HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text Full Text (PDF) All Versions of this Article: fj.05-5022fjev1 20/8/1209    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wang, T. Articles by Nath, A. Search for Related Content PubMed PubMed Citation Articles by Wang, T. Articles by Nath, A.

Granzyme B mediates neurotoxicity through a G-protein-coupled receptor

Tongguang Wang^*, Rameeza Allie^*, Katherine Conant^*, Norman Haughey^*, Jadwiga Turchan-Cholewo, Katrin Hahn^*, Antony Rosen, Joseph Steiner^*, Sanjay Keswani^*, Melina Jones^*, Peter A. Calabresi^* and Avindra Nath^*,1

^* Department of Neurology, Johns Hopkins University, Baltimore, Maryland, USA;

Department of Internal Medicine, Johns Hopkins University, Baltimore, Maryland, USA; and

Department of Anatomy and Neurobiology, University of Kentucky, Lexington, Kentucky, USA

¹Correspondence: Department of Neurology, Pathology Bldg. Rm. 509, 600 N. Wolfe St., Baltimore, MD 21287, USA. E-Mail: anath1{at}jhmi.edu

SPECIFIC AIMS

The purpose of this study was to determine 1) whether activated T cells release soluble factors such as granzyme B (GrB) that participate in T cell-induced neurotoxicity; 2) what were the mechanisms underlying the GrB-mediated neurotoxicity.

PRINCIPAL FINDINGS

1. Activated T cells release GrB, which is neurotoxic
To determine whether activated T cells induced neurotoxicity by releasing soluble factors, CD4 and CD8 T cells were isolated from normal donors and polyclonally activated by anti-CD3 and anti-CD28-conjugated beads. The conditioned T cell supernatants were then added onto cultured human fetal neurons. Both activated CD4 and CD8 T cells supernatants (1:10 dilution) caused significant increases in neuronal cell death when compared to supernatants from nonactivated T cells. Activated T cells released significant amounts of GrB into the supernatants, which could be detected by immunoprecipitation (Fig 1 A) and ELISA. Immunodepleting the supernatants of GrB reduced the activated T cells supernatant-induced neurotoxicity (Fig 1B ). These findings indicate that the GrB released by activated T cells is a critical soluble factor for mediating neurotoxicity.

View larger version (11K): [in this window] [in a new window]   Figure 1. Activated T cells release GrB into supernatants which cause neuronal toxicity. Peripheral blood mononuclear cells (PBMC) or sorted CD4+ or CD8+ cells were cultured for 3 days in Iscove’s modified Dulbecco’s Medium +5% human serum. In parallel, these cells were incubated with anti-CD3/CD28 (cells:beads, 10:1) mAb-conjugated magnetic beads to induce polyclonal activation (Ac). Supernatants (sups) were then collected for GrB detection. Both activated CD4+ and CD8+ T cells released significant high amounts of GrB into the sups as determined by GrB immunoblot (A). Cultured human fetal neurons were treated with culture supernatants (1:10 in Locke’s buffer) from unsorted human T cells. Control cultures were treated with supernatants from unactivated T cells. To determine the role of GrB, activated T cell supernatant (AcT) was firstly immunodepleted with Ab against GrB (GrB-Ab) or control mouse IgG bound to protein A beads before treatment. Neurotoxicity was determined by calculating the percentage of cells positive with trypan blue 48 h later (B). Data represent the mean ± SEM of 3 replicates from 3 independent experiments. Image representative for 3 independent experiments is shown.

