Harrison's Neurology in Clinical Medicine, 3rd Edition


Stephen L. Hauser Image M. Flint Beal

The human nervous system is the organ of consciousness, cognition, ethics, and behavior; as such, it is the most intricate structure known to exist. More than one-third of the 23,000 genes encoded in the human genome are expressed in the nervous system. Each mature brain is composed of 100 billion neurons, several million miles of axons and dendrites, and >1015 synapses. Neurons exist within a dense parenchyma of multifunctional glial cells that synthesize myelin, preserve homeostasis, and regulate immune responses. Measured against this background of complexity, the achievements of molecular neuroscience have been extraordinary. This chapter reviews selected themes in neuroscience that provide a context for understanding fundamental mechanisms that underlie neurologic disorders.


The landscape of neurology has been transformed by modern molecular genetics. More than 350 different disease-causing genes have been identified, and >1000 neurologic disorders have been genetically mapped to various chromosomal locations. Several hundred neurologic and psychiatric disorders now can be diagnosed through genetic testing (http://www.ncbi.nlm.nih.gov/sites/GeneTests/?db=GeneTests). The vast majority of these disorders represent highly penetrant mutations that cause rare neurologic disorders; alternatively, they represent rare monogenic causes of common phenotypes. Examples of the latter include mutations of the amyloid precursor protein in familial Alzheimer’s disease, the microtubule-associated protein tau (MAPT) in frontotemporal dementia, and α-synuclein in Parkinson’s disease. These discoveries have been profoundly important because the mutated gene in a familial disorder often encodes a protein that is also pathogenetically involved (although not mutated) in the typical, sporadic form. The common mechanism involves disordered processing and, ultimately, aggregation of the protein, leading to cell death (see “Protein Aggregation and Neurodegeneration”).

There is great optimism that complex genetic disorders that are caused by combinations of genetic and environmental factors have become tractable problems. Genome-wide association studies (GWAS) have been carried out in many complex neurologic disorders, with many hundreds of variants identified, nearly all of which confer only a small increment in disease risk (1.15–1.5 fold). GWAS are rooted in the “common disease, common variant” hypothesis, as they examine potential risk alleles that are relatively common (e.g., >5%) in the general population. More than 1000 GWAS have been carried out to date, with notable successes such as the identification of >50 risk alleles for multiple sclerosis. Furthermore, when bioinformatics tools are used, risk variants can be aligned in functional biologic pathways to identify novel pathogenic mechanisms as well as to reveal heterogeneity (e.g., different pathways in different individuals). Despite these successes, many experienced geneticists question the value of common disease-associated variants, particularly whether they are actually causative or merely mark the approximate locations of more important—truly causative—rare mutations.

This debate has set the stage for the next revolution in human genetics, made possible by the development of increasingly efficient and cost-effective high-throughput sequencing methodologies. It is currently possible to sequence an entire human genome in approximately an hour, at a cost of only $4000 for the entire coding sequence (“whole-exome”) or $10,000 for the entire genome; it is certain that these costs will continue to decline. This makes it feasible to look for disease-causing sequence variations in individual patients with the possibility of identifying rare variants that cause disease. The utility of this approach was demonstrated by whole-genome sequencing in a patient with Charcot-Marie-Tooth neuropathy in which compound heterozygous mutations were identified in the SH3TC2 gene that then were shown to co-segregate with the disease in other members of the family.

It is also increasingly recognized that not all genetic diseases or predispositions are caused by simple changes in the linear nucleotide sequence of genes. As the complex architecture of the human genome becomes better defined, many disorders that result from alterations in copy numbers of genes (“gene-dosage” effects) resulting from unequal crossing-over are likely to be identified. As much as 5–10% of the human genome consists of nonhomologous duplications and deletions, and these appear to occur with a much higher mutational rate than is the case for single base pair mutations. The first copy-number disorders to be recognized were Charcot-Marie-Tooth disease type 1A (CMT1A), caused by a duplication in the gene encoding the myelin protein PMP22, and the reciprocal deletion of the gene causing hereditary liability to pressure palsies (HNPP) (Chap. 45). Gene-dosage effects are causative in some cases of Parkinson’s disease (α-synuclein), Alzheimer’s disease (amyloid precursor protein), spinal muscular atrophy (survival motor neuron 2), the dysmyelinating disorder Pelizaeus-Merzbacher syndrome (proteolipid protein 1), late-onset leukodystrophy (lamin B1), and a variety of developmental neurologic disorders. It is now evident that copy-number variations contribute substantially to normal human genomic variation for numerous genes involved in neurologic function, regulation of cell growth, and regulation of metabolism. It is also already clear that gene-dosage effects will influence many behavioral phenotypes, learning disorders, and autism spectrum disorders. Deletions at ch1q and ch15q have been associated with schizophrenia, and deletions at 15q and 16p with autism. Interestingly, the 16p deletion also is associated with epilepsy. Duplications of the X-linked MeCP2 gene cause autism in males and psychiatric disorders with anxiety in females, whereas point mutations in this gene produce the neurodevelop-mental disorder Rett syndrome. The understanding of the role of copy-number variation in human disease is still in its infancy.

