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

Noebels JL, Avoli M, Rogawski MA, et al., editors. Jasper's Basic Mechanisms of the Epilepsies [Internet]. 4th edition. Bethesda (MD): National Center for Biotechnology Information (US); 2012.

Cover of Jasper's Basic Mechanisms of the Epilepsies

Jasper's Basic Mechanisms of the Epilepsies [Internet]. 4th edition.

Show details

On the Basic Mechanisms of Infantile Spasms

and .

Author Information

Infantile spasms is an unusual seizure disorder. The seizures themselves are only a few seconds in duration and consist of brief flexion and/or extension jerking movements. EEG abnormalities are also unique. The ictal event is dominated by an electrodecremental response – a flattening of the EEG. An interictal event, called hypsarrhythmia, is seen only in this syndrome and consists of a chaotic mixture of large asynchronous slow waves and spikes. Spasms in some children are responsive to ACTH or vigabatrin but usually not to other anticonvulsants. Over 200 disparate clinical conditions are associated with this seizure disorder. This has prompted investigators to speculate that there must be a shared common pathway or mechanism that these conditions converge on to produce this unique epileptic syndrome. This chapter outlines the clinical features of this syndrome and lists a series of questions concerning its basic mechanisms. Finally, features of seven proposed animal models are reviewed. We recommend that minimal criteria be developed in order for a model to be considered useful in studying the basic mechanisms of infantile spasms. It is anticipated that the study of such validated models will usher in a far deeper understanding of this devastating seizure disorder of early childhood.

The epileptic encephalopathies of early childhood are a challenge for both clinicians and basic scientists. For the clinician these disorders are difficult to manage and most often respond inadequately to available therapies – particularly commonly used anticonvulsants. Indeed, disorders such as infantile spasms, Lennox-Gastaut syndrome, early infantile epileptic encephalopathy (EIEE or Otahara’s syndrome) and severe myoclonic epilepsy of infancy (Dravet’s syndrome) commonly lead to persistent drug-resistant seizures and very poor neurodevelopmental outcomes including mental retardation. Thus there has long been a need for research scientists to address the underlying cause of these disorders so they can be understood and rational treatments can be developed based on an understanding of the basic mechanisms. Indeed, in 2007 the NINDS at the National Institutes of Health (NIH-USA) included creation of animal models of these disorders in their Epilepsy Research Benchmarks to encourage study of their basic mechanisms (


From the perspective of basic scientists there have been numerous barriers that have hampered study of these “catastrophic” epilepsies of childhood. Of the early epileptic encephalopathies, infantile spasms is the most common and occurs in approximately 1 in every 3225 live births. Numerous reviews have been written that describe the clinical features of this syndrome 1–6. However, these clinical descriptions provide few direct clues as to how these seizures are produced. The spasms do not arise immediately at birth but usually begin between 3 and 12 months of age – with peak incidence around 6 months. The spasms themselves almost always disappear with time but are replaced by other types of refractory epilepsy – such as focal seizures or Lennox-Gastaut syndrome. This observation has led to the notion that the spasms may be an age-dependent manifestation of an underlying abnormality and as the brain matures the neuronal substrate for the seizures changes and the behavioral and electrographic manifestation of the abnormality changes as well. As the name, infantile spasms, implies the seizures that characterize this disorder are unusual and typically consist of very brief (a few seconds), bilaterally symmetric contractions of muscles of the extremities, neck and/or trunk, which result in brief flexion and/or extension jerking movements – leading originally to the use of the term spasm. The spasms can vary in their intensity from jerking movements that are very dramatic to very subtle and at times can be hardly noticeable. Within a child, the intensity of the jerks can also wax and wan in their intensity. While the spasms can occur as single isolated events, they more commonly are observed in clusters, which can number more than 100 over a few minutes. In addition, these seizures commonly occur during periods of sleep-wake transitions.

The EEG correlates of the behavioral spasms are equally unique and distinguish this syndrome from all other forms of epilepsy. Similar to the accompanying behaviors, the ictal complex of spasms is only a few seconds in duration. Figure 1 shows an example of an ictal complex. Most typically it begins as a generalized slow wave that is followed by an electrodecremental response (temporary flattening of the EEG) and a subsequent run of fast spiking activity 7. However, there are often variations on this theme, when each of the components of an ictal event can be absent. Such variations can occur between patients and can occur from time to time in the same patient 7. EEG recordings during the interictal period show equally unusual electrographic events. Indeed, a defining and diagnostic feature of infantile spasms is a neurophysiological event called, hypsarrhythmia 8. Figure 2 shows an example of this highly abnormal EEG activity from a patient. This electrographic pattern is most often recorded during non-rapid eye movement (non-REM) sleep and consists of a seemingly chaotic mixture of asynchronous very high voltage slow waves intermixed with frequent large multifocal spikes. Classically, hypsarrhythmia has been described as random, high voltage slow waves and spikes that occur asynchronously in all cortical areas and change from one instant to the next. However as with the ictal complexes, variations in this pattern have been described 9.

Figure 1. Ictal EEG discharges recorded in a human infant with infantile spasms and a rat that was infused in neocortex with TTX.

Figure 1

Ictal EEG discharges recorded in a human infant with infantile spasms and a rat that was infused in neocortex with TTX. Infusion was initiated in infancy and persisted for 28 days. The top two panels show the original recordings. In the lower panels, (more...)

Figure 2. Comparison of interictal hypsarrhythmic patterns recorded in a human with infantile spasms and a rat whose neocortex was infused with TTX.

Figure 2

Comparison of interictal hypsarrhythmic patterns recorded in a human with infantile spasms and a rat whose neocortex was infused with TTX. The montage for the rat and human are the same as in Figure 1. (Taken with permission from Lee et al. 2008).

To study the basic mechanisms of infantile spasms, a very significant challenge is reproducing the distinct clinical features of this syndrome - i.e. brief behavioral flexor/extensor spasms, and the signature ictal complexes and hypsarrhythmia in experimental systems. However, since the features of the syndrome are so unique, if researchers were actually able to faithfully reproduce features of infantile spasms in the laboratory, the opportunities would be quite significant for advances an understanding of its pathophysiology.

