This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Neurophysiology is the science of neurons in action and deals with the way in which they generate and transmit electrical and chemical signals. This chapter will review the basic properties of neuronal membranes and the ion currents that contribute to the excitation and inhibition of neurons. We will move from excitatory and inhibitory synaptic potentials to the generation of action potentials and the signal transmission from neuron to neuron and neuron to muscle. We will review the principles of electrochemical transmission and the classes of neurotransmitters involved, including their basic chemistry. The boxes will focus on non-invasive ways of assessing brain function that can be applied in human studies and are therefore of particular relevance to psychiatric research.
2.1 Neurons transmit information
Although the ways in which the brain processes information are still clouded in mystery, and the translation of these neuronal processes into phenomenal awareness is largely a matter for philosophical debate, the basic principle of neural functioning seems clear enough. Any single neuron can be conceived as an input-output machine, which receives chemical and/or electrical signals and converts them into an output (normally NT release) in an all-or-nothing fashion. The mechanism behind this all-or-nothing mode is the action potential, which is a unitary electrical event that only occurs if the incoming signals exceed a certain threshold.
2.1.1 The reflex arc is a simple model of information processing
The monosynaptic input-output loop that leads to sensori-motor reflexes such as the knee jerk can serve as a simple model of information processing by neurons. This particular reflex arc (Fig. 2-1) starts in the muscle spindle of the quadriceps muscle, which is stretched by the tap on the tendon that connects it with the tibia bone. The muscle spindle is part of a bipolar neuron with its cell body in the dorsal root ganglion. It is connected with an axon that carries an action potential into the gray matter of the spinal medulla and forms a synapse with a motor neuron in the anterior horn. The action potential (AP) leads to release of the NT glutamate from the presynaptic (sensory) neuron, which in turn effects an excitatory postsynaptic potential (EPSP) in the postsynaptic (motor) neuron. This EPSP (in fact, the sum of a large number of EPSPs) triggers an action potential at the axon hillock of the motor neuron, which travels back to the muscle and leads to release of acetylcholine, another NT, at the neuromuscular endplate, which effects the contraction of the muscle and the extension of the lower leg. Another synapse of the afferent axon activates an inhibitory interneuron, which releases glycine and leads to an inhibitory postsynaptic potential (IPSP) in the motor neuron innervating the leg flexor muscle.
--- Fig. 2.1 ---
2.1.2 Changes in membrane potentials underlie signal transmission
In order to unravel the component processes of information transmission in and across neurons we need to distinguish three basic events that are largely separate in space and time, the EPSP or IPSP, the AP and the electro-chemical transmission, the release of NT molecules into the synaptic cleft and their reaction with post- and presynaptic receptors. These events are all enabled by the specific properties of the lipid bilayer membrane, especially its selective barrier function for molecules. Water, soluble gases like oxygen and carbon dioxide, small polar molecules like ethanol and lipophilic substances can all freely diffuse across the plasma membrane. Conversely, the membrane is almost impermeable for charged particles, even atomic ions like Na+ (sodium), K+ (potassium), Ca2+ (calcium)or Cl- (chloride), and larger polar molecules like sugars. These substances can only pass the membrane by way of selective transport through so called channels (e.g., sodium, potassium, chloride, or calcium channels), transmembrane proteins with microscopic (< 1 nanometer [nm] diameter) pores. Finally, large molecules, such as NTs, can enter and exit the neuron through the formation of vesicles that are coated with a lipid bilayer that fuses with or separates from the membrane, a process termed exo- or endocytosis.
The kinetic of the diffusion of molecules through the membrane is governed by the concentration gradient of the substance (i.e., the concentration difference between the intra- and extracellular space) and, if it is charged, by its electrostatic gradient (i.e., whether the charges inside and outside the membrane are positive or negative). For example, if the concentration of a positively charged ion is higher inside the cell, but the cell is overall negatively charged, the ion will be driven out of the cell by its concentration gradient (Fd) but pushed into it by the electrostatic gradient (Fe; see the example of K+ in Fig. 2-2). Because there are no channels for the negatively charged large organic ions (proteins) the membrane potential is mainly governed by the distribution of Na+, K+ and Cl-. Each of these ions has an equilibrium potential (Ei), at which the electrostatic and concentration gradient add to zero. The potential at which these gradients add to zero across all ions involved is called the membrane potential, Vm. The resting membrane potential Vr, which occurs when there is no additional input from synaptic or action potentials or opening of ion channels caused by NT binding, is ca. -70 millivolts (mV). By convention, the outside of the cell is the zero point, and the negative Vr thus denotes a preponderance of negative ions inside the cell.
