The neurotransmitter glutamate and human T cells: glutamate receptors and glutamate-induced direct and potent effects on normal human T cells, cancerous human leukemia and lymphoma T cells, and autoimmune human T cells.

J Neural Transm DOI 10.1007/s00702-014-1167-5

NEUROLOGY AND PRECLINICAL NEUROLOGICAL STUDIES - REVIEW ARTICLE

The neurotransmitter glutamate and human T cells: glutamate receptors and glutamate-induced direct and potent effects on normal human T cells, cancerous human leukemia and lymphoma T cells, and autoimmune human T cells Yonatan Ganor • Mia Levite

Received: 15 November 2013 / Accepted: 19 January 2014 Ó Springer-Verlag Wien 2014

Abstract Glutamate is the most important excitatory neurotransmitter of the nervous system, critically needed for the brain’s development and function. Glutamate has also a signaling role in peripheral organs. Herein, we discuss glutamate receptors (GluRs) and glutamate-induced direct effects on human T cells. T cells are the most important cells of the adaptive immune system, crucially needed for eradication of all infectious organisms and cancer. Normal, cancer and autoimmune human T cells express functional ionotropic and metabotropic GluRs. Different GluR subtypes are expressed in different T cell subtypes, and in resting vs. activated T cells. Glutamate by itself, at low physiological 10-8M to 10-5M concentrations and via its several types of GluRs, activates many key T cell functions in normal human T cells, among them adhesion, migration, proliferation, intracellular Ca2? fluxes, outward K? currents and more. Glutamate also protects activated T cells from antigen-induced apoptotic cell death. By doing all that, glutamate can improve substantially the function and survival of resting and activated human T cells. Yet, glutamate’s direct effects on T cells depend dramatically on its concentration and might be inhibitory at excess pathological 10-3M glutamate concentrations. The effects of glutamate on T cells also depend

Y. Ganor Department of Infection, Immunity and Inflammation, Cochin Institute, INSERM U1016, CNRS UMR8104, Paris Descartes University, Paris, France e-mail: [email protected] M. Levite (&) The School of Behavioral Sciences, The Academic College of Tel-Aviv Yaffo, Tel Aviv, Israel e-mail: [email protected]

on the specific GluRs types expressed on the target T cells, the T cell’s type and subtype, the T cell’s resting or activated state, and the presence or absence of other simultaneous stimuli besides glutamate. Glutamate also seems to play an active role in T cell diseases. For example, glutamate at several concentrations induces or enhances significantly very important functions of human T-leukemia and T-lymphoma cells, among them adhesion to the extracellular matrix, migration, in vivo engraftment into solid organs, and the production and secretion of the cancer-associated matrix metalloproteinase MMP-9 and its inducer CD147. Glutamate induces all these effects via activation of GluRs highly expressed in human T-leukemia and T-lymphoma cells. Glutamate also affects T cellmediated autoimmune diseases. With regards to multiple sclerosis (MS), GluR3 is highly expressed in T cells of MS patients, and upregulated significantly during relapse and when there is neurological evidence of disease activity. Moreover, glutamate or AMPA (10-8M to 10-5M) enhances the proliferation of autoreactive T cells of MS patients in response to myelin proteins. Thus, glutamate may play an active role in MS. Glutamate and its receptors also seem to be involved in autoimmune rheumatoid arthritis and systemic lupus erythematosus. Finally, T cells can produce and release glutamate that in turn affects other cells, and during the contact between T cells and dendritic cells, the latter cells release glutamate that has potent effects on the T cells. Together, these evidences show that glutamate has very potent effects on normal, and also on cancer and autoimmune pathological T cells. Moreover, these evidences suggest that glutamate and glutamatereceptor agonists might be used for inducing and boosting beneficial T cell functions, for example, T cell activity against cancer and infectious organisms, and that glutamate-receptor antagonists might be used for preventing

123

Y. Ganor, M. Levite

glutamate-induced activating effects on detrimental autoimmune and cancerous T cells. Keywords Glutamate Glutamate Receptor Normal human T cells Cancer human T cells Autoimmune human T cells

3.6.2

4

Table of Contents 1 2

3

Glutamate................................................................................... Glutamate receptors (GluRs) .................................................... 2.1 The ionotropic glutamate receptors (iGluRs) .................. 2.1.1 The NMDA type of iGluRs ............................. 2.1.2 The AMPA type of iGluRs.............................. 2.1.3 The KA type of iGluRs ................................... 2.2 Metabotropic glutamate receptors (mGluRs)................... Glutamate receptors and glutamate-induced direct effects on NORMAL human T cells.................................................................. 3.1 Introduction ....................................................................... 3.2 Normal human T cells express iGluRs and mGluRs ...... 3.2.1 iGluRs in normal human T cells ..................... 3.2.2 mGluRs in normal human T cells ................... 3.3 The direct effects of glutamate at different concentrations on normal human T cells, via iGluRs ............................. 3.3.1 T cell adhesion and chemotactic migration: Glutamate at physiological low nanomolar concentrations induces adhesion and chemotactic migration of naı¨ve/resting T cells, via AMPA iGluRs.... 3.3.2 Intracellular Ca2? fluxes: Glutamate at physiological mid micromolar concentrations increases intracellular Ca2? (iCa2?) in activated T cells, via AMPA, NMDA and KA iGluRs ............... 3.4 The direct effects of glutamate at different concentrations on normal human T cells, via mGluRs ............................. 3.4.1 Protection from apoptosis: Glutamate at physiological mid micromolar concentrations protects activated T cells from apoptotic ActivationInduced Cell Death (AICD), via group I mGluRs ............................................................... 3.4.2 Proliferation: Glutamate at pathological high millimolar concentrations inhibits the proliferation of activated T cells, via group I mGluRs ......... 3.5 The direct effects of glutamate at different concentrations on resting and activated normal human T cells, via a combined action of iGluRs and mGluRs.................................. 3.5.1 Outward potassium (K?) currents through Kv1.3 voltage-gated ion channels: Glutamate at physiological mid micromolar concentrations potentiates, while at pathologically high millimolar concentrations suppresses, Kv1.3 voltage-gated K? currents in naı¨ve/resting T cells, via group I/II mGluRs and non-NMDA iGluRs ................ 3.5.2 Cytokine secretion: Glutamate at pathological high millimolar concentrations affects T cell cytokine secretion, via group I mGluRs and NMDA iGluRs ............................................ 3.6 Proposed summary of the known glutamate-induced effects on resting and activated normal human T cells ................ 3.6.1 Condition 1: naı¨ve/resting T cells encountering physiological low micromolar–low nanomolar levels of glutamate .............................................

123

5

6

7

8

Condition 2: activated T cells encountering physiological mid micromolar levels of glutamate .. 3.6.3 Condition 3: activated T cells encountering very high and pathological millimolar levels of glutamate .................................................................... Glutamate receptors and glutamate-induced effects on CANCER human T cells .............................................................................. 4.1 Background: GluR antagonists block the growth of solid non-immune tumors............................................................ 4.2 Cancer human T cells express iGluRs and mGluRs......... 4.3 The direct effects of glutamate at different concentrations on cancer human T cells, via iGluRs and mGluRs .......... 4.3.1 Glutamate at physiological low nanomolar concentrations promotes in vivo spread of human T cell leukemia and T cell lymphoma, via AMPA iGluRs ................................................................. 4.3.2 Glutamate at physiological mid micromolar concentrations increases adhesion of human T cell leukemia, via NMDA iGluRs ............................ 4.3.3 Glutamate at pathological high millimolar concentrations promotes growth of human T cell leukemia and T cell lymphoma, via NMDA iGluRs ................................................................. 4.3.4 Glutamate at mid micromolar–high millimolar concentration increases the intracellular Ca2? (iCa2?) concentration in human T cell leukemia, via group I mGluRs ........................................... Glutamate receptors and glutamate-induced effects on AUTOIMMUNE human T cells ................................................................. 5.1 Multiple sclerosis (MS) ...................................................... 5.2 Rheumatoid arthritis (RA).................................................. 5.3 Systemic lupus erythematosus (SLE) ................................ T cells release glutamate which has potent effects on neuronal cells; DCs interacting with T cells release glutamate which has potent effects on the T cells................................................. Summary and concluding remarks: Glutamate is a ‘NeuroImmuno-Transmitter’ that at physiological concentrations has many direct, potent and activating effects on normal human resting and activated T cells, cancer T cells and autoimmune T cells, and on many other immune cells, all via its various ionotropic and metabotropic GluRs highly expressed in these cells .............................................................................................. References....................................................................................

1 Glutamate Glutamate is the most important excitatory neurotransmitter within the vertebrate nervous system (Meldrum 2000). Glutamate is involved in most aspects of the normal function and development of the central nervous system (CNS) (Foster and Fagg 1984; Mayer and Westbrook 1987; Komuro and Rakic 1993; Danbolt 2001). When the amino acid glutamic acid, which is acidic (and alike each amino acid, contains a central carbon atom, to which four different groups are bonded), loses a hydrogen from its side chain, it becomes glutamate, with a side chain composed of CH2CH2COO. In the human body, glutamic acid almost always exists as glutamate, because conditions in the body favor the loss of the hydrogen atom from glutamic acid,

The neurotransmitter glutamate and human T cells

Glutamate -O

NH3+

C5H9NO4

C-CH2-CH2-CH COO-

O Ionotropic glutamate receptors (iGluRs), i.e. glutamate-gated ion channel receptors

Metabotropic GluRs (mGluRs), i.e. glutamate-gated G protein-coupled receptors

Na+ and/or Ca2+

mGluR4,6,7,8

KA1/2

Group III

mGluR2,3

GluR5-7

Group II

mGluR1,5

GluR2

Group I

GluR5-7 KA1/2

GluR3

KA

GluR1-4

NR1

AMPA

NR1 NR2A/B/C/D NR3A/B

NR2A-D

NMDA

Fig. 1 Glutamate and glutamate receptors (GluRs). The amino acid glutamate is the major excitatory neurotransmitter in the mammalian CNS that is needed for, and beneficially involved in, most aspects of physiological brain function and development. Glutamate has also a signaling role in peripheral organs and tissues, and in endocrine cells. In contrast to the advantageous effects of glutamate when it is in physiological concentrations, elevated levels of glutamate, called ‘‘excess glutamate’’ lead to rapid and massive neuronal death–a process termed excitotoxicity–that plays a key pathological role in numerous acute and chronic neurological diseases and injuries. Glutamate has two major classes of receptors: (1) the ionotropic glutamate receptors (iGluRs) are membrane-spanning ion-channel

receptors gated by glutamate. They are multimeric receptors composed of four subunits, i.e., tetrameric receptors, and they are subdivided into three groups, based on their pharmacology and structural similarities. The iGluRs are subdivided to AMPA, NMDA and KA receptors; (2) the metabotropic glutamate receptors (mGluRs) are coupled to G proteins, and subdivided into three groups, termed group I, II and III mGluRs, based on sequence similarity, pharmacology and intracellular signaling mechanisms. Each of these classes and subclasses of iGluRs and mGluRs is further subdivided to specific receptor subtypes or groups, which contain multiple subunits (as shown in the figure, discussed very briefly in the text of this review, and in depth in numerous publications)

and it is glutamate that is the neurotransmitter that plays such a key role in the nervous system. But the function of glutamate extends beyond the nervous system, as it also plays signaling roles in peripheral organs and tissues, such as the heart, kidney, intestine, lungs, muscles, liver, ovary, testis, bone and pancreas, and in the adrenal, pituitary and pineal glands (Nedergaard et al. 2002; Hinoi et al. 2004; Gill and Pulido 2001). On top of all that, glutamate also induces direct, potent and important effects on most if not all cells of the immune system, as shown by a large body of evidence that accumulated in recent years. This topic is discussed in detail in our comprehensive and recently published book chapter entitled: ‘‘Glutamate in the immune system: glutamate receptors in immune cells, potent effects, endogenous production and involvement in disease’’ (Ganor and Levite 2012). Herein we will elaborate and extend these evidences only with regards to glutamate receptors and glutamateinduced effects on human T cells. T cells are the primary cells of the human immune system, which are responsible

for eradication of infectious organisms and cancer, and which help other immune cells perform their duties. Although highly regulated, glutamate levels are extremely different in the CNS and in the blood under physiological and pathological conditions. Under normal healthy conditions, the lowest levels of glutamate are found in the cerebrospinal fluid (CSF) and in the brain extracellular fluid, reaching a maximal concentration of 10-6M (1 lM) (Meldrum 2000). Glutamate levels are the highest within the synaptic cleft and may reach as much as 10-3M (1 mM) (Meldrum 2000). In the plasma of healthy individuals, glutamate is present at a concentration of 10-5 to 10-4M (10–100 lM) (Divino Filho et al. 1998; Graham et al. 2000; Meldrum 2000; Reynolds et al. 2002). While these are the normal values of glutamate in a healthy body, in numerous pathological conditions the concentration of glutamate increases substantially, both in the plasma and in the brain. Thus, glutamate levels may increase substantially, well above 10-4M, in the plasma in a kaleidoscope of pathologies conditions, such as immunodeficiency

123

Y. Ganor, M. Levite

(Droge et al. 1993; Eck et al. 1989b; Ferrarese et al. 2001) and cancer (Droge et al. 1988; Eck et al. 1989a; Ollenschlager et al. 1989). In addition, glutamate levels in the brain increase substantially, and then called ‘‘excess glutamate’’, in many neurological diseases and injuries, especially those that display a neuroinflammatory component, such as multiple sclerosis (MS), traumatic brain injury, acute brain anoxia/ischemia, epilepsy, glaucoma, meningitis, amyotrophic lateral sclerosis (ALS), Alzheimer’s disease, Huntington’s disease, Parkinson’s disease and others [see (Meldrum 2000; Sattler and Tymianski 2001) and (Pacheco et al. 2007) for few examples of elevation in glutamate levels]. Importantly, such excess glutamate is highly detrimental in the CNS, since it leads to ‘excitotoxicity’ (Sattler and Tymianski 2001), which stands for over-activation of specific types of glutamate receptors, resulting in a massive neuronal death, tissue damage and loss of brain function. The excitotoxicity mediated by excess glutamate plays a cardinal detrimental role in numerous neurological diseases and injuries. As for glutamate production and storage, neuronal cells produce glutamate and store it in vesicles, and the release of glutamate from the synaptic vesicles is a major source of extracellular glutamate (Nedergaard et al. 2002). Yet, glutamate production and release are not limited to neuronal cells. In fact, glutamate is produced also by immune cells, a topic also reviewed in our recent book chapter on glutamate in the immune system (Ganor and Levite 2012). In brief, immune cells that can produce glutamate include (but are not limited to) T cells (Garg et al. 2008), neutrophils (Collard et al. 2002), dendritic cells (DCs) (Affaticati et al. 2011; Pacheco et al. 2006), monocytes, macrophages and microglia (Kaul et al. 2005). In part 6 of this review we will discuss glutamate release by T cells as well as by DCs, and the functional consequences of such glutamate release on other immune cells and neuronal cells.