Another challenge in approaching infantile spasms is its highly variable etiology. Unlike temporal lobe epilepsy, where status epilepticus is thought to be a primary cause of this form of epilepsy and can be modeled in animals in both in vivo and in experimentally simpler in vitro models, infantile spasms are associated with a very large number (over 200) of clinically distinct conditions 3,10,11. Approximately 80% of patients are classified as having symptomatic infantile spasms, since they have a clearly identified underlying cause or are developmentally delayed. The remaining 20% have no identifiable cause and are referred to as cryptogenic. However, with advances in genetic, imaging and other diagnostic testing, this percentage is likely to decrease in coming years. Some of the broad categories associated with symptomatic infantile spasms are: 1) CNS infections such as meningitis, 2) developmental brain abnormalities or malformations such as cortical dysplasia, 3) hypoxic-ischemic encephalopathy and 4) genetic syndromes such as Down syndrome and Tuberous Sclerosis Complex. In terms of genetic causes, numerous single gene mutations, deletions or duplications have been linked to infantile spasms. These include: ARX, CDKL5 and MAG12 12–14. Interestingly, these have quite different functions in the brain which further amplifies the highly variable origins of infantile spasms.

The variable etiology of infantile spasms has lead to speculation on the existence of a common mechanism(s) shared by all patients even though the associated and presumed causative conditions are so different 3,15–18. Undoubtedly, it is quite remarkable that conditions as different as Down syndrome and meningitis can result in a seizure disorder with the same unique stereotyped clinical and electrographic features. How could a hypoxic-ischemic episode at birth, a mutation of a single gene such as TSC1, or a tumor all produce hypsarrhythmia and the ictal events of infantile spasms? Given that the product of more than 200 highly varied clinical conditions is the same (i.e. spasms), there must be a common final path that all these conditions converge on that produces this seizure disorder. There have been several suggestions as to what the shared mechanism might be and these will be discussed later in this chapter in concert with potential new animal models of infantile spasms.

Another challenge in understanding the origins of infantile spasms is identifying where in the brain these seizures are generated. The EEG abnormalities are recognized as being generalized events. However, similar to the generalized spike and wave discharges of absence seizures where thalamic nuclei are thought to play key synchronizing roles in these events, if researchers could identify the site(s) of spasm generation, this would greatly accelerate the discovery of underlying mechanisms. Based on clinical observations a general consensus has emerged in recent years that spasms likely result from abnormal interactions between cortical and brain stem neural networks 5,10,16,19. Observations that favor a cortical contribution are that some children with infantile spasms have cortical malformations or injuries and seizures in children can be eliminated by removing cortical abnormalities during epilepsy surgery. Moreover, PET imaging studies have reported regional hypometabolism in the cortex of a majority of children with this disorder 20–22. Hypsarrhythmia, since it is generalized across cortex also suggests significant abnormalities there. On the other hand, altered brain stem function is suggested by the occurrence of abnormal sleep patterns and the common occurrence of spasms during sleep-wake transitions 23. However, it is obvious even from this brief summary that there is a paucity of data directly implicating specific neural networks of the brain stem or cortex in seizure generation.

Finally, potential clues to the mechanisms underlying infantile spasms come from its unique profile of pharmacological responsiveness. As mentioned earlier, spasms in infants are usually unresponsive to conventional anticonvulsants. However, in 1958 spasms were reported to respond to treatment with adrenocorticotropic hormone (ACTH) 24. The rationale behind initiating this therapeutic strategy is unclear. However, since that time a number of clinical trials have been completed and it has been shown that, while not all children respond to ACTH, between 42–87% (depending on the study) of patients do respond by complete cessation of clinical spasms (both behavioral and EEG) and elimination of hypsarrhythmia on prolonged EEG recordings 25–28. The effects of ACTH are considered to be all-or-none and are in this respect unique in epilepsy therapeutics. Many have considered the complete elimination of hypsarrhythmia to be critical in order to prevent the continued progression of the disorder to other forms of epilepsy and impairments in cognition. Thus, an all-or none effect has been considered an important end-point in clinical trails – at least in the United States.

The site and mechanism of action of ACTH remains an enigma. In fact, the blood brain barrier is thought to be quite impermeable to ACTH so its action in the CNS has been questioned. However, high dosages of ACTH are often used therapeutically and it is thought by some that enough ACTH gets into the CNS to be effective 27. Alternatively, ACTH could be acting on the adrenal to stimulate the release of steroids which in turn could act on the brain to suppress spasms. There is evidence that steroids can be effective in suppressing spasms but in a head to head trial prednisone was found to be not as effective as high-dose ACTH 27. This has led some to conclude that ACTH may be acting centrally to suppress seizures. Yet, questions linger since prednisone or prednisolone is effective in some patients as the UKISS study has emphasized 28,29. Currently in the USA, ACTH therapy is considered “probably effective” in the treatment of infantile spasms by a joint commission of the AAP/AAN/CNS 30. Nonetheless, it is clear that current therapies are not ideal. Since ACTH, the recommended treatment can have significant side effects in some patients especially when high dosages are used.

It also remains a possibility that other treatments will be effective for subtypes of spasms – depending on their etiology. For instance vigabatrin appears to be quite effective in suppressing spasms in children with tuberous sclerosis complex (TSC) and has been used in Europe for well over a decade and was approved in the United States in 2009 for this purpose 31. However, it appears to be less effective in patients with other etiologies. In this regard, results of a UKISS study suggest that vigabatrin may be less effective than steroids 28. Much interest has been engendered by a recent preliminary report that suggests treatment of TSC patients with vigabatrin before the onset of seizures can be preventative 32.

In this regard, some reports have emphasized the importance of treating children - not necessarily before - but as soon as possible after the onset of spasms since results suggest that early treatments can lead to better long term outcomes including reductions in the proportion of patients with epilepsy and impaired cognition 33,34. It is thought that later treatments may still suppress spasms but may not be as effective in ameliorating the long-term sequelae of spasms. While these ideas remain controversial, they are in keeping with the notion that infantile spasms is a progressive disorder – akin to some other forms of epilepsy and that spasms themselves have consequences on normal brain development. However, it is important to note that even if the spasms are controlled by currently used therapies many infants will still develop epilepsy and/or cognitive deficits including autistic features regardless of the etiology of the condition (with the caveat that cryptogenic cases i.e, those without a clear pathology, usually have better outcomes). Whether this is because the child was treated too late in the course of the syndrome or the therapies were simply ineffective in blunting the progression of this disorder is unclear. However, until very recently the lack of animal models of infantile spasms has prevented a detailed examination of whether the syndrome can be progressive and the possible mechanisms underlying epileptogenesis.


