Cholera toxin B accelerates disease progression in lupus-prone mice by promoting lipid raft aggregation (original) (raw)

. Author manuscript; available in PMC: 2009 Sep 15.

Abstract

Infectious agents including bacteria and viruses are thought to provide triggers for the development or exacerbation of autoimmune diseases including systemic lupus erythematosus in the genetically predisposed individual. Molecular mimicry and engagement of Toll-like receptors (TLR) have been assigned limited roles that link infection to autoimmunity but additional mechanisms are suspected to be involved. Here we show that T cells from lupus prone mice display aggregated lipid rafts which harbor signaling, costimulatory, inflammatory, adhesion and TLR molecules. The percentage of T cells with clustered lipid rafts increases with age and peaks prior to the development of lupus pathology. We show that cholera toxin B, a component of Vibrio cholerae promotes autoantibody production and glomerulonephritis in lupus-prone mice by enhancing lipid raft aggregation in T cells. In contrast, disruption of lipid raft aggregation results in delay of disease pathology. Our results demonstrate that lipid rafts contribute significantly to the pathogenesis of lupus and provide a novel mechanism whereby aggregated lipid rafts represent a potential link between infection to autoimmunity.

Genetic and environmental factors contribute to the initiation and development of autoimmune diseases (1). Infectious agents, including bacteria and viruses, provide triggers for the initiation or exacerbation of autoimmune diseases in the genetically predisposed individual (2). Molecular mimicry and engagement of Toll-like receptor (TLR) have been assigned mechanistic roles whereby infectious agents instigate autoimmune diseases (3, 4). Molecular mimicry refers to shared structural homology between infectious agent components and proteins of the host (3). However, despite the fact that several infectious agents have been reported to promote autoimmune diseases, the number of identified molecular mimicry cases is quite limited (2). TLRs are key components of innate immune system and crucial regulators of both innate and adaptive immune responses. Autoimmune disease may be enhanced by several TLR ligands in mouse models, but TLR-dependent processes do not always explain the development of autoimmunity (4). The determination of additional mechanisms whereby pathogens provoke or promote autoimmune disease should be of scientific and clinical value.

In recent years, numerous studies demonstrate that a wide range of infectious agents including bacteria, viruses, parasites and prions infect mammalian cells only through intact lipid rafts (5–8). Lipid rafts have been shown to be involved in various cell processes including pathogen internalization, intracellular maturation of phagosomes, lysis and fusion of phagosomes, activation of intracellular signaling molecules, induction of cell death following infection and release of cytokines (7, 8).

Lipid rafts are enriched in sphingolipid and cholesterol microdomains on plasma membranes and serve as platforms that bring together signal proteins. As such they are pivotal in immune cell receptor-initiated signaling and its regulation both in terms of strength and duration (9). Besides the immune receptors that engage antigen on T, B and NK cells, a number of costimulatory molecules become invariable components of the lipid rafts and these include MHC class II, CD40, CD95, CD28, CTLA-4 and FcγRIIB1 (9–11). Lipid rafts are crucial in the regulation of T cell receptor (TCR) signaling (9).

Systemic lupus erythematosus (SLE) is a multi-system autoimmune disease characterized by inflammatory damage of various organs including the kidney and the skin, the production of autoantibodies against nuclear antigens and abnormalities in T cell function and receptor signaling (12). Current evidence suggests that T cells have an important role in the pathogenesis of SLE (13). Clustered lipid rafts have been found on the surface membrane of T cells from patients with SLE and shown to contribute to the aberrant CD3-mediated signaling (14, 15). Yet, it is unclear whether lipid rafts contribute to the pathogenesis of SLE. Patients with SLE are more prone to suffer various infections, which in turn may enhance disease activity and infections remain a major cause of morbidity and mortality in SLE (16). However, mechanisms whereby infections exacerbate SLE pathology remain unclear.

In this study we show that Vibrio Cholerae cholera toxin B (CTB) promotes disease progression in lupus-prone mice by enhancing T cell lipid raft aggregation, whereas disruption of lipid rafts delays disease progression. Clustered lipid rafts on T cells in MRL/lpr mice were found to contain diverse molecules including TCR signaling, inflammatory, costimulatory, adhesion and TLR molecules. The costimulatory T cell response mediated by these molecules following TCR ligation depends on the presence of intact lipid rafts. Our data strongly suggest that lipid rafts provide a potential link between infectious agents and autoimmune disease.