2 Glutamate receptors (GluRs) Glutamate acts via two classes of receptors: GluRs that are ligand-gated ion channels and called ionotropic glutamate receptors (iGluRs), and GluRs that are G protein-coupled and called metabotropic glutamate receptors (mGluRs) and that activate intracellular signal transduction pathways following binding of glutamate (Hollmann and Heinemann 1994; Kew and Kemp 2005; Masu et al. 1991; Monaghan et al. 1989; Tanabe et al. 1992). These two GluR families are shown schematically in Fig. 1, and are discussed in further detail in parts 2.1 and 2.2 below. Glutamate can activate all its ionotropic and metabotropic receptors. Yet, there are many selective GluR agonists and antagonists that

123

bind only, or preferentially, particular types of GluRs, and either activate or block their activity, respectively (Dingledine et al. 1999). Activation of GluRs in the nervous system is responsible for the basal excitatory synaptic transmission and many forms of synaptic plasticity, such as long-term potentiation (LTP) and long-term depression (LTD), mechanisms that are thought to underlie learning and memory. Thus, GluRs are also potential targets for therapies for several CNS disorders such as epilepsy and Alzheimer’s diseases. Various types of GluRs are expressed on the plasma membranes of the vast majority of neuronal and glial cells (Danbolt 2001). In addition, numerous GluRs are expressed also in peripheral tissues outside the CNS. Their role in these tissues is out of the scope of the current review, and for further reading on this topic the reader may refer to Nedergaard et al. (2002), Hinoi et al. (2004) and Gill and Pulido (2001). In addition, and as we discuss extensively later in this review, various other types of immune cells, especially T cells, also express a vast array of functional ionotropic and metabotropic receptors. 2.1 The ionotropic glutamate receptors (iGluRs) The iGluRs are membrane-spanning multimeric assemblies of four subunits, i.e., tetrameric receptors. They are subdivided into three groups according to their pharmacology, structural similarities, and the type of synthetic agonist that activates them: The N-methyl-D-aspartate (NMDA), Alpha-amino3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA), and 2-carboxy-3-carboxymethyl-4-isopropenylpyrrolidine (Kainate; KA) iGluRs. We will summarize the main characteristic features of the three iGluRs types in Sects. 2.1.1–2.1.3. While glutamate can activate all types of iGluRs, there are commercially available agonists, antagonists and antibodies (Abs) that bind preferentially selective iGluRs. On top of the following very short summary on the different types of iGluRs, the reader is referred to Mayer (2005), Midgett et al. (2012) and Wollmuth and Sobolevsky (2004) for recent reviews on the structure and function of iGluRs. A final note for this introduction: although a recent report suggested a novel nomenclature for ligand-gated ion channels, including iGluRs (Collingridge et al. 2009), we will use in this review the former names attributed to these receptors and their subunits, as in the original studies we cite. 2.1.1 The NMDA type of iGluRs The NMDA receptors are the principal receptors that control synaptic plasticity and memory (Li and Tsien


2009). NMDA iGluRs are hetero-oligomers composed of two obligatory NR1 subunits which are necessary for Ca2? conductivity of the receptor’s ion channel, and of two regulatory subunits of the four existing ones: NR2A–D, which determine the electrophysiological and pharmacological properties of the receptor. The two NR1 subunits and the two NR2 subunits co-assemble to form a tetramer (Rosenmund et al. 1998). Besides the NR1 and NR2 subunits, there is also a third subunit: NR3A/B, which is a regulatory subunit that decreases the receptor’s channel activity (Das et al. 1998). The assembly of NR3 with NR1 creates a functional glycine receptor that is not activated by glutamate (Chatterton et al. 2002). Importantly, the activation of the NMDA iGluRs requires the binding of two different agonists to two different subunits: the binding of glutamate to the NR2 subunit, and the binding of glycine to the NR1 subunit (Kew and Kemp 2005). In addition, membrane depolarization is needed to abrogate the blockade of these receptors by Mg2?, allowing the flow of the voltage-dependent Na? and small amounts of Ca2? ions into the cell (i.e., inward Ca2? [iCa2?] currents), and of K? ions out of the cell (i.e., outward K? currents) (Moriyoshi et al. 1991; Collingridge and Singer 1990). 2.1.2 The AMPA type of iGluRs The AMPA receptors are important for plasticity and synaptic transmission at many postsynaptic membranes. AMPA iGluRs are homo- or hetero-oligomers composed of the GluR1–GluR4 subunits (Keinanen et al. 1990) that assemble as functional tetramers (Rosenmund et al. 1998) containing four sites to which glutamate or AMPA iGluR agonists can bind (Mayer 2005). The ligand binding site of the AMPA iGluR is formed by the N-tail and the extracellular loop between the third and forth transmembrane domains (Armstrong et al. 1998), which move towards each other to open the channel pore following ligand binding. The iGluR channel opens when two sites are occupied (Platt 2007) and increases its current as more binding sites are engaged (Rosenmund et al. 1998). Permeability of AMPA iGluRs to Ca2? is governed by the GluR2 subunit, since the presence of a GluR2 subunit renders the channel impermeable to Ca2?. Native AMPA iGluRs contain GluR2 and are therefore impermeable to Ca2? ions (Kew and Kemp 2005). Gating of the AMPA iGluRs by glutamate is extremely fast, in contrast to the slow gating of the NMDA iGluRs (Dingledine et al. 1999). The different affinities of glutamate for AMPA vs. NMDA iGluRs have an important functional consequence: as both types of iGluRs are co-localized at neuronal synapses, the fast activation and opening of AMPA iGluRs alleviates the Mg2? block of NMDA iGluRs to facilitate their activation (Dingledine et al. 1999).

2.1.3 The KA type of iGluRs The KA receptors have a somewhat more limited distribution in the brain compared to the AMPA and NMDA receptors (Hollmann et al. 1989). KA iGluRs play a role in both pre- and postsynaptic neurons (Huettner 2003) and interestingly, they can lead to either excitation or inhibition, depending on their location. Thus, the presynaptic KA iGluRs are implicated in inhibitory neurotransmission, by modulating the release of the inhibitory neurotransmitter gamma-amino-butyric acid (GABA), while the postsynaptic KA iGluRs are involved in excitatory neurotransmission. The ion channel formed by KA iGluRs is permeable to Na? and K? ions, but impermeable to Ca2?. KA iGluRs are composed of tetrameric assemblies: GluR5–7 subunits can form homomeric functional receptors, as well as combine with KA1 and KA2 to form heteromeric receptors with distinct pharmacological properties. The KA1 and KA2 subunits themselves do not form homomeric functional receptors (Keinanen et al. 1990; Lerma 2006). 2.2 Metabotropic glutamate receptors (mGluRs) The mGluRs are involved in a variety of functions in the central and peripheral nervous system, such as learning, memory, anxiety and perception of pain (Ohashi et al. 2002). The mGluRs are found in pre- and postsynaptic neurons in synapses of the hippocampus, cerebellum and the cerebral cortex (Hinoi et al. 2001), as well as in other parts of the brain, and in peripheral tissues (Chu and Hablitz 2000). Similar to other types of metabotropic receptors, mGluRs have seven transmembrane domains that span the cell membrane (Platt 2007). Unlike iGluRs, the mGluRs are not ion channels, but rather lead, upon their ligation, to the subsequent activation of biochemical cascades that finally result in the modification of other proteins. This can lead to changes in the synapse’s excitability, for example, by presynaptic inhibition of neurotransmission (Sladeczek et al. 1993) or to the modulation and even induction of postsynaptic responses (Bonsi et al. 2005; Endoh 2004; Platt 2007). The mGluRs are also subdivided into three groups, termed group I, II and III mGluRs, based on sequence similarity, pharmacology and intracellular signaling mechanisms. Group I mGluRs, mGluR1 and 5, are associated with Gq proteins and coupled to phospholipase C (PLC), while group II mGluRs, mGluR2 and 3, and group III mGluRs, mGluR4, 6, 7 and 8, are associated with Gi and G0 proteins, and are negatively coupled to adenylate cyclase. These eight mGluRs are products of different genes (Masu et al. 1991; Pin and Duvoisin 1995; Tanabe et al. 1992). The mGluRs function as homodimers, with two glutamate molecules being required for full receptor activation (Kew and Kemp 2005).

123

Y. Ganor, M. Levite Table 1 Expression of a plethora of GluRs in human T cells Cells

GluRs identified

Methods used

References

Normal T cells T cells, naı¨ve/resting

iGluR AMPA: GluR3

RT-PCR, sequencing, Western blotting, flow cytometry, microscopy

Ganor et al. (2003, 2007)

iGluR NMDA: NR1

Flow cytometry

Mashkina et al. (2007)

iGluR NMDA: NR1, NR2B

RT-PCR, sequencing, flow cytometry

Miglio et al. (2005b)

mGluR group I: mGluR5,1

RT-PCR

Chiocchetti et al. (2006)

mGluR group I: mGluR5

RT-PCR, flow cytometry, microscopy

Pacheco et al. (2004, 2006)

RT-PCR

Poulopoulou et al. (2005a)

PBMCs, naı¨ve/resting PBLs Naı¨ve/resting PHA-activated PBMCs, naı¨ve/resting PBMCs Naı¨ve/resting PHA/Anti-CD3/antigenpulsed DCs-activated PBMCs from healthy individuals

iGluR NMDA: NR1, NR2A,B,D

mGluR group I: mGluR5,1 mGluR group I: mGluR1b mGluR group II: mGluR2,3 mGluR group III: mGluR8

Cancer T cells T leukemia (Jurkat)

iGluR AMPA: GluR3

RT-PCR, sequencing, flow cytometry

Ganor et al. (2003, 2009)

iGluR AMPA: GluR2,4

RT-PCR

Stepulak et al. (2009)

RT-PCR, sequencing, flow cytometry, microscopy

Miglio et al. (2005b, 2007)

iGluR KA: GluR6,7/KA1,2 iGluR NMDA: NR2A-D, NR3A,B iGluR NMDA: NR1, NR2B iGluR NMDA: NR1

Western blotting

Braun et al. (2010)


RT-PCR, flow cytometry

Pacheco et al. (2004), Chiocchetti et al. (2006)


RT-PCR

Stepulak et al. (2009)

RT-PCR


mGluR group II: mGluR2,3 mGluR group III: mGluR4,6,7 T leukemia (FRO, SUP-T1)

mGluR group I: mGluR1 (FRO, SUP-T1), mGluR5 (FRO)

T lymphoma (HuT-78, HP)

iGluR AMPA: GluR3 (HuT-78)

Flow cytometry

Ganor et al. (2009)

mGluR group I: mGluR1 (HuT-78), mGluR5 (HuT-78, H9)

RT-PCR


iGluR AMPA: GluR3

RT-PCR, Western blotting

Sarchielli et al. (2007)

mGluR group I: mGluR1b

RT-PCR

Poulopoulou et al. (2005a)

Autoimmune T cells Lymphocytes from MS patients PBMCs from ALS patients

mGluR group II: mGluR2,3 mGluR group III: mGluR8

3 Glutamate receptors and glutamate-induced direct effects on NORMAL human T cells 3.1 Introduction The vast majority of studies that investigated the expression of GluRs in human immune cells, and the effects induced by glutamate on these cells, focused on T cells. Hence, in this review we summarize, analyze and discuss the data accumulated thus far with regards to GluRs and

123

glutamate-induced effects in human T cells in health and disease. The relevant findings are presented in this review along three axes, dealing independently with normal T cells, cancer T cells and autoimmune T cells. Each functional outcome is described separately, while taking into account: (a) the concentrations of glutamate inducing the specific T cell effect, (b) the activation state of the target T cells, and (c) the specific GluR family and GluR subtype/s that mediate the respective glutamate-induced effect on the T cells. Finally, we also incorporate all these findings into a


suggested model that attempts to summarize all the effects of glutamate on human T cells in health and disease reported thus far. While this review will focus only on human T cells, for further reading on GluRs and glutamateinduced effects on non-human T cells, and also on other types of immune cells besides T cells, the readers are referred to our recently published comprehensive book chapter (Ganor and Levite 2012). Collectively, the studies described in this review provide clear evidence that glutamate on its own directly and potently affects many different human T cell features and functions. This is in line with the direct effects of various other neurotransmitters and neuropeptides on T cells (Levite 2008, 2012). Interestingly, while at low and physiological concentrations glutamate usually activates or elevates many T cell features and functions, when glutamate’s concentration raise markedly, such as in various pathological conditions, glutamate usually does the opposite and inhibits T cell function. Thus glutamate-induced effects depend on a group of factors that we collectively call ‘the context’. The most important ‘context factor’ dictating the functional outcome of the interaction between glutamate and T cells seems to be glutamate’s concentration. Accordingly, glutamate induces different immune effects at nanomolar, micromolar or millimolar concentrations. These different concentrations actually reflect the in vivo levels of glutamate under normal and healthy physiological conditions, in contrast to the excess glutamate in numerous pathological conditions (for reminder of the specific concentrations, see part 1 above). Another very important ‘context factor’ that crucially affects the responsiveness of T cells to glutamate is the T cell’s activation state, i.e., whether the T cells are naı¨ve/resting or rather already activated, primarily since different GluRs are expressed in naı¨ve/resting vs. activated T cells. 3.2 Normal human T cells express iGluRs and mGluRs The possible expression of GluRs in T cells was first provided by indirect evidence, showing that radio-labeled glutamate binds with high affinity to naı¨ve normal human T cells (Kostanyan et al. 1997). Such binding was inhibited by glutamatecontaining dipeptides, and was evident when glutamate was conjugated to dextran, demonstrating that GluRs are expressed on the T cell outer membrane (Kostanyan et al. 1997). Although the exact identification of the specific GluR subtypes expressed in these T cells was not reported, subsequent reports provided direct evidence for the expression of a plethora of iGluRs and mGluRs in human T cells (Table 1). 3.2.1 iGluRs in normal human T cells The first direct evidence of high expression of functional iGluRs on the cell surface of normal human T cells was in

fact discovered in our own studies (Ganor et al. 2003). Thus, using four different methodologies, which included RT-PCR, Western blotting, flow cytometry and immunofluorescent microscopy, we demonstrated for the first time that normal naı¨ve/resting human T cells that were freshly purified from the blood of healthy volunteers express high levels of a specific glutamate receptor subunit: the AMPA GluR3. Both GluR3 mRNA and GluR3 cell surface protein were detected (Ganor et al. 2003). Importantly, sequencing showed that the T cell-expressed GluR3 is identical to the brain’s GluR3. In that study, we also found that glutamate by itself, at low physiological concentrations, directly activates naı¨ve/resting human T cells via such AMPA iGluRs, and induces two very important T cell functions: first, T cell adhesion to fibronectin and laminin, and second, chemotactic migration towards the key chemokine CXCL12/SDF-1 (Ganor et al. 2003). Interestingly, the maximal glutamate-induced effects were seen when glutamate was at a low concentration of 10-8M. In a subsequent study we revealed another novel and interesting phenomena regarding GluR3 expression in human T cells: while naı¨ve/resting human T cells indeed express high levels of GluR3 on their cell surface, activated T cells do not (Ganor et al. 2007). The reason for this is the elimination/degradation of GluR3 from the T cell surface after the T cells have been activated via their T cell receptor (TCR). This GluR3 degradation following T cell activation is carried out by granzyme B, a proteolytic enzyme that is produced and secreted by TCR-activated T cells, and subsequently acts in an autocrine/paracrine manner to eliminate GluR3 from the cell surface of the human T cells (Ganor et al. 2007). Of note, in the experiment revealing this novel mechanism, the TCR activation, which normally occurs in vivo by a specific antigen presented to T cells by antigen presenting cells (APCs), was mimicked in vitro by anti-CD3/CD28 antibodies (Ganor et al. 2007). Interestingly, this process of GluR3 cleavage/elimination from the cell surface by granzyme B is not unique to T cells and seems to operate also in neurons, since the neuronal GluR3 also serves as a substrate for granzyme B-mediated cleavage (Gahring et al. 2001). The expression of GluR3 on the T cell surface is restored few days after TCR activation, when the cells revert to their naı¨ve/resting phenotype (Ganor et al. 2007). Collectively, these findings show that: (a) the expression of GluR3 in human T cells is a dynamic and fluctuating feature; (b) in T cells, GluR3 expression on the cell surface is directly or indirectly regulated by, or sensitive to, the TCR. This reveals a novel and exciting relationship in human T cells, whereby the cell surface expression of a neurotransmitter receptor, the AMPA GluR3, is regulated by the antigen receptor, the TCR.