--- Fig. 2.2 ---
Because the Ei for K+ is more negative than Vr (ca. - 75 mV), and the Ei for Na+ far more positive (ca. 60 mV), K+ ions diffuse out of the cell, whereas Na+ ions diffuse into the cell. Diffusion processes would thus ultimately lead to equal concentrations for these positive ions on both sides of the membrane, which would make the triggering of fast responses (action potentials) impossible. The concentration gradients are therefore maintained through an active transport process, the Na+/K+ pump, a transmembrane protein that receives the energy needed to move these ions against their respective gradients by splitting a phosphate group off from ATP molecules. This process is responsible for a large part of the energy consumption of neurons.
2.1.3 Postsynaptic potentials can lead to action potentials through spatial and temporal summation
Let us consider the case where a NT that docks onto the postsynaptic membrane opens a sodium channel, leading to influx of Na+ and a more positive Vm, and thus an EPSP. Because the EPSP still has a negative sign, just less so than the Vr, this process is termed depolarisation. Conversely, the process that leads to increased influx of Cl- ions and thus a more negative Vm is called hyperpolarisation. It makes firing of an action potential less likely and is thus inhibitory (IPSP). Because IPSPs are normally produced by a combination of Cl- entering and K+ leaving the neuron, the largest hyperpolarisation through this process is equivalent to the Ei for K+. Both EPSPs and IPSPs summate over time and space. Summation over time occurs if a further postsynaptic potential is generated before the previous one has been completely discharged. Whereas the postsynaptic currents (EPSC and IPSC) normally only last for 1-2 ms, the PSPs take longer to discharge. Their time constant (time until decrease to 1/e or ca. 37% of the peak) is in the range of 5 ms, which is also the approximate time window for temporal summation. Spatial summation occurs if PSCs from several dendrites meet downstream (e.g. at the axon hillock between soma and axon) to create a PSP that is larger than any of the original ones alone. Depolarisation of the axon hillock that exceeds a threshold of ca. -40mV triggers an AP (Fig. 2-3). The ion movements across the membrane that sustain the AP do not occur through chemically gated channels, as in the case of the PSPs, but through electrically gated channels. Such voltage dependent sodium channels that open when the threshold depolarisation is reached allow for the massive influx of Na+ into the neuron, leading to further depolarisation up to ca. 40mV, which is slightly below the Ei of Na+ at 60mV. This leads to compensatory K+ efflux with repolarisation and even a slight after-hyperpolarisation. Restitution of the resulting concentration imbalance to resting levels is one of the functions of the Na+/ K+-ATPase. The hyperpolarisation constitutes a relative refractory period, during which it takes a higher EPSC to reach the voltage threshold to open the sodium channels and trigger a new AP. During the repolarisation phase of the AP the sodium channels are shut and no new AP can be triggered (absolute refractory period). The whole AP/refractory period cycle takes only a couple of milliseconds, and thus the maximum firing frequency of a neuron is in the range of 500 Hz (Hz [Hertz] is the unit for frequency; 1Hz=1 cycle/s).
From the axon hillock the AP has to travel, often over many millimetres or even meters, in the case of very long motor or sensory axons, along the axon to the synapse. This propagation is achieved by bidirectional spreading of the electrotonic potential. However, the refractory sodium channels prevent a new AP at the site of its origin and the AP can thus only travel along the axon. Thus, a new AP will originate at the next non-refractory sodium channels and thus propagate all the way to the synapse at a speed of ca. 1mm/s. However, transmission speed can be much higher (up to 100ms/s) in myelinated axons. Here the myelin protein sheath works as insulation and prevents formation of APs. Thus, the AP has to spread electrotonically to the next node of Ranvier, where a new AP will be triggered. This process is, of course, limited by the inevitable loss of depolarisation over distance, but given a close enough spacing of nodes of Ranvier will reliably propagate APs, and much faster than in non-myelinated axons.
--- Fig. 2.3 ---
2.1.4 Electrochemical transmission occurs at the synapse
When it reaches the synapse, the AP induces opening of voltage-dependent Ca2+ channels. Ca2+ induces the release of neurotransmitters (NTs) from presynaptic membrane vesicles through exocytosis into the synaptic cleft. The released quantity of transmitters depends on the concentration of Ca2+ in the presynaptic terminal, which increases with the duration and frequency of the APs. A delay of a few milliseconds is involved in the electrochemical synaptic transmission because the Ca2+ influx only starts towards the end of the AP, and although the ensuing exocytosis occurs within fractions of a millisecond, the diffusion of NT molecules to the postsynaptic membrane takes some time. NT molecules travel through the synaptic cleft (ca. 20nm wide) and bind to postsynaptic receptors. The NT/receptor complex has an impact on postsynaptic membrane permeability and can thus trigger a new EPSP or IPSP. Then the cascade from postsynaptic to action potentials to NT release can start in a new neuron.