Based on this review of some of the challenges associated in treating and understanding the pathophysiology of infantile spasms, there are many questions that need to be answered if we are to begin to understand the cellular and molecular mechanisms responsible for this seizure disorder and develop new and hopefully rational therapies for these children. The following summarizes some of the questions.

  1. How can one interpret the natural course of this seizure disorder in terms of its underlying mechanisms? Why do spasms occur in clusters during infancy? Is the disorder progressive? And if so what are the mechanisms underlying the dramatic changes that occur over time? Are there underlying mechanisms of epileptogenesis? Or is the precipitating mechanism(s) unchanging and the alterations (e.g. in seizure phenotype – infantile spasms to Lennox-Gastaut syndrome or subsequent focal seizures) that occur with time a reflection of a relatively static mechanism(s) interacting with the on-going development of the brain, its neurons and maturing networks? Or is it a combination of interactions between a progressive epileptogenesis and brain development?
  2. If there is a shared final common path between the 200 conditions associated with the disorder and the products – the behavioral spasms, associated EEG abnormalities and impaired cognition– what is it? If there are unique treatments for subtypes of infantile spasms – e.g. based on different etiologies: i.e. vigabatrin for TSC – does this mean that there are parallel paths for generating infantile spasms. Or in this instance is vigabatrin acting upstream before convergence on the final pathway?
  3. Why do some but not all infants with the same underlying condition (i.e. perinatal asphyxia or TSC) develop spasms? Are there additional contributing factors? Is there a genetic predisposition?
  4. Where in the brain do the spasms originate? Do they arise from cortex and/or brain stem as some suggest or somewhere else? And if so where in cortex and which brain stem nuclei are responsible? And importantly what neuronal subtypes in each area not only participate in spasm generation but initiate the seizures? For instance, are principal neurons the key participants and initiators and do different subpopulations of interneurons (some possibly dysfunctional) contribute in unique ways to spasms or hypsarrhythmia.
  5. What is the basic mechanisms underlying hypsarrhythmia? Where is it generated and which neuronal subtypes participate in its generation? Do hypsarrhythmia and the ictal complexes of spasms share common mechanisms or are they independent electrophysiological phenomenon? What is more detrimental for long-term outcomes: hypsarrhythmia or the spasms themselves?
  6. Why do spasms wax and wane in their intensity? What does this say about the underlying mechanisms? How well does behavioral spasm intensity correlate with alterations in ictal complexes recorded simultaneously? If they correlate or not can this provide clues to underlying mechanisms?
  7. Why do children with infantile spasms become cognitively impaired - many severely so? Do the spasms themselves contribute in any way to cognitive decline? What is the role of hypsarrhythmia in these developmental disabilities?
  8. What is the mechanism of action of ACTH therapy? And what can this tell us about the mechanisms responsible for spasm generation?
  9. How will new therapies be developed to control not only the spasms but prevent the development of other epileptic phenotypes and improve cognitive outcome? Will one drug type be sufficient or a cocktail to target different processes?
  10. As many as 50% of cases of infantile spasms are currently refractory to all available therapies – including ACTH. Does this suggest that there are different mechanisms for spasm generation in these children and will they require different types of treatments?


It has long been thought that an understanding of the basic mechanisms of infantile spasms would be accelerated by the creation of animal models that would permit a detailed examination of it underlying cellular and molecular causes. However, until recently this has proved to be a difficult undertaking for several reasons. The rational approach for the creation of animal models would be to reproduce one of the clinical conditions most commonly associated with the disorder in immature laboratory animals with the hope that these animals would develop a disorder mimicking what is seen in humans. For example, there are numerous animal models of neonatal hypoxic-ischemic encephalopathy, which have been studied for many years but there are no reports of these animals having a condition like infantile spasms – although some can develop seizures later in life 35. Perhaps this result, although disappointing, may not be unanticipated since not every child that suffers a hypoxic-ischemic episode will develop infantile spasms. Perhaps a more considered approach would be to choose a disorder where the frequency of spasms in very high. The genetic disorder TSC is a good example since the frequency of infantile spasms in these children can be as high as 40 – 50% 36. Considering the advances made in human and mouse molecular genetics over the past decade creating mutant mouse models that reproduce genetic disorders in children have become common place. And indeed several mouse models of TSC have been created and yet while some of these mice have severe epilepsy, a condition similar to infantile spasms has never been reported despite extensive experimental investigations of these mice 37,38. Such findings have lead some investigators to speculate that perhaps the rodent brain is simply not sufficiently advanced in evolutionary terms to reproduce the highly stereotyped neurological abnormalities that characterize the more complex human infant brain. However, this does not appear to be the case because in the last few years, 7 different laboratories have reported models of infantile spasms either in mice or rats. Although these are very early days in the characterization and study of all of these animal models, there is every reason to be hopeful that at least one or more of them will provide unparalleled opportunities for the study the basic mechanisms of this disorder.

Before reviewing the features of the animal models, it is important to discuss the criteria a model needs to meet in order to be considered a valid model of this syndrome. Unfortunately at this early stage in model characterization, there appears to be little consensus on what features of the human condition an animal model must possess for it to be consider a valid model. In 1996, a report from a NIH/NINDS workshop on animal models of pediatric epilepsy was published. Based on the discussions at the workshop by leaders in the field a list of criteria for an “ideal model of infantile spasms” was developed 39. These included:

  1. Unprovoked spasms early in postnatal development that occur in clusters
  2. EEG correlates of the ictal complex (including the electrodecremental response)
  3. Abnormal interictal EEG – hypsarrhythmia
  4. Response to clinically relevant treatments (e.g. ACTH and/or vigabatrine)
  5. Behavioral/cognitive sequelae

