Oxidized phospholipids induce anergy in human peripheral blood T cells

Authors


Abstract

Lipids are key regulators of immune responses. In this study we investigated the direct impact of oxidized phospholipids (ox-PL) on T cell activation and function. We could demonstrate that ox-PL strongly inhibit proliferation of purified human T cells induced with anti-CD3/CD28 or anti-CD3/CD63 mAb, whereas proliferation of naive T cells from human cord blood was not affected by ox-PL. Unoxidized phospholipids showed no such effect. Inhibition of T cell proliferation by ox-PL was not due to cell death. Moreover, T cell proliferation triggered by PMA/ionomycin activation was not diminished by ox-PL. T cells activated in the presence of ox-PL produced and released low amounts of IFN-γ and IL-2, whereas IL-4 was only slightly diminished. Ox-PL prevented the expression of de novo synthesized activation markers (CD25, MHC class II) but not expression of CD63 or CD69. We further observed that T cells stimulated in the presence of ox-PL are poorly cytotoxic T cells. Most importantly, T cells activated in the presence of ox-PL failed to proliferate in response to restimulation. This hypo-proliferative state was accompanied with an up-regulation of early growth response gene 3 and Casitas B-lineage lymphoma protein B. Taken together, our results demonstrate that ox-PL are potent and specific regulators of T cell activation and function.

Abbreviations:
Cbl-b:

Casitas B-lineage lymphoma protein B

Egr-3:

early growth response gene 3

OxPAPC:

oxidation products derived from PAPC

ox-PL:

oxidized phospholipid

PAPA:

1-palmitoyl-2-arachidonoyl phosphatidic acid

PAPC:

1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine

PAPG:

1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphoglycerol

PAPS:

1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphatidylserine

PGPC:

1-palmitoyl-2-glutaroyl-sn-glycero-3-phosphorylcholine

POVPC:

1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-phosphorylcholine

Introduction

Inflammatory signals are typically considered to initiate adaptive immune responses. The endogenous mechanisms which regulate and terminate inflammatory reactions and consequently turn off immune reactions, are incompletely understood. Inflammatory processes are initiated by the innate immune system and involve the release of reactive oxygen species, which in addition to killing pathogens also cause oxidation of various surrounding molecules including phospholipids 1, 2.

Accumulating evidence suggests that oxidized phospholipids (ox-PL) are not merely side products of the inflammatory response, but can actively regulate inflammation. Most studies focused on the pro-inflammatory effects of ox-PL, which are thought to play a role in initiating and maintaining chronic inflammation such as atherosclerosis 1, 2. However, it is increasingly recognized that ox-PL at the same time possess potent anti-inflammatory properties, which include the direct antagonism of LPS recognition by cells of the innate immune system. Indeed, ox-PL effectively inhibit the interaction of LPS with LPS-binding protein, CD14, and TLR4 3. The biological significance of this finding was further underlined by the observation that exogenous administration of ox-PL could prevent mortality of mice exposed to high doses of LPS 3 but may contribute to mortality during Gram-negative sepsis in vivo via impairment of phagocytosis 4.

Moreover, ox-PL can also modulate the stimulatory function of professional antigen-presenting cells. We have recently demonstrated that oxidation products derived from 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine (OxPAPC) are specific regulators of DC activation 5. We could show that OxPAPC does not activate DC but efficiently prevents LPS-induced activation of DC, thereby limiting their capacity to stimulate T cells. OxPAPC inhibition was specific for LPS and did not affect DC maturation induced via CD40-CD40L interactions. Thus, ox-PL may represent negative feedback regulators of adaptive immune responses through limiting pathogen-induced activation of DC.

The potential direct effects of various types of ox-PL on cells of the adaptive immune system have not been analyzed so far. Therefore, this study was performed to investigate the direct impact of ox-PL on T cell activation and function. We found that ox-PL prevent the development of Th1-type responses, inhibit T cell proliferation and the induction as well as the effector phase of CTL induced through stimulation via the TCR/CD3 complex. Thus, ox-PL might represent novel immunosuppressive factors for adaptive immunity with unique characteristics.