2.2 NTs convey information between neurons
Not all synaptic molecules are NTs. In order to qualify as an NT a substance needs to
be synthesised and stored in the neuron
be released from the neuron upon electrical stimulation
have postsynaptic receptors
be deactivated after release and action
have selective inhibitors.
There are two main classes of NT receptors. In the case of ionotropic or ligand-gated receptors, the same protein acts as receptor and channel. Thus binding of an agonist will directly lead to opening of the channel and thus higher permeability for an ion and a postsynaptic current. Its action and termination are fast and it is thus involved in short-term information transmission and neural plasticity. Conversely, metabotropic receptors are not themselves ion channels, but indirectly control channels through G-proteins. G(GTP-binding)-proteins are central regulating molecules. When they are activated by a metabotropic receptor, they bind guanosine triphosphate (GTP) and split off a phosphate group from GTP. In this process G-proteins also release a subunit that activates other molecules, for example an ion channel. However, this subunit can also activate other proteins and thus trigger cascades of so-called second messengers (the NT molecules are the "first" messengers), which can influence gene expression or lead to other lasting biochemical changes in the postsynaptic neuron. These second messenger systems include cyclic adenosine monophosphate (cAMP), inositol triphosphate (IP3) and diacylglycerol (DAG). Because of their effect on second messenger systems the metabotropic receptors are supposed to support long-term neural plasticity. It is still a moot case whether psychotropic drugs such as antipsychotics or antidepressants, which all influence synaptic transmission, act through alteration of postsynaptic potentials or through more long-term plastic changes (the onset latency of most clinical effects would support the latter view). These issues will be discussed in more detail in Part 2.
Modulatory processes at the presynaptic membrane can inhibit or enhance NT release. Inhibition of calcium channels or hyperpolarisation of the membrane through increased K+- or Cl- permeability (leading to K+ efflux or Cl- influx, respectively, and thus more relative negative charge inside) decrease Ca2+ influx or sensitivity. Conversely, direct increase of Ca2+ permeability or closing of K+ channels, increased duration of the AP or high-frequency stimulation of the presynaptic neuron all lead to increased Ca2+ influx and thus enhance NT release.
2.2.1 NTs come from different chemical classes
The classical NTs are the monoamines dopamine, noradrenaline/norepinephrine, adrenaline/epinephrine (collectively termed catecholamines) and serotonin, the amino acids glutamic acid, glycine and gamma-amino-butyric acid (GABA), and acetylcholine. The monoamines derive their name from the single amine (one nitrogen and two or three hydrogen atoms) group at their tail. The catecholamines are named for the catechol ring (carbohydrate ring with two hydroxyl-groups), whereas serotonin or 5-hydroxy-tryptamine (5-HT) has the structure of an indolamine. All amino acids have an amine group at one end (the amino-terminal) and a carboxy (COOH) group at the other (the carboxy-terminal) (Fig. 2.4). Acetylcholine is chemically an ester (combination of two molecules via an oxygen bridge) of choline and acetic acid (Fig. 2.5). The enzyme that splits it into its components is therefore called acetylcholine-esterase, and drugs that inhibit this are called enzyme acetylcholine-esterase inhibitors (AChEI). AChEI are the main class of anti-dementia drugs, based on the assumption that cholinergic (acetylcholine-based) neurotransmission is disturbed in Alzheimer's disease.
--- Figs. 2.4, 2.5 ---
2.2.2 Dopamine, noradrenaline (norepinephrine) and adrenaline (epninephrine) are catecholamines
Catecholamines are synthesised in the synaptic terminal from a common precursor, the amino acid tyrosine, which shares a transport system into the brain with the other hydrophobic amino acids, for example tryptophan. Tyrosine reaches the synaptic terminal through active transport in the cytoskeleton of the axon. Depending on the availability of the enzymes that catalyse the respective reactions, tyrosine will be converted to dopamine (through the intermediary step 3,4-dihydroxyphenylalanine, DOPA), and further to noradrenaline/norepinephrine (NE) and adrenaline/epinephrine (E) (Fig. 2.6). Because dopamine does not pass through the blood-brain barrier, dopamine replacement therapy in Parkinson's disease, which is characterised by loss of dopaminergic neurons in the substantia nigra and elsewhere, uses DOPA. DOPA can be easily converted to dopamine by the enzyme DOPA-decarboxylase (also called Aromatic L-amino acid Decarboxylase [AAAD]). In order for this only to happen in the brain (and not in the general circulation, where dopamine would cause unwanted side effects, for example on blood pressure), AAAD inhibitors, which unlike DOPA do not pass through the blood brain barrier, are normally given at the same time (see also chapter 6.4).