2. Recombinant GrB induces neurotoxicity independent of perforin but via activation of Gi-coupled receptor
To determine the mechanism underlying GrB-induced neurotoxicity, recombinant GrB was used to treat the neuronal cultures. As low as 1 nM of GrB induced significant toxicity in neurons (Fig 2 A) but not in astroglia, as determined with trypan blue uptake assay. Interestingly, although cotreatment with perforin enhanced the GrB-induced neurotoxicity, GrB treatment alone was sufficient to cause toxicity in neurons (Fig 2A ). This observation indicates a perforin-independent pathway for the GrB-induced neurotoxicity. Furthermore, GrB-induced neurotoxicity could not be blocked by mannose-6-phosphate but was pertussis toxin (PTX) sensitive (Fig 2B ), demonstrating that the GrB-induced neurotoxicity is not related to mannose-6-phosphate receptor but is mediated via activation of Gi protein coupled receptors. As Gi coupled receptors are known to regulate adenylyl cyclase, which, in turn, regulates cAMP levels, we measured cAMP levels in the neuronal cultures after GrB treatment and found that GrB treatment decreased cAMP levels (Fig 2C ).

View larger version (10K): [in this window] [in a new window]   Figure 2. GrB-induced neurotoxicity is perforin and M6P receptor independent but mediated by Gi receptors and caspase dependent pathways. A) Human neuronal cultures were treated with GrB (1 nM), human perforin (50 ng/ml) or with the combination of the two. B) Human fetal neurons were treated with GrB in Locke’s buffer for 48 h. Mannose-6-phosphate (M6P, 1 mM), pertussis toxin (PTX, 100 ng/ml), or caspase inhibitor Z-VAD-FMK (Z-VAD, 10 µM) were added 1 h before GrB treatment. Neurotoxicity was determined with trypan blue uptake assay. C) cAMP levels were measured in neuronal cultures after GrB treatment. GrB treatment results in a significant decreases (P<0.01) in cAMP concentration at 5 min. Data represent the mean ± SEM of 3 replicates from 3 experiments.

3. Recombinant GrB causes neurotoxicity by increasing intracellular calcium and activating caspase-3
To characterize the downstream events involved in GrB-mediated neuronal cell death, we initially determined whether there was activation of the apoptotic cascade by measuring caspase-3 activity. Immunofluorescent staining showed that GrB treatment induced caspase-3 expression in neurons. GrB treatment also caused modest increases in intracellular calcium in neurons. The effect on intracellular calcium could be markedly potentiated by stromal-derived factor 1 (SDF-1). GrB-induced neurotoxicity could also be blocked by vitamin E and a neuroimmunophilin ligand, suggesting the free radical production may be an early event in GrB-mediated neurotoxicity.

CONCLUSIONS AND SIGNIFICANCE

T cell activation plays an important role in the pathogenesis of multiple sclerosis (MS) and other neuroinflammatory diseases. There is a significant correlation between the extent of axon damage and the numbers of T cells in MS plaques. Activated T cells can induce direct neuronal damage via cell-to-cell contact. Besides, activated T cells also release a variety of soluble factors, such as cytokines and proteases. However, little is known about whether these soluble factors cause neuronal damage. In the present study, we found that activated T cells could induce neuronal toxicity in a cell contact-independent manner by releasing soluble toxic factors. Among these soluble factors, GrB plays an important role in mediating neuronal toxicity.

GrB is stored in granules in restive T cells along with other cytotoxic factors, including perforin. On activation, T cells release GrB from the granules to act on target cells. Interestingly, although GrB was believed to play a more important role in mediating CD8 T cell-induced cytotoxicity, we observed that both activated CD4- and CD8-positive T cells release GrB extracellularly. This observation is consistent with reports that besides CD8 cells, subsets of CD4 positive cells can also express GrB. We found that the toxicity of GrB was specific for neurons, while no toxicity was observed in astrocytes. This is also consistent with a previous study that showed that astrocytes and oligodendrocytes are resistant to toxicity by cytotoxic T cells.