MATERIALS AND METHODS

Mice and materials

Female MRL/lpr/2J mice, NZB/W F1 mice, MRL/MPJ mice, B6.MRL/lpr mice, CD40 ligand knockout mice and C57BL/6 mice were purchased from the Jackson Laboratories (Bar Harbor, ME) and housed in the animal facility of Beth Israel Deaconess Medical Center. Methyl-β-cyclodextrin (MβCD), α-cyclodextrin (αCD), CTB, fluorescein isothiocyanate (FITC)-CTB, anti-actin antibody, double strand (ds) DNA, mouse IgG antibody, mouse complement 3 (C3) antibody and cytochalasin D were purchased from Sigma-Aldrich. Antibodies to TNFR2, IFN-γ, MCP-1, TNF-α, CD4 and CD8 for immunofluorescent staining were purchased from Santa Cruz (CA). Antibody of CD40L and TNFR2, recombinant TNF-α and MCP-1 were purchased from R & D system (MN). Syk and CD44 antibodies were purchased from ABCAM (MA). Antibodies to TLR4 and TLR9 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Immunofluorescent staining

T cells were isolated from spleen of MRL/lpr mice or C57BL/6 mice or other mouse strains using mouse T cell enrichment columns (R&D). T cells were fixed with 3% paraformaldehyde in phosphate-buffered saline (PBS) and cytospin onto slides, permeabilized with 0.05% Triton X-100 for 5 min at room temperature, and blocked with PBS containing 10% normal goat serum. Primary antibodies in 0.5% bovine serum albumin (BSA) was incubated for 45 min at room temperature. Cells were washed and incubated with a secondary antibody and FITC-conjugated CTB for 45 min. After three PBS washes, nuclei were stained with DAPI for 5 min. Slides were mounted with a coverslip using fluoromount-G. T cells were examined using a Zeiss LSM 510 Meta confocal microscopy. For enumeration of the percentage of cells with FITC conjugated CTB, 250 to 500 cells were scored visually by a single blinded observer. Cells in which staining with CTB-FITC did not display a homogenous ring staining pattern and displayed at least one aggregate covering one eighth of the cell circumference were identified as cells with aggregated lipid rafts.

Treatment of MRL/lpr mice by MβCD and CTB

Female MRL/lpr mice received MβCD (1 mg/mouse, i.p. n =12) or PBS (n=12) three times a week starting at age of 4 weeks for 16 weeks. Five MRL/lpr mice in each group were sacrificed at age of 16 weeks. The remaining mice were followed for survival to the end of the experimental period. Kidney and serum from sacrificed mice were collected for histological examination and measurement of serum IgG and anti-dsDNA antibody. Female MRL/lpr mice were treated intraperitoneally (i.p.) either with CTB (2 µg/mouse, n = 8) or PBS (n = 8) once a week starting at age of 4 weeks for 12 weeks or MβCD (1 mg/mouse) 3 times per week starting at age of 4 weeks for 4 months. Mice were sacrificed at end of the experiment and samples of serum, kidney were collected for examination. During the experimental period, urine protein content and mortality were monitored.

Kidney examination of histology and immunofluorence

Kidneys were fixed with 10% buffered formalin, embedded in paraffin, sectioned in 4-µm. Kidney tissues were stained with either the periodic acid-Schiff (PAS) reagent or with hematoxylin eosin. The slides were coded and blindly assessed. Renal pathology is graded by glomerular, interstitial and perivascular inflammation. Scores from 0 to 4 are assigned for each of the features. A minimum of 100 glomeruli were assessed to determine the glomerular scoring index in each mouse. For kidney immunofluorescence examination, cryostat-sectioned tissues were stained with fluorescein-conjugated antibodies detecting mouse IgG or mouse C3. Sections were examined with confocal microscopy.

Urine analysis

The mice in each group were placed overnight in a Nalgene metabolic cage to collect urine. Urine was measured with Multistix 10 SG and analyzed by Clinitek Status analyzer (Bayer Healthcare). Proteinuria is expressed as 0–4, 0 (none), 1 (30–100 mg/dl), 2 (100–300 mg/dl), 3 (300–2000 mg/dl), or 4 (> 2000 mg/dl).

Measurement of serum IgG and anti-DNA antibody

Serum IgG and anti-dsDNA antibody were detected by ELISA. 96-well plate was coated with double calf thymus DNA (1.5 mg/ml, Sigma) or mouse IgG antibody (0.1 mg/ml, Sigma) in PBS at 4°C overnight. After blocking with 1% BSA for 2 hours at room temperature (RT), serum samples were added to plate and incubated for 2 hours at RT. After washing 4 times with PBS, HRP-conjugated goat anti-mouse IgG antibody (Sigma) was added to plate and incubated for 1 hour at RT. Antibody binding was visualized using TMB (3’, 3’, 5’, 5’-tetramethylbenzidine, Sigma), and plates were read the optical density at 450 nm.

