect. Similar results were obtained when primary human DCs were used as APCs (data not shown). Therefore, MyD88 inhibition in human APCs induces a shift in the cytokine profile of responding T cells from IFNγ and IL-17 production to IL-4 production without significantly reducing T-cell proliferation.

Since MyD88 functions downstream to multiple receptors expressed by T cells, we next asked if inhibition of T-cell MyD88 would effectively reduce cytokine production. In contrast to THP-1 cells, we found that the MyDI peptide penetrated primary T cells efficiently (Figure S3 in Supplementary Material), obviating the need for shRNA. To investigate the effect of MyD88 inhibition on primary human T cell antigen-independent IFNγ production (16), cells were incubated with MyDI or MyDI-sc and then stimulated overnight with IL-18 + IL-12. hIFNγ secretion was abolished in the stimulated cells by the addition of MyDI but was not decreased in the cells treated with the MyDI-sc (Figure 5A). To test if similar effects could be obtained in autoimmune T cells, we treated MOG_35-55-specific murine T cells with MyDI and MyDI-sc prior to IL-18 stimulation. MyDI treatment significantly reduced IL-18-stimulated IFNγ secretion, in contrast to treatment with MyDI-sc (Figure 5B). We next tested the effect of MyDI on the cytokine response of MOG_35-55-specific T cells responding to antigen stimulation in the presence of IL-18 and MOG_35-55 peptide (Figure 5C). IL-18 treatment increased the IFNγ response of antigen-specific T cells. Here too, MyDI treatment significantly reduced the level of IFNγ when IL-18 was present (Figure 5C).

To investigate the in vivo effects of MyDI treatment on the T cell cytokine profile, we immunized mice with MOG_35-55 peptide emulsified in CFA and administered PTX on day 0 and at 48 h. The mice were treated i.p. on day 0 and every 48 h with 2 mg/kg MyDI or MyDI-sc or PBS. Eleven days following immunization, the DLNs were harvested and cells were activated ex vivo with increasing doses of MOG_35-55 or increasing doses of PPD, representing major antigens of the Mtb in the adjuvant. Cytokine secretion was analyzed 72 h after activation. As shown in Figure 6, MyDI treatment induced a shift in the cytokine profile of the LN T cell response to PPD. T cells from MyDI-treated animals secreted significantly less IFNγ and IL-17 (Figures 6A,B), but greater amounts of IL-5 (Figure 6C). Since MyDI treatment influenced the T cell cytokine profile, we next tested the effect of MyDI administration on the clinical outcome of MOG_35-55/CFA-induced EAE. Mice were treated with MyDI or MyDI-sc (2 mg/kg) or PBS (200 µL) i.p. three times a week beginning on the day of immunization with MOG_35-55/CFA. As shown in Figure 6D, MyDI treatment significantly reduced EAE disease severity.