SEARCH

SEARCH BY CITATION

Keywords:

  • arginase;
  • granulocyte;
  • immune suppression;
  • myeloid derived suppressor cells;
  • polymorphonuclear

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Human polymorphonuclear leucocytes (PMN) are thought to be immunosuppressive. The suppressive mechanism(s) used by PMN are, however, not well defined and in this study they were analysed using T-cell responses to CD3+ CD28 monoclonal antibodies (mAb) as a readout. We demonstrate that in vitro activated PMN (PMNact) can, without any T-cell interaction, induce apparent T-cell suppression by inhibiting the stimulatory capacity of the CD3 mAb. However, a cell-directed suppression of T-cell proliferation was observed when PMNact were added to pre-activated T cells that are already committed to polyclonal proliferation. This suppression was partially reversed by catalase addition (P < 0·01) and largely reversed by addition of exogenous interleukin-2 (P < 0·001) but was not significantly reduced by nitric oxide synthase inhibition, myeloperoxidase inhibition or addition of excess arginine. Following removal of PMNact, suppressed T cells could respond normally to further stimulation. In addition to suppressing proliferation, co-culture with PMNact also induced a significant decrease in T-cell viability that was reversed by catalase addition (P < 0·05). The addition of the arginase inhibitor N-hydroxy-nor-l-arginine induced both a further significant, catalase-sensitive, loss in T-cell viability and increased nitrite release (P < 0·001). These data demonstrate that PMN, when activated, can both induce T-cell death and reversibly inhibit proliferation of activated T cells. The mechanisms underlying these distinct processes and the effects of arginase inhibitors on PMN induced cytotoxicity merit further investigation.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

It is well established that a heterogeneous group of myeloid lineage cells play a central role in the suppression of immune responses during cancer and other pathological conditions. These cells are collectively termed myeloid derived suppressor cells (MDSC) and are comprised of two major subsets – monocytic MDSC and granulocytic MDSC (G-MDSC).[1] Elevated levels of one, or both, of these subsets have been reported in patients with malignant or inflammatory diseases, and these increases are thought to play a central role in the immune dysfunction underlying these conditions.[2]

There is increasing evidence that polymorphonuclear leucocytes (PMN) have immunosuppressive capabilities and can inhibit anti-tumour responses.[3] Human studies have reported that activated PMN (PMNact) can inhibit in vitro T-cell responses[4-14] and that systemic inflammation mobilizes increased numbers of circulating, immunosuppressive PMNact.[15] The circulating mature G-MDSC populations observed in association with some human cancers have been shown to closely resemble PMNact with respect to their phenotype, morphology, buoyant density and immunosuppressive activity.[6, 10, 13, 14, 16]

Mechanisms that have been linked to the suppressive activity of PMNact and G-MDSC include, but are not limited to, depletion of arginine and cysteine, production of reactive oxygen species (ROS) and reactive nitrogen species and cytokine release.[2, 3, 17] However, the majority of functional studies to date have been performed in murine models and the relative importance of these mechanisms in the suppression mediated by human PMNact and G-MDSC is unclear. In addition, it is unclear in many studies whether (i) observed suppression arises solely from reduced T-cell responses or whether increased T-cell death also contributes, (ii) the suppression is reversible, and (iii) PMN require activation to become suppressive.

The polyclonal T-cell response to CD3+ CD28 monoclonal antibody (mAb) stimulation has been widely used as a readout system to both functionally identify, and characterize, suppressive human PMN or G-MDSC populations.[7, 8, 10-13, 16, 18-21] However, it has not been demonstrated that this assay provides a robust measure of suppressive activity. In addition, only limited analysis of the mechanisms used in suppressing polyclonal responses has been performed, and many studies use media containing components such as high arginine levels or phenol red, which may mask some inhibitory pathways. Similarly it is unclear whether PMN/G-MDSC can suppress only naive T-cell responses or can similarly modulate T cells already committed to proliferation.

The characterization of robust assays for measuring PMN-mediated suppression and analysis of the suppressive mechanisms used is important for both understanding the role of these cells in disease and developing approaches to modulate their activity. In this study we have, in the setting of polyclonal T-cell responses to CD3+ CD28, analysed PMN-mediated suppression using methodology that enables discrimination between effects on proliferation and viability. Using this approach we analyse (i) the effectiveness of the polyclonal assay as a measure of PMN-mediated suppression, (ii) the ability of PMN to suppress T cells already committed to proliferation, (iii) the effect of a wide range of inhibitors directed against the putative suppressive mechanisms utsed by PMNact, (iv) the reversibility of PMN-mediated effects, and (v) the effects of PMN on T-cell viability.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Isolation of peripheral blood populations

Blood was collected from normal donors following informed consent according to Upper South B Ethical Committee (New Zealand) guidelines.

Cells were separated by centrifugation (20 min, 900 g) over Ficoll/Paque (Amersham Pharmacia Biotech, Uppsala, Sweden). The low-density fraction of peripheral blood mononuclear cells (PBMC) was removed and PMN obtained from the high density fraction by sedimentation (30 min) of the red blood cells with 1·5% Dextran/PBS and hypotonic water lysis of residual red blood cells.