Results

ox-PL inhibit T cell proliferation induced via TCR/CD3

In order to investigate the direct impact of ox-PL on T cell activation and function, we stimulated highly purified human peripheral blood T cells in the presence or absence of OxPAPC via TCR/CD3 cross-linking with plate-bound CD3 mAb. Results presented in Fig. 1A demonstrate that OxPAPC strongly inhibits T cell proliferation induced by CD3 mAb. To analyze whether the delivery of a costimulatory signal might overcome this inhibitory effect of OxPAPC, T cells were stimulated with CD3 plus CD28 mAb or CD3 plus CD63 mAb, an alternative T cell activation pathway 6. Activation of T cells via both routes resulted in vigorous proliferation (Fig. 1B and data not shown). However, in both cases addition of OxPAPC strongly inhibited T proliferation in a dose-dependent manner. In contrast to that, PAPC showed no inhibitory effect (Fig. 1C). Stimulation of T cells with a combination of PMA, which activates the protein kinase C pathway, and the calcium ionophore ionomycin was not affected by OxPAPC (Fig. 1D).

Figure 1.

OxPAPC inhibits peripheral blood T cell proliferation. (A, B, D) Purified human peripheral blood T cells were stimulated for 4 days with plate-bound CD3 mAb (A), CD3/CD28 mAb (B), or for 3 days with PMA/ionomycin (D) with or without OxPAPC. (C) Peripheral blood T cells were stimulated with plate-bound CD3/CD28 mAb in the presence or absence of PAPC. Mean values of triplicate determinations ± standard deviation for one representative of six independent experiments are shown. (E, F) Purified human naive T cells isolated from cord blood were stimulated with plate-bound CD3/CD28 mAb (E) or PMA/ionomycin (F) for 2 days with or without OxPAPC. Proliferation of T cells was monitored by adding [methyl-3H]thymidine followed by measuring [methyl-3H]thymidine incorporation 18 h later. Mean values of triplicate determinations ± standard deviation for one representative of four independent experiments are shown; ***p<0.001.

Next we analyzed the influence of ox-PL on proliferation of naive T cells isolated from cord blood. Interestingly, OxPAPC did not inhibit the proliferation of naive T cells stimulated via CD3/CD28 or PMA/ionomycin (Fig. 1E, F). OxPAPC also did not inhibit the proliferation of the T cell line Jurkat (data not shown).

These findings demonstrate that OxPAPC is not a general inhibitor of T cell proliferation and is not toxic for T cells at the concentrations used in this study. In line with that, monitoring of cell death in the T cell cultures by annexin V/PI staining revealed no increased cell death in the presence of OxPAPC (Fig. 2B), whereas T cell proliferation (Fig. 2A) and T cell numbers (Fig. 2C) were diminished.

Figure 2.

Viability of OxPAPC-treated T cells. (A, B) Purified human peripheral blood T cells were stimulated with plate-bound CD3/CD28 mAb (squares) or PMA/ionomycin (circles) over 6 days with (filled symbols) or without (open symbols) OxPAPC (5 µg/mL). (A) Proliferation of T cells was monitored by adding [methyl-3H]thymidine followed by measuring [methyl-3H]thymidine incorporation 18 h later. (B) Cell death of cultured T cells was assessed by staining with annexin V and PI in parallel. Mean values of triplicate determinations ± standard deviation for one representative of three independent experiments are shown. (C) Total cell numbers upon culture with plate-bound CD3/CD28 mAb in the presence of OxPAPC were assessed at day 5 of culture. The results of six independent experiments ± standard deviation are shown; ***p<0.001.

Characterization of the molecular moieties of OxPAPC required for the inhibitory feature

Oxidation of PAPC leads to the generation of a mixture of oxidation products (OxPAPC), some of which are potent bioactive substances 2, 7. 1-Palmitoyl-2-glutaroyl-sn-glycero-3-phosphorylcholine (PGPC) and 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-phosphorylcholine (POVPC) are the best-defined components of OxPAPC. These lipids as well as ox-PL with different head groups: oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphatidylserine (OxPAPS), oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphoglycerol (OxPAPG) and oxidized 1-palmitoyl-2-arachidonoyl phosphatidic acid (OxPAPA), were tested for their capacity to inhibit T cell proliferation. Results presented in Fig. 3 demonstrate that all lipid preparations tested were able to inhibit T cell proliferation, except POVPC (Fig. 3B). Also dimyristoylphosphatidylcholine, an unoxidized lipid containing saturated fatty acids, failed to inhibit T cell proliferation (Fig. 3A). Treatment of OxPAPC with triphenylphosphine or NaBH4 to reduce hydroperoxide groups or aldehyde and keto groups, respectively, did not remove the inhibitory capacity of OxPAPC (Fig. 3C).

Figure 3.