The role of splicing variation as a contributor to neurologic disease is another area of active investigation. Alternative splicing refers to the inclusion of different combinations of exons in mature mRNA, resulting in the potential for many different protein products encoded by a single gene. Alternative splicing represents a powerful mechanism for generation of complexity and variation, and this mechanism appears to be highly prevalent in the nervous system, affecting key processes such as neurotransmitter receptors and ion channels. Numerous diseases are known to result from abnormalities in alternative splicing. Increased inclusion of exon 10–containing transcripts of MAPT can cause frontotemporal dementia. Aberrant splicing also contributes to the pathogenesis of Duchenne’s, myotonic, and fascioscapulohumeral muscular dystrophies; ataxiatelangiectasia; neurofibromatosis; some inherited ataxias; and fragile X syndrome, among other disorders. It is also likely that subtle variations of splicing will influence many genetically complex disorders. For example, a splicing variant of the interleukin 7 receptor α chain, resulting in production of more soluble and less membrane-bound receptor, was found to be associated with susceptibility to multiple sclerosis (MS) in multiple different populations.

Epigenetics refers to the mechanisms by which levels of gene expression can be exquisitely modulated not by variations in the primary genetic sequence of DNA but rather by postgenomic alterations in DNA and chromatin structure, which influence how, when, and where genes are expressed. DNA methylation and the methylation and acetylation of histone proteins that interact with nuclear DNA to form chromatin are key mediators of these events. Epigenetic processes appear to be dynamically active even in postmitotic neurons. Imprinting refers to an epigenetic feature, present for a subset of genes, in which the predominant expression of one allele is determined by its parent of origin. The distinctive neurodevelopmental disorders Prader-Willi syndrome (mild mental retardation and endocrine abnormalities) and Angelman syndrome (cortical atrophy, cerebellar dysmyelination, Purkinje cell loss) are classic examples of imprinting disorders whose distinctive features are determined by whether the paternal or maternal copy of chromosome of the critical genetic region 15q11-13 was responsible. In a study of discordant monozygotic twins for MS in which the entire DNA sequence, transcriptome (e.g., mRNA levels), and methylome were assessed genomewide, tantalizing allelic differences in the use of the paternal, compared to maternal, copy for a group of genes were identified. Preferential allelic expression, whether due to imprinting, resistance to X inactivation, or other mechanisms, is likely to play a major role in determining complex behaviors and susceptibility to many neurologic and psychiatric disorders.

Another advance is the development of transgenic mouse models of neurologic diseases, which has been particularly fruitful in producing models relevant to Alzheimer’s disease, Parkinson’s disease, Huntington’s disease, and amyotrophic lateral sclerosis. These models are useful in both studying disease pathogenesis and developing and testing new therapies. Models in both Caenorhabditis elegans and Drosophila have also been extremely useful, particularly in studying genetic modifiers as well as therapeutic interventions.


The resting potential of neurons and the action potentials responsible for impulse conduction are generated by ion currents and ion channels. Most ion channels are gated, meaning that they can transition between conformations that are open or closed to ion conductance. Individual ion channels are distinguished by the specific ions they conduct; their kinetics; and whether they directly sense voltage, are linked to receptors for neurotransmitters or other ligands such as neurotrophins, or are activated by second messengers. The diverse characteristics of different ion channels provide a means by which neuronal excitability can be modulated exquisitely at both the cellular and subcellular levels. Disorders of ion channels—channelopathies—are responsible for a growing list of human neurologic diseases (Table 25-1). Most are caused by mutations in ion channel genes or by autoantibodies against ion channel proteins. One example is epilepsy, a syndrome of diverse causes characterized by repetitive, synchronous firing of neuronal action potentials. Action potentials normally are generated by the opening of sodium channels and the inward movement of sodium ions down the intracellular concentration gradient. Depolarization of the neuronal membrane opens potassium channels, resulting in outward movement of potassiumions, repolarization, closure of the sodium channel, and hyperpolarization. Sodium or potassium channel subunit genes have long been considered candidate disease genes in inherited epilepsy syndromes, and recently such mutations were identified. These mutations appear to alter the normal gating function of these channels, increasing the inherent excitability of neuronal membranes in regions where the abnormal channels are expressed.

TABLE 25-1



Whereas the specific clinical manifestations of channelopathies are quite variable, one common feature is that manifestations tend to be intermittent or paroxysmal, as occurs in epilepsy, migraine, ataxia, myotonia, or periodic paralysis. Exceptions are clinically progressive channel disorders such as autosomal dominant hearing impairment. The genetic channelopathies identified to date are all uncommon disorders caused by obvious mutations in channel genes. As the full repertoire of human ion channels and related proteins is identified, it is likely that additional channelopathies will be discovered. In addition to rare disorders that result from obvious mutations, it is likely that less penetrant allelic variations in channel genes or their pattern of expression might underlie susceptibility to some apparently sporadic forms of epilepsy, migraine, or other disorders. For example, mutations in the potassium channel gene Kir2.6 have been found in many individuals with thyrotoxic hypokalemic periodic paralysis, a disorder similar to hypokalemic periodic paralysis but precipitated by stress from thyrotoxicosis or carbohydrate loading.