T cell proliferation assay

T cells (1×105) isolated from spleens of C57BL/6 mice were stimulated with anti-CD3 antibody (0.1 µg/ml) in the presence or absence of CD28 antibody (10 µg/ml), or TNF-α (20 ng/ml), or TNFR2 antibody (10 µg/ml), or MCP-1 (2 µg/ml) with or without MβCD (20 mg/ml) and cultured for 72 hours at 37°C. After 60 h, cultures were pulsed with [3H] thymidine (0.5µCi/10µl/well) for 12 additional hours. Data presented are mean counts per min (c.p.m.)± s.e.m. of responses from three replicate cultures. IFN-γ levels in culture supernatants were measured using mouse IFN-γ ELISA kit (R&D, Minneapolis, MN).

Statistics

Statistical evaluations of clustered lipid rafts, renal pathology, serum IgG, anti-dsDNA antibody, T cell proliferation and IFN-γ data were performed using the Student's _t_-test._P_≤0.05 was considered statistically significant. Survival was assessed using Kaplan-Meier survival and long rank test.

RESULTS

CTB promotes disease progression in MRL/lpr mice

After we found that lipid rafts are aggregated on the surface membrane of T cells from patients with SLE and demonstrated that they contribute to increased CD3-mediated signaling responses (14) we wished to determine whether lipid raft clustering contributes to the pathogenesis of disease. CTB, a non-toxic component of Vibrio Cholerae, binds directly to cell surface gangliosides (GM1) (17, 18) present in lipid rafts and mediates lipid raft aggregation (18, 19). Accordingly, we considered that if lipid raft clustering contributes to disease progression, treatment of lupus prone mice with CTB should accelerate the appearance of autoimmunity and related pathology. MRL/lpr mice, a well established animal model of SLE, were treated weekly with CTB (2 µg/mouse, i.p.) starting at the age of 4 weeks for 12 weeks. Whereas, three mice died in the CTB-treated group, all mice in the control, PBS-treated group remained alive (Fig. 1A, P=0.08). There was increased proteinuria, hematuria, numbers of leukocytes in urine, levels of serum IgG and anti-DNA antibody titers in mice treated with CTB compared to the control group (Fig. 1B–F). Kidney pathology was significantly more severe in CTB-treated mice compared to control mice (Fig. 1G–H). We noted increased IgG and C3 deposition in glomeruli of mice treated with CTB compared to glomeruli of mice treated with PBS (Fig. 1I). We also observed that lung and liver injury was significantly more severe in CTB-treated mice compared to control mice (Fig. 1J). We did not observe clear effect of CTB on the size of lymph nodes and spleen in the treated mice. These data demonstrate that CTB promotes disease severity in MRL/lpr mice.

Figure 1. Effect of CTB treatment on disease progression in MRL/lpr mice.

Figure 1

MRL/lpr mice were treated with CTB (2 µg/mouse, once weekly, i.p.) starting at 4 weeks of age for 12 weeks. Disease severity was assessed by survival (A, P= 0.08); proteinuria (B), hematuria (C), leukocytes in urine (D) *P< 0.01 (B, C, D), at 28 weeks of age; OD450 values for serum IgG and anti-dsDNA (E–F), severity of glomerulonephritis (glomeruli), interstitial infiltrate (interstitial) and perivasculitis (perivascular) (G) at 32 weeks of age. Representative images of PAS-stained kidney sections glomerular IgG (H) or C3 (I) deposit, liver and lung section (J) from 32 week old MRL/lpr mice are shown.•P = 0.045 (E), ••P = 0.046 (F), *P = 0.0008, ** P = 0.016, ***P = 0.034 (G). P value means CTB versus control PBS group.

Lipid raft clustering on T cells from MRL/lpr mice

To understand how CTB promotes disease progression in MRL/lpr mice, we first determined whether T cells from MRL/lpr mice display aggregated lipid rafts in a manner similar to that seen in human SLE T cells (14, 15). T cells isolated from spleen of MRL/lpr mice and C57BL/6 mice were stained with FITC-CTB and the lipid raft distribution on the surface membrane was studied using confocal microscopy. Increased numbers of T cells from MRL/lpr mice displayed aggregated lipid rafts on the surface membrane compared to similarly stained T cells from normal C57BL/6 mice (Fig. 2A).

Figure 2. Lipid raft clustering on T cells from MRL/lpr mice.