Influence of different lipids on T cell proliferation. Peripheral blood T cells were stimulated with plate-bound CD3/CD28 mAb in the presence of the indicated lipids (1, 5 and 10 μg/mL) (A, B) or OxPAPC (1,5 and 10 μg/mL) treated with triphenylphosphine (TPP) or NaBH4 (C). Proliferation of T cells was monitored on day 4 of culture by adding [methyl-3H]thymidine followed by measuring [methyl-3H]thymidine incorporation 18 h later. Mean values of triplicate determinations ± standard deviation for one representative of three independent experiments are shown; **p<0.005, ***p<0.001.

Impact of OxPAPC on the expression of T cell activation marker

In order to investigate the inhibitory effect of OxPAPC on T cell activation in more detail, we next analyzed the expression of typical cell surface activation markers on T cells stimulated in the presence of OxPAPC. Results presented in Fig. 4 demonstrate that the up-regulation of MHC class II and CD25 molecules was reduced by OxPAPC, whereas the induction of CD63 and CD69 remained unchanged. Thus, ox-PL prevented the expression of de novo synthesized activation markers (CD25, MHC class II) in T cells, but not of CD63 or CD69, which originate largely from preformed intracellular pools 6.

Figure 4.

Regulation of activation markers on T cells by ox-PL. Peripheral blood T cells stimulated for 24 h via plate-bound CD3/CD28 mAb with or without OxPAPC (10 µg/mL). Expression of the activation markers MHC class II, CD25, CD63 and CD69 was analyzed by flow cytometry. The results of one representative of five independent experiments are shown.

OxPAPC inhibits type-1 cytokine production in T cells

Cytokine production in T cells plays an important role in the regulation of the immune response. Therefore, we wondered whether the proliferation-inhibiting effect of OxPAPC was accompanied with an altered cytokine expression profile.

Analysis of the cytokine expression profile of T cells stimulated with CD3/CD28 mAb in the presence of OxPAPC at the mRNA level revealed that OxPAPC strongly reduced the expression of IFN-γ, IL-2 and IL-10 (Fig. 5A). In contrast, IL-4 and IL-13 expression were only slightly reduced and cytokine production induced by PMA/ionomycin was not affected (Fig. 5A, C). The same response was detectable at the protein level. Using intracellular detection of cytokine production and measurement of released cytokines in the supernatant of T cells activated via CD3/CD28 mAb, we also observed that IFN-γ production was much more affected by OxPAPC than IL-4 (Fig. 5C, Table 1). Compared to peripheral blood T cells, OxPAPC has only a marginal inhibitory effect on IFN-γ production in naive T cells (Fig. 5B). In contrast to peripheral blood T cells, we found an increase of IL-4 production in naive T cells. Thus, OxPAPC seems to favor Th2 differentiation in naive T cells.

Figure 5.

OxPAPC inhibits Th1-type cytokine production. (A) Peripheral blood T cells were stimulated with plate-bound CD3/CD28 mAb or PMA/ionomycin for 24 h without (grey bars) or with (black bars) OxPAPC (5 μg/mL) and cytokine production was measured at the mRNA level via real-time PCR. Mean values ± standard deviation calculated from five independent experiments are shown. (B) Naive T cells from cord blood were stimulated with plate-bound CD3/CD28 mAb or PMA/ionomycin for 72 h without (grey bars) or with (black bars). OxPAPC (5 μg/mL) and cytokine production was measured at the mRNA level by real-time PCR. Mean values ± standard deviation calculated from two independent experiments are shown. (C) Cytokine levels in the supernatant of peripheral blood T cells stimulated with CD3/CD28 mAb or PMA/ionomycin in the presence or absence of OxPAPC for 24 h were measured by Luminex. The results of four independent experiments ± SEM are shown; **p<0.005.

Table 1. Intracellular detection of IFN-γ and IL-4
StimulationOxPAPCIFN-γa)IL-4a)
  1. a) Mean values ± standard error of the mean of four independent experiments are shown.

Mock0.4 ± 0.22.5 ± 1.8
CD3/CD2815.8 ± 1.415.9 ± 5.4
CD3/CD281 µg/mL12.5 ± 2.315.7 ± 7.3
CD3/CD285 µg/mL7.5 ± 1.419.9 ± 8.4
CD3/CD2810 µg/mL4.2 ± 1.59.4 ± 3.9

T cell activation in the presence of OxPAPC leads to anergy

The above findings that T cells stimulated via TCR/CD3 mAb in the presence of ox-PL do not proliferate but are still alive and show reduced cytokine levels, raised the question of whether these T cells can be restimulated with plate-bound CD3/CD28 mAb. Results presented in Fig. 6A indicate that T cells which were primed in the presence of OxPAPC failed to proliferate in response to restimulation. We also observed that proliferation of T cells preactivated via CD3/CD28 is strongly inhibited in the presence of OxPAPC during restimulation (Fig. 6B). This finding seems to be dependent of the amount of lipid used (Fig. 6A, B). T cells that were not treated with OxPAPC in the primary response proliferated due to restimulation with CD3/CD28 mAb. Inhibition of T cell proliferation by OxPAPC is reverted by exogenous IL-2, however only at low levels of OxPAPC (1 µg/mL) but not at 10 µg/mL (Fig. 6C).