Synaptic neurotransmission is the predominant means by which neurons communicate with each other. Classic neurotransmitters are synthesized in the presynaptic region of the nerve terminal; stored in vesicles; and released into the synaptic cleft, where they bind to receptors on the postsynaptic cell. Secreted neurotransmitters are eliminated by reuptake into the presynaptic neuron (or glia), diffusion away from the synaptic cleft, and/or specific inactivation. In addition to the classic neurotransmitters, many neuropeptides have been identified as definite or probable neurotransmitters; they include substance P, neurotensin, enkephalins, β-endorphin, histamine, vasoactive intestinal polypeptide, cholecystokinin, neuropeptide Y, and somatostatin. Peptide neurotransmitters are synthesized in the cell body rather than the nerve terminal and may colocalize with classic neurotransmitters in single neurons. A number of neuropeptides are important in pain modulation, including substance P and calcitonin gene-related peptide (CGRP), which causes migraine-like headaches in patients. As a consequence, CGRP receptor antagonists have been developed and shown to be effective in treating migraine headaches. Nitric oxide and carbon monoxide are gases that appear also to function as neurotransmitters, in part by signaling in a retrograde fashion from the post-synaptic to the presynaptic cell.

Neurotransmitters modulate the function of post-synaptic cells by binding to specific neurotransmitter receptors, of which there are two major types. Iono-tropic receptors are direct ion channels that open after engagement by the neurotransmitter. Metabotropic receptors interact with G proteins, stimulating production of second messengers and activating protein kinases, which modulate a variety of cellular events. Ionotropic receptors are multiple-subunit structures, whereas metabo-tropic receptors are composed of single subunits only. One important difference between ionotropic and metabotropic receptors is that the kinetics of ionotropic receptor effects are fast (generally <1 ms) because neurotransmitter binding directly alters the electrical properties of the postsynaptic cell, whereas metabotropic receptors function over longer periods. These different properties contribute to the potential for selective and finely modulated signaling by neurotransmitters.

Neurotransmitter systems are perturbed in a large number of clinical disorders, examples of which are highlighted in Table 25-2. One example is the involvement of dopaminergic neurons originating in the substantia nigra of the midbrain and projecting to the striatum (nigrostriatal pathway) in Parkinson’s disease and in heroin addicts after the ingestion of the toxin MPTP (1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine).

TABLE 25-2




A second important dopaminergic system arising in the midbrain is the mediocorticolimbic pathway, which is implicated in the pathogenesis of addictive behaviors including drug reward. Its key components include the midbrain ventral tegmental area (VTA), median fore-brain bundle, and nucleus accumbens (see Fig. 53-1). The cholinergic pathway originating in the nucleus basalis of Meynert plays a role in memory function in Alzheimer’s disease.

Addictive drugs share the property of increasing dopamine release in the nucleus accumbens. Amphetamine increases intracellular release of dopamine from vesicles and reverses transport of dopamine through the dopamine transporters. Patients prone to addiction show increased activation of the nucleus accumbens after administration of amphetamine. Cocaine binds to dopamine transporters and inhibits dopamine reuptake. Ethanol inhibits inhibitory neurons in the VTA, leading to increased dopamine release in the nucleus accumbens. Opioids also disinhibit these dopaminergic neurons by binding to μ receptors expressed by γ-aminobutyric acid (GABA)-containing interneurons in the VTA. Nicotine increases dopamine release by activating nicotinic acetylcholine receptors on cell bodies and nerve terminals of dopaminergic VTA neurons. Tetrahydrocannabinol, the active ingredient of cannabis, also increases dopamine levels in the nucleus accumbens. Blockade of dopamine in the nucleus accumbens can terminate the rewarding effects of addictive drugs.

Not all cell-to-cell communication in the nervous system occurs via neurotransmission. Gap junctions provide for direct neuron-neuron electrical conduction and also create openings for the diffusion of ions and metabolites between cells. In addition to neurons, gap junctions are widespread in glia, creating a syncytium that protects neurons by removing glutamate and potassium from the extracellular environment. Gap junctions consist of membrane-spanning proteins, termed connexins, that pair across adjacent cells. Mechanisms that involve gap junctions have been related to a variety of neurologic disorders. Mutations in connexin 32, a gap junction protein expressed by Schwann cells, are responsible for the X-linked form of CMT disease. Mutations in either of two gap junction proteins expressed in the inner ear—connexin 26 and connexin 31—result in autosomal dominant progressive hearing loss (Chap. 24). Glial calcium waves mediated through gap junctions also appear to explain the phenomenon of spreading depression associated with migraine auras and the march of epileptic discharges. Spreading depression is a neural response that follows a variety of different stimuli and is characterized by a circumferentially expanding negative potential that propagates at a characteristic speed of 20 m/s and is associated with an increase in extracellular potassium.