Figure 2

Confocal microscopy analysis of lipid raft clustering using FITC-conjugated cholera toxin B (FITC-CTB). Images (A); Kinetics of T cells with lipid raft clustering from female MRL/lpr mice and C57BL/6 mice (B). Percentage of T cells with clustered lipid rafts from female MRL/lpr (M/lpr), NZB/W F1 (B/W), MRL/MPJ (M/MPJ), B6MRL/lpr (B6.lpr) mice at 10 weeks of age (Ca) and NZB/W F1 mice at 8 months of age (Cb) and B6MRL/lpr mice at 10 months of age (Cb). * P < 0.01 (MRL/lpr versus other strains) (Ca). Percentage of T cells from MRL/lpr mice with clustered lipid rafts at 16 weeks of age and from C57BL/6 mice with or without stimulation of CD3+CD28 antibodies (70 µg/ml + 7 µg/ml) at 16 weeks of age (D). * P < 0.01 (CD3+CD28 or MRL/lpr versus control PBS) (D). T cells from C57BL/6 mice treated with CTB (2 µg/mouse, once a week) for four weeks starting at the age of 5 weeks were also stained with CTB (representative straining, left panel); T cells from C57BL/6 mice were treated with CTB (25 µg/ml) at 37°C (middle panel) or 4°C (right panel) and then stained with FITC-CTB (E). Percentage of T cells with clustered lipid rafts from MRL/lpr mice treated with CTB (2 µg, n = 3), or PBS (n = 3) once a week starting at 5 weeks of age for 4 weeks (F). Percentage of T cells from C57BL/6 with clustered lipid rafts 16 hours after the injection of CD3 Ab (70 µg/mouse) (G). Mice had been pre-treated with CTB (2 µg/mouse, once a week) for four weeks starting at the age of 5 weeks. •P = 0.034 (F), ••P = 0.0068 (G) (CTB versus control PBS group).

To determine the kinetics of appearance of lipid raft clustering, T cells from MRL/lpr mice were examined at different time points. We found that T cells with aggregated lipid raft appeared in 3 week old mice and peaked by the age of 6–7 weeks (Fig. 2B). To determine whether the appearance of T cells with clustered lipid rafts correlates with lupus pathology in MRL/lpr mice, we performed renal histology at different time points. We found that the appearance of T cells with clustered lipid rafts preceded the development of renal pathology. Eight week old mice did not display glomerulonephritis whereas glomerulonephritis was prevalent in 16-week old mice (not shown). There was a minimal (<5%) percentage of T cells with clustered lipid rafts on T cells from C57BL/6 mice throughout the study period (Fig. 2B). T cells from other lupus-prone strains such as NZB/WF1, MRL/MPJ and B6.MRL/lpr mice, which develop lupus pathology at later time point, did not display increased levels of T cells with clustered lipid rafts at 10 weeks of age in a manner similar to T cells from MRL/lpr mice (Fig. 2C, left panel). But we observed increased levels of T cells with clustered lipid rafts in NZB/W F1 at age of 8 months and B6.MRL/lpr mice at age of 10 months (Fig. 2C, right panel). Since MRL/lpr mice have increased numbers of spontaneously activated T cells, we asked whether the observed lipid raft clustering can be reproduced by TCR stimulation. C57BL/6 mice were administered intraperitoneally CD3 + CD28 antibodies and 16 hours later the percentage of T cells displaying aggregated lipid rafts was determined. We found high levels T cells displaying clustered lipid rafts in C57BL/6 mice following stimulation with CD3+CD28 antibodies (Fig. 2A, D).

Accelerated disease progression by CTB is due to enhanced lipid raft clustering

To further understand how CTB promotes disease progression in MRL/lpr mice, we performed following experiments: First, we determined whether treatment with CTB causes lipid raft clustering on normal T cells. T cells isolated from C57BL/6 mice treated with CTB (2 µg/mouse/week) for four weeks were stained with FITC-CTB. CTB treatment caused lipid raft clustering in T cells in vivo (Fig. 2E). We also found that the presence of CTB triggered lipid raft clustering in T cells from C57BL/6 mice when incubated at 37°C but not at 4°C in vitro (Fig. 2E). Next, we determined whether CTB treatment further enhanced lipid raft clustering on T cells in MRL/lpr mice. We found that treatment of MRL/lpr mice with CTB (2 µg/week for four weeks) further increased the percentage of T cells with clustered lipid rafts (Fig. 2F). In addition, we found that the numbers of T cells with clustered lipid rafts in C57BL/6 mice treated with CTB (2 µg/week for four weeks) followed by the administration of CD3 antibody (70 µg/mouse, 16 hours prior to the end of the experiment) increased significantly compared to the control group (Fig. 2G). Lastly, we excluded the possibility that CTB treatment caused kidney injury as we did not observe abnormal pathology in C57BL/6 mice treated with CTB (2 µg/day, i. p.) for one week, or weekly (2 µg, i.p.) for one month (data not shown). Also, we did not observe significant increase in the number of T cells in C57BL/6 mice treated daily with CTB (2 µg) for one week compared to control mice (data not shown), excluding thus the possibility that CTB treatment accelerates disease by causing lymphoproliferation. Together, these results suggest that accelerated lupus pathology progression in MRL/lpr mice following CTB treatment is due to enhanced lipid raft clustering on T cells.