Figure 6.

OxPAPC induces anergy. (A) Peripheral blood T cells were stimulated with plate-bound CD3/CD28 mAb or mock-stimulated for 5 days in the presence of 1, 5 and 10 μg/mL of OxPAPC. After washing and further culture in fresh media for 4 days without OxPAPC, cells were restimulated with plate-bound CD3/CD28 mAb for 4 days without OxPAPC. (B) Peripheral blood T cells were stimulated with plate-bound CD3/CD28 mAb for 5 days. After washing and further culture in fresh media for 4 days, T cells were restimulated with plate-bound CD3/CD28 mAb for 4 days in the presence of 1, 5 and 10 μg/mL of OxPAPC. (C) Peripheral blood T cells were stimulated with plate-bound CD3/CD28 mAb or mock-stimulated for 4 days in the presence of 1, 5 and 10 μg/mL of OxPAPC without (black bars) or with exogenous IL-2 (50 U/mL; open bars). Proliferation of T cells was monitored by adding [methyl-3H]thymidine followed by measuring [methyl-3H]thymidine incorporation 18 h later. Mean values from triplicate determinations ± standard deviation for one representative of four independent experiments are shown. (D, E) Peripheral blood or cord blood T cells were stimulated with plate-bound CD3/CD28 mAb or PMA/ionomycin for 24 h with or without OxPAPC (5 μg/mL) and production of Egr-3 and Cbl-b was measured at the mRNA level via real-time PCR, calculated from three independent experiments. Mean values ± standard deviation are shown; **p<0.005, ***p<0.001.

We then analyzed whether OxPAPC may up-regulate anergy-related genes and found that the transcription factor early growth response gene 3 (Egr-3) and the E3 ligase Casitas B-lineage lymphoma protein B (Cbl-b) are strongly induced by OxPAPC in peripheral blood T cells stimulated via CD3/CD28 (Fig. 6D, E). In contrast, OxPAPC did neither up-regulate the expression of both genes when peripheral blood T cells were stimulated with PMA, nor in naive T cells (Fig. 6D, E). Thus, induction of anergy in T cells by OxPAPC is accompanied with the well-defined negative regulators of T cell activation, Egr-3 and Cbl-b 8.

Next we analyzed whether T cells that had been anergized through OxPAPC treatment gain a suppressor function. Results presented in Fig. 7 demonstrate that T cells prestimulated in the presence of OxPAPC do not suppress proliferation of co-cultured T cells.

Figure 7.

OxPAPC-treated T cells are not suppressor cells. Peripheral blood T cells were preactivated with plate-bound CD3/CD28 mAb or mock-treated for 5 days in the presence of 1, 5 and 10 μg/mL of OxPAPC. Then the cells were irradiated, mixed at a 1:1 ratio with untreated T cells and stimulated with CD3/CD28 mAb. Proliferation of T cells was monitored on day 4 of culture by adding [methyl-3H]thymidine followed by measuring [methyl-3H]thymidine incorporation 18 h later. Mean values of triplicate determinations ± standard deviation for one representative of three independent experiments are shown.

The activity of cytotoxic T cells is reduced by OxPAPC

CD8+ effector cytotoxic T lymphocytes (CTL) recognize targets presenting an antigenic peptide on MHC class I and kill the target cells. CTL induction was determined in the presence or absence of OxPAPC. T cells were activated with plate-bound CD3/CD28 mAb 3 days before performing the cytotoxicity assay. Results presented in Fig. 8A demonstrate that OxPAPC inhibits differentiation of T cells into potent CTL.

Figure 8.