123

Y. Ganor, M. Levite

On top of all the above, we postulate that sometimes, maybe in the context of neuroinflammation, and/or on a specific immunogenetic background, GluR3 proteolysis from the cell surface of activated T cells may lead to detrimental and pathological effects in the brain. This is due to the release of GluR3-derived degradable autoantigenic peptides to the extracellular milieu, which in some conditions might subsequently lead to the generation of anti-GluR3 antibodies that damage the brain and contribute to ‘Autoimmune Epilepsy’ and to additional neurological problems. For further reading on this topic–the presence of anti-GluR3 antibodies in epilepsy patients, and their ability to activate AMPA receptors that contain the GluR3 subunit, lower seizure threshold, kill neurons, cause brain damage in vivo, and induce psycho/cognitive/motor/behavior abnormalities—the readers are refereed to Levite and Ganor (2008). Normal human T cells express not only AMPA iGluRs but also NMDA iGluRs, as shown in several studies. Indeed, using RT-PCR, sequencing and flow cytometry, normal human T cells were found to express the NMDA NR1 subunits (Mashkina et al. 2007; Miglio et al. 2005b) and the NMDA NR2B subunits (Miglio et al. 2005b). Interestingly, upon activation of the T cells with the potent mitogen phytohaemagglutinin (PHA), NR1 and NR2B levels increased, and the NR2A and NR2D subunits showed up (Miglio et al. 2005b). These observations strengthened the notion that iGluRs expression in normal human T cells is normally affected by, and/or coupled to, the activation state of the T cells that is mediated by the TCR. 3.2.2 mGluRs in normal human T cells The first mGluR subtypes identified in normal human T cells, using RT-PCR, flow cytometry and immunofluorescent microscopy, were mGluR5 and mGluR1 (Pacheco et al. 2004), which belong to the group I mGluRs. The mGluR5 is expressed constitutively both in naı¨ve/resting human T cells, and in activated T cells, i.e., resting T cells that were activated either by PHA, the OKT3 anti-CD3/ TCR antibodies, or antigen-pulsed DCs (Pacheco et al. 2004, 2006). In contrast, mGluR1 that also belongs to group I mGluRs, is expressed only in activated, but not in naı¨ve/resting normal human T cells (Pacheco et al. 2004, 2006). Another study confirmed, by RT-PCR, the constitutive expression of mGluR5 mRNA in both resting and activated human T cells, but showed the presence of mGluR1 mRNA also in resting T cells (Chiocchetti et al. 2006). A yet another study showed by RT-PCR that peripheral T cells of healthy individuals express the mRNA of mGluR1, 2, 3 and 8 (Poulopoulou et al. 2005a).

123

Taken together, the above studies show that normal human T cells definitely express several types of specific receptors for glutamate, both iGluRs and mGluRs, and that T cell activation often down- or up-regulates dramatically their expression. As a result, resting and activated normal human T cells express different GluR types and levels. This, once again, suggests a strong neuro-immune interaction in T cells between the specific receptors for glutamate and the TCR. 3.3 The direct effects of glutamate at different concentrations on normal human T cells, via iGluRs 3.3.1 T cell adhesion and chemotactic migration: Glutamate at physiological low nanomolar concentrations induces adhesion and chemotactic migration of naı¨ve/resting T cells, via AMPA iGluRs Following our discovery that normal resting human T cells express high levels of the iGluR AMPA GluR3 on their cell surface (Ganor et al. 2003), we further found that glutamate on its own (i.e., in the complete absence of any other stimuli like an antigen, mitogen, cytokine, etc.) induces adhesion of normal resting human T cells to two major glycoproteins of the extracellular matrix (ECM): fibronectin and laminin (Ganor et al. 2003). These glutamateinduced effects had a bell-shape dose-dependent relationship, were effective at a very broad, relatively low and physiological glutamate concentration range of 10-12 to 10-6M, and were maximal at 10-9 to 10-8M. T cell adhesion to the ECM is a very important T cell function, occurring rapidly and transiently only after activation of specific adhesion receptors (often being the specific integrins) expressed on the T cell outer surface. T cell adhesion to the ECM is critically needed for the migration, extravasation and homing of T cells from the blood and lymphatic system into solid organs and various tissues under physiological and pathological conditions. Only few physiological stimuli can activate the T cell’s adhesion receptors and induce adhesion of the T cells to the ECM’s glycoproteins, a fact that emphasizes how unique is glutamate that can do so. Glutamate induced T cell adhesion to fibronectin and laminin by activating its AMPA iGluRs expressed on the cell surface of these T cells, since this effect was mimicked by the AMPA iGluRs agonists AMPA and KA, and blocked by several AMPA iGluRs antagonists, among them CNQX and NBQX (Ganor et al. 2003). Moreover, glutamate induced T cell adhesion to laminin by upregulating the function of the T cell a6b1 laminin-binding integrins (Ganor et al. 2003).


Interestingly, in parallel to the TCR activation and the subsequent cleavage of GluR3 from the T cell surface by granzyme B, glutamate could no more induce adhesion of these activated T cells to laminin, like it did in resting T cells (Ganor et al. 2007). This suggested a possible mechanism by which activated T cells, via shedding their surface GluR3, can shut off also the pro-adhesive effect induced by glutamate. We speculate that this may occur in situations where the T cells have already reached their target/destiny and need to remain in site, without any further adhesion or migration, at least for a while. Glutamate on its own, at low 10-8M, induced an additional very important T cell function in naı¨ve/resting T cells, namely chemotactic migration. Thus, glutamate increased the chemotactic migration of naı¨ve/resting normal human T cells towards the very important chemokine CXCL12/SDF-1 (Ganor et al. 2003). CXCL12/SDF-1 is essential for recruiting T cells and various other cell types to the right place in the right time. Depending on the organ or tissue, CXCL12/SDF-1 serves also as a proliferation or cell survival factor, influences differentiation, induces adhesion and/or regulates cell migration. These CXCL12/ SDF-1-induced functions are mediated by the two chemokine receptors, CXCR4 and CXCR7 (Puchert and Engele 2014). Glutamate-induced T cell chemotaxis to CXCL12/SDF-1 was mediated by activation of CXCR4, proven by the fact that a CXCR4 monoclonal antibody abrogated this glutamate-induced effect. Together, these results demonstrated that glutamate, at very low nanomolar concentrations and by acting via its AMPA iGluRs that are highly expressed in normal naı¨ve/resting human T cells, induced both a pro-adhesive and a pro-migratory effects on such T cells.

the activation state of the target T cells. In specific, at mid micromolar concentrations, glutamate increased iCa2? in activated T cells, but not in naı¨ve/resting T cells (Lombardi et al. 2001). This glutamate-induced iCa2? potentiating effect had a bell-shape concentration-dependent relationship, was effective at a concentration range of 10-7 to 10-5M, and was maximal at 10-6M, alike glutamate’s physiological concentration in the brain’s extracellular fluids. In contrast, glutamate at a higher concentration range of 10-4 to 10-3M, alike present in the brain’s synapses and also in the blood under some pathological conditions (see part 1 above), failed to increase iCa2? (Lombardi et al. 2001). This suggests that while activated T cells are unresponsive to glutamate at a range of 10-4 to 10-3M, they respond favorably to much lower glutamate concentrations. It also suggests that the most effective interactions between glutamate and activated T cells take place after the T cells have crossed the blood brain barrier (BBB), enter the brain and encounter glutamate, which is present in the extracellular fluid at a concentration that can affect them (10-6M). The reported glutamate-mediated iCa2? increase in activated T cells was mediated by iGluRs, in view of the observation that three iGluRs agonists, NMDA, (S)-AMPA and KA, at a similar concentration range, mimicked the glutamate-induced effect and also caused iCa2? rise, while four iGluRs antagonist and blockers, D-AP5, (?)-MK801, NBQX and KYNA, inhibited it (Lombardi et al. 2001). Interestingly, the prototype mGluR agonist (1S,3R)-ACPD was ineffective at this concentration range (Lombardi et al. 2001), and so was the non-selective mGluR1/5 agonist (S)3,5-DHPG, tested at concentrations of up to 10-4M (Pacheco et al. 2004).

3.3.2 Intracellular Ca2? fluxes: Glutamate at physiological mid micromolar concentrations increases intracellular Ca2? (iCa2?) in activated T cells, via AMPA, NMDA and KA iGluRs

3.4 The direct effects of glutamate at different concentrations on normal human T cells, via mGluRs

The iCa2? ion fluxes and the Ca2?-mediated signaling in response to an antigenic stimulation of T cells via their TCR are very important events taking place in T cells. They are essential for the subsequent proliferation of the T cells (Guse 1998), and for various other functions that activated T cells ought to perform. On these grounds, it is clear that any molecule that augments iCa2? could significantly affect and improve T cell function. Several studies reported that glutamate can trigger Ca2? signaling in T cells. This specific glutamate-induced effect, alike the other effects glutamate triggers in T cells, is highly dependent on glutamate’s concentration, and also on

3.4.1 Protection from apoptosis: Glutamate at physiological mid micromolar concentrations protects activated T cells from apoptotic ActivationInduced Cell Death (AICD), via group I mGluRs It is well documented that activation of T cells via their TCR by appropriate antigens and MHC molecules, or by other less specific molecules that mimic this process, induces robust T cell proliferation, cytokine secretion, and the up-regulation of many other T cell features and functions. In contrast, chronic TCR activation, or reactivation of T cells within a short period after the first antigenic stimulation, induces apoptosis of the T cells via a

123

Y. Ganor, M. Levite

mechanism termed Activation-Induced Cell Death (AICD). AICD is crucial for maintenance of peripheral tolerance and for limiting an ongoing immune response (Green et al. 2003). Interestingly, glutamate at a broad concentration range of 10-8 to 10-4M protected human T cells from apoptotic AICD (Chiocchetti et al. 2006). This indicates that glutamate can contribute to improved/prolonged survival of activated T cells by protecting them from apoptosis induced by repeated/prolonged/chronic antigenic stimulation. Glutamate exerted its maximal anti-AICD effect at 10-6M. This effect was mimicked by the prototype mGluR agonist (1S,3R)-ACPD, as well as by the non-selective group I mGluR agonists quisqualate and (S)-3,5-DHPG, and by the selective mGluR5 agonist CHPG (Chiocchetti et al. 2006). In line with that, several group I mGluRs antagonists and blockers, AIDA, LY367385 and MPEP, antagonized glutamate-induced inhibition of AICD in activated T cells (Chiocchetti et al. 2006). The protection of activated T cells from AICD by glutamate was further shown to result from inhibition of FasL expression (Chiocchetti et al. 2006), which is known to be involved in AICD. The ability of glutamate on its own to prolong the survival of activated T cells by protecting them from AICD, if indeed occurring in vivo, may have important implications for fighting some chronic infectious organisms and cancer, since both can proliferate rapidly, and by doing so ‘force’ the T cells to deal with constantly growing amount of antigens, which are often present in the body for prolonged periods of time. 3.4.2 Proliferation: Glutamate at pathological high millimolar concentrations inhibits the proliferation of activated T cells, via group I mGluRs An early observation suggested that elevated levels of glutamate in the plasma correlate with a reduction in the mitogenic response of blood lymphocytes to pokeweed mitogen (Droge et al. 1988). Later studies confirmed this initial observation, and provided direct evidence that glutamate at a very high and pathological millimolar concentration range (10-3 to 10-2M) inhibited proliferation of T cells, when such T cell activation was induced by PHA or anti-CD3±CD28 antibodies. Interestingly, while glutamate suppressed the proliferation of activated T cells, it did not affect the proliferation of normal naı¨ve/resting human T cells (Lombardi et al. 2001, 2004; Pacheco et al. 2004), showing again the marked differences between glutamate-induced effects on naı¨ve/ resting and activated T cells. The iGluRs agonists (S)AMPA, NMDA and KA, tested at the same concentration range, did not affect the proliferation of PHA-activated T cells (Lombardi et al. 2004). Yet, the NMDA iGluR antagonists D-AP5 and (?)-MK801 inhibited PHAinduced (but not IL-2-induced) T cell proliferation (Miglio et al. 2005b). Hence, AMPA and KA iGluRs do not seem