T cells were isolated from the PBMC by immunomagnetic depletion using CD16, CD11b and HLA-DR mAb with goat anti-mouse IgG Dynabeads (GAM-Dynabeads) according to manufacturer protocols (Dynal AS, Oslo, Norway). CD14-depleted cells (CD14 PBMC) were prepared from the PBMC by immunomagnetic depletion using CD14 mAb and GAM-Dynabeads. T cells and CD14 PBMC were labelled with carboxyfluorescein diacetate, succinimidyl ester (CFSE; Invitrogen, Auckland, New Zealand) by incubation (5 min, room temperature) at 1 × 107/ml in 5% fetal calf serum/PBS with 1 μm CFSE before washing. The PMN were used directly in assays and responders were stored in liquid N2 until use.

Functional assays

All experiments were performed in RPMI+FCS [SILAC-RPMI (Sigma, St Louis, MO) supplemented with 100 μM arginine, 40 mg/L lysine, 50 mg/L leucine, 0.02 μM MnCl2 and 10% FCS (Invitrogen, Auckland, New Zealand)].

T cell proliferation assays were performed in 96-well round-bottom plates pre-coated (5 μg/ml, 2 hrs) with CD3 mAb (UCHT1, Beckman Coulter or OKT3, eBioscience, San Diego, CA). CFSE+T cells (1 × 105/well) and 1 μg/ml CD28 mAb (28.2, Beckman Coulter) were cultured (37°C, 5%CO2), with or without PMN (2 × 105/well) and combinations of fMLP and inhibitors. Following 72 h incubation, supernatants were harvested and cells labelled with either propidium iodide (PI, 5 μg/ml, Sigma), CD3-PE or CD4-PC7 + CD8-PE (Beckman Coulter). Following gating on viable cells, the proportion of mAb labelled T cells with a diluted CFSE signal (defined as proliferation) was determined by flow cytometry (Beckman Coulter FC500). The level of IFN-γ, IL-2 and IL-10 in supernatants was determined by ELISA (R&D systems, Minneapolis, MN). In a number of experiments CD3 mAb coated wells were pre-incubated (1 hr, 37°C) with combinations of media, PMN and fMLP prior to extensive washing (PBS). CFSE+T cells and CD28 mAb were then added and proliferation analysed at 72 hr as above.

CFSE-labelled CD14 PBMC were pre-activated by 24 hour culture (2 × 106/ml) with CD28 mAb in flat-bottom plates (12 well) pre-coated (5 μg/ml) with UCHT1, then harvested and washed (×2). For co-culture experiments, wells (96 well plate) containing media or PMN (2 × 105/well) were cultured 1h in the presence, or absence, of inhibitors prior to addition of pre-activated CD14- PBMC (1 × 105) and either media, fMLP or zymosan. Following 72 hr culture, supernatants were harvested for cytokine analysis and following T cell labelling proliferation was analysed as above.

Extra wells were also harvested for re-stimulation experiments. Following centrifugation over Ficoll/Paque to remove PMN and dead cells, cells were re-cultured with or without CD3/CD28 for 72 hr before analysis.

In a number of experiment PMN were cultured (2 × 106/ml) 3 hr with or without zymosan prior to collection of supernatant and addition (80% v/v) to pre-activated CD14 PBMC (1 × 105).

Reagents added to cell cultures and their final assay concentrations were: Formyl-Met-Leu-Phe-OH (fMLP, 1 μM), Catalase (Cat, 100 U/ml), Arginine (2 mM), Glutamate (2.5 mM), Diphenyleneiodonium chloride (DPI, 10 μM), c (ABAH, 100 μM) and Interleukin-2 (IL-2, 100 U/ml) obtained from Sigma (St Louis, MO). N-Monomethyl-L-arginine (NMMA, 500 μM) and N-Hydroxy-nor-L-arginine (nor-NOHA, 100 μM) were from Merck (Darmstadt, Germany). Zymosan (Sigma) was heat inactivated at 100°C (1 hr) then stored frozen until opsonised in fresh serum (45 min, 37°C) immediately prior to washing and use (Zym, 125 μg/ml).

Analysis of viability

The viability of CFSE-labelled T cells and CFSE-labelled, pre-activated CD14 PBMC was determined following 24 hr co-culture (as above) by adding PI and determining the PI+ CFSE+/CFSE+ ratio using flow cytometry.

Analysis of nitrite release

The PMN in RPMI-1640/fetal calf serum (5 × 105/well, 96-well plate) were cultured (37°, 5% CO2) for 1 hr in the presence or absence of inhibitors before the addition of zymosan and a further 90 min of culture (volume = 300 μl/well). Supernatant was harvested and nitrite levels were determined using a Griess reagent kit (Molecular Probes, Invitrogen, Auckland, New Zealand).

Statistical analysis

Statistical analysis was performed using graphpad prism 5.01. (GraphPad Software Inc., La Jolla, CA). Statistical analysis of treatment groups was performed as indicated using either (i) repeated measures analysis of variance of raw data in combination with Dunnett's post-hoc test or (ii) paired t-test. P < 0·05 was considered significant. *P < 0·05; **P < 0·01; ***P < 0·001.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

PMN effects on the proliferation of purified T cells

All experiments were performed in media containing physiological arginine levels (100 μm) and lacking phenol red to maximize detection of suppression due to arginine depletion and ROS generation.

Purified T-cell responses to CD3+ CD28 mAb are widely used to demonstrate the suppressive activity of PMN/G-MDSC and this approach was therefore investigated as a potential system for analysing immunosuppressive mechanisms.