OxPAPC inhibits CTL differentiation and function. (A) Peripheral blood T cells were stimulated with plate-bound CD3/CD28 mAb without or with OxPAPC (7.5 µg/mL) for 4 days. THP1 cells were loaded with 51Cr (50 µCi) for 1 h at 37°C and used as target cells (5×103) for decreasing numbers of preactivated T cells (1:40, 1:10, 1:2.5, 1:0.8). (B) OxPAPC (7.5 µg/mL) was directly added to the mixture of THP1 (target) cells and T cells preactivated with CD3/CD28 mAb. (C) Influence of OxPAPC on NK cells. K562 cells were used as target cells and co-cultured with decreasing numbers of freshly isolated PBMC for 4 h with or without directly added OxPAPC (7.5 µg/mL). Released 51Cr was analyzed in all experiments after 4 h. Mean values of triplicate determinations ± standard deviation for one representative of three independent experiments are shown.

To examine the effect of OxPAPC on the effector phase of CTL-mediated killing, OxPAPC was directly added to co-cultures of target cells and CD3/CD28 mAb-preactivated T cells. Also in this case, target cell lysis was inhibited (Fig. 8B). In contrast, lysis of target cells by NK cells was not diminished by OxPAPC (Fig. 8C). Thus, ox-PL do not block cytolytic mechanisms in general but selectively target TCR/CD3-mediated activation of the lytic machinery.

Discussion

Ox-PL are side products of collateral damage of surrounding tissue caused by free oxygen radicals in inflammatory processes 1, 2. Here, we demonstrate that such ox-PL selectively regulate the functional repertoire of human T cells. ox-PL were found to prevent the development of Th1-type responses, T cell proliferation and both the induction as well as the effector phase of CTL induced through stimulation via the TCR/CD3 complex. These observations suggest that direct inhibitory effects of ox-PL on T cells might contribute to control T cell function and to avoid overwhelming or even detrimental Th1-driven immune responses at sites of inflammation.

We observed that ox-PL regulate certain but not all activation parameters and functions of T cells induced by activation via the TCR/CD3 complex. The impact of ox-PL on T cell proliferation was particularly striking and resulted in a deep anergic state of peripheral blood T cells but not in naive T cells. This inhibitory effect was not overcome by costimulatory signals delivered by cross-linking of CD28 or CD63 cell surface receptors on T cells 6. Accordingly, ox-PL prevented the expression of de novo synthesized T cell activation markers (CD25, MHC class II), but not of CD63 or CD69, which originate from preformed intracellular pools 6, on the cell surface of anti-CD3/CD28 mAb-stimulated T cells. Thus, treatment of T cells with ox-PL rather modulates TCR/CD3 signaling than completely blocks it. Moreover, since T cell activation induced by PMA or PMA/ionomycin stimulation was not affected by ox-PL, the functional integrity of T cells is obviously not disturbed by ox-PL.

Our conclusion that ox-PL are selective regulators of T cell function, is also supported by the cytokine profile of T cells stimulated in the presence of ox-PL. We found that ox-PL strongly inhibit the production and release of Th1- but not Th2-type cytokines in T cells upon stimulation with CD3/CD28 mAb, whereas PMA/ionomycin-induced responses were not altered. These observations suggest that factors present at the site of inflammation can influence polarization by acting directly on T cells. Interestingly, ox-PL-treated T cells produced only low amounts of IL-10, a well-established immunosuppressive cytokine 9. Thus, it is unlikely that the inhibitory effects of ox-PL on T cells are due to elevated levels of IL-10 production. In addition, we have observed in this study that ox-PL efficiently repress the cytolytic function of CTL whereas IL-10 rather enhances CTL responses 10. Like IL-10, other prominent immunosuppressive drugs such as FK506, cyclosporine A, glucocorticoids or rapamycin also fail to repress the cytotoxic function of CTL 1113. Thus, ox-PL are novel modulators of adaptive immunity with unique immunosuppressive characteristics.

Oxidation of PAPC generates a mixture of oxygenated products (OxPAPC), some of which are potent bioactive substances 7. We also tested ox-PL with other head groups including OxPAPS, OxPAPG and OxPAPA and found very similar inhibitory effect. Thus, the head group of the ox-PL is apparently not the critical part of the molecule responsible for the suppression of T cell proliferation. Moreover, reduction of hydroperoxide, aldehyde and keto groups with triphenylphosphine or NaBH4 did not eliminate the inhibitory capacity of OxPAPC. Thus, further lipid peroxidation or electrophilic stress caused by OxPAPC are obviously also not the critical mechanisms involved.