123

to mediate the anti-proliferative effect that glutamate induced at the very high non-physiological concentration on activated T cells, while the exact contribution of NMDA iGluRs to this effect requires further studies. In fact, the inhibitory effect of glutamate on the proliferation of activated T cells, if occurring in vivo at all under pathological conditions, might be mediated by mGluRs, based on the observation that the selective mGluR5 agonist CHPG (5 9 10-4M) also inhibited the proliferation of CD3-activated T cells (Pacheco et al. 2004). Since such inhibitory effect was abrogated by the non-selective mGluR1/5 agonist (S)-3,5-DHPG (10-4M), and since both mGluR5 and mGluR1 are expressed in activated T cells, it could be that glutamate at very high concentration might play a dual role in human T cell function: on the one hand inhibiting T cell proliferation via mGluR5, and on the other hand reverting this effect via mGluR1 (Pacheco et al. 2004). Similar observations were indeed reported later, showing that glutamate released by antigen-pulsed DCs acted initially via mGluR5 to impair T cell proliferation, but later stimulation of mGluR1 abrogated the anti-proliferative effect mediated by mGluR5 to allow robust T cell proliferation (Pacheco et al. 2006). Together, the evidences discussed in the above parts suggest that glutamate at different concentration binds to mGluRs in activated T cells and by doing so induces two different and somewhat opposing effects: glutamate at 10-6M inhibits T cell apoptosis and prolongs survival, while glutamate at much higher and non-physiological concentration of 10-3 to 10-2M can inhibit T cell proliferation. Such unique ability of glutamate to induce contrasting effects on activated T cells in two different concentration ranges may be important for either maintaining or limiting ongoing immune responses, respectively. Yet, additional studies are needed to determine whether glutamate indeed exerts similar effects in physiological and/or pathological conditions. 3.5 The direct effects of glutamate at different concentrations on resting and activated normal human T cells, via a combined action of iGluRs and mGluRs 3.5.1 Outward potassium (K?) currents through Kv1.3 voltage-gated ion channels: Glutamate at physiological mid micromolar concentrations potentiates, while at pathologically high millimolar concentrations suppresses, Kv1.3 voltage-gated K? currents in naı¨ve/resting T cells, via group I/II mGluRs and non-NMDA iGluRs The Ca2? influxes that are necessary for the cell cycle progression, proliferation and other essential processes of


many cell types including T cells require a negative membrane potential as a driving force. In T cells, membrane depolarization is provided by voltage-gated K? (Kv) channels (Lin et al. 1993), and especially by the Kv1.3 channel (Cahalan et al. 2001). Interestingly, we previously discovered that the Kv1.3 and b1 integrins are physically and functionally coupled in normal human T cells, and that the mere opening of the Kv1.3 channels in resting T cells by depolarization leads by itself to the activation of the b1 integrins and to the subsequent T cell adhesion to fibronectin (Levite et al. 2000). Poulopoulou et al. found that glutamate at micromolar physiological concentrations of 10-6 to 10-5M (alike glutamate’s concentration in the CSF and plasma) potentiated Kv1.3 currents in normal resting human T cells (Poulopoulou et al. 2005b). The Kv1.3 potentiating effect of glutamate at 10-6M was mimicked by the prototype mGluR agonist (1S,3R)-ACPD, as well as by the non-selective group I mGluR agonist (S)-DHPG. At higher concentrations (10-4 to 10-3M) glutamate induced an opposite effect and suppressed Kv1.3 currents, as a result of a transition from the open to the inactivated state of the Kv1.3 channel (Poulopoulou et al. 2005b). The group II mGluR agonist DCG-IV mimicked the suppressive effect of high-dose glutamate, indicating that the block of Kv1.3 by non-physiological glutamate concentration of 10-4 to 10-3M was mediated by mGluRs. KA, the iGluR agonist that activates both KA and AMPA iGluRs, also caused a dose-dependent suppression of Kv1.3 currents, although its contribution to the overall effect was small (Poulopoulou et al. 2005b). These suppressive effects seem to correlate with glutamate’s above-mentioned anti-proliferative effect at a similar very high concentration (discussed above). Yet, since the ability of glutamate to modulate Kv1.3 currents was tested only in resting human T cells, additional studies are necessary to test the effects of glutamate at different concentrations on Kv1.3 currents in activated T cells. Together, these studies, and the ones described above and in the other related publications, show that glutamate at relatively low and physiological micromolar concentration acts on normal human T cells via iGluRs to potentiate Kv1.3 currents and promote T cell adhesion. Yet, at much higher and non-physiological millimolar concentrations glutamate acts on such cells via mGluRs to inhibit both Kv1.3 currents and T cell proliferation.

3.5.2 Cytokine secretion: Glutamate at pathological high millimolar concentrations affects T cell cytokine secretion, via group I mGluRs and NMDA iGluRs 3.5.2.1 Background on cytokine secretion by T cells All the T cell activities are completely dependent on various types of cytokines and cytokine receptors. Different T cell subpopulations, for example, CD4? T helper (Th) cells, produce and secrete many types of cytokines, and also have many types of cytokine receptors. Most cytokines act in several fashions: autocrine, affecting the generating cell (self); paracrine, affecting cells in the immediate vicinity; and endocrine, affecting cells remote from the secreting cell. The actions of all the cytokines are tightly regulated by several mechanisms to avoid undesired effects, but nevertheless, there are various perturbations in various cytokines in many diseases–immunological, neurological and others–and some of the cytokine abnormalities are believed to contribute directly to the tissue damage and overall pathology. Traditionally, Th cells were classified as either Th1 or Th2 cells: Th1 cells mainly secrete IFNc and IL-2 and promote immunity against intracellular pathogens, while Th2 cells secrete mainly IL-4, IL-5, IL-10 and IL-13 and promote humoral responses and defense against extracellular parasites. More recently, few additional and important T cell subsets have been identified, among them Th9 and Th17 cells, each producing and responding to a different composition of cytokines (Nakayama and Yamashita 2010; Zhou et al. 2009). On these grounds, it is clear that any physiological molecule, which is able to stimulate the beneficial secretion of specific cytokines in specific pathological conditions in which these cytokines may be in suboptimal concentration (e.g., immunodeficiency, infectious diseases, cancer and others), or suppress the detrimental secretion of some other cytokines (especially the pro-inflammatory ones) in other pathological conditions in which these cytokines are in excess (e.g., chronic inflammation, some autoimmune diseases, graft rejection, graft versus host disease and others), is expected to have profound impact on the type, efficiency and control of many immune reactions, and can even help overcome various types of diseases. This is the reason why the ability of glutamate on its own to induce secretion of various cytokines, discussed in the next paragraph, is of clear scientific and clinical interest and importance. Few final facts in this very minimalistic and superficial introduction on cytokine secretion by T cells are: First, it is well known now that the very same given cytokine

123

Y. Ganor, M. Levite Glutamate affects activated T cells in a direct and potent manner, that is highly dependent on: glutamate concentration, the activation state of the T cells, and the glutamate receptors they express. Therefore, glutamate can both activate and suppress various T cell features and functions in different contexts. At low physiological concentration, glutamate activates many key T cell functions and thereby is highly beneficial for the T cells.

NMDA iGluR

AMPA iGluR

Induction of 1) Adhesion to fibronectin and laminin 2) Chemotactic migration

AMPA iGluR

mGluR1 mGluR5

Induction of inward Ca2+ currents

Inhibition of apoptotic AICD, thereby prolongation of survival

NMDA iGluR

NR1

Glutamate

NR2A,B,D

GluR2

GluR1,4

NR2A,B,D

mGluR5

NR1

Glutamate

GluR2

GluR3

NR1

NR2B

Glutamate

3) Complex interaction of glutamate, at pathological excess millimolar concentration range, with normal activated T cells

NMDA iGluR

GluR2

2) Beneficial interaction of glutamate, at physiological mid micromolar concentration range, with normal activated T cells

GluR1,4

1) Beneficial interaction of glutamate, at physiological low nanomolar concentration range, with normal resting T cells

AMPA mGluR1 mGluR5 iGluR

Induction of IFNγ

TNFα

IL-10

IL-2

1) Inhibition of Proliferation 2) IL-6

Fig. 2 The direct effects of glutamate on human T cell function. Naı¨ve/resting and activated normal human T cells express specific receptors for glutamate (the GluRs) of various types and subtypes, which belong to the two large families of GluRs: the iGluRs and mGluRs. These GluRs were demonstrated by different methods (i.e., Western blotting, flow cytometry and microscopy) to be expressed on the cell surface of T cells, and many studies show that they are functional receptors. Naı¨ve/resting T cells express the GluR3 subunit of AMPA iGluRs, which probably combines with the GluR2 subunit, found in native AMPA iGluRs, to form a heterotetramer. Whether homotetramers containing only GluR3 are also expressed in human T cells is unknown. Naı¨ve/resting T cells express also heteromeric NMDA iGluR containing the NR1 and NR2B subunits, as well as group I mGluR5. Following TCR activation, T cells lose GluR3 from the cell surface, due to an autocrine/paracrine enzymatic cleavage of GluR3 by granzyme B that is secreted by these T cells. Activated T cells probably express AMPA iGluRs heterotetramers composed of GluR2 and GluR1,4, although such subunit composition still needs to be formally demonstrated. Activated T cells express also NMDA iGluRs heterotetramers composed of NR1 and NR2A,B,D subunits, as well as both group I mGluR5 and 1. Glutamate affects normal human T cells in a direct and potent manner that is highly dependent on: glutamate concentration, the activation state of the T cells, and the GluRs they express. Therefore, glutamate can both activate and suppress various T cell features and functions in different contexts. At low physiological concentration, glutamate activates many key T cell

functions and thereby is highly beneficial for the T cells. The figure shows three different conditions and outcomes for the direct interactions of glutamate with T cells: Condition 1: Glutamate at low physiological concentrations, in the nanomolar range, acts via AMPA iGluRs containing GluR3 expressed by these cells, to increase many important T cell functions: T cell adhesion to ECM glycoproteins, among them fibronectin and laminin, and chemotactic migration to CXCL12/SDF-1 via its CXCR4 chemokine receptor. These two glutamate-induced effects are beneficial for the naı¨ve/resting T cells. Such interactions between glutamate and naı¨ve/resting T cells may take place in the normal CNS and in the brain extracellular fluids, and may serve to assist T cells in their survey or exit of the CNS. Condition 2: Glutamate at physiological levels in the micromolar range, acts via both iGluRs and mGluRs expressed on activated T cells, to increase their inward Ca2? currents and inhibit their apoptotic AICD, respectively. These two glutamate-induced effects are beneficial for the activated T cells. Such interactions may take place in blood under normal physiological conditions, and may be part of the normal process promoting T cell proliferation following antigen binding. Condition 3: Glutamate at excess levels in the millimolar range, such as those present in blood and/or CNS in a variety of pathological conditions, may act via mGluR5 to decrease proliferation, and via mGluR1 and NMDA iGluRs to increase cytokine secretion of activated T cells. Such interactions may control the expansion of the activated cells and help combating the disease

sometimes has very different specific effects on different target cells and in different physiological and pathological contexts; Second, there are very potent and complex

interactions between different cytokines, so that a specific cytokine often up-regulates substantially the secretion of another cytokine, or rather suppress it; Third, specific

123


cytokines can of course affect both ‘good and normal T cells’–T cells responding to infectious organisms or cancer, and ‘bad T cells’–autoimmune T cells and cancerous T-leukemia and T-lymphoma cells, as discussed herein later, in parts 5 and 6. 3.5.2.2 The effects of glutamate on T cell cytokine secretion Several studies provided evidence that glutamate can affect cytokine secretion by T cells. At very high nonphysiological concentration of 10-3M, glutamate increased IFNc and IL-10 secretion by CD3-activated T cells. Strangely, at an even higher concentration of 5 9 10-3M, glutamate had an opposite effect, and decreased IFNc, IL10 and IL-5 secretion by these T cells. NMDA at 5 9 10-4M also suppressed IFNc secretion by IL-2-activated T cells (Mashkina et al. 2007), showing that stimulation of the NMDA iGluRs in these activated T cells by excess glutamate can lead to IFNc inhibition. Yet, since (as mentioned already before) glutamate concentrations of 1–5 9 10-3M are much higher than glutamate’s concentrations in the plasma and in the brain’s extracellular fluid, and present only in neurological synapses, we believe these glutamate-induced effects on IFNc, IL-10 and IL-5 are irrelevant to physiological situations. Once again, glutamate had very different effects on naı¨ve/ resting vs. activated T cells, since it did not affect the cytokine secretion of naı¨ve/resting cells (Lombardi et al. 2004). Interestingly, and in relevance to T cells in vivo under physiological conditions, glutamate at 1,000 fold lower concentration of 10-6M (alike the normal glutamate concentration in the brain’s extracellular fluid) may operate via mGluRs to modulate IL-6 production and enhance the secretion of TNFa, IFNc, IL-2 and IL-10. Thus, glutamate released by antigen-pulsed DCs acted on mGluR5 expressed in T cells, in the context of a DC-T cell co-culture, to impair early IL-6 production. At later time points, when antigen-pulsed DCs induced T cell activation and expression of mGluR1, glutamate operated via mGluR1 to counteract the suppressive effect on IL-6 production, and also enhanced the secretion of IFNc, IL-10, TNFa and IL-2 (Pacheco et al. 2006). Of note, although the levels of glutamate released by DCs to the co-culture media were estimated at the 10-6M range, the actual concentration of glutamate within the DC-T cell synapse might in fact be much higher. Collectively, these studies show clearly that glutamate has a dual ability either to increase or to decrease the secretion of several T cell cytokines, and that the exact glutamate-induced effect once again depends on the specific context, which is composed of many factors: glutamate’s concentration, the specific GluRs involved, the activation state of the T cells being exposed to glutamate, the specific T cell subtype, the specific cytokine involved,

and whether or not the T cells are exposed to other stimuli besides glutamate at the same time. This explains why it is so difficult to predict the specific effect of glutamate on a certain cytokine without knowing all these factors, and why further studies are of course needed on this topic, to unveil what exactly happens at each situation. Having said that, in the next part we will try to combine and summarize what we have learned thus far, and draw few suggested guidelines and predictive clues to make the story clearer. 3.6 Proposed summary of the known glutamateinduced effects on resting and activated normal human T cells Based on the multitude of studies described in this review on the effects of glutamate on human T cells, we propose a model for the different dialogs and their functional outcomes, between three ‘players’: glutamate, GluRs, and normal human T cells that express functional GluRs on their cell surface. Each of these ‘players’ can present in different versions/modalities: glutamate can be present at different concentrations; the GluRs can be of different iGluRs or mGluR types and subtypes; and the T cells can be either in a naı¨ve/resting or activated state, and also of different types and subtypes, e.g., CD4? T helper cells– Th1, Th2 or Th17 and others, CD8? cytotoxic T cells, regulatory T cells, memory T cells, natural killer T cells and others. Also, the T cells can be either normal/healthy T cells, or rather pathological T cells–autoimmune T cells or cancerous T-leukemia/lymphoma cells. The effects of glutamate on the later pathological T cells will be discussed in parts 4 and 5. The proposed model, shown in Fig. 2, was drawn along three different conditions that lead to three different outcomes. And on top of this model and the data discussed in this review on which this model is based, the reader is encouraged to read an interesting review describing a role for glutamate as a key immunomodulator in the initiation and development of T cell-mediated immunity (Pacheco et al. 2007). 3.6.1 Condition 1: naı¨ve/resting T cells encountering physiological low micromolar–low nanomolar levels of glutamate This situation and its outcomes are illustrated schematically in Fig. 2, condition 1. Under normal physiological conditions, naı¨ve/resting human T cells patrol the body, migrate in and out of the lymphoid organs, and also enter various other body organs. The T cells remain in their resting condition until they recognize, via their TCR and in the context of appropriate MHC molecules, the specific antigen they are directed to, and become activated. But

123

Y. Ganor, M. Levite

even when this robust (yet relatively rare) event of TCR activation takes place, the activation state of the T cells is transient, lasting some 48–72 h, after which the T cells revert gradually to their naı¨ve/resting condition. Thus, in their ‘everyday life’, T cells are mostly naı¨ve/resting, and in this condition they express high levels of the glutamate/ AMPA GluR3 (Ganor et al. 2003), but not mGluR1 which only activated T cells express. If some of the naı¨ve/resting T cells manage to cross the BBB and enter the CNS (since it is mainly the activated T cells that do so), or if activated T cells revert back to a naı¨ve/resting state within the CNS itself, they could be affected by glutamate at low levels present in the healthy brain fluids. In such situations, glutamate at this low micromolar–low nanomolar concentration range could stimulate its AMPA iGluRs expressed on these naı¨ve/ resting T cells, and by doing so: (a) induce T cell adhesion to fibronectin and laminin (possibly by opening the Kv1.3 channels), (b) induce T cell chemotactic migration towards key chemokines present in the CNS. These two glutamateinduced effects could assist T cells in their patrol and survey of the CNS, and maybe also in their exit back to the periphery, if the T cells indeed do so. As the levels of glutamate in blood are higher under normal physiological conditions (at the micromolar range), the interactions between glutamate and naı¨ve/resting T cells in the periphery may not necessarily be productive. In other words, naı¨ve/resting T cells may be ‘blind’ and unresponsive to glutamate in the blood, since glutamate is present in a too high concentration to activate its receptors expressed on these T cells. 3.6.2 Condition 2: activated T cells encountering physiological mid micromolar levels of glutamate This situation and its outcomes are illustrated schematically in Fig. 2, condition 2. According to the findings revealed thus far by several groups, activated normal human T cells express on their cell surface: (1) NMDA iGluR subunits NR1 and NR2A/B/D; (2) AMPA iGluRs that do not contain the GluR3 subunit; (3) mGluR5 and mGluR1. The expression of these GluRs permits these activated T cells to respond to the physiological levels of glutamate at a mid micromolar concentration range of 10-6 to 10-4M present in blood. Such interactions lead to two effects: (a) an increase in iCa2? via iGluRs, (b) a decrease in apoptosis via mGluRs. These two different glutamateinduced effects can be very important for the survival and function of the activated T cells, since they may contribute to a longer life time of circulating T cells after they become activated by a specific antigen originating from the infectious organism or cancer that they ought to eradicate.