--- Fig. 2.6, 2.7 ---
2.2.3 Serotonin is an indolamine
The conversion of tryptophan to 5-HT (serotonin), through the intermediary step 5-Hydroxytryptophan (Fig. 2.7), occurs in the soma. In this case thus, the final NT rather than the precursor is transported through the axon. Its degradation starts, as with the catecholamines, through oxidation by MAO, in this case to 5-Hydroxyindolacetaldehyde, which is further oxidated to 5- hydroxy-indolacetic acid (5-HIAA), the main end-product of indolamine metabolism, and a surrogate marker of serotonin concentrations that can be measured in urine or CSF. Serotonin itself can be measured in blood platelets. Serotonin levels seem crucially to depend on availability of tryptophan. Tryptophan depletion has been used as an experimental model of depression, based on the theory that depression is caused by monoamine deficiency (monoamine deficit model of depression, which will be discussed in more detail in chapter 11).
2.2.4 The action of NTs can be terminated through enzymatic degradation
Enzymatic degradation is one of the main mechanisms for the termination of NT action. Other enzymes that are targets of psychotropic drugs include monoamine-oxidase (MAO) and catechol-O-methyltransferase (COMT). MAO removes the amine group from monoamines and thus converts them into aldehydes, which can be oxidated further to the corresponding acid (e.g., 5-HIAA, see Fig. 2.7) or reduced to the corresponding alcohol. COMT adds a methylene (CH3) group to the catechol ring. Through these enzymatic processes catecholamines can be converted into methylated alcohols or methylated acid. In neurons, the main metabolic pathway for dopamine leads to the methylated acid 3-Methoxy-4-hydroxyphenyl acetic acid or homovanillic acid (HVA). For NE, the main metabolic pathway in neurons yields 3-Methoxy-4-hydroxyphenylglycol (MHPG). Both HVA and MHPG can be measured in urine as surrogate markers of catecholamine production. MAO inhibitors are used in depression in order to maintain high levels of monoamines in the synaptic cleft, and COMT inhibitors are used in Parkinson's disease with the same rationale.
2.2.5 Glutamate and GABA are produced by central metabolic pathways
The amino acid L-glutamate is the product of the transamination of alpha-ketoglutarate, a component of the Krebs cycle, the common pathway of glucose, protein and lipid catabolism. Humans therefore do not depend on the glutamate that is ingested with food, unlike for the "essential" amino acids, for which they do not have a pathway of biosynthesis, for example tryptophan or phenylalanine, the precursor of tyrosine. Decarboxylation of L-glutamate, which is catalysed by the enzyme glutamate decarboxylase I in nervous tissue and needs pyridoxal-phosphate as coenzyme, yields GABA (Fig. 2.8). Its action is terminated by reuptake into the presynaptic neuron or further enzymatic conversion to succinyl-CoA, another ubiquitous component of the Krebs cycle. It is therefore impossible to draw any conclusions about glutamate or GABA activity in the brain from measuring its metabolic products. The same is true for glycine, which is the simplest amino acid and a basic building block for many more complex biomolecules. One of its catabolic reactions is catalysed by D-aminoacid-oxidase (DAO), which is also involved in the metabolism of the amino acid D-serine, which is a co-factor required for the activation of the NMDA receptor by glutamate. Some post-mortem evidence suggests that the activity of DAO may be increased in schizophrenia.
--- Fig. 2.8 ---
Glutamate is the main excitatory NT of the CNS, GABA the main inhibitory NT of the brain, and glycine the main inhibitory NT of the spinal cord and the PNS. Acetylcholine and the monoamines are the NTs of specific (cholinergic, dopaminergic, serotonergic, noradrenergic) neurons and can have a range of inhibitory and excitatory effects. They are therefore sometimes classified as "neuromodulatory". Another group of NTs with a largely modulatory effect are the peptide NTs. They seem to modulate the function of the classical NTs through poorly known mechanisms, but also have their own receptors, for example the enkephalins, which bind to morphine receptors. The neuropeptides play a major role in the transmission and modulation of pain perception, especially the enkephalins and substance P. Some of the neuropeptides, for example angiotensin II, vasoactive intestinal peptide (VIP), somatostatin and cholecystokinin (CCK), also have functions outside the nervous system and straddle the boundary between NTs and hormones. We will revisit them in chapter 3 on neuroendocrinology.