The fundamental issue of how memory, learning, and thinking are encoded in the nervous system is likely to be clarified by identification of the signaling pathways involved in neuronal differentiation, axon guidance, and synapse formation and by an understanding of how these pathways are modulated by experience. Many families of transcription factors, each consisting of multiple individual components, are expressed in the nervous system. Elucidation of these signaling pathways has begun to provide insights into the causes of a variety of neurologic disorders, including inherited disorders of cognition such as X-linked mental retardation. This problem affects ~1 in 500 males, and linkage studies in different families suggest that as many as 60 different X-chromosome-encoded genes may be responsible. Rett syndrome, a common cause of (dominant) X-linked progressive mental retardation in females, is due to a mutation in a gene (MECP2) that encodes a DNA-binding protein involved in transcriptional repression. As the X chromosome accounts for only ~3% of germ-line DNA, by extrapolation, the number of genes that potentially contribute to clinical disorders affecting intelligence in humans must be potentially very large. As discussed later in the chapter, there is increasing evidence that abnormal gene transcription may play a role in neurodegenerative diseases such as Huntington’s disease, in which proteins with polygluta-mine expansions bind to and sequester transcription factors. A critical transcription factor for neuronal survival is CREB (cyclic adenosine monophosphate responsive element-binding) protein, which also plays an important role in memory in the hippocampus.


Myelin is the multilayered insulating substance that surrounds axons and speeds impulse conduction by permitting action potentials to jump between naked regions of axons (nodes of Ranvier) and across myelinated segments. Molecular interactions between the myelin membrane and the axon are required to maintain the stability, function, and normal life span of both structures. A single oligodendrocyte usually ensheathes multiple axons in the central nervous system (CNS), whereas in the peripheral nervous system (PNS) each Schwann cell typically myelinates a single axon. Myelin is a lipid-rich material formed by a spiraling process of the membrane of the myelinating cell around the axon, creating multiple membrane bilayers that are tightly apposed (compact myelin) by charged protein interactions. Several inhibitors of axon growth are expressed on the innermost (periaxonal) lamellae of the myelin membrane (see “Stem Cells and Transplantation”). A number of clinically important neurologic disorders are caused by inherited mutations in myelin proteins of the CNS or PNS (Fig. 25-1). Constituents of myelin also have a propensity to be targeted as autoantigens in autoimmune demyelinating disorders (Fig. 25-2). Specification to oligodendrocyte precursor cells (OPCs) is transcriptionally regulated by the Olig 2 and Yin Yang 1 genes, whereas myelination mediated by postmitotic oligodendrocytes depends on a different transcription factor, myelin gene regulatory factor (MRF). It is noteworthy that in the normal adult brain large numbers of OPCs (expressing platelet-derived growth factor receptor alpha [PDGFR-α] and NG2) are widely distributed but do not myelinate axons, even in demyelinating environments such as lesions of MS. The characterization of these cells, including an understanding of their transcriptional regulation and functional roles, could result in novel approaches to remyelination and brain repair.



The molecular architecture of the myelin sheath illustrating the most important disease-related proteins. The illustration represents a composite of CNS and PNS myelin. Proteins restricted to CNS myelin are shown in green, proteins of PNS myelin are lavender, and proteins present in both CNS and PNS are red. In the CNS, the X-linked allelic disorders Pelizaeus-Merzbacher disease and one variant of familial spastic paraplegia are caused by mutations in the gene for proteolipid protein (PLP) that normally promotes extracellular compaction between adjacent myelin lamellae. The homologue of PLP in the PNS is the P0 protein, mutations in which cause the neuropathy Charcot-Marie-Tooth disease (CMT) type 1B. The most common form of CMT is the 1A subtype caused by a duplication of the PMP22 gene; deletions in PMP22 are responsible for another inherited neuropathy termed hereditary liability to pressure palsies (Chap. 45).

In multiple sclerosis (MS), myelin basic protein (MBP) and the quantitatively minor CNS protein myelin oligodendrocyte glycoprotein (MOG) are probably T cell and B cell antigens, respectively (Chap. 39). The location of MOG at the outermost lamella of the CNS myelin membrane may facilitate its targeting by autoantibodies. In the PNS, autoantibodies against myelin gangliosides are implicated in a variety of disorders, including GQ1b in the Fisher variant of Guillain-Barré syndrome, GM1 in multifocal motor neuropathy, and sulfatide constituents of myelin-associated glycoprotein (MAG) in peripheral neuropathies associated with monoclonal gammopathies (Chap. 46).



A model for experimental allergic encephalomyelitis (EAE). Crucial steps for disease initiation and progression include peripheral activation of preexisting autoreactive T cells; homing to the CNS and extravasation across the blood-brain barrier; reactivation of T cells by exposed autoantigens; secretion of cytokines; activation of microglia and astrocytes and recruitment of a secondary inflammatory wave; and immune-mediated myelin destruction. ICAM, intercellular adhesion molecule; LFA-1, lymphocyte function-associated antigen-1; VCAM, vascular cell adhesion molecule; IFN, interferon; IL, interleukin; TNF, tumor necrosis factor.