Purified peripheral blood T cells proliferated strongly in response to continuous stimulation with solid-phase CD3 mAb (spCD3) and soluble CD28 mAb (Fig. 1a). The addition of PMN to these cultures significantly reduced proliferation (7–55% reduction, median = 25%). Activation of PMN with N-formyl-L-methionyl-L-leucyl-L-phenylalanine (fMLP) resulted in a significantly larger reduction in proliferation (60–91% reduction, median = 76%). Previous studies have implicated myeloperoxidase (MPO) in the suppressive activity of PMN. The presence of the MPO inhibitor 4-aminobenzoic acid hydrazide (ABAH) prevented the suppression induced by activated PMN (PMNact) (Fig. 1b).

image

Figure 1. Polymorphonuclear leucocytes (PMN) suppress the proliferation of CD3+ CD28-activated T cells. CFSE-labelled T cells were stimulated with CD28 monoclonal antibody (mAb) in wells pre-coated with anti-CD3 mAb for 72 hr then analysed. (a) Scatter plot of the T-cell proliferation observed in nine individual experiments following culture of T cells with either medium alone (Nil), PMN, or PMN + fMLP (PMNact). (b) Histogram of proliferation in cultures containing combinations of PMN, PMNact and a myeloperoxidase inhibitor (4-aminobenzoic acid hydrazide; ABAH). Data from individual experiments (n = 5) were normalized relative to the proliferation of T cells alone (defined as 100%) and pooled data are shown as mean ± SEM. (c) T cells, either alone (Nil) or with PMNact, were cultured in CD3-coated wells that had been either untreated (Nil) or pre-incubated (1 hr, 37°) with medium or PMNact before extensive washing. Data from individual experiments (n = 4) were normalized relative to the proliferation of T cells alone (defined as 100%) and pooled data are shown as mean ± SEM. Statistical analysis was performed using repeated measures analysis of variance of raw data and asterisks indicate significant differences between treatment groups linked by bars.

Download figure to PowerPoint

The use of this assay as a valid measure of suppressive activity requires that the suppression arises from a PMNact–T-cell interaction and not from PMNact–mAb interactions reducing the ability of the CD3 and CD28 mAb to stimulate proliferation. The ability of PMNact to modulate mAb function was investigated by analysing the effect of PMNact on the stimulatory spCD3 mAb. Pre-incubation (1 hr) of spCD3 mAb (UCHT1) coated wells with media before washing and addition of T cells did not significantly reduce proliferation (Fig. 1c). However, pre-culture of spCD3 mAb with PMNact before removal of PMN by washing and addition of T cells significantly reduced subsequent T-cell proliferation, indicating a direct effect on the spCD3. PMNact had a similar effect on the spCD3 mAb OKT3 (n = 3, data not shown). Hence, PMNact are able to induce apparent suppression in this assay system, independently of any effect that they may have on the T-cell, by reducing the stimulatory capacity of CD3 mAb.

PMN effects on the proliferation of pre-activated T cells

The ability of PMN to suppress the responses of T cells that are already committed to polyclonal proliferation was analysed. This approach was also investigated as an alternative assay for suppressor activity, as the addition of PMN subsequent to polyclonal stimulation would avoid the non-specific suppressive effects arising from PMNact–mAb interaction. Short-term stimulation with CD3+ CD28 mAb can commit T cells to subsequent mAb-independent but antigen-presenting-cell (APC) -dependent polyclonal proliferation. Monocyte-depleted PBMC (CD14 PBMC) were used as the responder population for this assay because these preparations contain both T cells and the required APC but lack monocytes, which are reportedly able to protect lymphocytes from some PMN-mediated effects.[4] CD14 PBMC were therefore activated for 24 hr with spCD3 mAb + CD28 mAb and then, following washing, cultured for a further 72 hr in fresh media, either alone, or in the presence of PMN. Substantial arginine-dependent proliferation of pre-activated T cells was observed following culture (Fig. 2a) with 34–86% (median = 71%) undergoing one or more divisions and 18–73% (median = 38%) undergoing two or more divisions (Fig. 2b). The addition of PMN alone did not significantly reduce T-cell proliferation (Fig. 2c). However, significantly reduced T-cell proliferation was observed following addition of PMN in combination with fMLP or zymosan (Fig. 2a,c), which are potent activators of PMN[22] and have been reported to induce their suppressor activity.[4, 5, 9, 14] There was a significantly larger suppressive effect on the proportion of T cells undergoing two or more divisions than on the proportion undergoing one or more divisions (Fig. 2c) using both fMLP-activated PMN [median reduction (range), 32% (20–43) versus 25% (7–21)] and zymosan-activated PMN [median reduction (range), 48% (26–74) versus 25% (7–21)]. Zymosan had a significantly greater effect than fMLP (Fig. 2c) and was used in all subsequent experiments. The zymosan-activated PMN had similar suppressive effects on both the CD4+ and CD8+ T cells present (Fig. 2d).