BOX 2-1: Non-invasive techniques: EEG, MEG,
Although even single channels can be recorded in preparations in vitro, and single cells in invasive recordings from animals, most neurophysiological research in humans has to rely on non-invasive techniques. Electroencephalography (EEG) measures the cortical potential changes on the scalp, but requires the spatial summation of large numbers of synchronous postsynaptic potentials for a sufficient signal-to-noise ratio. EEG uses electrodes (made of silver, lead, zinc, or platinum, for example) and amplifiers, which were originally connected to an oscillograph and nowadays to a computer. The result is a visual picture of brain wave. Already the inventor of the EEG, the German psychiatrist Hans Berger (1873-1941) observed that these brain waves changed dramatically if the subject engaged in mental activity, compared to rest. The resting rhythm was in the alpha frequency (8-12Hz), also termed "Berger rhythm", whereas cognitive activity and attention were accompanied by faster activity (beta: 12-30Hz). Slow waves in the theta (3.5-7.5Hz) and delta (<3.5Hz) ranges occur during deep relaxation and sleep, but also during certain pathological states and as a consequence of psychotropic or narcotic drugs. Frequencies even higher than beta (gamma range: 30-100Hz) have also been associated with cognitive activities. In the time before modern neuroimaging with computed tomography and magnetic resonance imaging (both available for clinical use since the 1970s), EEG was an important diagnostic tool to determine presence and location of space occupying lesions. In this role it has been completely superseded by the more accurate and reliable imaging techniques. EEG still has an important clinical role in the diagnosis and classification of seizure disorders and some neuropsychiatric disorders, for example Creutzfeldt-Jacob's disease (CJD). There is also an EEG renaissance in cognitive and clinical research because the observed neuroelectric patterns can be compared with those found in vitro and in animal research. Many prescription and illicit drugs lead to changes in the EEG; for example benzodiazepines increase beta activity, anticonvulsants can slow down the background rhythm from alpha to theta frequencies and antipsychotics are also associated with slowing of the EEG and increases in epileptiform activity. This activity resembles the synchronized sharp wave activity observed in patients with epilepsy, and sometimes even the characteristic spike-wave patterns. This epileptiform activity, which is most common (in about a third of treated patients) under clozapine and olanzapine, is thought to reflect a reduced seizure threshold. Antipsychotics and antidepressants indeed have a small risk of inducing seizures (commonly thought to be below 1%) but in most patients these EEG changes will remain subclinical.
A more complicated (and far more expensive) way of measuring changes in synaptic activity non-invasively is magnetoencephalography (MEG). MEG systems consist of arrays of sensors that pick up the magnetic field changes produced by the synaptic currents. Like EEG, MEG needs synaptic changes to occur synchronously in large numbers of neurons (at least in the order of 10,000s), but it has the advantage that the attenuation of magnetic signals depends only on the distance from the source, and not on the type of surrounding tissue. MEG thus allows for a more reliable reconstruction of cortical sources of scalp signals than EEG.
Sensory stimuli evoke synchronous activity in the central nervous system that can be measured on the scalp by EEG electrodes. These so-called evoked potentials (EPs) have mainly been described for the visual, auditory and tactile domain, but can in principle be measured for all sensory channels. EPs are normally described as positive or negative going (denoted with the letters P or N), according to the direction of the deflection in standard referencing procedures. This convention is unrelated to the contribution from excitatory vs. inhibitory neural activity, though. The letter is followed by a number that denotes the latency in milliseconds or the position in a sequence of positive or negative deflections. For example, the P100, generated in primary and higher visual areas, is a positive deflection with latency from stimulus onset of about 100ms (Fig. 2.9). Even subtle changes in the physical stimulus can evoke large changes in neural activity, for example through violations of expectancy. The classical way of studying such "event-related potentials" (ERPs) is through so-called oddball paradigms, where a train of regular "standard" stimuli (for example tones of a particular frequency) is disrupted by a deviant "oddball" (a tone at a different frequency). These oddballs are associated with the P300 response, which can be elicited in all sensory modalities, and the auditory "mismatch negativity". The P300 is probably the most widely studied neurophysiological biomarker of mental disorders (see, for example, chapter 10 on its uses in schizophrenia research). The MMN is also of great neuropsychiatric interest, for example because of its modulation by glutamate antagonists at the NMDA receptor, and its potential use as a prognostic marker for coma patients.