Disruption of lipid rafts delays disease development in MRL/lpr mice

We considered that if CTB enhances lipid raft clustering and advances lupus pathology, then depletion of lipid rafts should delay disease progression in MRL/lpr mice. MRL/lpr mice were treated with MβCD, which is known to disrupt lipid rafts by depleting cholesterol (9, 14, 15). We found that disease progression, as manifested by proteinuria, kidney pathology, deposition of immunoglobulin and complement in glomeruli, serum IgG and anti-dsDNA antibody levels was significantly reduced in MRL/lpr mice treated with MβCD compared to PBS-treated control group at 16 weeks of age (Fig. 3A–G). We also observed that MβCD treatment (for 4 months) extended survival of MRL/lpr mice compared to control PBS treatment as recorded at 13 months of age (Fig. 3H). To confirm our results, we treated MRL/lpr mice with atorvastatin which is known to inhibit cholesterol synthesis. We found that inhibition of cholesterol synthesis by atorvastatin also inhibited disease progression in MRL/lpr mice (unpublished data). We did not observe obvious effect of MβCD on lymph node and spleen size in treated mice. These data provide novel proof that delayed disease progression in MRL/lpr mice by MβCD or atorvastatin is due to reduced T cell lipid raft clustering.

Figure 3. Effect of MβCD treatment on disease progression in MRL/lpr mice.

Figure 3

MRL/lpr mice were treated with MβCD (1 mg/mouse three times a week, i.p.). Treatment started at 4 weeks of age and lasted for 16 weeks. Proteinuria (A) and leukocytes in urine (B) in MRL/lpr mice treated with PBS (n = 7) or MβCD (n = 7) were recorded (*P< 0.01). At 16 weeks of age, OD450 values of serum IgG (C) and anti-dsDNA Ab (D), severity of glomerulonephritis (glomeruli), interstitial infiltrates (interstitial) and perivasculitis (perivascular) were determined (E). Representative images of PAS-stained tissue sections of kidney (F) and confocal images of glomerular IgG or C3 deposits (G) are shown. n = 5, PBS treatment; n=5, MβCD treatment. Survival of MRL/lpr mice treated with MβCD (n= 4, 1 mg/per mouse for 16 weeks, i.p) or PBS (n= 4) at age of 13 months (H). •P = 0.024 (C), ••P = 0.016 (D), *P = 0.02, **P = 0.005, ***P = 0.046 (E) (MβCD versus control PBS-treated groups).

Delayed disease progression by MβCD is due to reduced lipid raft clustering

To understand how MβCD delayed disease progression in MRL/lpr mice, we performed the following experiments: First, we investigated whether lipid raft clustering is reduced in MRL/lpr mice treated with MβCD. We found that the percentage of T cells with aggregated lipid rafts was significantly reduced in MRL/lpr mice treated with MβCD compared to T cells from mice treated with PBS or control αCD (Fig. 4A). Second, to determine whether MβCD inhibits TCR-mediated lipid raft clustering, C57BL/6 mice were treated with anti-CD3 antibody following the administration of MβCD for two weeks. The results show that TCR-mediated lipid raft clustering was reduced in mice treated with MβCD compared to control treated mice (Fig. 4B). Third, we investigated whether MβCD treatment induces apoptosis of T cells because delayed disease progression by MβCD might have resulted from increased T cell apoptosis. The number of T cells was similar between C57BL/6 mice treated with MβCD (1 mg) daily for one week and control PBS-treated mice (data not shown). Lastly, we found that the numbers of T cells with aggregated lipid rafts was also reduced in MRL/lpr mice treated with atorvastatin (not shown). These data indicate that delayed disease progression in MRL/lpr mice by MβCD is due to reduced lipid raft clustering on T cells.

Figure 4. Modulation of lipid raft clustering by MβCD.

Figure 4

Percentage of T cells from MRL/lpr mice with clustered lipid rafts after treatment with MβCD (1 mg, n = 3), αCD (1 mg, n = 3), or PBS (n = 3) for three times a week starting at 5 weeks of age for 2 weeks (A). Percentage of T cells from C57BL/6 mice with clustered lipid rafts following treatment with PBS (n = 3), MβCD (1 mg, n = 3), or αCD (1 mg, n = 3) three times a week starting at 5 weeks of age for two weeks and subsequently injected with CD3 antibody (Ab) (70 µg/mouse) (B). Confocal studies were performed 16 hours later. *P = 0.0019, **P = 0.01(MβCD versus control PBS group).