In this regard, the use of the term “ideal” needs to be emphasized. In numerous reviews and discussions of models, it is often commented that it is unlikely that any one animal model will meet all of these criteria but that does not prevent it from advancing our understanding of the mechanisms of infantile spasms 40. Perhaps a more useful list of criteria needs to be agreed upon by the scientific community that lists the minimal criteria that are required to model this disorder. And if these criteria are not meet it would not be considered a model of infantile spasms. Of course this would not mean that the model is not important in the study of the basic mechanisms of epilepsy – if the animals had other types of seizures – or in the study of the consequences of gene mutations in genetically engineered mouse that mimics gene mutations associated with human spasms. It would just mean that the model would not be useful in answering questions like those listed earlier concerning this syndrome. Recently one review mentioned minimum criteria consisting of: 1) An abnormal EEG; 2) Spontaneous seizures that begin in the postnatal period but not necessarily spasms and 3) Cognitive deficits 41 While another considered: 1) Seizures in infancy and 2) ACTH responsiveness as minimal criteria 42. Clearly, significant debate and discussion will be necessary to resolve such differences of opinion. However, ultimately the utility of any animal model will be dependent on its usefulness in being able to answer basic questions concerning the pathophysiological mechanisms of this disorder. It is hard to envision how an animal model would be useful in studying infantile spasms if it does not at least have behavioral spasms and the EEG correlate of ictal complexes observed clinically. It would also be preferable if the spasms were initiated in infancy. It may also be necessary to construct models of infantile spasms refractory to the current clinically used treatments in order to identify new approaches. In the reviews that follow, we will attempt to highlight the features of the proposed animal models and their advantages in terms of addressing pressing questions concerning the pathophysiology of this disorder.


The Conditional Deletion of Aristaless-related-homeobox gene (ARX) Mouse Model

The X-linked Arx gene encodes a transcription factor that is likely involved in early cortical development 43. Mutations of this gene can be associated with a variety of clinical phenotypes including mental retardation, epilepsy and the infantile spasm syndrome, X-linked lissencephaly with abnormal genitalia (XLAG) 41. Arx appears to be critical in neurodevelopmental processes such as progenitor cell proliferation and neuronal migration. Based on observations that patients with XLAG and mice with similar mutations of Arc have deficits in interneuron migration from the basal ganglion 43,44, Dobyns and colleagues hypothesized that infantile spasms – at least in XLAG - could be due to deficits in cortical interneuron function and thus coined the phrase “interneuronopathy” as the basis for infantile spasms in these patients 18. Indeed, the loss of interneurons from cortex could be a mechanism that many conditions converge on to produce infantile spasm. Thus “interneuronopathies” could be critical component of a “final common path” producing spasms.

To test the interneuronopathy hypothesis, Marsh and colleagues selectively eliminated ARX from interneurons arising from the basal ganglion by crossing mice carrying a floxed Arx allele with Dlx5/6CIG mice 45. Neuroanatomical studies have confirmed deficits in subpopulations of interneurons in the cortex. Importantly, EEG recordings have shown that male mice carrying this mutation have epilepsy. Hypsarrhythmia was not observed, However, spasm-like behaviors were observed in adulthood and these were reported to be correlated with a spike and slow wave complex followed by an electrodecrement on EEG recordings. Surprisingly, video-EEG recordings from younger mice did not show similar spasms but instead limbic seizures. There is no easy explanation for the late-onset of spasm-like events in this animal model. EEG recordings from infant mice can be challenging and the authors were only able to record for 24 hours in 3 mutant male mice. Perhaps more prolonged recordings would reveal the presence of earlier spasms in these mice. Nonetheless, the direct link between human genetic abnormalities and the creation of relevant mutant mice is a powerful approach. Discoveries made with this model and the development of experimental tools to study interneuron deficits could lead to testing the interneuronopathy hypothesis in other animal models of infantile spasms. The development of new technologies that allow routine video-EEG recordings over extended periods from infant mice would accelerate progress in characterizing such mouse models.

The Triple Repeat Expansion Model of ARX

This mouse model is based on the discovered link between mutations in the Arx gene and infantile spasms reviewed above. Since knock-out of Arx in mice results in perinatal lethality 43,44, Price and colleagues used a knock-in strategy to engineer a targeted expansion of the first polyalanine tract in Arx 46. This is the mutation very commonly associated with infantile spasms in humans 47. Multiple neurobehavioral phenotypes were discovered in the resultant mice. As anticipate, based on Arx expression in the precursors of neocortical interneurons, the loss of subpopulations of interneurons was observed. Behavioral spasms were observed in both control and mutant mice between postnatal days 7 and 11. Two types were described: low and high amplitude spasm-like movements. There were no differences between the frequency of the low amplitude myoclonic events in mutant mice compared to controls. However, there were twice as many high amplitude movements in mutant mice. While EEG recordings on postnatal days 7 –11 were not possible, it was reported that EEG recordings on postnatal days 16–20 showed a high amplitude slow wave followed by attenuation of the background EEG activity concurrent with brief myoclonic jerks. In older mice (24 to 70 days of age), spasms were not observed instead limbic motor seizures as well as 6 sec spike wave bursts - accompanied by behavioral arrest – were seen. Hysarrhythmia was not reported in this mouse model. However, cognitive impairment and abnormal social interactions suggestive of an autism phenotype were observed.

Thus this animal model has several desirable features for a model of infantile spasms. Again quality EEG recordings from infant mice are a significant barrier to a full description of the spasm phenotype in infant mice. It would be interesting to know what the EEG correlates are of the high-amplitude spasm-like movements in control mice. Study of this mouse model in concert with the ARX conditional knock-out described above should lead to a better understanding of the role interneuron loss may play in the pathophysiology of seizures and in particular infantile spasms. Study of the origins of the neurobehavioral abnormalities in these mice could address the roles of interneuron loss, and/or seizures play in these cognitive and autistic phenotypes that are often seen in patients with a history of infantile spasms.

The NMDA model

Intraperitoneal injections of the NMDA, a glutamate receptor agonist, in rats on postnatal days 12 to 18 produce hyperflexion and tonic spasms of the body 48,49. These have been reported to be concurrent with a slow wave and subsequent electrodecrement on EEG recordings from cortex – although other neurophysiological discharges can occur 50. Recently, this model has been modified by prenatal treatment with betamethasone on gestational day 15 50. Although the rationale underlying this experimental strategy has not been fully articulated, it appears that it is an attempt to mimic prenatal stress. This treatment sensitized rats to NMDA since they responded at lower doses of the drug. In addition, rats became responsive to ACTH in that the latency to onset of NMDA-induced spasms was increased. Since the seizures were not blocked, this is not fully in keeping with the all-or-none effects of ACTH seen clinically. Additionally, ACTH usually takes days to become effective in children and commonly does not act immediately as reported for this model. The draw back of this model is that the spasms are evoked by NMDA and do not occur spontaneously. However, as with other models described below its utility could be to rapidly screen drugs for efficacy against spasms - assuming that the scientific community as a whole concludes that the spasms evoked by NMDA are relevant to those that occur clinically.