image

Figure 2. Polymorphonuclear leucoytes (PMN) inhibit the responses of pre-activated T cells. CFSE+ CD14 peripheral blood mononuclear cells (PBMC) were pre-activated, washed, then cultured (72 hr) with combinations of medium alone (Nil), PMN and PMN stimuli [fMLP, zymosan (Zym)] and analysed. All data are from the same set of seven independent experiments (a) Histograms of T-cell proliferation following culture in the presence or absence of arginine and activated (zymosan or fMLP) PMN. Markers indicate the proportion of T cells undergoing one or more divisions or two or more divisions. Data are from a representative experiment of the seven performed. (b) Scatter plot of the proportion of T cells that underwent one or more divisions or two or more divisions following culture in medium alone. Data from seven experiments are shown and the solid bar indicates the median (c) Relative T-cell proliferation in cultures containing PMN and the indicated stimuli. Data are shown as the proportion undergoing one or more divisions or two or more divisions and for each experiment was normalized relative to the proliferation in matching cultures without PMN. Pooled data from seven independent experiments are shown as mean ± SEM and were analysed by repeated measures analysis of variance. Asterisks indicate values significantly different from those observed in the respective control cultures or significant differences between treatment groups linked by bars. (d) Histogram of the relative proliferation of the CD4+ and CD8+ T-cell subsets following culture of CD14 PBMC with PMN + zymosan. Data from three separate experiments were normalized relative to the proliferation in matching cultures without PMN and shown as mean ± SEM. (e) Interferon-γ (IFN-γ) levels in supernatants collected from cultures containing CD14 PBMC and the indicated combinations of stimuli and PMN. Pooled data from four separate experiments are shown as mean ± SEM.

Download figure to PowerPoint

Suppressor populations can modify not only proliferation but also the increase in interleukin-2 (IL-2), interferon-γ (IFN-γ) and IL-10 release that is induced following continuous activation of resting T cells. Analysis of the proliferation assay supernatants from pre-activated CD14 PBMC demonstrated that levels of IL-2 and IL-10 were, as expected,[23, 24] below detection limits irrespective of the presence of PMNact (n = 4, data not shown).

Low levels of IFN-γ were detected and these levels were significantly increased by addition of IL-2 (Fig. 2e). The IFN-γ release in either the presence or absence of IL-2 was not significantly altered by the presence of PMNact.

Effect of inhibitors on PMN-mediated suppression

The mechanism by which zymosan-activated PMN suppress the proliferation of pre-activated CD14 PBMC was analysed using inhibitors of potential suppressive pathways (Fig. 3a). PMN activation generates superoxide from which numerous ROS are derived. Hydrogen peroxide, formed from the dismutation of superoxide, is used by myeloperoxide as a substrate in the further generation of reactive species. The enzyme catalase catalyses the decomposition of H2O2 to water and oxygen. The addition of catalase to cultures resulted in a small, but significant, reduction in suppression. The addition of the MPO-specific inhibitor ABAH[25] alone resulted in substantial cell death. The MPO is a major consumer of H2O2 and therefore catalase was added together with ABAH to prevent potentially toxic H2O2 build-up. This combination prevented cell death but did not increase the effect observed with catalase alone.

image

Figure 3. Inhibition of polymorphonuclear leucocyte (PMN) -mediated suppression. CFSE+ CD14 peripheral blood mononuclear cells (PBMC) were pre-activated, washed, then cultured (72 hr) under the indicated conditions before analysis or re-culture. (a) Relative proliferation of T cells in cultures containing PMN + zymosan (PMNact) in combination with either interleukin-2 (IL-2) or the indicated inhibitors. Proliferation (two or more divisions) was normalized relative to that observed in matching cultures without PMN. Data for each inhibitor (mean ± SEM) are from the indicated number of seven experiments and were analysed by a paired t-test. Asterisks indicate significant differences between control and treatment groups linked by bars. (b) Relative proliferation in cultures supplemented with either medium alone, PMN+ zymosan (PMNact) or conditioned medium (80% v/v) obtained following activation (3 hr) of PMN with either medium alone (control) or zymosan (PMNact). Data from three individual experiments were normalized relative to the proliferation of CD14 PBMC cultured alone (defined as 100%) and pooled data are shown as mean ± SEM. Statistical analysis was performed using repeated measures analysis of variance and asterisks indicate values significantly different from the nil treatment group. (c) Pre-activated CD14 PBMC were cultured for 72 hr, either alone or with PMNact, then, following PMN or mock depletion, underwent a secondary culture (72 hr) in either medium alone or in the presence of spCD3/CD28. T-cell proliferation (two or more divisions) was analysed before and after secondary culture and data from three separate experiments are shown as histograms.

Download figure to PowerPoint

The addition of exogenous IL-2 significantly increased the proliferation of pre-activated T cells (1·2–2·4-fold, P = 0·003) and its presence substantially reversed the suppressive effect of PMNact. The addition of the inducible nitric oxide synthase (iNOS) -specific inhibitor N-monomethyl l-arginine did not significantly reverse suppression. Similarly, supplementation with excess arginine or glutamate, to counter potential media depletion of arginine and cystine, respectively, did not significantly reverse suppression. Results obtained using the ROS scavengers N-acetylcysteine and methionine could not be interpreted because of the suppressive effects they had on the proliferation of pre-activated T cells cultured alone. During preliminary experiments it was observed that although the arginase inhibitor nor-N-ω-hydroxy-nor-l-arginine (NOHA) did not modulate the viability of CD14 PBMC either cultured alone or with PMN, it induced increased cell death when present together with activated PMN. The effects of nor-NOHA on proliferation were therefore not analysed and its effects on cell viability are detailed later in the Results.

The conditioned media collected following 3 hr culture of PMN with zymosan, did not, when added to CD14 PBMC, significantly inhibit T-cell proliferation indicating that longer-lived activation-induced soluble factors are not involved (Fig. 3b).

