Figure 1.
Squalene administration leads to accumulation of membrane cholesterol in resting CD4 T-cells.
(A) F1 hybrid mice (n = 5/group) were injected i.p. or not (purple line) with a single dose (black line) or 4 doses of squalene (red line) within a week interval (180 µg/dose/mouse). Seven days after the last injection, negatively-sorted splenic CD4 cells from individual mice were co-stained with CD3-PE, CD4-FITC and Filipin III. Shown is the amount of cholesterol in plasma cell membrane of gated CD3+CD4+ splenic cells as measured by MFI of Filipin III in FACS at single-cell level in one representative mouse from each group (left panel). Right panel, F1 hybrid mice (n = 7/group) were injected i.p. (black line) or not (red line) with a single dose of squalene (180 µg/mouse) and 7 days later negatively-sorted splenic CD4 cells from individual mice were co-stained with CD3-PE, CD4-FITC, and Filipin III. Shown is the percentage of gated CD4+ T-cells ± standard deviation (SD) and MFI values of Filipin III ± SD before and after squalene injection as collected among 700 cell events in gated population of CD3+4+ T-cells for one of three representative experiments. (B) Cholesterol accumulation in the spleen was identified by Oil Red O (ORO) staining of frozen spleen sections, counter-stained with hematoxylin from untreated or squalene treated mice (180 µg/mouse) given one or four doses, and analyzed 7 days post-injection (n = 3/group). Left panel, spleen section from untreated mouse. Middle panels, spleen section from squalene treated mice. Right panel, positive control for ORO lipid droplet staining in adipocytes. Shown is one representative ORO stained section in each group. Dark arrows indicate ORO stain. (C) Quantitative real-time RT-PCR of HMG-CoA reductase mRNA and Squalene epoxidase mRNA extracted from negatively-sorted CD4 splenocytes isolated from individual F1 hybrid mice (n = 5/group) that were treated (light bars) or not treated (dark bars) with a single dose of squalene (180 µg/mouse) and analyzed 7 days post-injection. Y axis indicates the mean fold increase in mRNA expression level relative to the endogenous 18S rRNA expression level (control ± standard deviation (SD). Shown are two combined separate experiments (*p value<0.05). (D) FACS measurements of CD3+4+Foxp3+ T-reg cells from negatively-sorted CD4+ splenic cells of the same F1 mouse groups analyzed in panel A that were co-stained with CD3-PE and Filipin III. Shown is the percentage of gated CD4+Foxp3+ T-reg cells ± SD and MFI values of Filipin III ± SD collected among 500 cell events in the gated population of GFP+-Foxp3/GFP+ cells from one mouse in each group from two representative experiments.
Figure 2.
Squalene induced accumulation of membrane cholesterol in resting CD4 T-cells alters the partitioning of lipid rafts.
Negatively-sorted CD4+ splenocytes were isolated from the spleen of untreated or squalene treated (single 180 µg dose/mouse) F1 mice 7 days post-injection. Cells were co-stained with CD4-PE, CTB-FITC and DAPI and analyzed by CLSM. Upper panels, quadrants indicating single-color stained cells at X40 magnification from untreated (left column) and squalene treated F1 mouse (right column): upper left quadrant, CTB-FITC staining (green); upper right quadrant, CD4-PE stain (red); lower left quadrant, DAPI staining of nuclei (blue), and lower right quadrant, merged channels for CTB-FITC, CD4-PE and DAPI staining. Middle panels, merged CTB-FITC, CD4-PE and DAPI staining of clusters of splenocytes from untreated and squalene treated F1 mouse at X220 magnification. Bottom panels, color intensity quantification for cell clusters shown in the middle panels using the using the ZEISS ZEN 2009 analysis software. Of note, the amount of GM1 moiety revealed by CTB-FITC (green) in merged channel 2 is significantly higher for the CD4 T-cells from squalene treated mouse than that of the untreated mouse. Shown is one of three mice analyzed in each group.
Figure 3.
Tyrosine phosphorylation patterns of major signaling modules of T helper differentiation upon enrichment of membrane cholesterol.