The Down syndrome –Ts65Dn Mouse Model

The incidence of epilepsy in Down syndrome is significant -approximately 8% and a third of these patients have infantile spasms. The Ts65Dn mouse, which is segmentally trisomic for the distal end of chromosome 16, is an accepted model of Down syndrome 51. While spontaneous seizures have never been reported in these mice (with the caveat that audiogenic seizures can be triggered in these animals by intense noise), these mice have increased GABAB receptor-mediated currents, which likely result from overexpression of the GIRK2 channel due to the extra copy of this gene on chromosome 16. Since GABAB receptor agonists have convulsant properties in rodents, Cortez and colleagues thought it is possible that treatment with these drugs might induce seizures and possibly even spasms in this mouse model 52. Results showed that such treatments produce absence seizures in control mice but clusters of extensor spasms in the Ts65Dn mouse. Each spasm in a cluster appeared to be associated with burst discharges on EEG recordings which were separated by a marked attenuation in cortical activity. This type of activity was observed from 1 week to 2 months postnatally. Mice were also used to screen the effectiveness of a number of anticonvulsants. ACTH, valproic acid, ethosuximide, and vigabatrin were all reported effective in suppressing spasms. Like the clinically-reported effectiveness of vigabatrin in TSC patients, results from this mouse model suggest that infantile spasms in Down patients may have a unique pharmacological responsiveness – which could include more conventional anticonvulsants. However, it would be important to know if the enhanced GABAB receptor function also occurs in Down patients. The advantage of this mouse model is it ease of use in drug screening. A draw back is that seizures are not spontaneous and are evoked by drug treatment. The consequences of the spasms in terms of epilepsy or neurobehavioral deficits have not yet been reported.

The CRH Hypothesis

Since ACTH is effective is suppressing infantile spasms, Baram and colleagues have pursued the molecular mechanisms by which it acts 53. It is known from neuroendocrinology studies that systemic ACTH treatment can result through classic negative feedback systems to downregulate the expression of corticotrophin releasing hormone (CRH) in the central nervous system 54. This can be mediated in part by steroids that are released from the adrenal or by ACTH itself. Importantly, CRH has been shown to be a potent convulsant drug and is particularly effective in infant rats 55. Thus a reasonable hypothesis has been proposed that ACTH acts in infantile spasms to suppress CRH expression and thereby decrease neuronal excitability. This hypothesis has been extended in suggesting that prenatal or perinatal stress may be a common factor in etiology of infantile spasms and may be important part of a final common path that leads to the generation of spasms 17. It is likely that many infants with spasms have undergone some form of stress earlier in their lives. Stress is thought to increase CRH levels in the brain and ACTH treatment may act to normalize CRH levels. While investigators continue to actively pursue these ideas, there are several shortcomings that limit enthusiasm for them at this time. It is well known that CRH produces severe limbic motor seizures in infant rats and have never been reported to produce spasm-like behaviors. Moreover EEG recordings during CRH-induced seizures show rhythmic sharp activity consistent with seizures arising from the limbic system 56. Neither ictal complexes nor hypsarrhythmia have been reported in recordings from CRH treated mice. Thus despite the strengths of the rationale underlying the CRH/stress hypothesis, the link between infantile spasms and CRH remains strictly hypothetical at this time. However, it is possible that treatment of the brain with exogenous CRH is unable to mimic the effects of its release locally from neurons. And if a more physiological condition could be produced where the effects of endogenous CRH were evoked, spasm-type seizures and EEG correlates may be seen.

The Multiple Hit Model

Based on the notion reviewed earlier that infantile spasms results from abnormal interactions between cortical and subcortical networks – particularly those in the brain stem, Scantlebury and colleagues developed a multiple-hit model of infantile spasms 57. In this model investigators have simultaneously produced pathological abnormalities in cortical and subcortical brain areas of infant rats. To accomplish this on postnatal day 3 Doxorubicin is injected intraventricularly and lipopolysaccharide is injected into the right cerebral cortex. Doxorubicin is an anthracycline chemotherapeutic agent that injures and kills neurons through oxidative stress, while lipopolysaccharide is a toxin released from gram-negative bacteria which is able to damage white matter and activate inflammatory cells. Two days after these treatments, rats are injected (ip) with p-chlorophenylalanine to deplete serotonin in the brain by blocking the synthesizing enzyme, tryptophan hydroxylase. This is used in an attempt to increase neuronal excitability. Rats display clusters of flexor or extensor spasms from postnatal day 4 to 13 and other seizure types begin to emerge after postnatal day 9 that resemble limbic motor seizures. The EEG correlates of spasms are complex. Similar to humans, spasms have electrographic correlates in only 49% of events. Electrodecremental responses (i.e. attenuation of background activity) were observed in 27% of spasms that had electrographic correlates. The remaining 73% of spasms with electrographic correlates coincided with spike and sharp wave discharges, and/or fast rhythmic activity. The interictal EEG was also abnormal and consisted of high amplitude spikes and slow waves. ACTH treatment was found not be effective in this model and vigabatrine suppressed seizures but only transiently on postnatal day 5. The rats also displayed numerous neurobehavioral deficits including learning and memory deficits, stereotypes and impaired socialization suggesting a possible autism-like phenotype. As would be expected from the treatments used, neuroanatomical examination of the brain of these rats showed marked neuropathologies including a reduction in the thickness of the right cortex and diffuse damage to the corpus callosum, striatum, hippocampus, thalamus among other structures. The advantage of this model is that it may reproduce the drug-resistant symptomatic refractory spasms that are encountered clinically and therefore could be useful in screening for new treatments for this patient population.