Neurotrophic factors (Table 25-3) are secreted proteins that modulate neuronal growth, differentiation, repair, and survival; some have additional functions, including roles in neurotransmission and in the synaptic reorganization involved in learning and memory. The neurotrophin (NT) family contains nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), NT3, and NT4/5. The neurotrophins act at TrK and p75 receptors to promote survival of neurons. Because of their survival-promoting and antiapoptotic effects, neurotrophic factors are in theory outstanding candidates for therapy for disorders characterized by premature death of neurons as occurs in amyotrophic lateral sclerosis (ALS) and other degenerative motor neuron disorders. Knockout mice lacking receptors for ciliary neurotrophic factor (CNTF) or BDNF show loss of motor neurons, and experimental motor neuron death can be rescued by treatment with various neurotrophic factors, including CNTF, BDNF, and vascular endothelial growth factor (VEGF). However, in phase 3 clinical trials, growth factors were ineffective in human ALS. The growth factor glial-derived neurotrophic factor (GDNF) is important for survival of dopaminergic neurons. Direct infusions of GDNF showed initial promise in Parkinson’s disease (PD), but the benefits were not replicated in a larger clinical trial.

TABLE 25-3




The nervous system is traditionally considered to be a nonmitotic organ, particularly with respect to neurons. These concepts have been challenged by the finding that neural progenitor or stem cells exist in the adult CNS that are capable of differentiation, migration over long distances, and extensive axonal arborization and synapse formation with appropriate targets. These capabilities also indicate that the repertoire of factors required for growth, survival, differentiation, and migration of these cells exists in the mature nervous system. In rodents, neural stem cells, defined as progenitor cells capable of differentiating into mature cells of neural or glial lineage, have been experimentally propagated from fetal CNS and neuroectodermal tissues and also from adult germinal matrix and ependyma regions. Human fetal CNS tissue is also capable of differentiation into cells with neuronal, astrocyte, and oligodendrocyte morphology when cultured in the presence of growth factors.

Once the repertoire of signals required for cell type specification is better understood, differentiation into specific neural or glial subpopulations could be directed in vitro; such cells also could be engineered to express therapeutic molecules. Another promising approach is to utilize growth factors such as BDNF to stimulate endogenous stem cells to proliferate and migrate to areas of neuronal damage. Administration of epidermal growth factor with fibroblast growth factor replenished up to 50% of hippocampal CA1 neurons a month after global ischemia in rats. The new neurons made connections and improved performance in a memory task.

A major advance has been the development of induced pluripotent stem cells. Using this technique, adult somatic cells such as skin fibroblasts are treated with four pluripotency factors (SOX2, KLF4, cMYC, and Oct4), and this generates induced pluripotent stem cells (iPSCs). These adult-derived stem cells sidestep the ethical issues of utilizing stem cells derived from human embryos. The development of these cells has tremendous promise for both studying disease mechanisms and testing therapeutics. There is no consensus on the best way to generate and differentiate iPSCs; however, techniques to avoid using viral vectors and the use of Crelox systems to remove reprogramming factors result in a better match of gene expression profiles with those of embryonic stem cells. Thus far, iPSC cells have been made from patients with all the major human neurode-generative diseases, and studies utilizing them are under way.

Although stem cells hold tremendous promise for the treatment of debilitating neurologic diseases such as Parkinson’s disease and spinal cord injury, it should be emphasized that medical application is in its infancy. Major obstacles are the generation of position- and neurotransmitter-defined subtypes of neurons and their isolation as pure populations of the desired cells. This is crucial to avoid persistence of undifferentiated embryonic stem (ES) cells, which can generate tumors. The establishment of appropriate neural connections and afferent control is also critical. For instance, human ES motor neurons will have to be introduced at multiple segments in the neuraxis, and then their axons will have to regenerate from the spinal cord to distal musculature.

Experimental transplantation of human fetal dopaminergic neurons in patients with Parkinson’s disease has shown that these transplanted cells can survive within the host striatum; however, some patients developed disabling dyskinesias, and this approach is no longer in clinical development. Human ES cells can be differentiated into dopaminergic neurons, which reverse symptoms of Parkinson’s disease in experimental animal models. Studies of transplantation for patients with Huntington’s disease have reported encouraging, although very preliminary, results. Oligodendrocyte precursor cells transplanted into mice with a dysmyelinating disorder effectively migrated in the new environment, interacted with axons, and mediated myelination; such experiments raise hope that similar transplantation strategies may be feasible in human disorders of myelin such as MS. The promise of stem cells for treatment of both neurodegenerative diseases and neural injury is great, but development has been slowed by unresolved concerns over safety (including the theoretical risk of malignant transformation of transplanted cells), ethics (particularly with respect to use of fetal tissue), and efficacy.

In developing brain, the extracellular matrix provides stimulatory and inhibitory signals that promote neuronal migration, neurite outgrowth, and axonal extension. After neuronal damage, reexpression of inhibitory molecules such as chondroitin sulfate proteoglycans may prevent tissue regeneration. Chondroitinase degraded these inhibitory molecules and enhanced axonal regeneration and motor recovery in a rat model of spinal cord injury. Several myelin proteins, specifically Nogo, oligodendrocyte myelin glycoprotein (OMGP), and myelin-associated glycoprotein (MAG), also may interfere with axon regeneration. Sialidase, which cleaves one class of receptors for MAG, enhances axonal outgrowth. Antibodies against Nogo promote regeneration after experimental focal ischemia or spinal cord injury. Nogo, OMGP, and MAG all bind to the same neural receptor, the Nogo receptor, which mediates its inhibitory function via the p75 neurotrophin receptor signaling.