The permanence of the suppressed proliferative responses induced by PMNact was analysed in re-culture experiments (Fig. 3c). Following culture of pre-activated CD14 PBMC with either nil or PMNact the CD14 PBMC were enriched (> 95%) over a density gradient and then re-cultured in either medium alone or in the presence of spCD3/CD28 mAb. Before re-culture those T cells obtained from co-cultures with PMNact had undergone considerably less proliferation. Re-culture of these cells in medium alone induced only a small increase in these cells' proliferation. However re-culture in the presence of CD3/CD28 mAb induced similarly strong proliferation in all cultures irrespective of the T cells' initial exposure to PMNact. These results show that the suppressive effects of PMNact on T-cell proliferation are not permanent.

Effects of co-culture on T cell and PMN viability

Analysis of viability following 24-hr culture of pre-activated CD14 PBMC (Fig. 4a) demonstrated that addition of either PMN, the arginase inhibitor nor-NOHA or PMN + nor-NOHA, did not significantly alter the proportion of non-viable (PI+) cells present. However, PMN activation significantly increased the proportion of PI+ CD14 PBMC (3·2-fold to 4·7-fold) and a significantly larger increase (3·7-fold to 6·3-fold) was observed in the presence of both PMNact and nor-NOHA (Fig. 4a).

image

Figure 4. Effects on T-cell and polymorphonuclear cell (PMN) viability. CFSE labelled (a) pre-activated CD14 peripheral blood mononuclear cells (PBMC) or (b, c) T cells were incubated (24 hr) with the indicated combinations of medium alone (nil), zymosan, nor-NOHA and catalase (Cat) either alone or in the presence of PMN. In CD14 PBMC experiments (n = 4) the ratio of PI+ CD14 PBMC was determined (a) and in the T-cell experiments (n = 4) the ratio of both the PI+ T cells (b) and PI+ PMN (c) was determined. Data from individual experiments were normalized [relative to the PI+ ratio in the matching control cultures of either CD14 PBMC only (9 ± 3%), T cells only (21 ± 13%) or T cell + PMN (10 ± 3% PI+ PMN)] and pooled data (mean ± SEM) shown as histograms. Statistical analysis was performed using repeated measures analysis of variance of raw data and asterisks indicate values significantly different from those observed in the T-cell only cultures or significant differences between treatment groups linked by bars.

Download figure to PowerPoint

image

Figure 5. Nitrite release by polymorphonuclear leucocytes (PMN) The. PMN were cultured with combinations of zymosan, nor-NOHA and inhibitors before collection of supernatant and analysis of nitrite concentration (μm) using Griess reagent. Data from three individual experiments are shown as mean ± SEM and statistical analysis was performed using repeated measures analysis of variance. Asterisks indicate values significantly different from those observed in the cultures containing only PMN.

Download figure to PowerPoint

Because CD14 PBMC contain multiple cell types, co-cultures of purified T cells and PMN were analysed to determine the effects of PMN activation and nor-NOHA on the viability of both the T cells (Fig. 4b) and PMN (Fig. 4c) present. The presence of PMN or nor-NOHA either alone, or in combination, did not alter T-cell viability (Fig. 4b). However, the addition of PMNact significantly increased the proportion of PI+ T cells (1·2-fold to 3·3-fold) and a significantly larger increase (1·8-fold to 4·9-fold) was observed in the presence of both nor-NOHA and PMNact. The presence of catalase significantly reduced the effects of both PMNact and PMNact + nor-NOHA on T-cell viability. Analysis of PMN viability in these same PMN–T-cell co-cultures demonstrated that the low levels of PI+ PMN present (7–14%) following co-culture with T cells alone were not significantly increased by addition of nor-NOHA (Fig. 4c). Activation of PMN significantly increased the numbers of PI+ PMN but the further addition of nor-NOHA or catalase did not significantly alter the proportion of PI+ PMN.

Nitrite release by PMN

The generation of NO and its metabolites represents a potential suppressive and cytotoxic mechanism and therefore levels of the NO metabolite nitrite in PMN cultures was analysed (Fig. 5). Low levels of nitrite release were detected following culture of PMN in the presence of either medium, nor-NOHA or zymosan alone. However, a significant increase in nitrite levels (fold increase 3–19, median = 8) was observed in cultures containing both zymosan and nor-NOHA. NO and nitrites may be generated from arginine as a result of iNOS activity or alternatively as a result of the oxidation of nor-NOHA by other enzymes such as peroxidases. NADPH oxidase generates superoxide from which numerous ROS are derived. The H2O2 formed from the dismutation of superoxide, is used by myeloperoxide as a substrate in the further generation of reactive species. Diphenyleneiodonium chloride (DPI) inhibits NADPH oxidase and other flavoproteins, the enzyme catalase catalyses the decomposition of H2O2 to water and oxygen, while ABAH specifically inhibits MPO. The zymosan + nor-NOHA-induced release of nitrite was not significantly reduced by the iNOS-specific inhibitor nMMA. Nitrite release was significantly reduced by DPI (reduction = 50–69%, median = 36%) and ABAH (reduction = 16–38%, median = 22%) but was not significantly reduced by catalase (Fig. 5).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

This study demonstrates that activated PMN can modulate T-cell responses by both inducing cell death and suppressing the responses of T cells already committed to proliferation.