The TTX Model of Focal Neocortical Inactivation

In contrast to many of the models of infantile spasms, the rationale behind using local infusion of tetrodotoxin (TTX) in the brain to induce infantile spasms in rats pups is not immediately apparent. Indeed the experimentally strategy was initially employed to examine the role of neuronal activity in postnatal development of the hippocampus not to develop a model of childhood epilepsy 58,59. Unexpectantly however, rats treated in this way developed very brief seizures and later it was discovered that these seizures had a very close resemblance to infantile spasms 60. In this model, TTX (10 μM) is chronically (for 28 days) infused by an osmotic minipump either into the hippocampus or neocortex of rats beginning on postnatal day 10–12. Identical infusions only a few days later do not produce spasms or seizures. Two-deoxyglucose studies showed that the treatment produces a region of neuronal inactivation approximately 2.5 cm in diameter centered on the tip of the infusion cannula. Thus this treatment may mimic the regional neocortical hypometabolism reported from PET studies of infantile spasms patients 20–22. Initial behavioral observations showed that seizures begin sometime before postnatal day 21 and consist of brief (1–2 sec) flexion or extensions of the trunk and/or forelimbs. Multichannel video-EEG recordings show that behavioral spasms are concurrent with a generalized ictal complex that is virtually identical to that seen in humans. Figure 1 compares an ictal complex from a rat to that recorded from a patient. The complex begins with a large slow wave followed by an electrodecremental response on which runs of spikes are subsequently superimposed. The spasms frequently occur in clusters of as many as 50 spasms and most often occur during sleep-wake transitions. Interictally, patterns of hypsarrhythmia are recorded in the majority of animals having spasms and these have a frequency spectrum very similar to that recorded in human infants. Figure 2 shows an example of hypsarrhythmia from a rat in comparison to that from a human. The spasms usually persist for 2 months but eventually rats develop severe limbic motor seizures by 6 months of age. Pharmacological responsiveness to ACTH and vigabatrin have not been evaluated nor has learning and memory deficits been fully characterized in these animals. The advantage of this model is the robust nature of the EEG recordings and how closely they mimic those of humans. Thus, the model could be exploited to study the cellular neurophysiology of infantile spasms. Indeed, a recent study using high digital sampling rates followed by compressed spectral array analysis and band-pass filtering showed that high frequency EEG activity occurs throughout ictal events in these animals and prominently during the electrodecrement that is recorded using conventional EEG filter settings 61. Studies are needed to determine the age of onset of spasms in the animals and to characterize their evolution to the condition seen on postnatal day 21 – a process possibly akin to epileptogenesis.

Recently, Frost and Hrachovy 3 proposed the “developmental desynchronization model” for infantile spasms. Like the “interneuropathy” and the “CRH/stress” models discussed above, this an attempt to describe a common pathophysiological mechanism that could result from the myriad of etiologies associated with this disorder. Put simply, this model suggests that infantile spasms result from a temporal desynchronization of two or more on-going neurodevelopmental processes. The desynchronization could be caused by injury to the brain or a mutation in a gene important for neuronal development among the many other etiologies for this disorder. At the systems, cellular and molecular levels, normal brain development is though to be dependent on reinforcing interactions between networks of cells and networks of molecules. It is hypothesized that when one developing system is delayed or accelerated relative to its maturational partners, a developmental desychronization results which produces infantile spasms. Features of the TTX model are consistent with the concept of developmental desynchronization since it is well known that neuronal activity plays an important role in the normal course of neuron and network maturation 62. By suppressing neuronal activity with TTX in a portion of the neocortex, one would expect that the maturation of neurons in that region would be altered and possibly delayed compared to its synaptic partners at distant sites. Thus the development of TTX exposed cells should be out of sync with unexposed cells that continue developing at a normal pace. At this time, it is unclear how such a maturational desynchronization could lead to infantile spasms but different types of desynchonization resulting from different etiologies would have to converge on a “critical system” which would initiate the pathophysiological mechanisms in a shared or common pathway leading to spasms in children. Validated animal models could be important tools used to support or refute such theoretical models of pathogenesis and should help identify hypothesized critical systems and common pathway – if they exist.


We are at a very early stage in studying the basic mechanisms of infantile spasms. However, features of several promising animal models have been described and more may be forthcoming in the near future. Technical challenges are substantial for studying seizures in infant rodents but not insurmountable. The basic science and clinical communities need to evaluate each of the proposed animal models and develop minimal criteria for what they consider a valid animal model of this disorder. Once validated, models should provide an unprecedented opportunity to study the mechanisms responsible for this catastrophic epilepsy and hopefully provide new avenues for effective therapies.


Dr Moshe has received research support from NIH: R01 NS20253 (PI), R01-NS43209 (Investigator),2UO1-NS45911 (Investigator), and the Heffer Family Foundation and Focus Autism/Segal Family Foundation. He is Associate Editor of Neurobiology of Disease and in the Editorial Board of Epileptic Disorders, Brain and Development and Physiological Research. He has received a consultancy fee from Eisai and a speaker1s fee from GSK.

Dr Swann's laboratory is supported by grants from the NIH (NS18309, NS062992), the Vivian L. Smith Foundation and an investigator initiated grant from Questcor.