Clustered lipid rafts represent the assembly site for diverse molecules

Next we analyzed the composition of clustered lipid rafts on T cells in MRL/lpr mice. We found that clustered lipid rafts contain T cell signaling molecules including, CD3ε, ZAP-70, Lck, PI3K, Syk and LAT molecules (Fig. 5A). Lipid raft clusters existed on CD4+ and CD8+ T cells in MRL/lpr mice, but not on the surface membrane of double negative CD4−CD8−T cells (Fig. 5B). Costimulatory molecules, such as CD28 and CD40 ligand (CD40L) were also present in the clustered lipid rafts (Fig. 5B, C). Inflammatory molecules such as TNF-α, TNFR2, MCP-1 and IFN-γ which are known to play important roles in the pathogenesis of autoimmune diseases (20–23) colocalized with CTB on the surface membrane of MRL/lpr T cells (Fig. 5C). The adhesion molecule CD44, which is overexpressed in human SLE T cells (24), was found to localize in the lipid rafts of MRL/lpr T cells (Fig. 5C). TLR molecules including TLR4 and TLR9 were also found in lipid rafts of MRL/lpr T cells (Fig. 5D). Finally, we found that T cells infiltrating kidney tissues expressed Syk and CD44 (Fig. 5E). Together, these data demonstrate that clustered lipid rafts represent the assembly site for diverse signal molecules on T cells in MRL/lpr mice.

Figure 5. Confocal microscopic analysis of the composition of clustered lipid rafts on T cells in MRL/lpr mice.

Figure 5

T cells isolated from spleen of MRL/lpr mice or C57BL/6 mice at 6 weeks of age were stained with the indicated primary and secondary antibodies conjugated with Texas Red or PE and FITC-CTB and examined using confocal microscopy. TCR signaling molecules including CD3, ZAP70, Syk, LAT, Lck, PI3K, CD4 and CD8 localized to clustered lipid rafts in MRL/lpr mice, no TCR signaling molecules localized to lipid rafts in normal C57BL/6 mice (A–B). Costimulatory molecules CD28 and CD40L localized to clustered lipid rafts in MRL/lpr mice (B–C). Inflammatory molecules (TNFR2, TNF-α, IFN-γ and MCP-1) and adhesion molecules (CD44) localized to clustered lipid rafts in MRL/lpr mice (C). TLR4 and TLR9 localized to clustered lipid rafts in MRL/lpr mice (D). T cells infiltrating kidney tissue express Syk and CD44 molecules (E).

TCR costimulatory function depends on intact lipid rafts

To further understand how lipid raft clustering modulates disease progression in MRL/lpr mice, we determined whether diverse signaling molecules recruited in clustered lipid rafts contribute to TCR signaling since lipid raft clustering may amplify and sustain TCR signaling (9). We determined whether these recruited molecules can provide costimulation to TCR-mediated T cell proliferation since T cells from lupus-prone _Fas_-intact MRL mice display increased TCR-mediated proliferation (25). We found that TCR-induced T cell proliferation was significantly enhanced in the presence of anti-CD28 antibody, anti-TNFR2 antibody, TNF-α or MCP-1 compared to stimulation of TCR antibody alone and it was abolished in the presence of MβCD (Fig. 6A). These data demonstrate that molecules localized in the clustered lipid rafts have costimulatory effects on TCR signaling which depends on lipid raft clustering.

Figure 6. Importance of lipid raft clustering in TCR costimulation by diverse molecules.

Figure 6

Figure 6

Lipid raft clustering enhances TCR-mediated T cell proliferation (A). T cells from C57BL/6 mice were stimulated by CD3 Ab (0.1 µg/ml) with or without MCP-1 (2 µg/ml), TNFR2Ab (10 µg/ml), TNF-α (20 ng/ml) and CD28Ab (10 µg/ml) in the presence or absence of MβCD (20 µg/ml). *P = 0.0013, •P = 0.0007, **P = 0.0146, ***P = 0.0032 (when compared to the CD3-stimulated cells). Percentage of T cells with clustered lipid rafts in C57BL/6 mice treated with MβCD (1 mg/mouse, n = 3) or αCD (1 mg/mouse, n = 3) or PBS (n = 6) daily for two weeks prior to the injection of CD3 Ab (70 µg) with or without CD28 Ab (35 µg) (B). *P = 0.0062 versus CD3 alone (16h). **P = 0.0068 versus CD3+ CD28 (16h). Percentage of T cells with clustered lipid rafts from CD40L deficient mice (n = 6) or wild type mice (n = 6) 12 hours or 12 days after they were injected with CD3+CD28 Abs (70 µg + 7 µg) at 6 weeks of age (C).•P = 0.02, ••P = 0.0006 (CD40L−/− versus CD40L+/+ mice). IFN-γ levels in supernatants of T cells from C57BL/6 mice stimulated with CD3Ab (0.1 µg/ml) or CD3Ab (0.1 µg/ml) + CD28Ab (10 µg/ml) or CD3Ab (0.1 µg/ml) + CD28Ab (10 µg/ml) in the presence of MβCD (20 µg/ml) (D). *P = 0.0043 (compared to CD3 alone). IFN-γ levels in sera of CD40L deficient mice (n = 3) or wild type mice (n = 3) 3 days after stimulation with CD3+CD28 antibodies (70 µg + 7 µg). *P = 0.046 versus CD40L+/+ mice (E). T cells (1 ×105 cells/200 µl) from C57BL/6 and MRL/lpr mice (age of 9 weeks) were treated with CTB (1.25 µg/well) and then with an anti-CTB antibody (4.5 µg/well). Control T cells were only treated with the anti-CTB antibody. Supernatants were collected after T cells were incubated for 72 hours at 37°C. IFN-γ levels in supernatants were measured using mouse IFN-γ ELISA kit. **P=0.0024 (C57BL/6), *P=0.0004 (MRL/lpr) (F).