Excitotoxicity refers to neuronal cell death caused by activation of excitatory amino acid receptors (Fig. 25-3). Compelling evidence for a role of excitotoxicity, especially in ischemic neuronal injury, is derived from experiments in animal models. Experimental models of stroke are associated with increased extracellular concentrations of the excitatory amino acid neurotransmitter gluta-mate, and neuronal damage is attenuated by denervation of glutamate-containing neurons or the administration of glutamate receptor antagonists. The distribution of cells sensitive to ischemia corresponds closely with that of N-methyl-D-aspartate (NMDA) receptors (except for cerebellar Purkinje cells, which are vulnerable to hypoxemia-ischemia but lack NMDA receptors), and competitive and noncompetitive NMDA antagonists are effective in preventing focal ischemia. In global cerebral ischemia, non-NMDA receptors (kainic acid and α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate [AMPA]) are activated, and antagonists to these receptors are protective. Experimental brain damage induced by hypoglycemia also is attenuated by NMDA antagonists.



Involvement of mitochondria in cell death. A severe excitotoxic insult (A) results in cell death by necrosis, whereas a mild excitotoxic insult (B) results in apoptosis. After a severe insult (such as ischemia), there is a large increase in glutamate activation of NMDA receptors, an increase in intracellular Ca2+ concentrations, activation of nitric oxide synthase (NOS), and increased mitochondrial Ca2+ and superoxide generation followed by the formation of ONOO. This sequence results in damage to cellular macromolecules including DNA, leading to activation of poly-ADP-ribose polymerase (PARS). Both mitochondrial accumulation of Ca2+ and oxidative damage lead to activation of the permeability transition pore (PTP) that is linked to excitotoxic cell death. A mild excitotoxic insult can occur due either to an abnormality in an excitotoxicity amino acid receptor, allowing more Ca2+ flux, or to impaired functioning of other ionic channels or of energy production, which may allow the voltage-dependent NMDA receptor to be activated by ambient concentrations of glutamate. This event can then lead to increased mitochondrial Ca2+and free radical production, yet relatively preserved ATP generation. The mitochondria may then release cytochrome c (Cytc), caspase 9, apoptosis-inducing factor (Aif), and perhaps other mediators that lead to apoptosis. The precise role of the PTP in this mode of cell death is still being clarified, but there does appear to be involvement of the adenine nucleotide transporter that is a key component of the PTP.

Excitotoxicity is not a single event but rather a cascade of cell injury. Excitotoxicity causes influx of calcium into cells, and much of the calcium is sequestered in mitochondria rather than in the cytoplasm. Increased cytoplasmic calcium causes metabolic dysfunction and free radical generation; activates protein kinases, phospholipases, nitric oxide synthase, proteases, and endonucleases; and inhibits protein synthesis. Activation of nitric oxide synthase generates nitric oxide (NO·), which can react with superoxide (O·2) to generate peroxynitrite (ONOO), which may play a direct role in neuronal injury. Another critical pathway is activation of poly-ADP-ribose polymerase, which occurs in response to free radical–mediated DNA damage. Experimentally, mice with knockout mutations of neuronal nitric oxide synthase or poly-ADP-ribose polymerase, or those which overexpress superoxide dismutase, are resistant to focal ischemia.

Another aspect of excitotoxicity is that it has been demonstrated that stimulation of extrasynaptic NMDA receptors mediates cell death, whereas stimulation of synaptic receptors is protective. This has been shown to play a role in excitotoxicity in transgenic mouse models of Huntington’s disease, in which the use of low-dose memantine to selectively block the extrasynaptic receptors is beneficial.

Although excitotoxicity is clearly implicated in the pathogenesis of cell death in stroke, to date treatment with NMDA antagonists has not proved clinically useful. Transient receptor potentials (TRPs) are calcium channels that are activated by oxidative stress in parallel with excitotoxic signal pathways. In addition, glutamate-independent pathways of calcium influx via acid-sensing ion channels have been identified. These channels transport calcium in the setting of acidosis and substrate depletion, and pharmacologic blockade of these channels markedly attenuates stroke injury. These channels offer a potential new therapeutic target for stroke.

Apoptosis, or programmed cell death, plays an important role in both physiologic and pathologic conditions. During embryogenesis, apoptotic pathways operate to destroy neurons that fail to differentiate appropriately or reach their intended targets. There is mounting evidence for an increased rate of apoptotic cell death in a variety of acute and chronic neurologic diseases. Apoptosis is characterized by neuronal shrinkage, chromatin condensation, and DNA fragmentation, whereas necrotic cell death is associated with cytoplasmic and mitochondrial swelling followed by dissolution of the cell membrane. Apoptotic death and necrotic cell death can coexist or be sequential events, depending on the severity of the initiating insult. Cellular energy reserves appear to have an important role in these two forms of cell death, with apoptosis favored under conditions in which ATP levels are preserved. Evidence of DNA fragmentation has been found in a number of degenerative neurologic disorders, including Alzheimer’s disease, Huntington’s disease, and ALS. The best characterized genetic neurologic disorder related to apoptosis is infantile spinal muscular atrophy (Werdnig-Hoffmann disease), in which two genes thought to be involved in the apoptosis pathways are causative.