Riikonen R. Epidemiological data of West syndrome in Finland. Brain Dev. 2001;23:539–541. [PubMed: 11701251]
Hrachovy RA. West’s syndrome (infantile spasms). Clinical description and diagnosis. Adv Exp Med Biol. 2002;497:33–50. [PubMed: 11993738]
Frost JD Jr, Hrachovy RA. Pathogenesis of infantile spasms: a model based on developmental desynchronization. J Clin Neurophysiol. 2005;22:25–36. [PubMed: 15689710]
Shields WD. Infantile Spasms: Little Seizures, BIG Consequences. Epilepsy Curr. 2006;6:63–69. [PMC free article: PMC1464162] [PubMed: 16761063]
Dulac O, Soufflet C, Chiron C, et al. What is West syndrome. Int Rev Neurobiol. 2002;49:1–22. [PubMed: 12040888]
Lacy JR, Penry JK. Infantile Spasms. New York: Raven Press; 1976.
Kellaway P, Hrachovy RA, Frost JD Jr, et al. Precise characterization and quantification of infantile spasms. Ann Neurol. 1979;6:214–218. [PubMed: 534418]
Gibbs EL, Fleming MM, Gibbs FA. Diagnosis and prognosis of hypsarhythmia and infantile spasms. Pediatrics. 1954;13:66–73. [PubMed: 13133555]
Hrachovy RA, Frost JD Jr, Kellaway P. Hypsarrhythmia: variations on the theme. Epilepsia. 1984;25:317–325. [PubMed: 6539199]
Frost JD Jr, Hrachovy RA. Infantile Spasms. Boston: Kluber Academic Publishers; 2003.
Baram TZ. Pathophysiology of massive infantile spasms: perspective on the putative role of the brain adrenal axis. Ann Neurol. 1993;33:231–236. [PMC free article: PMC3786774] [PubMed: 8388675]
Kitamura K, Yanazawa M, Sugiyama N, et al. Mutation of ARX causes abnormal development of forebrain and testes in mice and X-linked lissencephaly with abnormal genitalia in humans. Nat Genet. 2002;32:359–369. [PubMed: 12379852]
Weaving LS, Christodoulou J, Williamson SL, et al. Mutations of CDKL5 cause a severe neurodevelopmental disorder with infantile spasms and mental retardation. Am J Hum Genet. 2004;75:1079–1093. [PMC free article: PMC1182143] [PubMed: 15492925]
Marshall CR, Young EJ, Pani AM, et al. Infantile spasms is associated with deletion of the MAGI2 gene on chromosome 7q11.23-q21.11. Am J Hum Genet. 2008;83:106–111. [PMC free article: PMC2443840] [PubMed: 18565486]
Dulac O, Chiron C, Robain O, et al. Infantile spasms: a pathophysiological hypothesis. Semin Pediatr Neurol. 1994;1:83–89. [PubMed: 9422224]
Lado FA, Moshe SL. Role of subcortical structures in the pathogenesis of infantile spasms: what are possible subcortical mediators. Int Rev Neurobiol. 2002;49:115–140. [PubMed: 12040889]
Baram TZ. Models for infantile spasms: an arduous journey to the Holy Grail. Ann Neurol. 2007;61:89–91. [PMC free article: PMC2917175] [PubMed: 17315207]
Kato M, Dobyns WB. X-linked lissencephaly with abnormal genitalia as a tangential migration disorder causing intractable epilepsy: proposal for a new term, “interneuronopathy” J Child Neurol. 2005;20:392–397. [PubMed: 15921244]
Chugani HT, Shewmon DA, Sankar R, et al. Infantile spasms: II. Lenticular nuclei and brain stem activation on positron emission tomography. Ann Neurol. 1992;31:212–219. [PubMed: 1575460]
Chugani HT, Shields WD, Shewmon DA, et al. Infantile spasms: I. PET identifies focal cortical dysgenesis in cryptogenic cases for surgical treatment. Ann Neurol. 1990;27:406–413. [PubMed: 2353794]
Metsahonkala L, Gaily E, Rantala H, et al. Focal and global cortical hypometabolism in patients with newly diagnosed infantile spasms. Neurol. 2002;58:1646–1651. [PubMed: 12058093]
Itomi K, Okumura A, Negoro T, et al. Prognostic value of positron emission tomography in cryptogenic West syndrome. Dev Med Child Neurol. 2002;44:107–111. [PubMed: 11848106]
Hrachovy RA, Frost JD Jr, Kellaway P. Sleep characteristics in infantile spasms. Neurol. 1981;31:688–693. [PubMed: 6264348]
Sorel L, Dusaucy-Bauloye A. Findings in 21 cases of Gibbs’ hypsarrhythmia; spectacular effectiveness of ACTH. Acta Neurol Psychiatr Belg. 1958;58:130–141. [PubMed: 13532578]
Hrachovy RA, Frost JD Jr, Kellaway P, et al. Double-blind study of ACTH vs prednisone therapy in infantile spasms. J Pediatr. 1983;103:641–645. [PubMed: 6312008]
Hrachovy RA, Frost JD Jr, Glaze DG. High-dose, long-duration versus low-dose, short-duration corticotropin therapy for infantile spasms. J Pediatr. 1994;124:803–806. [PubMed: 8176573]
Baram TZ, Mitchell WG, Tournay A, et al. High-dose corticotropin (ACTH) versus prednisone for infantile spasms: a prospective, randomized, blinded study. Pediatrics. 1996;97:375–379. [PMC free article: PMC3100715] [PubMed: 8604274]
Lux AL, Edwards SW, Hancock E, et al. The United Kingdom Infantile Spasms Study comparing vigabatrin with prednisolone or tetracosactide at 14 days: a multicentre, randomised controlled trial. Lancet. 2004;364:1773–1778. [PubMed: 15541450]
Kossoff EH, Hartman AL, Rubenstein JE, et al. High-dose oral prednisolone for infantile spasms: an effective and less expensive alternative to ACTH. Epilepsy Behav. 2009;14:674–676. [PubMed: 19435579]
Mackay MT, Weiss SK, ms-Webber T, et al. Practice parameter: medical treatment of infantile spasms: report of the American Academy of Neurology and the Child Neurology Society. Neurol. 2004;62:1668–1681. [PMC free article: PMC2937178] [PubMed: 15159460]
Willmore LJ, Abelson MB, Ben-Menachem E, et al. Vigabatrin: 2008 update. Epilepsia. 2009;50:163–173. [PubMed: 19230067]
Jozwiak S, Domanska-Pakiela D, Kotulska K, et al. Treatment before seizures: new indications for antiepileptic therapy in children with tuberous sclerosis complex. Epilepsia. 2007;48:1632–1634. [PubMed: 17692054]
Lombroso CT. A prospective study of infantile spasms: clinical and therapeutic correlations. Epilepsia. 1983;24:135–158. [PubMed: 6299719]
Kivity S, Lerman P, Ariel R, et al. Long-term cognitive outcomes of a cohort of children with cryptogenic infantile spasms treated with high-dose adrenocorticotropic hormone. Epilepsia. 2004;45:255–262. [PubMed: 15009227]
Kadam SD, White AM, Staley KJ, et al. Continuous electroencephalographic monitoring with radio-telemetry in a rat model of perinatal hypoxia-ischemia reveals progressive post-stroke epilepsy. J Neurosci. 2010;30:404–415. [PMC free article: PMC2903060] [PubMed: 20053921]