Classically, GrB is a serine protease that induces apoptosis by caspase activation after crossing the plasma membrane of target cells. In several systems/cell types, the ability of GrB to enter the target cells is dependent on perforin. In other systems, GrB entry is facilitated by expression of the mannose-6-phosphate receptor. We demonstrate that although perforin did enhance GrB-mediated neurotoxicity, recombinant GrB itself was sufficient to induce neurotoxicity, and the effect was independent of the mannose-6-phosphate receptor. This is consistent with recent observations that GrB binding to the cell membrane is cell-type specific, and the toxicity in these cells can be independent of both perforin and mannose-6-phosphate. Further, GrB-mediated effects were PTX sensitive. This implicates the stimulation of the Gi-coupled receptors. Conceivably, these interactions may occur at the cell membrane and may not require GrB uptake. Alternatively, GrB may generate a ligand by cleavage of extracellular matrix, which then interacts with the Gi coupled receptors. Moreover, it is becoming increasingly recognized that a number of cell surface receptors may be cleaved by proteinases with a consequent change in the activity of intracellular signaling molecules. Gi coupled receptors are known to regulate adenylyl cyclase, which, in turn, regulates cAMP levels. However, adenylyl cyclase is differentially regulated by the G_i family of heterotrimeric G-proteins and is stimulus dependent. Hence, we measured cAMP levels in the neuronal cultures after GrB treatment and found a decrease in cAMP levels. This may have important implications for neuronal function since as a second messenger cAMP plays a critical role in multiple forms of synaptic plasticity and in long-term memory formation.

GrB also disrupted calcium homeostasis by increasing resting levels of [Ca²⁺]_c, and enhancing IP3-mediated ER calcium release to attain concentrations of [Ca²⁺]_c that are sufficient to activate calcium-dependent death effectors, including the caspases. Because SDF-1 levels are elevated in the cerebrospinal fluid of patients with neuroinflammatory disorders, in areas of inflammation, GrB-mediated endoplasmic reticulum calcium overload may result in a synergistic and lethal response to SDF-1, suggesting that, even physiological or slightly elevated levels of such chemokines in the appropriate setting may be sufficient to cause neuronal injury and death. We also found that GrB-mediated neuronal cell death involved caspase-3 activation and could be prevented by MnTMPyP, a superoxide dismutase/catalase mimetic, vitamin E and neuroimmunophilins, suggesting an important role for oxidative stress upstream of caspase activation. These observations may have important therapeutic implications.

In conclusion, activated T cells release GrB, which causes neurotoxicity via novel membrane-mediated interactions. GrB causes stimulation of Gi receptors, leading to decrease in cAMP levels, increase in intracellular calcium, and induction of oxidative stress and activation of caspases-3. These observations may have important implications for T cell-mediated neuroinflammatory diseases such as MS.

View larger version (36K): [in this window] [in a new window]   Figure 3. Schematic flow chart depicting the possible mechanism for GrB-mediated activated T cells supernatant-induced neuronal damage. CD3/CD28 activated T cells release GrB into the supernatants. GrB induces neurotoxicity by both perforin-dependent or -independent mechanisms. As for the latter, GrB activates PTX-sensitive Gi-coupled receptors by either direct binding, or by cleaved fragments binding or by direct cleavage. The activation of Gi-coupled receptor then results in decreased cAMP levels and elevated intracellular calcium levels, which may induce caspase-3 activation, leading to neuronal apoptosis. It is likely that oxidative stress also participates in the neurotoxicity, and it is upstream of caspase activation.

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-5022fje

This article has been cited by other articles:

				L. Hu, M. Pennington, Q. Jiang, K. A. Whartenby, and P. A. Calabresi Characterization of the Functional Properties of the Voltage-Gated Potassium Channel Kv1.3 in Human CD4+ T Lymphocytes J. Immunol., October 1, 2007; 179(7): 4563 - 4570. [Abstract] [Full Text] [PDF]

				V. W. Yong, F. Giuliani, M. Xue, A. Bar-Or, and L. M. Metz Experimental models of neuroprotection relevant to multiple sclerosis Neurology, May 29, 2007; 68(22_suppl_3): S32 - S37. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text Full Text (PDF) All Versions of this Article: fj.05-5022fjev1 20/8/1209    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wang, T. Articles by Nath, A. Search for Related Content PubMed PubMed Citation Articles by Wang, T. Articles by Nath, A.