(A) SDS-PAGE silver stain of pooled, negatively-sorted CD4+ splenic T-cell lysates from F1 mice (n = 4/group) untreated (left lane) or squalene treated (single dose of 180 µg, right lane) 3 days post-injection shows no detectable quantitative alterations in the protein bands between the two groups of mice. (B) Tyrosine phosphorylation patterns of the same samples in panel A were blotted with anti-phosphorylated tyrosine Ab-HRP conjugate. Of note, the amount of 55–100 kDa phosphorylated protein bands is increased in mice treated with squalene. (C) Immunoprecipitation of pooled splenic CD4+ T-cell lysates from F1 mice treated or not with squalene (180 µg/mouse, n = 4/group) 3 days post-injection was carried out for IL-12Rβ2, IL-2Rα, IL-4Rα, CD28, or CD3 receptors using specific Abs, and probed with specific anti-phospho Abs for STAT4, STAT5, STAT6, PI3K, and ZAP-70 kinases. Only the phosphorylated STAT-4, STAT-5, and ZAP-70 in squalene treated mice were significantly enhanced. Shown is one of two representative experiments.
Figure 4.
Co-localization of cytokine receptors with lipid rafts of resting CD4 T-cells before and after squalene treatment.
Resting CD4 T-cells from individual F1 mice (n = 5/group) before and 7 days after squalene treatment (180 µg/mouse) were analyzed for interleukin receptor expression and distribution at the single-cell level by CLSM. Cells were stained with IL-4Rα-, IL-12Rβ2-, or IL-2Rα-PE conjugates, and co-stained for GM1 ganglioside by CTB- FITC conjugate and for nuclei with DAPI. First column indicates single-channel color for DAPI staining (blue), second column indicates GM1 staining (green), third column indicates interleukin receptor (IL-Rs) staining (red), and last column indicates merged channels at X63 magnification. Top-two rows, indicate cells from untreated (upper row) and squalene treated mice (lower row) stained for IL-4Rα. Middle-two rows, indicate cells from untreated (upper row) and squalene treated mice (lower row) stained for IL-12Rβ2. Bottom-two rows, indicate cells from untreated (upper row) and squalene treated mice (lower row) stained for IL-2Rα. Arrows indicate presence of IL-Receptor co-expression with the GM1 resident of lipid rafts. Enlargements of the merged channels are depicted to the right along with two different angles of the membrane for each IL-Receptor at X220 magnification. Shown are representative images in one of three experiments.
Figure 5.
Alteration in cytokine receptors mRNA expression after squalene enrichment of membrane cholesterol in resting lymphocytes.
(A) Quantitative real-time RT-PCR of IL-4Rα, IL-12Rβ2, and IL-2Rα mRNA extracted from peripheral blood lymphocytes of individual F1 mice (n = 5/group) analyzed before squalene treatment (dark bars) and 7 days after squalene injection (180 µg/mouse) (light bars). Y axis indicates the mean fold increase in mRNA expression level relative to the endogenous 18S rRNA expression level (control ± SD). Shown are two combined separate experiments (*p values<0.05). (B) Aliquots samples in panel A were stained with CD4-FITC conjugate, co-stained either with IL-4Rα-PE or IL-12Rβ2-PE or IL-2Rα-PE conjugates, and analyzed by FACS at the single-cell level for the surface IL-Rs expression level based on MFI measurements. Shown are the IL-Rs MFI values ± SD measured in individual mice before and after squalene treatment. Of note, no significant changes occurred in the IL-Rs expression on cell surface after squalene treatment (*p values>0.05).
Figure 6.
Squalene induced accumulation of membrane cholesterol in resting CD4 T cells favors Th1 polarization under antigen-specific or non antigen specific stimulation.
(A) Isolated adherent splenocytes (APCs, 5×105) from individual F1 mice treated or not with 180 µg of squalene, were pulsed (+) or not pulsed (-) with HA110–120 synthetic peptide (40 µg/mL/106 cells) in vitro 7 days after squalene injection, and co-cultured with negatively-sorted CD4 splenic T-cells (106 cells) from the same groups of mice, treated or not with squalene (Sq) (n = 4/group). Various cell co-culture combinations are shown on the X-axis, where (+) indicates presence and (–) indicates absence from the culture. Cell-culture supernatants were collected 24–48 h later, and secretion of IL-2, IL-4, and IFN-γ (Y-axis) was measured in pg/ml by Luminex. Bars represent average ± SD. Differences among groups were highly significant (p<0.001) for cytokines with the following exceptions denoted as §: For IL-4, co-culutre in lane 6 differed significantly from lane 5 (p = 0.047) but not from lane 3 (p = 0.097). No significant difference was observed between co-culture in lane 3 and 5 for any of the three cytokines. (B) Intracellular cytokine staining for IFN-γ (left panels) and IL-4 (right panels) in splenic cells and CD4-gated splenic cells from individual untreated (top panels) and squalene treated F1 mice (bottom panels) (n = 3/group) were stimulated for 48 h with anti-CD3/CD28 Abs (2.5 µg each/106 cells). Shown are the overlapped FACS histograms of gated CD4+ T cells synthesizing IL-4 or IFN-γ (red cell events), and total splenic cells (dark cell events) from squalene treated or untreated F1 mice, 7 days after squalene treatment. R1 gate indicates the low-proliferating cell population whereas the R2 gate indicates the high proliferating cell population in each experiment. Dead cells are shown in the un-gated cell population below the R1 gate. Shown is one of two representative experiments.