There is increasing interest in the immunosuppressive activity of human PMN and G-MDSC. Functional studies are central to both defining these populations and analysing their suppressive capabilities. T-cell responses to CD3+ CD28 mAb are widely used, often in isolation, to demonstrate the suppressive activity of both PMN and G-MDSC. The validity of this assay is dependent on the suppressor population not disrupting the level of mAb-mediated stimulation. However, to our knowledge this has not been demonstrated to date, and controls addressing this possibility are absent in published studies using this assay. The observation in this study that PMNact can induce suppression by directly acting on CD3 mAb thereby compromises the use of this assay to demonstrate suppressor activity. These results do not negate the possibility that PMNact can also modulate T-cell function. They do, however, suggest that considerable caution should be applied to the use of assays that involve exposure of T-cell stimuli such as mAb or mitogens to suppressor populations capable of inducing extensive chemical modifications.

The proliferation of T cells can occur through both IL-2-dependent and IL-2-independent pathways.[26-29] Standard proliferation assays in which T cells are continuously stimulated without media exchange maintain IL-2 at high levels and result in mainly IL-2-driven proliferation.[27] However, transient T-cell generation of IL-2 and a later, more IL-2-independent, cell cycle progression appears to more accurately reflect in vivo T-cell activation.[26, 30] It is now clear that short periods of activation can commit T cells to continued, APC-dependent, but IL-2-independent, proliferation.[27-29] In the current study, T cells were committed to polyclonal proliferation by short term CD3–CD28 activation and then subsequently re-cultured in fresh medium to make their continued proliferation more reliant on APC–T-cell interactions.[27] The observed pattern of cytokine release by these cells was consistent with a low IL-2 environment.[23, 24] The addition of PMN subsequent to antibody activation in this system has the advantage of avoiding any confounding PMN–mAb interactions. This study provides the first evidence that PMNact can suppress the proliferation of T cells already committed to cell cycle progression. The observation that suppression was not observed in the presence of exogenous IL-2 suggests that PMNact do not modulate IL-2 responsiveness. Furthermore, the demonstration in re-culture experiments that suppressed T-cell populations are able to respond to secondary activation indicates that the PMNact do not induce a global or irreversible loss of T-cell proliferative capacity. The possibility that the suppression observed in both this study and other studies of human PMN may reflect, at least in part, modulation of APC function cannot be excluded.

Human PMN and G-MDSC populations are thought to use ROS generation or arginine metabolism as their predominant suppressive mechanisms, but the exact pathways used are unclear. The use in many studies of [3H]thymidine uptake or cytokine release to measure responses also makes it difficult to determine whether observed suppression reflects PMN-mediated effects on the viability rather than responses of T cells. Early studies implicated H2O2 generation and MPO activity in the suppressive activity of PMNact, although other mechanisms were not analysed.[4-6] However, the requirement, in a number of these studies, for protein-free conditions to observe suppression makes the in vivo relevance of these findings unclear. Many later studies used continuous CD3–CD28 mAb stimulation to demonstrate suppressive activity in PMN and G-MDSC populations and it is unclear what contribution PMN–mAb interactions made to the observed suppression.[8, 10, 11, 13, 16, 18-21] Only limited analyses of inhibitory pathways were performed in those studies but arginine depletion and catalase were reported to have little or no effect on suppression.[8, 10, 16, 20] It has been reported that G-MDSC can inhibit allo-antigen-specific responses[31] and that PMN can both inhibit allo-antigen-specific and lectin-induced responses,[14, 15] and induce T-cell hypo-responsiveness.[7] The PMN-mediated suppression was reported to be arginase-dependent in one study[14] but arginase-independent and partly H2O2-dependent in the others. This may reflect the considerable differences between these studies with respect to methodologies, the responder populations used and the numbers/activation status of the PMN used.

In the current study flow cytometry, rather than [3H]thymidine or cytokine release, was used to measure proliferation to minimize the impact of viability changes on the assay results. In addition, assays were performed in medium containing physiological arginine levels and lacking phenol red to increase the possibility of detecting effects due to arginase activity or ROS release. Because of the large number of potential suppressive pathways a wide range of inhibitors was tested. The results of the current study demonstrate that the suppressive activity of PMN is dependent on the presence of activating stimuli. Furthermore, the observed suppression of the proliferation of pre-activated T cells was not due to arginine or cysteine depletion, IL-10 release, iNOS activity or MPO activity and was only partially dependent on H2O2. Supernatants from PMNact also did not suppress, indicating that longer-lived soluble factors released during activation were not involved. It is well established that human PMNact can, through arginase release, eventually deplete arginine to levels that impair T-cell function.[11, 12] Although arginine depletion did not appear to play a role in the suppression observed in this study, longer incubation times may be required for sufficient depletion to occur. The results of this study suggest that the observed PMNact-induced suppression involves other mechanisms than those tested. However, it has been reported that H2O2 release by PMN is localized to the area in direct cellular contact with lymphocytes and therefore inhibitors such as catalase and ABAH may not be able to prevent activity at these contact points.[15]