123

3.6.3 Condition 3: activated T cells encountering very high and pathological millimolar levels of glutamate This situation and its outcomes are illustrated schematically in Fig. 2, condition 3. Under few pathological conditions, and only in those, glutamate concentration in the plasma and in the brain’s extracellular fluid increases to the 10-3M range. If/when activated T cells that circulate in the periphery or in the CNS, in the context of some infection or cancer (that caused their activation), encounter such very high and non-physiological 10-3M levels of glutamate, their mGluRs and iGluRs may become activated by glutamate. This may lead to decreased T cell proliferation, and to increased production of some cytokines, respectively. These glutamate effects under pathological conditions could be important for preventing or controlling the expansion of the activated cells on the one hand, and for improving their activity against infectious organisms and cancer on the other.

4 Glutamate receptors and glutamate-induced effects on CANCER human T cells 4.1 Background: GluR antagonists block the growth of solid non-immune tumors Early studies reported on elevated plasma glutamate levels in patients with various types of cancer (Droge et al. 1988; Eck et al. 1989a; Ollenschlager et al. 1989). Many studies have also shown that GluRs are expressed in a variety of neuronal and non-neuronal cancer cell lines and tumors (Kalariti et al. 2005; Stepulak et al. 2009). On top of all that, many observations suggest that glutamate promotes tumor growth, since antagonists to both iGluRs and mGluRs were shown to block the growth of various tumors. In their review, Nicoletti et al. (2007) discuss the ability of mGluR3 and mGluR4 to control the proliferation of brain tumor cells, while mGluR1 have been implicated in the development of melanomas. Additional evidences for the ability of glutamate to promote tumor growth, or for the ability of its receptor antagonists to block it, are the following: (1) blocking Ca2?-permeable AMPA iGluRs in human glioblastoma cells inhibited their locomotion and induced their apoptosis (Ishiuchi et al. 2002); (2) treatment of gliomas with the NMDA iGluR antagonists MK801 or memantine slowed tumor growth (Braun et al. 2010; Takano et al. 2001); (3) the NMDA iGluR antagonists dizocilpine and ketamine, and the AMPA iGluR antagonist GYKI52466, exerted an anti-proliferative effect on colon adenocarcinoma, astro-


cytoma, neuroblastoma, breast carcinoma and lung carcinoma, as a result from decreased cell division and increased cell death (Rzeski et al. 2001; Stepulak et al. 2005); (4) blocking group II mGluRs reduced the growth of glioma cells in vivo (Arcella et al. 2005). Together, all the above observations suggest an important role for glutamate and it various receptors in tumor biology by promoting tumor growth, and suggest that iGluRs and mGluRs can be attractive therapeutic targets for several malignant tumors, including those of the brain (Nicoletti et al. 2007). 4.2 Cancer human T cells express iGluRs and mGluRs Various cancerous T cells express GluRs, as shown in Table 1. In our own studies on the expression of GluRs in human cancer T cells, and on the effects of glutamate on these cells, we found that human T-leukemia (Jurkat) and T-lymphoma (HuT-78) cell lines express on their cell surface high levels a specific iGluR: the AMPA GluR3, which is the very same iGluR that we found in high levels in normal human resting T cells (Ganor et al. 2003, 2009). Studies of other groups have shown that human T-leukemia cells (Jurkat) express also many other types of iGluRs: the AMPA GluR2 and GluR4 (Stepulak et al. 2009); the KA GluR6, GluR7, KA1 and KA2 (Braun et al. 2010; Stepulak et al. 2009); and the NMDA NR1, NR2A-D and NR3A,B (Miglio et al. 2005b, 2007; Stepulak et al. 2009). Several cancer human T cell lines also express group I mGluRs: T-leukemia (Jurkat, FRO) and T-lymphoma (H9, HuT-78) express mGluR5; T-leukemia (Jurkat, FRO, SUPT1) and T-lymphoma (HuT-78) express mGluR1 (Chiocchetti et al. 2006; Pacheco et al. 2004). Another study confirmed the expression of group I mGluRs in T-leukemia (Jurkat) cells, and further demonstrated that these cells also express group I and II mGluRs (Stepulak et al. 2009). 4.3 The direct effects of glutamate at different concentrations on cancer human T cells, via iGluRs and mGluRs 4.3.1 Glutamate at physiological low nanomolar concentrations promotes in vivo spread of human T cell leukemia and T cell lymphoma, via AMPA iGluRs Our own studies revealed that glutamate promotes various cancerous T cells functions via AMPA GluR3 expressed on the cell surface of these cells. One of glutamate-induced effects on such cancerous T cells was to increase their in vivo migration and extravasation into solid organs

(Ganor et al. 2009). Thus, ex vivo exposure of Jurkat human T-leukemia cells to glutamate at low and physiological concentration of 10-8M enhanced significantly their subsequent in vivo engraftment into the liver and chorioallantoic membrane of a chick embryo (Ganor et al. 2009). Importantly, and in correlation with this pro-metastatic in vivo effect, glutamate also induced in vitro a significant elevation of the cancer-associated matrix metalloproteinase (MMP) inducer CD147, and also increased the secretion of the cancer-associated MMP-9, in both Jurkat T-leukemia and HuT-78 T-lymphoma cancer cells (Ganor et al. 2009). These glutamate-induced effects were mediated by AMPA iGluRs, since they were mimicked by AMPA, the selective AMPA iGluR agonist, and blocked by CNQX, the selective AMPA iGluR antagonist (Ganor et al. 2009). These findings show that glutamate may facilitate the metastatic spread of T cell leukemia and T cell lymphoma in vivo, and their penetration from the circulation into solid organs. Moreover, glutamate induced its pro-metastatic effect at a low physiological concentration of 10-8M, such as that found primarily in the CSF and in the brain extracellular fluids (Meldrum 2000). Based on this finding we postulate that the increased engraftment of T cell cancers induced by glutamate would probably reach its optimum within the CNS (Fig. 3). 4.3.2 Glutamate at physiological mid micromolar concentrations increases adhesion of human T cell leukemia, via NMDA iGluRs Glutamate by itself induced adhesion of human T cell leukemia line (Jurkat) to fibronectin. This effect was mimicked by NMDA, and blocked by the NMDA iGluRs antagonists (?)-MK801 and D-AP5 (Miglio et al. 2007). The pro-adhesive effect induced by glutamate and NMDA showed a bell-shape concentration-dependent relationship: it was effective at a broad concentration range of 10-8 to 10-5M, and reached a maximum at 10-6M glutamate and at 10-5M NMDA. As we previously found that glutamate on its own, at a similar concentration range, induces adhesion of normal (i.e., non-cancer) resting human T cells (Ganor et al. 2003), it seems that glutamate can activate its iGluR in T cells regardless of whether they are normal/healthy T cells, or rather cancerous T cells, and by doing so lead to the activation of their specific adhesion receptors, and to their subsequent adhesion to the respective glycoproteins of the ECM. Having said that, it is important to keep in mind the different consequences of glutamate-induced effects on human T cells on the overall well being: when the T cells are normal/healthy cells, glutamate-induced effects on

123

Y. Ganor, M. Levite Glutamate interacts directly with human cancer T cells: T-leukemia and T-lymphoma, and with human autoimmune T cells of patients with Multiple Sclerosis, and induces many key functions for the benefit of the cells

AMPA iGluR

NMDA iGluR

Glutamate at physiological low-mid nanomolar-micromolar

GluR3

NR2B

Glutamate at pathological excess millimolar

NR1

Glutamate at physiological mid micromolar

GluR2

GluR3

Glutamate at physiological low nanomolar

2) Autoimmune T cells in Multiple Sclerosis

mGluR1

mGluR5

GluR2

1) Cancer T cells: T cell leukemia / lymphoma

AMPA iGluR

Augmentation of:

Induction of:

Induction of:

Induction of:

1) Extravasation into

1) Adhesion to

1) Inward Ca2+

1) Chemotactic migration to CXCL12/SDF-1

solid organs

fibronectin

2) c-jun and c-fos

2) Proliferation in response to

2) CD147 production

2) Adhesion to

gene expression

myelin-derived proteins

3) MMP9 production

laminin

Fig. 3 Glutamate and its receptors may play an important role in T cell-mediated pathologies. (1) Glutamate, at physiological nanomolar range, such as that found primarily in the CSF and brain extracellular fluids, activates GluR3-containing AMPA iGluRs expressed in few types of human cancer T cells: T leukemia and lymphoma. Glutamate-induced effects on the human T leukemia and lymphoma cells are for the benefit of these detrimental cancer T cells, as it is for normal non-cancer T cells, since glutamate triggers or elevates several important T cell features. Glutamate’s effects on human T leukemia and lymphoma known thus far include (but are not restricted to) the following: (a) increase of in vivo extravasation into solid organs; (b) increase in the expression of MMP-9, a very important cancer-associated matrix metalloproteinase; (c) increase in the expression level of CD147–the MM-9 inducer that is also a cancerrelated molecule that promotes cancer spread; (d) increase of the adhesion of cancer T cells (an effect induced by glutamate at its physiological micromolar concentration range, and via NMDA

iGluRs). Finally, glutamate at its pathological excess millimolar range, acts via group I mGluR1,5 to increase inward Ca2? flow and expression of early Ca2? inducible genes. (2) Glutamate and AMPA iGluRs seem to play an important role in multiple sclerosis (MS), and glutamate seems to improve significantly the function of the autoimmune encephalitogenic T cells, based on all the observations discussed in the text, in part 5.1 and in the cited papers. In brief, the main observations are: (a) the iGluR AMPA GluR3 is expressed in T cells of MS patients, and its expression is upregulated during relapse and in patients with neurological evidence of disease activity; (b) glutamate or AMPA enhance the proliferation of the autoreactive T cells in response to myelin-derived proteins; (c) glutamate or AMPA augment the chemotactic migration of the MS autoreactive T cells; (d) there is an abnormal response to glutamate of T cells from MS patients; (e) iGluR AMPA GluR3 is highly expressed not only in normal human T cells but also in mouse encephalitogenic EAEinducing anti-myelin basic protein T cell clones (Ganor et al. 2003)

them are most probably beneficial for these T cells by improving their adhesion, migration, survival and function, and such glutamate-induced effects are also beneficial for health. Yet, when the T cells are cancerous or autoimmune, glutamate-induced effects may still be beneficial for these T-leukemia and T-lymphoma cells and improve their migration, penetration into organs and function, but in such cases these effects are detrimental for health and promote the T cell cancer or the T cell mediated autoimmune disease (Fig. 3).

4.3.3 Glutamate at pathological high millimolar concentrations promotes growth of human T cell leukemia and T cell lymphoma, via NMDA iGluRs

123

There are indirect evidences showing that GluR antagonists inhibit the growth of cancer T cells, which suggest that glutamate itself does the opposite to what its antagonists do, and in fact promotes the growth of these cells. The specific indirect evidences supporting this conclusion are the following: Miglio et al. (2007) showed that the NMDA


iGluR antagonists (?)-MK801 and D-AP5, at concentrations of 1–5 9 10-4M, inhibited the growth of Jurkat T-leukemia cells by promoting their apoptosis. These results were confirmed by a later study, showing that the noncompetitive NMDA iGluR antagonist Ketamine in a very high millimolar concentrations induced apoptosis of Jurkat T-leukemia cells, via the mitochondrial pathway, and independent of death receptor signaling (Braun et al. 2010). These observations, supporting the ability of glutamate to promote growth of cancer T cells, and its selective GluR antagonists to block it, are in line with the findings demonstrating that GluR antagonists limit growth of several types of non-immune tumors, and with the multitude of early studies showing clearly that glutamate levels increase in the plasma of patients with malignancies (Droge et al. 1988; Eck et al. 1989a; Ollenschlager et al. 1989). 4.3.4 Glutamate at mid micromolar–high millimolar concentration increases the intracellular Ca2? (iCa2?) concentration in human T cell leukemia, via group I mGluRs Miglio et al. (2005a) showed that stimulation of the group I mGluRs, mGluR5 and mGluR1, in cancer human T cell leukemia line (Jurkat) evoked calcium signals and c-jun and c-fos gene expression. The corresponding experiments showed that at a concentration range of 10-5 to 10-3M, the prototype mGluR agonist (1S,3R)-ACPD, the non-selective mGluR1/5 agonist (S)-3,5-DHPG, and the selective mGluR5 agonist CHPG increased iCa2?. Moreover, several selective group I mGluR antagonists (i.e., AIDA, LY367385, MPEP) antagonized the effect of DHPG (Miglio et al. 2005a). Both the extracellular Ca2? and the Ca2? released from intracellular stores contribute to the iCa2? increase. The reported rise in iCa2? resulted also in an upregulation of the early Ca2?-inducible genes c-fos and cjun, thus activating multiple downstream signaling regulating important T cell functions (Miglio et al. 2005a). As the protein products of these genes play important roles in the regulation of the cell cycle (Tay et al. 1996), these findings suggest that glutamate promotes T cell leukemia, like it does for normal human T cells, by activating multiple downstream signaling events that are involved in cell proliferation and cytokine mRNA transcription (Fig. 3).