--- Fig. 2.9 ---
EPs/ERPs have a place in the clinical diagnosis, for example of demyelinating disorders such as multiple sclerosis, where information transmission from the periphery to the CNS is delayed because, as explained above, speed of AP propagation depends on the insulation from the myelin sheath. They are also widely used in psychiatric research to pinpoint the locus of disrupted information processing. In the example above, the P1 component of the visual ERP was used to highlight disruptions of visual processing in schizophrenia. That information processing may be disrupted in sensory (rather than only in frontal and limbic) systems in schizophrenia is a relatively outcome of this type of non-invasive research.
Further reading: (Ford et al., 2007, Thaker, 2008)
BOX 2.2-2: TMS
Whereas EEG measures changes in synaptic activity, transcranial magnetic stimulation (TMS) can induce them. TMS works according to the laws of electromagnetic induction. The TMS apparatus is a bank of capacitors that discharge a strong (up to 10000 Ampere) and very brief (ca. 200 microseconds) current into a coil that is held over the head. This current generates a magnetic field that induces an electric field in the tissue under the coil. If this tissue is conductive (as is the case for nervous tissue) this electric field will lead to an electric current, which can affect the Vm and cause local membrane de- or hyperpolarisation. Depending on the orientation of the coil and the stimulated neurons, and of the stimulation parameters, the effects of TMS can be excitatory or inhibitory. For example, single pulse stimulation over the hand are of the primary motor cortex normally leads to a movement in the contralateral hand, an excitatory effect. Similarly, stimulation over the visual cortex may induce phosphenes. Conversely, repetitive stimulation at intermediate frequencies (e.g. 1Hz) or the more recently developed "theta burst" (3 cycles per second of fast bursts of 5 pulses) are thought to have inhibitory effects, e.g. to slow down cognitive operations that rely on the stimulated area.
In cognitive neuroscience, TMS is used as "functional lesion" method. Whereas neuropsychologists traditionally had to base their studies on opportunistic samples of patients with clinical lesions, experimental neuropsychology with TMS can now manipulate brain activity as the independent variable and assess the resulting cognitive deficits (or enhancements). TMS is thus a unique method to assess brain structure-function relationship systematically. With the use of single pulse (sp) TMS researcher can disrupt or facilitate neural processing at particular points of time, whereas repetitive TMS (rTMS) disrupts or enhances neural processing during a train of stimuli and beyond.
In a psychiatric context rTMS is also used as experimental treatment method for hallucinations and depression. In both cases the aim is to target an area that is supposed to be under- or overactive. In the case of depression, the model is that frontal cortex is underactive (hypofrontality), which leads to problems with executive function and motivation and generally inability to rein in the uncontrollable emotions presented by the limbic system. Thus, the aim is to enhance prefrontal function with high-frequency (10Hz) rTMS. There are some promising reports of clinical improvement (and some evidence for a biochemical correlate, increased dopamine release, but only in animal experiments), but rTMS has yet to pass the standard tests for an evidence-based treatment method. Based on neuroimaging findings of a role of temporal lobe activity in hallucinations, rTMS of left temporoparietal cortex has been attempted for a reduction in severity and frequency of hallucinations. Again, the first clinical results are promising, but problems like the design of a placebo condition, quantification of clinical effects, and sample size of studies make it difficult to demonstrate improvement definitively.
Further reading: (Sack and Linden, 2003, Floel and Cohen, 2006, George and Aston-Jones, 2010)
BOX 2.2-3: Metabolic imaging
EEG and MEG have exquisite temporal resolution (in the range of the underlying synaptic events) and can thus trace neural activation in real time. However, their spatial resolution and localization accuracy are limited. Functional magnetic resonance imaging (fMRI) can trace neural activation at a spatial resolution that is about one order of magnitude higher (in the millimeter range), but suffers from limitations in its temporal resolution because it relies on a sluggish vascular response. However, with the echo-planar imaging (EPI) technique, functional images of the whole brain can be acquired within 1-2 seconds, which is sufficient to capture complex cognitive processes or changes in mental states. The hallucination study described below is an example of such an application. PET has even lower spatial (and temporal) resolution (Fig. 2.10), but can yield excellent molecular resolution (discussed in Chapter 7). The temporal resolution of TMS ranges from tens of milliseconds in the case of single pulse (sp) TMS to several minutes in the case of virtual lesion protocols with repetitive TMS (rTMS). An invasive method that, like fMRI, relies on vascular signals and has an excellent spatial resolution is optical imaging. In animal studies it can be combined with direct intracerebral recordings to yield both global activation maps and information about fine-grained temporal activation patterns, but this is not possible in human studies. However, even in humans it may be possible to explore the blind spot in the left lower corner of Fig. 2.10 and trace brain correlates of perception or thought at millisecond and millimeter resolution through the combination of fMRI and EEG.