The induction of T-cell death represents a potential immunosuppressive mechanism and it has been reported that human PMN can induce T-cell apoptosis/death. To date, however, the mechanism underlying this process has not been determined.[8, 10] In the current study we demonstrate that activated, but not resting, PMN induce cell death in a proportion of T cells via an H2O2-dependent mechanism. Furthermore, if arginase activity is inhibited at the time of PMN activation, the proportion of non-viable T cells is further increased via a process also involving H2O2 generation. Differential sensitivity of T-cell subsets to H2O2-mediated cell death has been reported previously and may explain why only a proportion of T cells were susceptible to PMNact-mediated cell death.[32] This differential sensitivity may also explain why the presence of ABAH, which inhibits a major pathway of H2O2 consumption and may therefore increase its levels,[33] induced increased cell death in pre-activated cells but not purified circulating T cells. Inhibition of arginase activity has not previously been associated with a direct effect on cell viability but could potentially modulate iNOS activity and thereby increase the release of pro-apoptotic NO species.[17] The observation that nitrite release by PMNact is increased in the presence of an arginase inhibitor supports this possibility. However, this increase was not reduced with an iNOS inhibitor, but was reduced by flavoprotein and MPO inhibitors, suggesting that it may arise from the iNOS-independent oxidation of the inhibitor molecule itself by other PMN oxidases.[34, 35] This pathway is consistent with the observed effect that catalase had on the arginase inhibitor-induced cytotoxicity as PMNact have been reported to have increased, H2O2-dependent, cytotoxicity in the presence of NO generation.[36] These results do not preclude either arginase or iNOS activity playing a role in modulating human PMN cytotoxicity but further underline the potent ability of PMN to chemically modify compounds. The current study provides the first evidence that arginase inhibitors can impact on PMN-mediated nitrite release and cytotoxicity. As these inhibitors are regarded as potential therapeutic inhibitors of MDSC activity these findings merit further investigation.[11, 12]

Studies to date have focused predominantly on the ability of PMNact and G-MDSC to suppress overall T-cell proliferation and cytokine release. However, both the early induction of T-cell death and the subsequent inhibition of the responses of remaining viable cells may impact on the overall number of effector T cells generated and the relative contributions of these two pathways requires further analysis. The generation of H2O2 by PMN appears to play an important role in both of these pathways and further study is required to determine whether H2O2 itself or downstream metabolites of H2O2 such as those generated by MPO are responsible.

The results of this study clearly demonstrate that PMN require activation to induce their suppressive activity. The type of activating stimuli, together with the PMN : T cell ratio[8, 15] may alter the nature and magnitude of this activity and this may impact on the immunosuppressive capability of PMN both in vitro and in vivo. The possibility that different PMN subsets may respond differently to activation must also be considered.

Understanding the mechanisms underlying the suppressive activity of PMN is important for the development of approaches to modulate their activity in the context of tumour growth and inflammation. In the current study we demonstrate that PMN, when activated, are able to not only induce T-cell death but also suppress the proliferation of activated T cells. The observation that activated PMN can both modulate CD3 mAb and, in the presence of an arginase inhibitor, induce increased T-cell death raises a cautionary note for PMN studies using these reagents.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