Next, we investigated whether recruited molecules promote lipid raft clustering. We determined whether CD28 promotes TCR-mediated lipid raft clustering since CD28 deficient mice are resistant to the development of lupus (26). C57BL/6 mice were treated with CD3 antibody alone or in combination with CD28 antibody. The results demonstrate that CD28 significantly enhanced lipid raft clustering on T cells in C57BL/6 mice in the presence of CD3 antibody compared to mice treated with CD3 antibody alone and MβCD inhibited this costimulation (Fig. 6B). These results indicate that recruited molecules further contribute to TCR-mediated lipid raft clustering. Along the same line, we investigated whether a deficiency of recruited molecules affects lipid raft clustering on T cells. Specifically, we asked whether a deficiency in CD40L reduces lipid raft clustering since CD40L-deficient are resistant to the development of lupus in MRL/lpr mice (27, 28). We found that lipid raft clustering was significantly reduced and lasted for a short time in CD40L-deficient mice treated with CD3+CD28 antibodies compared to wild type mice (Fig. 6C). These data demonstrate that CD40L deficiency affected lipid raft clustering on T cells.

Lastly, we investigated whether IFN-γ production depends on lipid raft clustering since IFN-γ is known to promote lupus pathology in MRL/lpr mice (22). We found that IFN-γ levels were markedly increased in supernatants from T cells treated with CD3+CD28 antibodies compared to CD3 antibody alone and increased IFN-γ levels were abolished by MβCD (Fig. 6D). Also, serum IFN-γ levels were reduced in CD40L-deficient mice following stimulation of CD3+CD28 antibodies compared to wild type mice (Fig. 6E).

Lastly, we asked whether aggregation of lipid rafts would affect T cell function. To this end, we purified T cells from C57BL/6 mice and MRL/lpr mice (age of 9 weeks) treated them with CTB and then with an anti-CTB antibody for 72 hours. Control T cells were only treated with the anti-CTB antibody alone. As shown in Fig. 6F, cross-linking of lipid rafts with CTB resulted in significantly increased IFN-γ production by both C57BL/6 (P=0.0024) and MRL/lpr (P=0.0004). It should be noted that the increase was more pronounced in the MRL/lpr T cells.

Lipid raft clustering depends on actin polymerization in MRL/lpr mice

To understand how lipid raft clustering occurs, we investigated whether actin polymerization is important for the assembly of diverse molecules (29, 30). Confocal microscopy results showed that actin colocalized with lipid rafts on T cells from MRL/lpr mice (Fig. 7A). To further determine the role of actin in lipid raft clustering, we investigated whether inhibition of actin polymerization limits lipid raft clustering on T cells from MRL/lpr mice. MRL/lpr mice were treated with or without cytochalasin D, an inhibitor of actin polymerization (31, 32). Cytochalasin D (1 mg/kg) was administered daily for one week. We found that lipid raft clustering was significantly inhibited in MRL/lpr mice following treatment with cytochalasin D compared to control (Fig. 7B). These results suggest that lipid raft clustering on T cells in MRL/lpr mice localizes diverse molecules through actin polymerization.

Figure 7. Lipid raft clustering depends on actin polymerization.

Figure 7

Confocal microscopic image of actin co-localized with clustered lipid raft in T cells from MRL/lpr mice (A). Actin co-localization was abolished in T cells from MRL/lpr mice treated with cytochalasin D. Percentage of T cells with lipid raft clustering from MRL/lpr mice treated with cytochalasin D (20 mg/kg, n = 4), vehicle (n = 4) or PBS (n = 4) daily starting at 6 weeks of age for one week (B). *P = 0.0074 (compared to the vehicle group).

DISCUSSION

T cells with clustered lipid rafts appear in MRL/lpr mice as early as 3 weeks of age and their numbers peak by the 7th week of age. Clustered lipid rafts contain signaling, adhesion and costimulatory molecules. Our studies demonstrate that CTB accelerates disease progression in MRL/lpr mice by enhancing aggregation of T cell lipid rafts, whereas disruption of lipid rafts delayed disease progression in MRL/lpr mice.