Chu-Shore CJ, Major P, Camposano S, et al. The natural history of epilepsy in tuberous sclerosis complex. Epilepsia. 2010;51:1236–1241. [PMC free article: PMC3065368] [PubMed: 20041940]
Zeng LH, Xu L, Gutmann DH, et al. Rapamycin prevents epilepsy in a mouse model of tuberous sclerosis complex. Ann Neurol. 2008;63:444–453. [PMC free article: PMC3937593] [PubMed: 18389497]
Meikle L, Pollizzi K, Egnor A, et al. Response of a neuronal model of tuberous sclerosis to mammalian target of rapamycin (mTOR) inhibitors: effects on mTORC1 and Akt signaling lead to improved survival and function. J Neurosci. 2008;28:5422–5432. [PMC free article: PMC2633923] [PubMed: 18495876]
Stafstrom CE, Moshe SL, Swann JW, et al. Models of pediatric epilepsies: strategies and opportunities. Epilepsia. 2006;47:1407–1414. [PubMed: 16922889]
Stafstrom CE. Infantile spasms: a critical review of emerging animal models. Epilepsy Curr. 2009;9:75–81. [PMC free article: PMC2685885] [PubMed: 19471616]
Marsh ED, Golden JA. Developing an animal model for infantile spasms: pathogenesis, problems and progress. Dis Model Mech. 2009;2:329–335. [PMC free article: PMC2953562] [PubMed: 19553693]
Stafstrom CE, Holmes GL. Infantile spasms: criteria for an animal model. Int Rev Neurobiol. 2002;49:391–411. [PubMed: 12040904]
Kitamura K, Yanazawa M, Sugiyama N, et al. Mutation of ARX causes abnormal development of forebrain and testes in mice and X-linked lissencephaly with abnormal genitalia in humans. Nat Genet. 2002;32:359–369. [PubMed: 12379852]
Colombo E, Collombat P, Colasante G, et al. Inactivation of Arx, the murine ortholog of the X-linked lissencephaly with ambiguous genitalia gene, leads to severe disorganization of the ventral telencephalon with impaired neuronal migration and differentiation. J Neurosci. 2007;27:4786–4798. [PMC free article: PMC4916654] [PubMed: 17460091]
Marsh E, Fulp C, Gomez E, et al. Targeted loss of Arx results in a developmental epilepsy mouse model and recapitulates the human phenotype in heterozygous females. Brain. 2009;132:1563–1576. [PMC free article: PMC2685924] [PubMed: 19439424]
Price MG, Yoo JW, Burgess DL, et al. A triplet repeat expansion genetic mouse model of infantile spasms syndrome, Arx(GCG)10+7, with interneuronopathy, spasms in infancy, persistent seizures, and adult cognitive and behavioral impairment. J Neurosci. 2009;29:8752–8763. [PMC free article: PMC2782569] [PubMed: 19587282]
Poirier K, Eisermann M, Caubel I, et al. Combination of infantile spasms, non-epileptic seizures and complex movement disorder: a new case of ARX-related epilepsy. Epilepsy Res. 2008;80:224–228. [PubMed: 18468866]
Mares P, Velisek L. N-methyl-D-aspartate (NMDA)-induced seizures in developing rats. Brain Res Dev Brain Res. 1992;65:185–189. [PubMed: 1572063]
Kabova R, Liptakova S, Slamberova R, et al. Age-specific N-methyl-D-aspartate-induced seizures: perspectives for the West syndrome model. Epilepsia. 1999;40:1357–1369. [PubMed: 10528930]
Velisek L, Jehle K, Asche S, et al. Model of infantile spasms induced by N-methyl-D-aspartic acid in prenatally impaired brain. Ann Neurol. 2007;61:109–119. [PubMed: 17315208]
Salehi A, Faizi M, Belichenko PV, et al. Using mouse models to explore genotype-phenotype relationship in Down syndrome. Ment Retard Dev Disabil Res Rev. 2007;13:207–214. [PubMed: 17910089]
Cortez MA, Shen L, Wu Y, et al. Infantile spasms and Down syndrome: a new animal model. Pediatr Res. 2009;65:499–503. [PubMed: 19190545]
Brunson KL, vishai-Eliner S, Baram TZ. ACTH treatment of infantile spasms: mechanisms of its effects in modulation of neuronal excitability. Int Rev Neurobiol. 2002;49:185–197. [PMC free article: PMC3092432] [PubMed: 12040892]
Brunson KL, Khan N, Eghbal-Ahmadi M, et al. Corticotropin (ACTH) acts directly on amygdala neurons to down-regulate corticotropin-releasing hormone gene expression. Ann Neurol. 2001;49:304–312. [PMC free article: PMC2849730] [PubMed: 11261504]
Baram TZ, Hirsch E, Snead OC III, et al. Corticotropin-releasing hormone-induced seizures in infant rats originate in the amygdala. Ann Neurol. 1992;31:488–494. [PMC free article: PMC3153947] [PubMed: 1596084]
Baram TZ, Schultz L. Corticotropin-releasing hormone is a rapid and potent convulsant in the infant rat. Brain Res Dev Brain Res. 1991;61:97–101. [PMC free article: PMC3153946] [PubMed: 1914160]
Scantlebury MH, Galanopoulou AS, Chudomelova L, et al. A model of symptomatic infantile spasms syndrome. Neurobiol Dis. 2009
Galvan CD, Hrachovy RA, Smith KL, et al. Blockade of neuronal activity during hippocampal development produces a chronic focal epilepsy in the rat. J Neurosci. 2000;20:2904–2916. [PubMed: 10751443]
Galvan CD, Wenzel JH, Dineley KT, et al. Postsynaptic contributions to hippocampal network hyperexcitability induced by chronic activity blockade in vivo. Eur J Neurosci. 2003;18:1861–1872. [PubMed: 14622219]
Lee CL, Frost JD Jr, Swann JW, et al. A new animal model of infantile spasms with unprovoked persistent seizures. Epilepsia. 2008;49:298–307. [PubMed: 17941850]
Frost JD Jr, Lee CL, Hrachovy RA, et al. High frequency EEG activity associated with ictal events in an animal model of infantile spasms. Epilepsia. 2011;52:53–62. [PMC free article: PMC3233379] [PubMed: 21204817]
Katz LC, Shatz CJ. Synaptic activity and the construction of cortical circuits. Science. 1996;274:1133–1138. [PubMed: 8895456]
Copyright © 2012, Michael A Rogawski, Antonio V Delgado-Escueta, Jeffrey L Noebels, Massimo Avoli and Richard W Olsen.

All Jasper's Basic Mechanisms of the Epilepsies content, except where otherwise noted, is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported license, which permits copying, distribution and transmission of the work, provided the original work is properly cited, not used for commercial purposes, nor is altered or transformed.

Bookshelf ID: NBK98133PMID: 22787673


Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Similar articles in PubMed

See reviews...See all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...