FMRI provides an indirect measure of neural activity. Synaptic activity leads to the release of vasoactive substances such as nitric oxide that lead to local vasodilatation. The ensuing influx of fresh blood leads to over-supply of oxygen because it exceeds the increased metabolism of oxygen due to the increased need for aerobic glycolysis. The ratio of oxygenated (oxy-) to deoxygenated (deoxy-) haemoglobin will therefore be shifted in favour of the oxy-haemoglobin. Because deoxy-haemoglobin is paramagnetic, meaning that it distorts local magnetic fields, the result is an increase of MRI signal. In colour-coded statistical maps, this will normally be denoted in warm colours. For example, when we measure brain activity in patients experiencing auditory hallucinations with fMRI (Fig. 2.11), we can detect a hotspot in their auditory cortex (even without any change in external auditory stimulation).
--- Fig. 2.11 ---
Another technique for metabolic imaging is positron emission tomography (PET) with radioactively labeled glucose (flurodeoxyglucose, FDG) or water. FDG-PET picks up the increased consumption of glucose in activated brain areas. Although it was the most widely technique of metabolic imaging into the late 1990s PET has now been largely superseded by fMRI because of its ease of administration and absence of radioactivity. PET still very much has a place in psychiatric research as a tool for receptor mapping.
FMRI has already contributed significantly to our understanding of psychopathology (e.g., hallucinations). It provides information that cannot be obtained from structural imaging and/or neuropsychology alone. FMRI can also reveal the networks involved in cognitive operations that are affected by neuropsychiatric disorders. Although fMRI has no diagnostic use yet, clinical applications in the monitoring of pharmacological and psychological interventions are currently being explored.
Further reading: (Linden and Fallgatter, 2009)
2.3 Learning points
Mental disorders are disorders of perception, though, action and emotions. All these functions are supported by complex interplay of information processing by neurons in the CNS. Although very little is known about the mechanisms of higher cognitive functions, the basic elements of neuronal signal processing, which are likely to provide their foundations as well, are relatively well understood. A neuron convert chemical input (docking of a receptor to their dendritic membrane) into chemical output (NT release from synaptic boutons) through electrical signals. NT binding to a receptor can lead to excitatory or inhibitory postsynaptic potentials. The excitatory potentials can sum up (temporal and spatial summation) to yield an action potential, which travels along the axon to the synaptic terminal, where it triggers influx of calcium, which results in release of NT. This NT can in turn bind to postsynaptic receptors, starting a new cycle.
NTs come from different chemical classes, including amino acids, monoamines (molecules with one amino-group, -NH2) and peptides. The amino acids glutamic acid and GABA are the main excitatory and inhibitory NTs of the human brain. The monoamines can be further subdivided into catecholamines (derived from the amino acid tyrosine) - dopamine, norepinephrine, epinephrine - and indolamines (derived from the amino acide tryptophan) - serotonin. Each monoamine has its specific brainstem nuclei that project onto cortical excitatory or inhibitory neurons. The monoamines and acetylcholine are also regarded as "neuromodulators".
Although we cannot study neuronal excitation directly in humans, several non-invasive techniques are available for the investigation of the neural correlates of information processing and its dysfunctions. These include techniques with high temporal (EEG, MEG) and high spatial resolution (fMRI). TMS is a method that allows testing brain-behaviour relationships through virtual lesions and also has potential clinical applications, for example in depression. These techniques can be complemented with neurochemical approaches, such as the measurement of NT metabolites in CSF, blood or urine, or radioligand imaging of NT receptors, in order to further the understanding of the molecular basis of information processing and its disturbance in mental disorders.
2.4 Revision and discussion questions
Describe a full cycle of electrochemical transmission starting with the generation of an action potential in the presynaptic neuron and ending with an action potential in the postsynaptic neuron.
What are the main biochemical classes of NTs and how does the human body get them?
What are the advantages and disadvantages of the non-invasive techniques for the investigation of human brain function?
2.5 Further reading:
(Kandel et al., 2000), Parts II and III
Figure legends :
Fig. 2.1: The reflex arc of the knee jerk illustrates basic principles of excitatory and inhibitory neuronal activity.
Fig. 2.2: The resting membrane potential.
Fig. 2.3: The action potential starts with depolarisation, followed by repolarisation and after-hyperpolarisation.
