NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Forum on Neuroscience and Nervous System Disorders. Glutamate-Related Biomarkers in Drug Development for Disorders of the Nervous System: Workshop Summary. Washington (DC): National Academies Press (US); 2011.

Cover of Glutamate-Related Biomarkers in Drug Development for Disorders of the Nervous System

Glutamate-Related Biomarkers in Drug Development for Disorders of the Nervous System: Workshop Summary.

Show details

4Treatment Implications of Biomarkers

Biomarker development has the potential to increase the efficiency of drug development, refine or enhance clinical trial data, and speed access to safe drugs (IOM, 2008). For nervous system disorders, these goals might be accomplished through new target development, patient stratification, and side-effect reduction. This section highlights three central nervous system (CNS) disorders—ischemia, autism spectrum disorders, and chronic pain—where current glutamate biomarker research has the potential to advance drug development.


One of the tenets of glutamate dysfunction is that increased extracellular glutamate levels, under conditions of ischemia and trauma, set in motion a cascade of events that lead to intense calcium influx into glutamate neurons. With large participation by astrocytes, calcium influx into post-synaptic glutameteric neurons leads to widespread cell death from excitotoxicity and necrosis (Choi, 1994). Focal ischemia accounts for 80 percent of stroke damage. But this basic tenet of glutamate's dominance is more nuanced, as a result of a decade or more of research.

Dennis Choi, executive vice president at the Simons Foundation, described current thinking about glutamate as having lost momentum for new drug development to block calcium influx by N-methyl D-aspartate (NMDA) receptor blockade. One of the main problems, Choi said, was that animal studies with positive results did not translate to humans. With the benefit of newer research, Choi explained that, contrary to expectations, ischemia is not wholly accountable for neuronal necrosis, nor are excess calcium (Ca2+) ions. Other cations and receptors do play a profound role, suggesting potential new targets for drug development.

Further research has revealed roles for glutamate's metabotropic receptors, as well as other ions and receptors beyond glutamate's. Although more is known, Choi emphasized that the lack of biomarkers has been highly detrimental to progress. Calcium excess in ischemia still triggers necrotic cell death in the first hours, so blocking NMDA receptors during that window is important. But afterward, sufficient calcium release also causes release of zinc (Zn2+) ions into the extracellular space, which, in turn, blocks the NMDA receptor. At that point, NMDA blockade should cease. But there are no biomarkers to determine when that point occurs. Meanwhile, ischemia is known to cause acidosis, with proton release from ATP hydrolysis (Xiong et al., 2008). The pH levels of the brain fall to 6.5. That pH drop activates receptors throughout the brain known as acid-sensing ion channels (ASICs), which are proton-gated cation channels widely distributed in peripheral sensory neurons and the CNS.

Acidosis through activation of ASICs also is responsible for substantial neuronal injury. The greater the acidosis with calcium and sodium ions acting through ASICs, the greater the infarct is (Xiong et al., 2008). Using biomarkers within the CNS, were they to exist, researchers could understand how the movements and timing of zinc, calcium, and hydrogen increases help to mitigate the impact of ischemic stroke. Choi concluded that although glutamate is an important player, biomarkers for ASIC's action and cations are essential to pave the way for new target development.


The potential for patient stratification became clear in the discussion of glutamate biomarkers for autism spectrum disorders (ASDs). ASD, as its name implies, covers a broad spectrum of symptoms. Although most genetic causes of autism are unknown, several single-gene disorders are associated with high rates of ASD. The best understood genetic subtype is caused by a genetic mutation in the Fragile X gene, but Fragile X only accounts for 2 to 5 percent of those with ASD (Kelleher and Bear, 2008). Although those with the Fragile X gene have symptoms that overlap with other ASD cases, the underlying basis may be fundamentally different. The next most common single-gene disorder causing autism is tuberous sclerosis complex (TSC).

The single gene defect in Fragile X syndrome silences FMR1, the gene encoding the Fragile X protein, which normally represses protein translation. When silenced by the mutation, Fragile X is responsible for mental retardation (Bagni and Greenough, 2005). The silencing of the FMR1 protein causes excess signaling through mGluR5. The receptor itself is not defective; the defect is in the heightened rate of translation, at the presynaptic terminal, of up to 400 distinct mRNAs that FMR normally represses (Brown et al., 2001). The lack of repression markedly increases the rate of protein synthesis at axonal terminals, including proteins for glutamate signaling. As further confirmation, studies of Fragile X knockout mice found a decrease in mGluR5 signaling reversed the syndrome's manifestations (Dölen and Bear, 2008). These findings led to ongoing clinical trials of several metabotropic glutamate antagonists. But one group of autism patients is expected to be adversely affected by the treatment—those with tuberous sclerosis.

Tuberous sclerosis, an autosomal dominant disorder, is caused either by mutations of hamartion (TSC1) or tuberin (TSC2). Those defective proteins act in the brain to inhibit protein synthesis at axonal terminals. Decreased protein synthesis is the polar opposite of the effects of Fragile X mutation. Consequently, a metabotropic antagonist given to patients with tuberous sclerosis would likely block glutamate signaling to such a great extent that it would be deleterious. The awareness of opposing functions of two genetic causes of autism highlights the need for stratifying patients by genotype. But the vast majority of autism cases have no genotype biomarkers. Developing biomarkers of increased or decreased rates of protein synthesis at synaptic terminals have been largely unsuccessful, stated Mark Bear, Picower Professor of Neuroscience at Massachusetts Institute of Technology.