5 Glutamate receptors and glutamate-induced effects on AUTOIMMUNE human T cells 5.1 Multiple sclerosis (MS) MS and its animal model–experimental autoimmune encephalomyelitis (EAE)–are demyelinating autoimmune

diseases mediated by autoreactive T cells that enter the CNS and attack the nerve enwrapping myelin sheath. In addition, myelin-producing cells in the CNS and some axons are lost as a result of the autoimmune and inflammatory attack on the CNS. Interestingly, two previous studies reported that when mice (Pitt et al. 2000) or rats (Smith et al. 2000) sensitized for EAE were treated with the AMPA iGluR antagonist NBQX, the result was a substantial amelioration of disease, increased oligodendrocyte survival and reduced dephosphorylation of neurofilament H, an indicator of axonal damage. Despite the clinical differences, treatment with NBQX had no effect on lesion size, and did not reduce the degree of CNS inflammation (Pitt et al. 2000). In addition, NBQX did not alter the proliferative activity of antigenprimed T cells in vitro (Pitt et al. 2000). The researchers concluded that NBQX ameliorated EAE by blocking the AMPA iGluRs expressed in neuronal or glial cells, and by doing so protected them from the excitotoxic effects of excess glutamate. In addition to this interpretation and conclusion, we proposed in our own paper on glutamate effects on T cells (Ganor et al. 2003) an additional mechanism of action by which AMPA antagonists can block EAE/MS. We suggested that NBQX suppressed EAE in the studies performed by Pitt et al. and Smith et al. because it blocked AMPA iGluRs expressed in the autoaggressive encephalitogenic T cells, not only by blocking such receptors in neurons. Hence, we believe that by blocking the T cellexpressed AMPA iGluRs, NBQX could have prevented the activation of the autoaggressive T cells by glutamate originating from nerve endings at the sites of inflammation/ damage in the CNS, thereby reducing their pathogenic EAE-inducing activity. Our proposal is based on: (1) our demonstration that the iGluR AMPA GluR3 is highly expressed not only in normal human T cells but also in mouse encephalitogenic EAE-inducing anti-myelin basic protein T cell clones (Ganor et al. 2003); (2) the evidences discussed in this review, showing that glutamate at lowmid concentrations has several beneficial effects on activated T cells, such as an increase in iCa2? and protection from apoptotic AICD. One can thus envision that the very same GluR antagonist that under normal conditions would have negative consequences on immune functions, since it would prevent the beneficial effects of glutamate on normal activated T cells, would have the opposite consequences in the context of a T cell-mediated autoimmune disease such as MS, as in the latter case the activated T cells are the detrimental autoimmune and autoaggressive T cells. In line and confirmation with the above suggestions, a later very interesting and important study by Sarchielli et al. (2007) demonstrated that iGluR AMPA GluR3 is expressed in T cells of MS patients (Table 1), and that

123

Y. Ganor, M. Levite

GluR3 expression is upregulated in T cells of MS patients during relapse, and in patients with neurological evidence of disease activity. Another very interesting finding made in this study was that both glutamate and AMPA (10-8 to 10-5M) enhanced the proliferation of the autoreactive T cells in response to myelin-derived proteins: MBP and MOG, and also augmented the chemotactic migration towards CXCL12/SDF-1 of T cells derived from both MS patients and control individuals (Sarchielli et al. 2007). In the MS patients, significantly higher proliferation values in response to glutamate were found in patients assessed during relapse and in those with gadolinium (Gd)? enhancing lesions on MRI (Sarchielli et al. 2007). In contrast to these glutamate-induced effects exerted in a concentration range of 10-8 to 10-5M, higher glutamate concentrations above 10-5M appeared to inhibit the MBP and MOG-specific T cell proliferation, as well as the chemotactic response of the T cells of both MS patients and control individuals. Sarchielli et al. (2007) concluded that the higher GluR3 expression and the higher activating effect of glutamate on T cells of MS patients during relapse and with evidence of disease activity on MRI, suggest the involvement of glutamate-mediated mechanisms in the T-cell detrimental autoimmune effects. In MS patients, glutamate within physiological ranges in the CSF and brain extracellular space might enhance myelin antigen-specific proliferation and chemotactic migration via activation of AMPA receptors, which can be relevant for myelin and neuronal damage in MS. Excess glutamate levels seem to induce an inhibitory effect on lymphocyte function, and therefore the detrimental effect of glutamate in this case could be attributed to a direct toxicity on glial and neuronal cells. Another study also reported that the inhibition of PHAinduced cell proliferation caused by glutamate at very high 10-3M concentration (see above) was lower in T cells of MS patients (Lombardi et al. 2003). Together, these findings support the idea that glutamate may upregulate the function of encephalitogenic autoimmune T cells in MS, and that GluR3 plays a key role in mediating this effect. This pathological scenario is drawn schematically in Fig. 3. If glutamate-induced activation of MS-associated T cells indeed plays a vital role in the disease, blocking glutamate signaling in MS may turn out clinically beneficial. 5.2 Rheumatoid arthritis (RA) RA is a chronic disease characterized by severe joint inflammation and progressive destruction. Although the cause of RA is to a large extent still unknown, autoimmunity plays a pivotal role in both its chronicity and progression, and RA is considered a systemic autoimmune disease.

123

T cells of several types play a major role in RA, and the reader is referred to numerous publications on the topic, for example Weyand et al. (2000), Miossec and Kolls (2012), Carvalheiro et al. (2013) and Maddur et al. (2012). The T cells in RA infiltrate into the synovial membrane, where they initiate and maintain activation of macrophages and synovial fibroblasts, transforming them into tissuedestructive effector cells (Weyand et al. 2000). A recent review discusses the possible involvement of glutamatergic signaling machineries in the pathophysiology of RA (Hinoi and Yoneda 2011). Interestingly, in collagen-induced arthritis (CIA), an animal model of RA: (a) glutamate stimulated proliferation of synovial fibroblasts that contribute to bone and cartilage destruction (Hinoi and Yoneda 2011); (b) there was a drastic increase in the endogenous levels of glutamate in the synovial fluid; (c) treatment with the NMDA iGluR antagonist Memantine significantly improved the course of CIA and upregulated the expression of Foxp3 in spleen CD4? T cells, which was followed by an increase in CD4?CD25? regulatory T cells, which decrease disease severity and prevent its progression (Lindblad et al. 2011). In addition, Boettger et al. (2010) showed that intrathecally applied ketamine, a non-competitive NMDA iGluR antagonist into animals with chronic antigen-induced arthritis (AIA) significantly reduced arthritis severity as indicated by reduced joint swelling, but even more intriguingly by reduced infiltration with inflammatory cells and joint destruction in the acute and the chronic phase of arthritis. While these observations clearly suggest an active role of glutamate and GluRs in the pathophysiology of RA, further studies are needed to extend these findings to human RA patients, and reveal the direct in vivo effects of glutamate on the autoimmune T cells of RA patients. 5.3 Systemic lupus erythematosus (SLE) SLE is a systemic connective-tissue autoimmune disease that can affect any part of the body. SLE most often harms the heart, joints, skin, lungs, blood vessels, liver, kidneys, and nervous system. As occurs in other autoimmune diseases, the immune system attacks the body’s cells and tissues, resulting in inflammation and tissue damage. It is a type III hypersensitivity reaction caused by antibodyimmune complex formation. Yet, T cell abnormalities, among them aberrant T helper cytokine profiles, have been implicated in the loss of immune tolerance to nuclear and cytoplasmic antigens in SLE, and were linked to a variety of clinical manifestations of the disease (La Cava 2009; Zlotnik et al. 2012). Recent reviews discusses the role of T cells in promoting and maintaining SLE in relation to their cellular and molecular abnormalities, and provide an


update on recent T cell-targeted therapeutic approaches for the restoration of T cell homeostasis in this disease (La Cava 2009; Zlotnik et al. 2012). As to glutamate and T cells in SLE: in a recent study, Poulopoulou et al. (2008) measured serum glutamate concentrations and Kv1.3 channel activity in patients with SLE. Their logic for examining these two features is based on the following previously published independent findings: (1) alterations in glutamate homeostasis on the one hand, and in Kv1.3 voltage-gated potassium channel function on the other, have been independently associated with T cell dysfunction; (2) selective blockade of Kv1.3 channels leads to inhibition of T cell activation and to improvement of T cell-mediated manifestations in animal models of autoimmunity; (3) opening of the Kv1.3 channel in normal human T cells leads to activation of the T cell’s b1 integrins and to the subsequent adhesion of the T cells to the ECM (Levite et al. 2000); (4) low extracellular glutamate concentrations enhance ex vivo the activity of the Kv1.3 channel in normal T cells. In the study of Poulopoulou et al. the authors used high-performance liquid chromatography for glutamate measurements, and used the whole-cell patch-clamp technique for electrophysiologic studies performed in freshly isolated, non-cultured peripheral T cells. The researchers found that the mean ± SD serum concentrations of glutamate were significantly lower in patients with either clinically quiescent SLE (77 ± 27 microM [n = 18]) or active SLE (61 ± 36 microM [n = 16]) than in healthy controls (166 ± 64 microM [n = 24]) (both P \ 0.0001). The intrinsic gating properties of the Kv1.3 channels in SLE patient-derived T cells were found to be comparable with those in healthy control-derived T cells. Yet, interestingly, the electrophysiologic data from the SLE patient-derived T cells exposed to low extracellular glutamate concentrations similar to their respective serum levels (50 microM) demonstrated Kv1.3 current responses that were enhanced by almost 20 % (P \ 0.01) compared with those subsequently obtained from the same T cells in the presence of glutamate concentrations within control serum levels (200 microM). The researchers concluded that SLE patients have lower glutamate levels in their serum than healthy individuals, and that such lower glutamate levels enhance the functional activity of the Kv1.3 channel in T cells of SLE patients. Based on these observations, and also on the previously published evidences for the key role of Kv1.3 channel activity in T cell activity, the authors suggested that the low glutamate levels in the serum of SLE patients enhance the in vivo functional activity of the Kv1.3 channel of the patient’s T cells and by doing so contribute to SLE T cell hyperactivity (Poulopoulou et al. 2008). The researchers also suggested that further elucidation of glutamate levels and Kv1.3 responses in SLE, as well as the possible pathogenic role of this unsuspected metabolic

abnormality, may have therapeutic implications for SLE patients (Poulopoulou et al. 2008). Based on the interesting observations and suggestions raised in the study of Poulopoulou et al., and until the findings of this single study will hopefully be confirmed and extended in subsequent studies, we ourselves wish to raise here several questions for further thinking: (a) what could be the reason for the abnormally low levels of glutamate in the serum of SLE patients? which glutamateproducing cells are the ones that release it in abnormally lower levels–the T cells themselves that can produce glutamate (as shown in (Garg et al. 2008; Melzer et al. 2013) and discussed below) and/or other cells?; (b) could the lower levels of glutamate in the serum of SLE patients induce/activate functions of their resting yet autoimmuneassociated T cells, which higher glutamate levels normally do not, such as adhesion, migration and iCa2? fluxes and also extend the survival of their activated T cells by protecting them from apoptotic AICD (Chiocchetti et al. 2006)?; (c) could all the effects induced by the abnormally low glutamate levels in the serum genuinely contribute to the overall pathology of SLE?; (d) could administration of GluR antagonists be of benefit in SLE by virtue of blocking glutamate-induced effects on the autoimmune-associated T cells? We hope that in the coming years, answers will be found to these clinically-relevant questions.

6 T cells release glutamate which has potent effects on neuronal cells; DCs interacting with T cells release glutamate which has potent effects on the T cells Several studies show that T cells can release low levels of glutamate, which can then affect other cells in the nervous system. In addition, when DCs interact with T cells, they release high amounts of glutamate, which in turn affect the T cells in a potent manner. Thus, DCs seem to use their own glutamate to affect the T cell they interact with. In the below paragraphs we summarize the main studies showing glutamate release by T cells and by DCs and its functional consequences. (1) Garg et al. (2008) showed that during culture of murine primary T cells that were purified from lymph nodes and spleens, high levels of glutamate accumulate: 463 ± 136 lM after 24 h of culture. Interestingly, the glutamate released by the T cells is efficiently cleared if T cells are co-cultured with astrocytes. The T cell-derived glutamate elicits in turn the release of neuroprotective thiols (cysteine, glutathione, and cysteinyl-glycine) and lactate from astrocytes. This study also revealed that the T cells improve glutamate clearance capacity of astrocytes under normoxic and toxic conditions (Garg et al. 2008).

123

Y. Ganor, M. Levite

(2) Melzer et al. (2013) recently showed that CD8? T cells release glutamate by a TCR-stimulus-dependent mechanism, and that the T cell-derived glutamate has potent effects on neurons. In further detail, this study showed that following TCR stimulation, mouse CD8? T cells acquire the molecular repertoire for vesicular glutamate release: (i) they upregulate expression of glutaminase required to generate glutamate via deamination of glutamine, (ii) they upregulate expression of vesicular protonATPase and vesicular glutamate transporters required for filling of vesicles with glutamate. Subsequently, CD8? T cells release glutamate in a strictly stimulus-dependent manner. Supernatant glutamate concentrations increased exponentially with time after stimulation by anti-CD3/ CD28 antibodies, and reached levels of about 0.5 mM after 72 h of stimulation. Upon repetitive TCR stimulation, CD25high CD8? T effector cells exhibited higher estimated single cell glutamate release rates compared to CD25low CD8? T memory cells. Moreover, and importantly, glutamate liberation by oligodendrocyte-reactive CD25high CD8? T effector cells was capable of eliciting collateral excitotoxic cell death of neurons (despite glutamate re-uptake by glia cells and neurons) in intact CNS gray matter. Based on these observations, the authors concluded that glutamate release may represent a crucial effector pathway of neural-antigen reactive CD8? T cells, contributing to excitotoxicity in CNS inflammation. We hope that subsequent studies on glutamate release by human T cells will clarify if all these occur in humans as well, and what are the physiological and pathological, autocrine and paracrine, immune and neuronal consequences of the T cell derived glutamate. (3) Pacheco et al., who studied human DCs and T cells purified from PBMC of healthy individuals (Pacheco et al. 2006), report on: (a) what they consider only ‘‘background levels’’ of glutamate release by T cells, at the range of 10 lM (see Fig. 2C in their paper), (b) the release of low levels of glutamate by immature DCs, and (c) much higher glutamate levels released by activated mature DCs (Pacheco et al. 2006). During the maturation process, the DCs release physiologically relevant glutamate amounts (2–25 mM per 105 cells), which are at the range of glutamate released by brain macrophages: 6–40 mM per 105 cells, and which are high enough to activate glutamate receptors. Interestingly, Pacheco et al. further showed that while mGluR5 was constitutively expressed in resting T cells, mGluR1 was expressed in the T cells only after 24–48 h of co-culture with superantigen-pulsed DCs. (4) Affaticati et al. (2011) also studied glutamate release by DCs and its effects on T cells, and showed that DCs express glutamate and glutamate-specific vesicular glutamate transporters, and that they are capable of fast glutamate release through a Ca(2?)-dependent mechanism.