The critical role of T cells in the development of autoimmune disorders in MRL/lpr mice has been demonstrated in previous studies (33–36). Enhancement of CD3-mediated signaling by costimulatory molecules facilitates T cell activation and promotes the development of autoimmunity. Clustered lipid rafts presumably provide more stable, efficient signaling platforms for TCR signaling molecules and amplify and sustain TCR signaling. Our studies demonstrate that recruited molecules, such as CD28, TNFα, TNFR2, MCP-1 and CD40L have costimulatory activity on TCR-mediated signaling. The costimulatory function to TCR-mediated signaling depends on intact lipid rafts because disruption of lipid rafts abolished the costimulatory effects to TCR-mediated proliferation and production of IFN-γ (Fig. 6). Cross-linked components of lipid rafts such as CD28 significantly enhanced TCR-mediated lipid raft clustering. Loss of clustered lipid raft components, such as CD40L, resulted in defective TCR-mediated lipid raft clustering. Actually, T cells which lack TNFR2 are defective in TCR-mediated activation (37). Blocking of recruited signaling molecules to lipid rafts may inhibit lupus pathology. Indeed, inhibition of PI3K has been shown to inhibit lupus pathology in MRL/lpr mice (38, 39). Lipid raft clustering may promote activated T cells extravasation from blood vessels to sites of pathological organ tissue because components, such as CD44, facilitate this process (24).

Cholera toxin (CT) from V. cholerae, a gram negative bacterium, may cause massive secretory diarrhea. CT is composed of an enzymatic A-subunit and a homo-pentameric B-subunit (40). CT exerts its effect through CTB which is non-toxic and binds to GM1(41). CTB binding to GM1 induces lipid raft clustering on mammalian cells such as T cells and epithelial cells (14, 18, 19). CTB-mediated lipid raft clustering depends on actin cytoskeleton (42). Our results demonstrate that actin polymerization plays an important role in aggregation of lipid rafts on T cells in MRL/lpr mice. Therefore, CTB treatment may promote lipid raft clustering on T cells through actin polymerization. Clustered lipid rafts represent the assembly site of diverse signaling molecules including TCR. TCR for specific autoantigen is also recruited to clustered lipid rafts and therefore, autoantigens binding TCR localized in clustered lipid rafts amplify their signaling transduction and promote autoimmunity and development of autoimmune disease. Since lipid rafts also play important role in B cell receptor (BCR) signaling and B cells are critical in pathogenesis of SLE, it is possible that CTB or MβCT treatment enhances or decreases lipid raft clustering on B cells and affects disease progression in MRL/lpr mice.

SLE patients are more prone to suffer infection-related morbidity and mortality (16). Lipid rafts from host cells are exploited by a wide range of infectious agents including Brucella, Chlamydia, Legionella, Listeria, Pseudomonas, Salmonella, Shigella, Streptococcus, EBV, Ebola, hepatitis C, HIV, HSV and influenza viruses and some protozoa such as toxoplasma, plasmodium to enter, survive and replicate within the cell (5, 7). Bacterial pathogens by activating G proteins they employ the cytoskeleton to invade a host cell and to gain motility in the cell. G-protein activation, in turn, induces the generation of actin-rich membrane ruffles to internalize the bacteria (43). The G-protein and actin cytoskeleton exert crucial role in T cell activation and TCR signaling (44, 45). Other studies and our study demonstrate that TLRs including TLR4, 7 and 9 (46–48) are localized to lipid rafts. TLR binding to pathogen-associated molecular patterns may promote lipid raft clustering which in turn provide a stable platform to amplify TLR signaling transduction. TLR-mediated cell signaling also depends on intact lipid rafts (46–48).

Poorly understood interactions between environmental and genetic factors lead to the development of autoimmune diseases. Genetic factors may account for increased production of molecules including CD3-associated signaling molecules and TLRs, whereas components of infectious agents provoke the appearance of SLE by enhancing lipid raft clustering. Patients treated with immunosuppressive agents such as prednisone and cytotoxic drugs become prone to suffer infections which in turn may cause disease reactivation. Our results demonstrate that lipid raft clustering contributes to the pathogenesis of SLE because their dissolution delays disease whereas acceleration of their formation promotes disease pathology. Clustered lipid rafts in MRL/lpr mice T cells harbor various molecules which appear to amplify and maintain TCR mediated signaling. Our data strongly suggest that lipid rafts provide a potential link though which infections instigate or promote autoimmune disease.

ACKNOWLEDGMENTS

We thank Dr. P.H. Lapchak for critically reading the manuscript. This work was supported in part by NIH R01 AI42269.

REFERENCE