Mitochondria are essential in controlling specific apoptosis pathways. The redistribution of cytochrome c, as well as apoptosis-inducing factor (AIF), from mitochondria during apoptosis leads to the activation of a cascade of intracellular proteases known as caspases. Caspase-independent apoptosis occurs after DNA damage, activation of poly-ADP-ribose polymerase, and translocation of AIF into the nucleus. Redistribution of cytochrome c is prevented by overproduction of the apoptotic protein BCL2 and is promoted by the proapoptotic protein BAX. These pathways may be triggered by activation of a large pore in the mitochondrial inner membrane known as the permeability transition pore, although in other circumstances they occur independently. Recent studies suggest that blocking the mitochondrial pore reduces both hypoglycemic and ischemic cell death. Mice deficient in cyclophilin D, a key protein involved in opening the permeability transition pore, are resistant to necrosis produced by focal cerebral ischemia.


The possibility that protein aggregation plays a role in the pathogenesis of neurodegenerative diseases is a major focus of current research. Protein aggregation is a major histopathologic hallmark of neurodegenerative diseases. Deposition of β-amyloid is strongly implicated in the pathogenesis of Alzheimer’s disease. Genetic mutations in familial Alzheimer’s disease cause increased production of β-amyloid with 42 amino acids, which has an increased propensity to aggregate, compared with β-amyloid with 40 amino acids. Mutations in genes encoding the MAPT lead to altered splicing of tau and the production of neurofibrillary tangles in fronto-temporal dementia and progressive supranuclear palsy. Familial Parkinson’s disease is associated with mutations in leucine-rich repeat kinase 2 (LRRK2), α-synuclein, parkin, PINK1, and DJ-1. PINK1 is a mitochondrial kinase (see later), and DJ-1 is a protein involved in protection from oxidative stress. Parkin, which causes autosomal recessive early-onset Parkinson’s disease, is a ubiquitin ligase. The characteristic histopathologic feature of Parkinson’s disease is the Lewy body, an eosinophilic cytoplasmic inclusion that contains both neurofilaments and α-synuclein. Huntington’s disease and cerebellar degenerations are associated with expansions of polyglutamine repeats in proteins, which aggregate to produce neuronal intranuclear inclusions. Familial ALS is associated with superoxide dismutase mutations and cytoplasmic inclusions that contain superoxide dismutase. An important finding was the discovery that the ubiquinated inclusions observed in most cases of ALS and the most common form of frontotemporal dementia are composed of TAR DNA binding protein 43 (TDP-43). Subsequently, mutations in the TDP-43 gene and in the fused in sarcoma gene (FUS) were found in familial ALS. These two proteins are involved in transcription regulation as well as RNA metabolism. In autosomal dominant neurohypophyseal diabetes insipidus, mutations in vasopressin result in abnormal protein processing, accumulation in the endoplasmic reticulum, and cell death.

Another key mechanism linked to cell death is mitochondrial dynamics, which refer to the processes involved in movement of mitochondria, as well as in mitochondrial fission and fusion, which play a critical role in mitochondrial turnover and in replenishment of damaged mitochondria. Mitochondrial dysfunction is strongly linked to the pathogenesis of a number of neurodegenerative diseases, such as Friedreich’s ataxia, which is caused by mutations in an iron-binding protein that plays an important role in transferring iron to iron-sulfur clusters in aconitase and complex I and II of the electron transport chain. Mitochondrial fission is dependent on the dynamin-related proteins (Drp1), which bind to its receptor Fis, whereas mitofuscins 1 and 2 (MF 1/2) and optic atrophy protein 1 (Opa1) are responsible for fusion of the outer and inner mitochondrial membrane, respectively. Mutations in Mfn2 cause Charcot-Marie-Tooth neuropathy type 2A, and mutations in Opa1 cause autosomal dominant optic atrophy. Both β-amyloid and mutant huntingtin protein induce mitochondrial fragmentation and neuronal cell death associated with increased activity of Drp1. In addition, mutations in genes causing autosomal recessive Parkinson’s disease, parkin and PINK1, cause abnormal mitochondrial morphology and result in impairment of the ability of the cell to remove damaged mitochondria by autophagy.

The current major scientific question is whether protein aggregates contribute to neuronal death or whether they are merely secondary bystanders. A major focus in all the neurodegenerative diseases is now on small protein aggregates termed oligomers. These aggregates may be the toxic species of β-amyloid, α-synuclein, and proteins with expanded polyglutamines such as those that are associated with Huntington’s disease. Protein aggregates are usually ubiquinated, which targets them for degradation by the 26S component of the protea-some. An inability to degrade protein aggregates could lead to cellular dysfunction, impaired axonal transport, and cell death by apoptotic mechanisms.