Figure 7.
Squalene enrichment of membrane cholesterol fosters the autoreactivity of antigen-specific Th1 diabetogenic cells.
(A) Diabetogenicity of antigen (HA110–120)-specific CD4 splenic Th1 cells was determined in RAG2 KO, RIP-HA mouse model for inducible type 1 diabetes (T1D). The HA110–120-specific CD4 splenic T-cells (2×105 cells) from either F1 mice (Foxp3-GFP+/− TCR-HA+/− mice) (filled circle), or from parental Foxp3-GFP+/+ mice (open circle) were infused alone, or in combination with Foxp3-GFP sorted splenic cells (2×105 cells) from Fl mice (filled triangle) or Foxp3-GFP sorted cells (2×105 cells) from parental Foxp3-GFP+/+ mice (open triangle) into RAG2 KO, RIP-HA Tg recipient mice (n = 5 mice/group). Shown is the mean glycemia values for individual mice in each group ± SD. Horizontal dark line indicates the upper limit of euglycemia (200 mg/dL) previously determined in a large cohort of non manipulated RAG2 KO RIP-HA Tg recipients (n = 20). Of note, only the antigen (HA110–120)-specific CD4 splenic T-cells, but not T-regs or other non-antigen (HA110–120)-specific T-cells were able to induce hyperglycemia in RAG2 KO, RIP-HA Tg recipients. (B) Diabetogenicity of antigen (HA110–120)-specific CD4 splenic Th1 cells from squalene treated F1 mice was tested in the same RAG2 KO, RIP-HA mouse model for inducible T1D described in panel A. The HA110–120-specific CD4 splenic T-cells (2×104 cells) from squalene treated (180 µg/mouse) or untreated F1 mice were first depleted of Foxp3+ splenic cells 7 days post-squalene treatment, and then infused i.p. in RAG2 KO, RIP-HA Tg recipients. Shown is the cumulative incidence of hyperglycemia (X-axis) from two separate experiments, n = 5 mice/group/experiment. Cumulative incidence of hyperglycemia was calculated by dividing the number of mice per group that developed two consecutive readings of hyperglycemia (>200 mg/dL) at various intervals of time by the total number of mice and then multiplied by 100.
Figure 8.
Squalene enrichment of membrane cholesterol in CD4+Foxp3+ T-reg cells does not alter their suppressogenic function.
(A) T-reg suppression of HA110–120 -specific CD4 T-effector (T-eff) cells from untreated or squalene treated (single dose of 180 µg/mouse) F1 mice (n = 4/group) has been tested in an in vitro suppression assay. Isolated HA110–120-pulsed APCs (5×105) were co-cultured for 48 h with FACS-sorted HA-specific CD4+ T-eff cells (106 cells) and FACS-sorted Foxp3-GFP+ T-regs (106 cells) from either untreated or squalene treated (Sq) F1 mice. Various cell co-culture combinations from individual mice are shown in the X-axis, where (+) indicates the presence and (–) indicates the absence of cells in the culture system. Cell-culture supernatants from each Treg/T-eff combination were then measured for secreted IFN-γ (Th1 cytokine) and IL-4 (Th2 cytokine) by Luminex (Y-axis). No significant changes were observed when compared different co-cultures as specified (brackets, *p values>0.05). (B) Quantitative real-time RT-PCR measured the T-bet mRNA extracted from triplicate wells of each T-reg/T-eff cells combination described in panel A. The Y axis indicates the mean fold increase in mRNA expression level relative to endogenous 18S rRNA control ± SD. The mRNA relative values were normalized to the untreated T-eff co-cultures with HA-pulsed APCs (reference control sample). Shown is one of two representative experiments. No significant changes were observed when compared different co-cultures as specified (brackets, *p values>0.05).