The two best-characterized phospholipids within OxPAPC are POVPC and PGPC 7. Both substances are generated by oxidative fragmentation of arachidonic acid in PAPC. POVPC contains a 5-carbon residue with an ω-aldehyde group at the sn-2 position, in contrast to the ω-carboxylic group in PGPC. Despite grossly similar chemical structure, these two compounds act on T cells in very different ways. Whereas PGPC inhibited T cell proliferation, POVPC showed no such effect. Different functional effects of PGPC and POVPC have been reported previously for other cell types including endothelial cells, where both lipids induced adhesion of monocytes to endothelial cells but only PGPC induced adhesion of neutrophils 14. Previously, Leitinger et al.2, 14 hypothesized that POVPC and PGPC may act through different types of membrane receptors. Thus, human T cells may express a specific receptor for the 5-carbon residue fragments from arachidonic acid with an ω-carboxylic group and ox-PL might trigger a negative signal via this receptor.

Interestingly, OxPAPC and PGPC, but not POVPC, have been found to up-regulate tissue factor in endothelial cells and this effect was partially mediated by the transcription factor NFAT 15. NFAT plays multiple roles in T cell activation and recent data suggest that activation of NFAT in the absence of appropriate co-factors, such as AP-1, causes formation of anergic T cells 16. We observed that ox-PL induce anergy in T cells. So it is intriguing to assume that engagement of the specific receptor for PGPC stimulates the anergy-inducing capacity of the NFAT pathway. This idea is further supported by our finding that OxPAPC induces the negative-regulatory transcription factor Egr-3 in T cells, which might be responsible for the induction of Cbl-b, a critical inhibitor of T cell activation 8, 17, 18. We did not observe that ox-PL induce regulatory T cell function. Thus, signaling via the putative ox-PL receptor on T cells is unlikely to up-regulate the transcription factor FoxP3, which is characteristic for regulatory T cells 16.

Ox-PL are gaining increasing importance as regulators of innate immune responses 3, 4, 19. Our recent findings already demonstrated that ox-PL represent negative feedback regulators of adaptive immune responses through limiting pathogen-induced activation of DC 5. In this study we show for the first time that ox-PL can directly modulate activation of T cells induced via the TCR/CD3 complex. Our data demonstrate that ox-PL represent a novel immunosuppressive factor for adaptive immunity with unique inhibitory characteristics. T cells have been implicated in mediating many aspects of autoimmune inflammation 20. Thus, compounds like PGPC might be suitable tools to control undesired T cell responses in autoimmune diseases including rheumatoid arthritis.

Materials and methods

Media, reagents and chemicals

Cells were maintained in RPMI 1640 (Gibco Ltd., Paisley, Scotland), supplemented with 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, and 10% FCS (HyClone, Logan, UT). PMA (final concentration 100 nM), ionomycin (100 nM), PI and monensin (5 µM) were obtained from Sigma-Aldrich (St. Louis, MO). IL-2 was obtained from Peprotech (Rocky Hill, NJ). Annexin V was purchased from Caltag Laboratories (Burlingame, CA).

Lipid oxidation

PAPA, PAPC, PAPG, PAPS, PGPC and POVPC were obtained from Avanti Polar Lipids (Alabaster, AL). Lipids were oxidized by exposure of dry lipid to air for 72 h. The extent of oxidation was monitored by positive ion electrospray ionization mass spectrometry as described previously 7. Determination of phospholipid content was performed by a standard phosphorus quantification procedure 21.

Reduction of OxPAPC with NaBH4

Solution of NaBH4 in isopropanol (150 µg in 2.3 mL) was added to a flask containing 1 mg of dry OxPAPC. The reaction mixture was kept at 0°C for 60 min. The excess of NaBH4 was destroyed by careful addition of 0.1 N HCl to achieve neutral pH. The lipids were extracted two times with 10 mL of chloroform. Lower organic layer was evaporated under reduced pressure. The residue was applied on normal-phase HPLC column eluted with acetonitril-methanol-water (77:8:15 v/v/v) containing 0.02% triethylamine. Fractions eluting earlier than lyso-PC were combined and evaporated. After their analysis by thin-layer chromatography and mass spectrometry, lipid phosphorus content was determined. Mock-treated lipids were processed identically but without NaBH4.

Reduction of OxPAPC with triphenylphosphine

Fifty-fold molar excess of triphenylphosphine (Sigma, Vienna, Austria) dissolved in chloroform-methanol (2:1 v/v) to a concentration of 64 mg/mL was added to a flask containing dry OxPAPC. The reaction mixture was kept for 60 min at room temperature. The solvents were removed by evaporation with argon. The residue was dissolved in chloroform and applied onto a silica gel column (70-230 mesh, 60 Å; Aldrich, Vienna, Austria). The column was washed with chloroform, then with chloroform-methanol (3:1 v/v), and finally phospholipids were eluted with chloroform-methanol-water (65:25:4 v/v/v). Fractions containing phospholipids were combined and evaporated under a stream of argon. The dried residue was dissolved in chloroform. Lipid peroxides were determined by the ferric thiocyanate method 22. After centrifugation of samples for 1 min at 2000 rpm, optical density at 500 nm was measured. Mock-treated lipids were processed identically but without triphenylphosphine.