Research on autism has brought to the fore a major challenge for drug development: predicting outcomes for patients with the same diagnosis, but with different subtypes, noted Bear. Recognition of this challenge began with excitement surrounding novel treatments now being tested in clinical trials. However, Bear indicated, although one group of autism patients has responded well, another group of autism patients, it could be reasoned, might be harmed because of the lack of biomarkers to stratify patients with different subtypes of disease. What has been learned about autism pathophysiology over the past decade has pointed to the importance of stratifying patients with the same diagnosis but a different genotype, such as Fragile X and tuberous sclerosis, to predict treatment outcome.

Most neurological and psychiatric diseases are diagnosed by symptoms, history, and course of illness without the benefit of biomarkers. If more were known, patient subtypes would likely emerge. Grouping patients by subtypes maximizes the opportunity to understand causation and find new treatments. A clinical trial could be designed strictly for individuals with that subtype rather than all patients with the same diagnosis. Lack of homogeneity in a treatment group may decrease the chances of finding a robust effect or even preclude finding an effect, stated William Potter, cochair emeritus of the Neuroscience Steering Committee at the FNIH Biomarkers Consortium. The drug that may be effective for one subtype may be ineffective or possibly harmful for another. The clinical trial may be stopped prematurely or investment in research might be halted unnecessarily—to the detriment of patients with a particular subtype.


Chronic pain syndromes are highly prevalent, affecting up to 30 percent of the U.S. population (Johannes et al., 2010). Of two marketed glutamate-targeted drugs, one (ketamine) is an anesthetic, thereby indicating the prominent role that glutamate plays in pain. The problem with ketamine is that the CNS action is associated with side effects, including confusion, drowsiness, learning and memory impairment, and ataxia. This and the inaccessibility of the CNS and the lack of consensus on what is the host of the sensitization mechanisms are impeding progress. To obviate part of the problem, Brian Cairns, associate professor at the University of British Columbia–Vancouver, raised the importance of looking for pain-related glutamate biomarkers in the peripheral nervous system, where few have looked in any detail. The concept of peripheral sensitization biomarkers arose from his knowledge of the adverse effects of CNS treatments and his growing understanding of pain pathophysiology.

Glutamate is the primary neurotransmitter for sensory neurons carrying pain information from the periphery to the CNS and within the CNS. Much of the difficulty of finding pain biomarkers arises from research pointing to CNS sensitization as the main driver of chronic pain. In many chronic pain syndromes, including temporomandibular disorders (TMDs) with muscular pain, the chronic pain is experienced1 as a result of plasticity in the form of central sensitization in the somatosensory cortex. Central sensitization can occur after prolonged increase in excitation of pain neurons in the CNS triggered by sustained, repetitive, and high-frequency input from nociceptors (i.e., pain sensory receptors in the periphery). Central pain sensitization is exaggerated pain signaling. It is the pathological equivalent of long-term potentiation. One manifestation of central sensitization is that painful stimuli that would normally cause minor pain instead induce exaggerated pain (hyperalgesia).

Hyperalgesia is manifest in TMD and in many other pain disorders. In animal models, its molecular basis begins with high and prolonged exposure to a painful stimulus, which triggers release of inflammatory cytokines by the immune system. Cytokines and other proinflammatory agents trigger spontaneous discharge of peripheral nociceptors, leading to transmission of pain signals to the CNS and induction of pain behaviors (Cairns, 2010).

NMDA antagonists have been used in research and clinical practice to attenuate pain. Cairns and his colleagues sought to identify an NMDA antagonist that was active primarily in the periphery. NMDA receptors are diheteromeric, consisting of two subunits: two NR2A or two NR2B units. The latter are found preferentially in the periphery (Collingridge et al., 2004). The finding that 40 to 60 percent of masseter nociceptive neurons in the periphery express the NR2B subunit provided the rationale for testing a peripherally acting NMDA antagonist to protect against central side effects of NMDA antagonists (Gazerani et al., 2010). Cairns found that glutamate-evoked masticatory muscle afferent discharge is mediated through peripheral NR2B subunits. In a rodent model, the NR2B antagonist ifenprodil reduced the glutamate-evoked masticatory muscle sensory afferent discharge. In a human trial, ketamine reduced TMD pain approximately one hour after a single injection into the masseter muscle (Castrillon et al., 2008). Research also revealed that patients with muscular pain in TMD were found to have elevated levels of glutamate in their masseter muscles (Castrillon et al., 2010). Direct intramuscular injection of glutamate induced pain that was mediated through activation of peripheral NMDA receptors, Cairns explained.

Taken together, these and other findings are interpreted as supporting peripheral sensitization of nociceptive afferent fibers in TMDs, Cairns said. Peripheral sensitization due to increased glutamate levels in tissues and activated glutamate receptors have been underestimated as part of the pathophysiology of chronic pain. Both peripheral and central sensitizations are likely at play. Based on the studies with ketamine or ifenprodil, he reaffirmed the need to develop and test peripherally restricted NMDA receptor antagonists. Finally, he concluded that elevated interstitial glutamate levels in the periphery might be a potential target that would allow development of drugs that treat pain, but avoid the side effects found in current drugs that target the CNS.



The International Association for the Study of Pain does not define pain as a sensation. Rather, pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage.

Copyright © 2011, National Academy of Sciences.
Bookshelf ID: NBK62177


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.0M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...