Autophagy is the degradation of cystolic components in lysosomes. There is increasing evidence that autophagy plays an important role in degradation of protein aggregates in the neurodegenerative diseases, and it is impaired in Alzheimer’s disease (AD), Parkinson’s disease, and Huntington’s disease (HD). Autophagy is particularly important to the health of neurons, and failure of autophagy contributes to cell death. In Huntington’s disease, a failure of cargo recognition occurs, contributing to protein aggregates and cell death. Rapamycin, which induces autophagy, exerts beneficial therapeutic effects in transgenic mouse models of AD, PD, and HD.

In experimental models of Huntington’s disease and cerebellar degeneration, protein aggregates are not well correlated with neuronal death and may be protective. A substantial body of evidence suggests that the mutant proteins with polyglutamine expansions in these diseases bind to transcription factors and that this contributes to disease pathogenesis. In Huntington’s disease there is dysfunction of the transcriptional co-regulator PGC-1α, a key regulator of mitochondrial biogenesis. There is evidence that impaired function of PGC-1α is also important in both Parkinson’s disease and Alzheimer’s disease, making it an attractive target for treatments. Agents that upregulate gene transcription are neuroprotective in animal models of these diseases. A number of compounds have been developed to block β-amyloid production and/or aggregation, and these agents are being studied in early clinical trials in humans. Another approach under investigation is immunotherapy with antibodies that bind β-amyloid, tau, or α-synuclein.

Another emerging theme is the role of chronic inflammation, and in particular of activated microglia and innate immunity, in the pathogenesis of many neurodegenerative diseases. Activation of Toll-like receptors (TLR) in response to pattern-recognition signals from damaged or aging cells, including those mediated by heat shock proteins or aggregated proteins, can trigger or amplify pro-inflammatory responses. Familial frontotemporal degeneration (Chap. 29) is caused by mutations in the gene encoding progranulin, a growth factor that regulates inflammation via binding to tumor necrosis factor (TNF) receptors.


Systems neuroscience refers to study of the functions of neurocircuits and the way they relate to brain function, behavior, motor activity, and cognition. Brain imaging techniques, primarily functional MRI (fMRI) and positron emission tomography (PET), have made it possible to investigate cognitive processes such as perception, making judgments, paying attention, and thinking. This has allowed insights into how networks of neurons operate to produce behavior. Many of these studies at present are based on determining the connectivity of neural circuits and how they operate and how this can be modeled to produce improved understanding of physiologic processes. fMRI uses contrast mechanisms related to physiologic changes in tissue, and brain perfusion can be studied by observing the time course of changes in brain water signal as a bolus of injected paramagnetic gadolinium contrast moves through the brain. More recently, to study intrinsic contrast-related local changes in blood oxygenation with brain activity, blood-oxygen-level-dependent (BOLD) contrast has been used to provide a rapid noninvasive approach for functional assessment. These techniques have been used reliably in both behavioral and cognitive sciences. One example is the use of fMRI to demonstrate mirror neuron systems, imitative pathways activated when observing actions of others (Fig. 25-4). Mirror neurons are thought to be important for social conditioning and for many forms of learning, and abnormalities in mirror neurons may underlie some autism disorders. Data also suggest that enhancement of mirror neuron pathways might have potential for rehabilitation after stroke. Other examples of the use of fMRI include the study of memory. Recent studies have shown that not only is hippocampal activity correlated with declarative memory consolidation, it also involves activation in the ventral medial prefrontal cortex. Consolidation of memory over time results in decreased activity of the hippocampus and progressively stronger activation in the ventral medial prefrontal region associated with retrieval of consolidated memories. fMRI also has been utilized to identify sequences of brain activation involved in normal movements and alterations in their activation associated with both injury and recovery and to plan neurosurgical operations. Diffusion tensor imaging is a recently developed MRI technique that can measure macroscopic axonal organization in nervous system tissues; it appears to be useful in assessing myelin and axonal injuries as well as brain development. Advances in understanding neural processing have led to the development of the ability to demonstrate that humans have on-line voluntary control of human temporal lobe neurons.



Mirror neuron systems are bilaterally activated during imitationA. Bilateral activations (circled in yellow) in inferior frontal mirror neuron areas during imitation, as measured by BOLD fMRI signal changes. In red, activation during right hand imitation. In blue, activation during left hand imitation. B. In contrast, there is lateralized (contralateral) primary visual activation of the primary visual cortex for imitated actions presented to the right visual field (in red, left visual cortex) and to the left visual field (in blue, right visual cortex). C. Lateralized primary motor activation for hand actions imitated with the right hand (in red, left motor cortex) and with the left hand (in blue, right motor cortex). (From L Aziz-Zadeh et al: J Neurosci 26:2964, 2006.)

A further advance that has far-reaching implications for the development of novel interventions for neurologic, including behavioral, conditions has been the development of deep-brain stimulation as a highly effective therapeutic intervention for treating excessively firing neurons in the subthalamic nucleus of patients with Parkinson’s disease and the precingulate cortex in patients with depression.