Antibodies

The following murine mAb were generated in our laboratory: negative-control mAb VIAP (calf intestinal alkaline phosphatase-specific), 1/47 (MHC class II), 11C9 (CD63), and 3G10 (CD25). 15E8 (CD28) mAb was purchased from Caltag Laboratories. OKT3 (CD3) was obtained from Jansen-Cilag (Vienna, Austria). FN50 (CD69), MP4–25D2 (IL-4) and UL6–13D2 (IFN-γ) were purchased from BD PharMingen (San Diego, CA).

Isolation of T cells

PBMC were isolated from heparinized whole blood of normal healthy donors by standard density gradient centrifugation with Lymphoprep (Nycomed, Oslo, Norway). Subsequently, T cells were separated by magnetic sorting using the MACS technique (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) as described previously 23, 24. Purified T cells were obtained through negative depletion of CD11b+, CD14+, CD16+, CD19+, CD33+ and MHC class II+ cells with the respective mAb. The purity of T cells used in this study was 96 ± 1% as determined by FACS analysis.

Naive T cells were isolated from cord blood as described above. Cord blood samples from healthy donors were collected during healthy full-term deliveries. Approval was obtained from the Medical University of Vienna institutional review board for these studies. Cord blood T cells used in this study were CD45RA+ (92 ± 3%) and CD45RO.

T cell proliferation assay

Purified T cells (1×105/well) were incubated in 96-well cell culture plates (Corning-Costar Europe, Badhoevedorp, The Netherlands) coated with either CD3 mAb (OKT3, 1 μg/mL) and CD28 mAb (15E8, 2 μg/mL) or OKT3 plus CD63 mAb (11C9, 2 µg/mL) as previously described 6. PMA and ionomycin were used at 100 nM. Lipids were added at the beginning of T cell stimulation. The assay was performed in triplicates. Proliferation of T cells was monitored by measuring [methyl-3H]thymidine (ICN Pharmaceuticals Inc., Irvine, CA) incorporation, added at the indicated time points. Cells were harvested 18 h later and incorporated [methyl-3H]thymidine was detected on a microplate scintillation counter (Topcount; Packard, Meriden, CT). Viability of T cells after cultivation was assessed by staining with FITC-labeled annexin V and PI and flow cytometric analysis.

Restimulation of T cells

T cells (1×105) were stimulated with anti-CD3/CD28 mAb in the presence of 1, 5 and 10 µg/mL of OxPAPC or without lipid. After 5 days, activated T cells were harvested, counted and incubated for 4 days in fresh media. T cells were then harvested again, counted and restimulated with plate-bound anti-CD3/CD28 mAb with or without OxPAPC. Proliferation of T cells was analyzed 4 days later.

Suppressor activity

Stimulation of purified T cells with anti-CD3/CD28 mAb was performed in the presence of 1–10 µg/mL of OxPAPC. After 5 days, preactivated T cells were harvested and cultured for 4 days in fresh media. The cells were then irradiated (30 Gy, 137Cs source) and T cells (1×105) preactivated or not with ox-PL were mixed with primary T cells (1×105) and stimulated with anti-CD3/CD28 mAb. Proliferation was determined on day 4, as described above.

Flow cytometric analysis

For membrane staining, 5×105 cells were incubated for 30 min at 4°C with unconjugated mAb. After washing twice with PBS, Oregon Green-conjugated goat anti-mouse Ig antibodies from Molecular Probes Inc. (Eugene, OR) was used as a second-step reagent. For cytoplasmic staining, cells were harvested, fixed with FIX (An der Grub Bio Research, Kaumberg, Austria) for 20 min, washed twice with PBS, and permeabilized for 20 min with PERM solution (An der Grub Bio Research) in the presence of the primary mAb. Oregon Green-conjugated goat anti-mouse Ig was used as second-step reagent. Flow cytometric analysis was performed using a FACScalibur flow cytometer (Becton Dickinson, Franklin Lakes, NJ).