123

On top of all these studies showing the release of glutamate by T cells and DCs and its functional outcomes, in our recently published book chapter on ‘‘Glutamate in the immune system’’ (Ganor and Levite 2012) in the ‘NerveDriven Immunity’ book, we summarize and discuss the evidences for the release of glutamate by additional immune cells, and cite specifically the following observations: (a) the release of glutamate by monocytes/macrophages and activated microglia, and the subsequent glutamate-induced neuronal death (Piani et al. 1991, 1992). The macrophages/microglia-derived glutamate leads to neuronal death contributing to neuronal damage in human immunodeficiency virus type 1 infection [for review see (Kaul et al. 2005)]; (b) The release of glutamate by neutrophils, and the subsequent glutamate-induced decrease of endothelial cell permeability (Collard et al. 2002).

7 Summary and concluding remarks: Glutamate is a ‘Neuro-Immuno-Transmitter’ that at physiological concentrations has many direct, potent and activating effects on normal human resting and activated T cells, cancer T cells and autoimmune T cells, and on many other immune cells, all via its various ionotropic and metabotropic GluRs highly expressed in these cells The numerous studies discussed in this review show clearly that glutamate, by binding to its specific receptors–the various types of iGluRs and mGluRs–that are expressed in human T cells, induces many direct, potent and important effects on these T cells. As such, glutamate and its receptors seems to play a key role in T cell mediated immunity under physiological and some pathological conditions. At low physiological concentrations, glutamate clearly activates many key functions of naı¨ve/resting beneficial normal human T cells but also of cancerous T cells (Tleukemia and T-lymphoma) and of autoimmune detrimental T cells. Based on the role of glutamate and its GluRs in T cells, and also on many other immune cells [reviewed in much detail for many different types of immune cells in (Levite 2012)], we suggest to change from now onwards glutamate’s title from a ‘Neurotransmitter’ to ‘Neuro-Immuno-Transmitter’. Glutamate clearly deserves now this novel title of ‘Neuro-Immuno-Transmitter’ since all the criteria we set recently for this title (Ganor and Levite 2012) are met: (1)

Glutamate has specific receptors–both iGluRs and mGluRs–that are highly expressed in normal human T cells (Table 1) and in many other types of immune cells. In addition, GluRs are expressed also in cancer and autoimmune T cells;


(2)

(3)

(4)

Glutamate by itself binds its GluRs expressed in normal human T cells, cancerous human T cells and autoimmune human T cells, and induces or rather suppresses (depending on the context) various key T cell functions (Fig. 2). At low physiological concentrations glutamate clearly activates many key T cell functions in naı¨ve/resting normal human T cells, cancerous T cells and autoimmune T cells, thereby improving substantially their function. Glutamate also affects many other immune cell types (Levite 2012); Glutamate seems to play an active role in various human T cell-mediated diseases (Fig. 3), and in immune diseases mediated by other immune cells and cancers; Glutamate is produced by T cells, DCs and various other types of immune cells.

Together, glutamate unequivocally affects human T cell function and represents a crucial intersection between the immune system and the nervous system. In fact, glutamate is clearly not the only molecule that deserves now to be renamed as ‘Neuro-Immuno-Transmitter’, since the very same new title is valid for adrenaline, noradrenaline, dopamine, acetylcholine, gamma-amino butyric acid (GABA), serotonin, somatostatin, substance P, neuropeptide Y (NPY), vasointestinal peptide (VIP), calcitonin gene-related peptide (CGRP), opioids and cannabinoids. For further details on the ‘immune face’ of each of these neurotransmitters, the reader is referred to our recently published book on this topic, entitled ‘‘Nerve-Driven Immunity–Neurotransmitters and Neuropeptides in the Immune System’’ (Levite 2012). Finally, on the basis of all the data discussed in this review we propose the novel use of either glutamate or the few approved glutamate receptor agonists available as drugs, for the beneficial activation of normal/healthy human T cells in patients with solid cancers or unresolved infectious diseases. In such cases, exposing the patient’s T cells to glutamate or to its agonists could lead to their beneficial activation, improved function, and subsequently improved eradiation of the cancer and the infectious organisms. Likewise, approved glutamate receptor antagonists available as drugs may be found useful for blocking the undesired effects of glutamate on autoimmune T cells and cancerous T-leukemia and T-lymphoma cells. But all these suggestions should of course be tested carefully in each context separately. Conflict of interest The authors have no financial relationship with the organization that sponsored the research.

References Affaticati P, Mignen O, Jambou F, Potier MC, Klingel-Schmitt I, Degrouard J, Peineau S, Gouadon E, Collingridge GL, Liblau R,

Capiod T, Cohen-Kaminsky S (2011) Sustained calcium signalling and caspase-3 activation involve NMDA receptors in thymocytes in contact with dendritic cells. Cell Death Differ 18(1):99–108 Arcella A, Carpinelli G, Battaglia G, D’Onofrio M, Santoro F, Ngomba RT, Bruno V, Casolini P, Giangaspero F, Nicoletti F (2005) Pharmacological blockade of group II metabotropic glutamate receptors reduces the growth of glioma cells in vivo. Neuro Oncol 7(3):236–245 Armstrong N, Sun Y, Chen GQ, Gouaux E (1998) Structure of a glutamate-receptor ligand-binding core in complex with kainate. Nature 395(6705):913–917 Boettger MK, Weber K, Gajda M, Brauer R, Schaible HG (2010) Spinally applied ketamine or morphine attenuate peripheral inflammation and hyperalgesia in acute and chronic phases of experimental arthritis. Brain Behav Immun 24(3):474–485 Bonsi P, Cuomo D, De Persis C, Centonze D, Bernardi G, Calabresi P, Pisani A (2005) Modulatory action of metabotropic glutamate receptor (mGluR) 5 on mGluR1 function in striatal cholinergic interneurons. Neuropharmacology 49(Suppl 1):104–113 Braun S, Gaza N, Werdehausen R, Hermanns H, Bauer I, Durieux ME, Hollmann MW, Stevens MF (2010) Ketamine induces apoptosis via the mitochondrial pathway in human lymphocytes and neuronal cells. Br J Anaesth 105(3):347–354 Cahalan MD, Wulff H, Chandy KG (2001) Molecular properties and physiological roles of ion channels in the immune system. J Clin Immunol 21(4):235–252 Carvalheiro H, da Silva JA, Souto-Carneiro MM (2013) Potential roles for CD8(?) T cells in rheumatoid arthritis. Autoimmun Rev 12(3):401–409 Chatterton JE, Awobuluyi M, Premkumar LS, Takahashi H, Talantova M, Shin Y, Cui J, Tu S, Sevarino KA, Nakanishi N, Tong G, Lipton SA, Zhang D (2002) Excitatory glycine receptors containing the NR3 family of NMDA receptor subunits. Nature 415(6873):793–798 Chiocchetti A, Miglio G, Mesturini R, Varsaldi F, Mocellin M, Orilieri E, Dianzani C, Fantozzi R, Dianzani U, Lombardi G (2006) Group I mGlu receptor stimulation inhibits activationinduced cell death of human T lymphocytes. Br J Pharmacol 148(6):760–768 Chu Z, Hablitz JJ (2000) Quisqualate induces an inward current via mGluR activation in neocortical pyramidal neurons. Brain Res 879(1–2):88–92 Collard CD, Park KA, Montalto MC, Alapati S, Buras JA, Stahl GL, Colgan SP (2002) Neutrophil-derived glutamate regulates vascular endothelial barrier function. J Biol Chem 277(17): 14801–14811 Collingridge GL, Singer W (1990) Excitatory amino acid receptors and synaptic plasticity. Trends Pharmacol Sci 11(7):290–296 Collingridge GL, Olsen RW, Peters J, Spedding M (2009) A nomenclature for ligand-gated ion channels. Neuropharmacology 56(1):2–5 Danbolt NC (2001) Glutamate uptake. Prog Neurobiol 65(1):1–105 Das S, Sasaki YF, Rothe T, Premkumar LS, Takasu M, Crandall JE, Dikkes P, Conner DA, Rayudu PV, Cheung W, Chen HS, Lipton SA, Nakanishi N (1998) Increased NMDA current and spine density in mice lacking the NMDA receptor subunit NR3A. Nature 393(6683):377–381 Dingledine R, Borges K, Bowie D, Traynelis SF (1999) The glutamate receptor ion channels. Pharmacol Rev 51(1):7–61 Divino Filho JC, Hazel SJ, Furst P, Bergstrom J, Hall K (1998) Glutamate concentration in plasma, erythrocyte and muscle in relation to plasma levels of insulin-like growth factor (IGF)-I, IGF binding protein-1 and insulin in patients on haemodialysis. J Endocrinol 156(3):519–527

123

Y. Ganor, M. Levite Droge W, Eck HP, Betzler M, Schlag P, Drings P, Ebert W (1988) Plasma glutamate concentration and lymphocyte activity. J Cancer Res Clin Oncol 114(2):124–128 Droge W, Murthy KK, Stahl-Hennig C, Hartung S, Plesker R, Rouse S, Peterhans E, Kinscherf R, Fischbach T, Eck HP (1993) Plasma amino acid dysregulation after lentiviral infection. AIDS Res Hum Retroviruses 9(9):807–809 Eck HP, Drings P, Droge W (1989a) Plasma glutamate levels, lymphocyte reactivity and death rate in patients with bronchial carcinoma. J Cancer Res Clin Oncol 115(6):571–574 Eck HP, Frey H, Droge W (1989b) Elevated plasma glutamate concentrations in HIV-1-infected patients may contribute to loss of macrophage and lymphocyte functions. Int Immunol 1(4):367–372 Endoh T (2004) Characterization of modulatory effects of postsynaptic metabotropic glutamate receptors on calcium currents in rat nucleus tractus solitarius. Brain Res 1024(1–2):212–224 Ferrarese C, Aliprandi A, Tremolizzo L, Stanzani L, De Micheli A, Dolara A, Frattola L (2001) Increased glutamate in CSF and plasma of patients with HIV dementia. Neurology 57(4): 671–675 Foster AC, Fagg GE (1984) Acidic amino acid binding sites in mammalian neuronal membranes: their characteristics and relationship to synaptic receptors. Brain Res 319(2):103–164 Gahring L, Carlson NG, Meyer EL, Rogers SW (2001) Granzyme B proteolysis of a neuronal glutamate receptor generates an autoantigen and is modulated by glycosylation. J Immunol 166(3):1433–1438 Ganor Y, Levite M (2012) Glutamate in the immune system: glutamate receptors in immune cells, potent effects, endogenous production and involvement in disease. In: Levite M (ed) Nervedriven immunity: neurotransmitters and neuropeptides in the immune system. Springer, Vienna, pp 121–161 Ganor Y, Besser M, Ben-Zakay N, Unger T, Levite M (2003) Human T cells express a functional ionotropic glutamate receptor GluR3, and glutamate by itself triggers integrin-mediated adhesion to laminin and fibronectin and chemotactic migration. J Immunol 170(8):4362–4372 Ganor Y, Teichberg VI, Levite M (2007) TCR activation eliminates glutamate receptor GluR3 from the cell surface of normal human T cells, via an autocrine/paracrine granzyme B-mediated proteolytic cleavage. J Immunol 178(2):683–692 Ganor Y, Grinberg I, Reis A, Cooper I, Goldstein RS, Levite M (2009) Human T-leukemia and T-lymphoma express glutamate receptor AMPA GluR3, and the neurotransmitter glutamate elevates the cancer-related matrix-metalloproteinases inducer CD147/EMMPRIN, MMP-9 secretion and engraftment of T-leukemia in vivo. Leuk Lymphoma 50(6):985–997 Garg SK, Banerjee R, Kipnis J (2008) Neuroprotective immunity: T cell-derived glutamate endows astrocytes with a neuroprotective phenotype. J Immunol 180(6):3866–3873 Gill SS, Pulido OM (2001) Glutamate receptors in peripheral tissues: current knowledge, future research, and implications for toxicology. Toxicol Pathol 29(2):208–223 Graham TE, Sgro V, Friars D, Gibala MJ (2000) Glutamate ingestion: the plasma and muscle free amino acid pools of resting humans. Am J Physiol Endocrinol Metab 278(1):E83–E89 Green DR, Droin N, Pinkoski M (2003) Activation-induced cell death in T cells. Immunol Rev 193:70–81 Guse AH (1998) Ca2? signaling in T-lymphocytes. Crit Rev Immunol 18(5):419–448 Hinoi E, Yoneda Y (2011) Possible involvement of glutamatergic signaling machineries in pathophysiology of rheumatoid arthritis. J Pharmacol Sci 116(3):248–256 Hinoi E, Ogita K, Takeuchi Y, Ohashi H, Maruyama T, Yoneda Y (2001) Characterization with [3H]quisqualate of group I