This work was funded by research grants from The New Zealand Lotteries Grant Board, Maurice & Phyllis Paykel Trust, The Bone Marrow Cancer Trust and the Cancer Society of NZ (Canterbury/Westland Division).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References
  • 1
    Gabrilovich DI, Nagaraj S. Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol 2009; 9:16274.
  • 2
    Filipazzi P, Huber V, Rivoltini L. Phenotype, function and clinical implications of myeloid-derived suppressor cells in cancer patients. Cancer Immunol Immunother 2012; 61:25563.
  • 3
    Muller I, Munder M, Kropf P, Hansch GM. Polymorphonuclear neutrophils and T lymphocytes: strange bedfellows or brothers in arms? Trends Immunol 2009; 30:52230.
  • 4
    el-Hag A, Clark RA. Immunosuppression by activated human neutrophils. Dependence on the myeloperoxidase system. J Immunol 1987; 139:240613.
  • 5
    Zoschke DC, Messner RP. Suppression of human lymphocyte mitogenesis mediated by phagocyte-released reactive oxygen species: comparative activities in normals and in chronic granulomatous disease. Clin Immunol Immunopathol 1984; 32:2940.
  • 6
    Schmielau J, Finn OJ. Activated granulocytes and granulocyte-derived hydrogen peroxide are the underlying mechanism of suppression of T-cell function in advanced cancer patients. Cancer Res 2001; 61:475660.
  • 7
    Cemerski S, Cantagrel A, Van Meerwijk JP, Romagnoli P. Reactive oxygen species differentially affect T cell receptor-signaling pathways. J Biol Chem 2002; 277:1958593.
  • 8
    Thewissen M, Damoiseaux J, van de Gaar J, Tervaert JW. Neutrophils and T cells: bidirectional effects and functional interferences. Mol Immunol 2011; 48:2094101.
  • 9
    McKenna KC, Beatty KM, Vicetti Miguel R, Bilonick RA. Delayed processing of blood increases the frequency of activated CD11b+ CD15+ granulocytes which inhibit T cell function. J Immunol Methods 2009; 341:6875.
  • 10
    Choi J, Suh B, Ahn YO, Kim TM, Lee JO, Lee SH, Heo DS. CD15+/CD16low human granulocytes from terminal cancer patients: granulocytic myeloid-derived suppressor cells that have suppressive function. Tumour Biol 2011; 33:1219.
  • 11
    Rotondo R, Bertolotto M, Barisione G et al. Exocytosis of azurophil and arginase 1-containing granules by activated polymorphonuclear neutrophils is required to inhibit T lymphocyte proliferation. J Leukoc Biol 2011; 89:7217.
  • 12
    Munder M, Schneider H, Luckner C et al. Suppression of T-cell functions by human granulocyte arginase. Blood 2006; 108:162734.
  • 13
    Hock BD, Mackenzie KA, Cross NB, Taylor KG, Currie MJ, Robinson BA, Simcock JW, McKenzie JL. Renal transplant recipients have elevated frequencies of circulating myeloid-derived suppressor cells. Nephrol Dial Transplant 2012; 27:40210.
  • 14
    Sippel TR, White J, Nag K, Tsvankin V, Klaassen M, Kleinschmidt-Demasters BK, Waziri A. Neutrophil degranulation and immunosuppression in patients with GBM: restoration of cellular immune function by targeting arginase I. Clin Cancer Res 2011; 17:69927002.
  • 15
    Pillay J, Kamp VM, van Hoffen E et al. A subset of neutrophils in human systemic inflammation inhibits T cell responses through Mac-1. J Clin Invest 2011; 122:110.
  • 16
    Rodriguez PC, Ernstoff MS, Hernandez C, Atkins M, Zabaleta J, Sierra R, Ochoa AC. Arginase I-producing myeloid-derived suppressor cells in renal cell carcinoma are a subpopulation of activated granulocytes. Cancer Res 2009; 69:155360.
  • 17
    Bronte V, Zanovello P. Regulation of immune responses by l-arginine metabolism. Nat Rev Immunol 2005; 5:64154.
  • 18
    Solito S, Falisi E, Diaz-Montero CM et al. A human promyelocytic-like population is responsible for the immune suppression mediated by myeloid-derived suppressor cells. Blood 2011; 118:225465.
  • 19
    Brandau S, Trellakis S, Bruderek K et al. Myeloid-derived suppressor cells in the peripheral blood of cancer patients contain a subset of immature neutrophils with impaired migratory properties. J Leukoc Biol 2011; 89:3117.
  • 20
    Ko JS, Zea AH, Rini BI et al. Sunitinib mediates reversal of myeloid-derived suppressor cell accumulation in renal cell carcinoma patients. Clin Cancer Res 2009; 15:214857.
  • 21
    Zea AH, Rodriguez PC, Atkins MB et al. Arginase-producing myeloid suppressor cells in renal cell carcinoma patients: a mechanism of tumor evasion. Cancer Res 2005; 65:30448.
  • 22
    Turner NC, Wood LJ, Foster M, Gueremy T. Effects of PAF, FMLP and opsonized zymosan on the release of ECP, elastase and superoxide from human granulocytes. Eur Respir J 1994; 7:93440.
  • 23
    Rafiq K, Charitidou L, Bullens DM, Kasran A, Lorre K, Ceuppens J, van Gool SW. Regulation of the IL-10 production by human T cells. Scand J Immunol 2001; 53:13947.
  • 24
    McDyer JF, Li Z, John S, Yu X, Wu CY, Ragheb JA. IL-2 receptor blockade inhibits late, but not early, IFN-γ and CD40 ligand expression in human T cells: disruption of both IL-12-dependent and -independent pathways of IFN-γ production. J Immunol 2002; 169:273646.
  • 25
    Kettle AJ, Gedye CA, Hampton MB, Winterbourn CC. Inhibition of myeloperoxidase by benzoic acid hydrazides. Biochem J 1995; 308(Pt 2):55963.
  • 26
    Malek TR, Castro I. Interleukin-2 receptor signaling: at the interface between tolerance and immunity. Immunity 2010; 33:15365.
  • 27
    Bonnevier JL, Yarke CA, Mueller DL. Sustained B7/CD28 interactions and resultant phosphatidylinositol 3-kinase activity maintain G1–>S phase transitions at an optimal rate. Eur J Immunol 2006; 36:158397.
  • 28
    Appleman LJ, Berezovskaya A, Grass I, Boussiotis VA. CD28 costimulation mediates T cell expansion via IL-2-independent and IL-2-dependent regulation of cell cycle progression. J Immunol 2000; 164:14451.
  • 29
    Shi M, Lin TH, Appell KC, Berg LJ. Cell cycle progression following naive T cell activation is independent of Jak3/common γ-chain cytokine signals. J Immunol 2009; 183:4493501.
  • 30
    Sojka DK, Bruniquel D, Schwartz RH, Singh NJ. IL-2 secretion by CD4+ T cells in vivo is rapid, transient, and influenced by TCR-specific competition. J Immunol 2004; 172:613643.
  • 31
    Nagaraj S, Youn JI, Weber H et al. Anti-inflammatory triterpenoid blocks immune suppressive function of MDSCs and improves immune response in cancer. Clin Cancer Res 2010; 16:181223.
  • 32
    Gupta S, Young T, Yel L, Su H, Gollapudi S. Differential sensitivity of naive and subsets of memory CD4+ and CD8+ T cells to hydrogen peroxide-induced apoptosis. Genes Immun 2007; 8:5609.
  • 33
    Test ST, Weiss SJ. Quantitative and temporal characterization of the extracellular H2O2 pool generated by human neutrophils. J Biol Chem 1984; 259:399405.
  • 34
    Boucher JL, Genet A, Vadon S, Delaforge M, Mansuy D. Formation of nitrogen oxides and citrulline upon oxidation of N-ω-hydroxy-l-arginine by hemeproteins. Biochem Biophys Res Commun 1992; 184:115864.
  • 35
    Beranova P, Chalupsky K, Kleschyov AL et al. N-ω-hydroxy-l-arginine homologues and hydroxylamine as nitric oxide-dependent vasorelaxant agents. Eur J Pharmacol 2005; 516:2607.
  • 36
    Andonegui G, Trevani AS, Gamberale R, Carreras MC, Poderoso JJ, Giordano M, Geffner JR. Effect of nitric oxide donors on oxygen-dependent cytotoxic responses mediated by neutrophils. J Immunol 1999; 162:292230.