Cytotoxicity assay

Cytotoxicity was measured in a standard 51Cr (ICN) release assay 25. Briefly, 2×106 target cells (THP1, K562) were resuspended in 100 µL of PBS and labeled with 51Cr (50 µCi) for 1 h at 37°C. After washing two times with medium, 5×103 target cells/well were added to triplicates of decreasing numbers of T cells in 96-well U-bottom plates. To determine the spontaneous release, only medium was added to the target cells in three wells and to define the maximal release, 100 µL of a Triton X-100 solution (2%) was added to a separate triplicate of wells with target cells. The cells were then centrifuged, except when mAb inhibition studies were performed, and incubated at 37°C for 4 h. Released 51Cr was then harvested with a filter harvesting system (Skatron Ltd., Oslo, Norway) and measured on a γ-counter (Topcount Model 5000 from Packard). The percentage of specific lysis was calculated by the formula: (value of the probe – spontaneous release) / (maximal release – spontaneous release) × 100.

Determination of cytokine production

T cells were activated with CD3/CD28 mAb with or without OxPAPC in 96-well plates (Corning-Costar). After 24 h the supernatants of primary stimulation cultures were harvested and IL-2 and IFN-γ were measured via the Luminex100 System (R&D Systems Inc., Minneapolis, MN) as described in the manufacturer's protocol.

RNA isolation and cDNA preparation

Up to 1×107 cells/mL were lysed in 1 mL TRI reagent (Sigma). Isolation of total RNA was performed according to the manufacturer's instructions. Total RNA (2 µg) was reverse-transcribed with murine leukemia virus reverse transcriptase (Fermentas, St. Leon-Rot, Germany) using oligo(dT)18 primers, according to the manufacturer's protocol. The mixture was incubated for 5 min at 37°C, 60 min at 42°C and 10 min at 70°C. Finally cDNA was diluted 1:3 to use it for real-time PCR.

Real-time PCR

Quantitative RT-PCR was performed using a LightCycler (Roche Molecular Biochemicals, Indianapolis, IN) using SYBR Green I detection. In all assays, cDNA was amplified using a standard program (10-min denaturizing step; 55 cycles of 5 s at 95°C, 15 s at 65°C and 15 s at 72°C; melting point analysis in 0.1°C steps; final cooling step at 4°C). Each LightCycler capillary (20 µL; Roche Diagnostics, Mannheim, Germany) was loaded with 1.5 µL cDNA, 1.8 µL MgCl2 (25 mM), 10.1 µL double-distilled H2O and 0.4 µL of each primer (5 mM). Relative quantification of target gene expression was performed using a mathematical model 26 recommended by Roche Molecular Biochemicals. β2-Microglobulin was used as a housekeeping gene. The following forward and reverse primers were used: IL-2: (forward) GAATCCCAAACTCACCAGGA, (reverse) ATGGTTGCTCTCTCATCAGC; IL-4: (forward) GCCACCATGAGAAGGACACT, (reverse) ACTCTGGTTGGCTTCCTTCA; IL-10: (forward) TGCCTTCAGCAGAGTGAAGA, (reverse) GGTCTTGGTTCTCAGCTTGG; IL-13: (forward) GTACTGTGCAGCCCTGGAAT, (reverse) TTTACAAACTGGGCCACCTC; IFN-γ: (forward) TTCAGCTCTGCATCGTTTTG, (reverse) TCTTTTGGATGCTCTGGTCA; Cbl-b: (forward) ATGCTGAATGGAACACATGG, (reverse) ACTATGCCTTGCAGGAGGTG; Egr-3: (forward) CAACTGCCTGACAATCTGTACC, (reverse) AGTAGGTCACGGTCTTGTTGC.

Statistical analysis

Statistical analysis was performed using a two-tailed Student's t-test using unpaired nonparametric test (Mann–Whitney). Significance is represented as *p<0.05, **p<0.01 and ***p<0.001.

Acknowledgements

The authors thank M. Merio, K. Wenhardt and P. Kohl for expert technical assistance. This work was supported by a grant of the Wiener Wissenschafts-, Forschungs- und Technologiefonds (WWTF) and Fonds zur Förderung wissenschaftlicher Forschung (FWF, P20266). V.N.B. and O.O. are supported by a grant of the European Union Molstroke project LSHM-CT-2004–005206 and a grant from the Fonds zur Förderung wissenschaftlicher Forschung (S9407-B11).

Appendix

Conflict of interest: The authors declare no financial or commercial conflict of interest.

Footnotes

  1. 1

    WILEY-VCH

  2. 2

    WILEY-VCH

  3. 3

    WILEY-VCH

  4. 4

    WILEY-VCH

  5. 5

    WILEY-VCH

  6. 6

    WILEY-VCH

  7. 7

    WILEY-VCH

  8. 8

    WILEY-VCH

Ancillary