123

metabotropic glutamate receptor subtype in rat central and peripheral excitable tissues. Neurochem Int 38(3):277–285 Hinoi E, Takarada T, Ueshima T, Tsuchihashi Y, Yoneda Y (2004) Glutamate signaling in peripheral tissues. Eur J Biochem 271(1):1–13 Hollmann M, Heinemann S (1994) Cloned glutamate receptors. Annu Rev Neurosci 17:31–108 Hollmann M, O’Shea-Greenfield A, Rogers SW, Heinemann S (1989) Cloning by functional expression of a member of the glutamate receptor family. Nature 342(6250):643–648 Huettner JE (2003) Kainate receptors and synaptic transmission. Prog Neurobiol 70(5):387–407 Ishiuchi S, Tsuzuki K, Yoshida Y, Yamada N, Hagimura N, Okado H, Miwa A, Kurihara H, Nakazato Y, Tamura M, Sasaki T, Ozawa S (2002) Blockage of Ca(2?)-permeable AMPA receptors suppresses migration and induces apoptosis in human glioblastoma cells. Nat Med 8(9):971–978 Kalariti N, Pissimissis N, Koutsilieris M (2005) The glutamatergic system outside the CNS and in cancer biology. Expert Opin Investig Drugs 14(12):1487–1496 Kaul M, Zheng J, Okamoto S, Gendelman HE, Lipton SA (2005) HIV-1 infection and AIDS: consequences for the central nervous system. Cell Death Differ 12(Suppl 1):878–892 Keinanen K, Wisden W, Sommer B, Werner P, Herb A, Verdoorn TA, Sakmann B, Seeburg PH (1990) A family of AMPAselective glutamate receptors. Science 249(4968):556–560 Kew JN, Kemp JA (2005) Ionotropic and metabotropic glutamate receptor structure and pharmacology. Psychopharmacology 179(1):4–29 Komuro H, Rakic P (1993) Modulation of neuronal migration by NMDA receptors. Science 260(5104):95–97 Kostanyan IA, Merkulova MI, Navolotskaya EV, Nurieva RI (1997) Study of interaction between L-glutamate and human blood lymphocytes. Immunol Lett 58(3):177–180 La Cava A (2009) Lupus and T cells. Lupus 18(3):196–201 Lerma J (2006) Kainate receptor physiology. Curr Opin Pharmacol 6(1):89–97 Levite M (2008) Neurotransmitters activate T-cells and elicit crucial functions via neurotransmitter receptors. Curr Opin Pharmacol 8(4):460–471 Levite M (2012) Nerve-driven immunity: neurotransmitters and neuropeptides in the immune system, First edition edn edn. Springer, Berlin Levite M, Ganor Y (2008) Autoantibodies to glutamate receptors can damage the brain in epilepsy, systemic lupus erythematosus and encephalitis. Expert Rev Neurother 8(7):1141–1160 Levite M, Cahalon L, Peretz A, Hershkoviz R, Sobko A, Ariel A, Desai R, Attali B, Lider O (2000) Extracellular K(?) and opening of voltage-gated potassium channels activate T cell integrin function: physical and functional association between Kv1.3 channels and beta1 integrins. J Exp Med 191(7):1167– 1176 Li F, Tsien JZ (2009) Memory and the NMDA receptors. N Engl J Med 361(3):302–303 Lin CS, Boltz RC, Blake JT, Nguyen M, Talento A, Fischer PA, Springer MS, Sigal NH, Slaughter RS, Garcia ML et al (1993) Voltage-gated potassium channels regulate calcium-dependent pathways involved in human T lymphocyte activation. J Exp Med 177(3):637–645 Lindblad SS, Mydel P, Hellvard A, Jonsson IM, Bokarewa MI (2011) The N-methyl-D-aspartic acid receptor antagonist memantine ameliorates and delays the development of arthritis by enhancing regulatory T cells. Neurosignals 20(2):61–71 Lombardi G, Dianzani C, Miglio G, Canonico PL, Fantozzi R (2001) Characterization of ionotropic glutamate receptors in human lymphocytes. Br J Pharmacol 133(6):936–944

The neurotransmitter glutamate and human T cells Lombardi G, Miglio G, Canonico PL, Naldi P, Comi C, Monaco F (2003) Abnormal response to glutamate of T lymphocytes from multiple sclerosis patients. Neurosci Lett 340(1):5–8 Lombardi G, Miglio G, Dianzani C, Mesturini R, Varsaldi F, Chiocchetti A, Dianzani U, Fantozzi R (2004) Glutamate modulation of human lymphocyte growth: in vitro studies. Biochem Biophys Res Commun 318(2):496–502 Maddur MS, Miossec P, Kaveri SV, Bayry J (2012) Th17 cells: biology, pathogenesis of autoimmune and inflammatory diseases, and therapeutic strategies. Am J Pathol 181(1):8–18 Mashkina AP, Tyulina OV, Solovyova TI, Kovalenko EI, Kanevski LM, Johnson P, Boldyrev AA (2007) The excitotoxic effect of NMDA on human lymphocyte immune function. Neurochem Int 51(6–7):356–360 Masu M, Tanabe Y, Tsuchida K, Shigemoto R, Nakanishi S (1991) Sequence and expression of a metabotropic glutamate receptor. Nature 349(6312):760–765 Mayer ML (2005) Glutamate receptor ion channels. Curr Opin Neurobiol 15(3):282–288 Mayer ML, Westbrook GL (1987) The physiology of excitatory amino acids in the vertebrate central nervous system. Prog Neurobiol 28(3):197–276 Meldrum BS (2000) Glutamate as a neurotransmitter in the brain: review of physiology and pathology. J Nutr 130(4S Suppl):1007S–1015S Melzer N, Hicking G, Bittner S, Bobak N, Gobel K, Herrmann AM, Wiendl H, Meuth SG (2013) Excitotoxic neuronal cell death during an oligodendrocyte-directed CD8? T cell attack in the CNS gray matter. J Neuroinflammation 10:121 Midgett CR, Gill A, Madden DR (2012) Domain architecture of a calcium-permeable AMPA receptor in a ligand-free conformation. Front Mol Neurosci 4:56 Miglio G, Varsaldi F, Dianzani C, Fantozzi R, Lombardi G (2005a) Stimulation of group I metabotropic glutamate receptors evokes calcium signals and c-jun and c-fos gene expression in human T cells. Biochem Pharmacol 70(2):189–199 Miglio G, Varsaldi F, Lombardi G (2005b) Human T lymphocytes express N-methyl-D-aspartate receptors functionally active in controlling T cell activation. Biochem Biophys Res Commun 338(4):1875–1883 Miglio G, Dianzani C, Fallarini S, Fantozzi R, Lombardi G (2007) Stimulation of N-methyl-D-aspartate receptors modulates Jurkat T cell growth and adhesion to fibronectin. Biochem Biophys Res Commun 361(2):404–409 Miossec P, Kolls JK (2012) Targeting IL-17 and TH17 cells in chronic inflammation. Nat Rev Drug Discov 11(10):763–776 Monaghan DT, Bridges RJ, Cotman CW (1989) The excitatory amino acid receptors: their classes, pharmacology, and distinct properties in the function of the central nervous system. Annu Rev Pharmacol Toxicol 29:365–402 Moriyoshi K, Masu M, Ishii T, Shigemoto R, Mizuno N, Nakanishi S (1991) Molecular cloning and characterization of the rat NMDA receptor. Nature 354(6348):31–37 Nakayama T, Yamashita M (2010) The TCR-mediated signaling pathways that control the direction of helper T cell differentiation. Semin Immunol 22(5):303–309 Nedergaard M, Takano T, Hansen AJ (2002) Beyond the role of glutamate as a neurotransmitter. Nat Rev Neurosci 3(9):748–755 Nicoletti F, Arcella A, Iacovelli L, Battaglia G, Giangaspero F, Melchiorri D (2007) Metabotropic glutamate receptors: new targets for the control of tumor growth? Trends Pharmacol Sci 28(5):206–213 Ohashi H, Maruyama T, Higashi-Matsumoto H, Nomoto T, Nishimura S, Takeuchi Y (2002) A novel binding assay for metabotropic glutamate receptors using [3H] L-quisqualic acid and recombinant receptors. Z Naturforsch C 57(3–4):348–355

Ollenschlager G, Karner J, Karner-Hanusch J, Jansen S, Schindler J, Roth E (1989) Plasma glutamate—a prognostic marker of cancer and of other immunodeficiency syndromes? Scand J Clin Lab Invest 49(8):773–777 Pacheco R, Ciruela F, Casado V, Mallol J, Gallart T, Lluis C, Franco R (2004) Group I metabotropic glutamate receptors mediate a dual role of glutamate in T cell activation. J Biol Chem 279(32):33352–33358 Pacheco R, Oliva H, Martinez-Navio JM, Climent N, Ciruela F, Gatell JM, Gallart T, Mallol J, Lluis C, Franco R (2006) Glutamate released by dendritic cells as a novel modulator of T cell activation. J Immunol 177(10):6695–6704 Pacheco R, Gallart T, Lluis C, Franco R (2007) Role of glutamate on T-cell mediated immunity. J Neuroimmunol 185(1–2):9–19 Piani D, Frei K, Do KQ, Cuenod M, Fontana A (1991) Murine brain macrophages induced NMDA receptor mediated neurotoxicity in vitro by secreting glutamate. Neurosci Lett 133(2):159–162 Piani D, Spranger M, Frei K, Schaffner A, Fontana A (1992) Macrophage-induced cytotoxicity of N-methyl-D-aspartate receptor positive neurons involves excitatory amino acids rather than reactive oxygen intermediates and cytokines. Eur J Immunol 22(9):2429–2436 Pin JP, Duvoisin R (1995) The metabotropic glutamate receptors: structure and functions. Neuropharmacology 34(1):1–26 Pitt D, Werner P, Raine CS (2000) Glutamate excitotoxicity in a model of multiple sclerosis. Nat Med 6(1):67–70 Platt SR (2007) The role of glutamate in central nervous system health and disease—a review. Vet J 173(2):278–286 Poulopoulou C, Davaki P, Koliaraki V, Kolovou D, Markakis I, Vassilopoulos D (2005a) Reduced expression of metabotropic glutamate receptor 2mRNA in T cells of ALS patients. Ann Neurol 58(6):946–949 Poulopoulou C, Markakis I, Davaki P, Nikolaou C, Poulopoulos A, Raptis E, Vassilopoulos D (2005b) Modulation of voltage-gated potassium channels in human T lymphocytes by extracellular glutamate. Mol Pharmacol 67(3):856–867 Poulopoulou C, Papadopoulou-Daifoti Z, Hatzimanolis A, Fragiadaki K, Polissidis A, Anderzanova E, Davaki P, Katsiari CG, Sfikakis PP (2008) Glutamate levels and activity of the T cell voltagegated potassium Kv1.3 channel in patients with systemic lupus erythematosus. Arthritis Rheum 58(5):1445–1450 Puchert M, Engele J (2014) The peculiarities of the SDF-1/CXCL12 system: in some cells, CXCR4 and CXCR7 sing solos, in others, they sing duets. Cell Tissue Res 355(2):239–253 Reynolds JD, Amory DW, Grocott HP, White WD, Newman MF (2002) Change in plasma glutamate concentration during cardiac surgery is a poor predictor of cognitive outcome. J Cardiothorac Vasc Anesth 16(4):431–436 Rosenmund C, Stern-Bach Y, Stevens CF (1998) The tetrameric structure of a glutamate receptor channel. Science 280(5369): 1596–1599 Rzeski W, Turski L, Ikonomidou C (2001) Glutamate antagonists limit tumor growth. Proc Natl Acad Sci USA 98(11):6372–6377 Sarchielli P, Di Filippo M, Candeliere A, Chiasserini D, Mattioni A, Tenaglia S, Bonucci M, Calabresi P (2007) Expression of ionotropic glutamate receptor GLUR3 and effects of glutamate on MBP- and MOG-specific lymphocyte activation and chemotactic migration in multiple sclerosis patients. J Neuroimmunol 188(1–2):146–158 Sattler R, Tymianski M (2001) Molecular mechanisms of glutamate receptor-mediated excitotoxic neuronal cell death. Mol Neurobiol 24(1–3):107–129 Sladeczek F, Momiyama A, Takahashi T (1993) Presynaptic inhibitory action of a metabotropic glutamate receptor agonist on excitatory transmission in visual cortical neurons. Proc Biol Sci 253(1338):297–303

123

Y. Ganor, M. Levite Smith T, Groom A, Zhu B, Turski L (2000) Autoimmune encephalomyelitis ameliorated by AMPA antagonists. Nat Med 6(1):62–66 Stepulak A, Sifringer M, Rzeski W, Endesfelder S, Gratopp A, Pohl EE, Bittigau P, Felderhoff-Mueser U, Kaindl AM, Buhrer C, Hansen HH, Stryjecka-Zimmer M, Turski L, Ikonomidou C (2005) NMDA antagonist inhibits the extracellular signalregulated kinase pathway and suppresses cancer growth. Proc Natl Acad Sci USA 102(43):15605–15610 Stepulak A, Luksch H, Gebhardt C, Uckermann O, Marzahn J, Sifringer M, Rzeski W, Staufner C, Brocke KS, Turski L, Ikonomidou C (2009) Expression of glutamate receptor subunits in human cancers. Histochem Cell Biol 132(4):435–445 Takano T, Lin JH, Arcuino G, Gao Q, Yang J, Nedergaard M (2001) Glutamate release promotes growth of malignant gliomas. Nat Med 7(9):1010–1015 Tanabe Y, Masu M, Ishii T, Shigemoto R, Nakanishi S (1992) A family of metabotropic glutamate receptors. Neuron 8(1):169–179

123

Tay DL, Hoffbrand AV, Wickremasinghe GR (1996) Expression of c-fos and c-jun proteins and AP-1 binding activity during cell cycle progression of HL60 cells and phytohemagglutininstimulated lymphocytes. Exp Hematol 24(2):277–284 Weyand CM, Bryl E, Goronzy JJ (2000) The role of T cells in rheumatoid arthritis. Arch Immunol Ther Exp (Warsz) 48(5):429–435 Wollmuth LP, Sobolevsky AI (2004) Structure and gating of the glutamate receptor ion channel. Trends Neurosci 27(6):321–328 Zhou L, Chong MM, Littman DR (2009) Plasticity of CD4? T cell lineage differentiation. Immunity 30(5):646–655 Zlotnik A, Tsesis S, Gruenbaum BF, Ohayon S, Gruenbaum SE, Boyko M, Sheiner E, Brotfain E, Shapira Y, Teichberg VI (2012) Relationship between glutamate, GOT and GPT levels in maternal and fetal blood: a potential mechanism for fetal neuroprotection. Early Hum Dev 88(9):773–778

Relationship between E receptors and a T-specific surface antigen on human T cells.

Dopamine and T cells: dopamine receptors and potent effects on T cells, dopamine production in T cells, and abnormalities in the dopaminergic system in T cells in autoimmune, neurological and psychiatric diseases.

Donor Unrestricted T Cells: A Shared Human T Cell Response.

Withania somnifera Induces Cytotoxic and Cytostatic Effects on Human T Leukemia Cells.

Immunosuppressive effects of erythropoietin on human alloreactive T cells.

CD4-T-cell antigen receptor complexes on human leukemia T cells.

Ia determinants on stimulated human T lymphocytes. Occurrence on mitogen- and antigen-activated T cells.

Antigen presenting, contrasuppressor human T cells.

Regulatory T cells: human skin residency.

Interindividual variation in human T regulatory cells.

Synergistic targeting of breast cancer stem-like cells by human γδ T cells and CD8+ T cells.

Resident memory T cells in human health and disease.

CD1 and mycobacterial lipids activate human T cells.

Induction of Cell Death in Growing Human T-Cells and Cell Survival in Resting Cells in Response to the Human T-Cell Leukemia Virus Type 1 Tax.

Regulatory T cells and B cells: implication on autoimmune diseases.

A new effect of IL-4 on human γδ T cells: promoting regulatory Vδ1 T cells via IL-10 production and inhibiting function of Vδ2 T cells.

Human breast milk T lymphocytes display the phenotype and functional characteristics of memory T cells.

HIF-1α- and hypoxia-dependent immune responses in human CD4+CD25high T cells and T helper 17 cells.

CD33-specific chimeric antigen receptor T cells exhibit potent preclinical activity against human acute myeloid leukemia.

Engineered antigen-specific human regulatory T cells: immunosuppression of FVIII-specific T- and B-cell responses.

Human IgE(+) B cells are derived from T cell-dependent and T cell-independent pathways.

Human T lymphocytes and T-cell lines as target cells for lymphocyte chemotaxis.

ArtinM Mediates Murine T Cell Activation and Induces Cell Death in Jurkat Human Leukemic T Cells.

Avidity of human T cell receptor engineered CD4(+) T cells drives T-helper differentiation fate.