Fig. 2.4: The basic structure of amino acids consists of a carbon atom with a carboxylic acid group, an amino group and a side chain. The twenty amino acids that make up human proteins are all alpha amino acids, meaning that the same (alpha) carbon atom carries the carboxy- and amino groups. The negatively charged salts of amino acids that result from the loss of the proton of the carboxy-group are often termed with the suffix "ate", for example "glutamate".
Fig. 2.5: The neurotransmitter acetylcholine is an ester.
Fig. 2.6: Catecholamine biosynthesis, starting from the amino acid tyrosine. The respective enzymes are listed to the right of the arrows. The step to DOPA, catalysed by tyrosine hydroxylase, is rate limiting, meaning that a dysfunction of this enzyme would result in a deficit of catecholamines. DOPA decarboxylase is of clinical importance because inhibitors of this enzyme are added to DOPA in the therapy of Parkinson's disease in order to avoid conversion to dopamine in the periphery. Dopamine (DA) is further coverted to noradrenaline by addition of a hydroxyl group to its beta carbon atom, and noradrenaline to adrenaline/epinephrine (E) by addition of a methyl group. Noradrenaline is a member of the group of phenylethanolamines.
Fig. 2.7: Serotonin biosynthesis and metabolism. The first two steps are analogous to the conversion of tyrosine to dopamine. Serotonin (5-hydroxytryptamine, 5-HT) can then be metabolised through MAO and in a further step aldehyde dehydrogenase to 5-hydroxy-indolacetic acid (5-HIAA), which can be measured in blood or CSF as a surrogate marker of serotonergic activity. Another pathway leads from 5-HT to melatonin.
Fig. 2.8: Glutamic acid and GABA (a gamma amino acid) are the most important amino acid NTs in the brain. Their reaction is catalysed by glutamic acid decarboxylase (GAD).
Fig. 2.9 The P1 component of the visual ERP from a visual working memory task with one (black), two (green) or three (red) objects, recorded at the central occipital electrode Oz (over visual cortex). Patients with schizophrenia showed overall smaller amplitude and less load modulation than control participants. Note that positive deflections go down, following a convention in ERP research. Modified from (Haenschel et al., 2007), with kind permission of the American Medical Association.
Fig. 2.10: Temporal and spatial resolution of the most widely used non-invasive neuroimaging/ neurophysiology techniques. Adapted from(Sack and Linden, 2003). For an explanation see text.
Fig. 2.11: Example of fMRI mapping of brain correlates of psychopathology - activity in auditory cortex during auditory hallucinations (a) and while listening to speech (b) (adapted from (Dierks et al., 1999)).
DIERKS, T., LINDEN, D. E., JANDL, M., FORMISANO, E., GOEBEL, R., LANFERMANN, H. & SINGER, W. 1999. Activation of Heschl's gyrus during auditory hallucinations. Neuron, 22, 615-21.
FLOEL, A. & COHEN, L. 2006. Translational studies in neurorehabilitation: from bench to bedside. Cogn Behav Neurol, 19, 1-10.
FORD, J., KRYSTAL, J. & MATHALON, D. 2007. Neural synchrony in schizophrenia: from networks to new treatments. Schizophr Bull, 33, 848-52.
GEORGE, M. & ASTON-JONES, G. 2010. Noninvasive techniques for probing neurocircuitry and treating illness: vagus nerve stimulation (VNS), transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS). Neuropsychopharmacology, 35, 301-16.
HAENSCHEL, C., BITTNER, R. A., HAERTLING, F., ROTARSKA-JAGIELA, A., MAURER, K., SINGER, W. & LINDEN, D. E. 2007. Contribution of impaired early-stage visual processing to working memory dysfunction in adolescents with schizophrenia: a study with event-related potentials and functional magnetic resonance imaging. Arch Gen Psychiatry, 64, 1229-40.
KANDEL, E. R., SCHWARTZ, J. H. & JESSELL, T. M. 2000. Principles of neural science, New York ; London, McGraw-Hill, Health Professions Division.
LINDEN, D. E. & FALLGATTER, A. J. 2009. Neuroimaging in psychiatry: from bench to bedside. Front Hum Neurosci, 3, 49.
SACK, A. T. & LINDEN, D. E. 2003. Combining transcranial magnetic stimulation and functional imaging in cognitive brain research: possibilities and limitations. Brain Res Brain Res Rev, 43, 41-56.
THAKER, G. 2008. Neurophysiological endophenotypes across bipolar and schizophrenia psychosis. Schizophr Bull, 34, 760-73.