Human Molecular Genetics

Human Molecular Genetics Advance Access originally published online on November 17, 2004
Human Molecular Genetics 2005 14(2):205-220; doi:10.1093/hmg/ddi016

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 14/2/205    most recent ddi016v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (48) Request Permissions Google Scholar Articles by Moretti, P. Articles by Zoghbi, H. Y. Search for Related Content PubMed PubMed Citation Articles by Moretti, P. Articles by Zoghbi, H. Y. Social Bookmarking          What's this?

Human Molecular Genetics, Vol. 14, No. 2 © Oxford University Press 2005; all rights reserved

Abnormalities of social interactions and home-cage behavior in a mouse model of Rett syndrome

Paolo Moretti^1,2,, J. Adriaan Bouwknecht^1,,, Ryan Teague¹, Richard Paylor^1,3 and Huda Y. Zoghbi^1,2,3,4,5,*

¹Department of Molecular and Human Genetics, ²Department of Neurology, ³Department of Neuroscience, ⁴Department of Pediatrics and ⁵Howard Hughes Medical Institute, Baylor College of Medicine, Houston, TX 77030, USA

^* To whom correspondence should be addressed. Tel: +1 7137986558; Fax: +1 7137988728; Email: hzoghbi{at}bcm.tmc.edu

Received July 28, 2004; Accepted November 4, 2004

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Rett syndrome (RTT) is an autistic spectrum disorder with a known genetic basis. RTT is caused by loss of function mutations in the X-linked gene MECP2 and is characterized by loss of acquired motor, social and language skills in females beginning at 6–18 months of age. MECP2 mutations also cause non-syndromic mental retardation in males and females, and abnormalities of MeCP2 expression in the brain have been found in autistic spectrum disorders. We studied home-cage behavior and social interactions in a mouse model of RTT (Mecp2^308/Y) carrying a mutation similar to common RTT causing alleles. Young adult mutant mice showed abnormal home-cage diurnal activity in the absence of motor skill deficits. Nesting, a phenotype related to social behavior, and social interactions were both impaired in these animals. Mecp2^308/Y mice showed deficits in nest building and decreased nest use. Although there were no differences in aggression or exploration of novel inanimate stimuli, mutant mice took less initiative and were less decisive approaching unfamiliar males and spent less time in close vicinity to them in several social interaction paradigms. The abnormalities of diurnal activity and social behavior in Mecp2^308/Y mice are reminiscent of the sleep/wake dysfunction and autistic features of RTT. These data suggest that MECP2 regulates the expression and/or function of genes involved in social behavior. The study of Mecp2^308/Y mice will allow the identification of the molecular basis of social impairment in RTT and related autistic spectrum disorders.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Classic Rett syndrome (RTT, MIM 312750), a leading cause of mental retardation with autistic features in females, is an X-linked disorder caused by mutations in MECP2, the gene encoding methyl-CpG binding protein 2 (MeCP2) (1). In its classic form, the disorder is characterized by a period of normal development followed by loss of acquired cognitive, motor and social skills. Affected girls achieve normal developmental milestones until 6–18 months of age, when they fail to attain new skills and begin a period of regression. Patients develop receptive and expressive aphasia, and manifest a constellation of characteristic abnormalities of motor function. These include loss of purposeful hand movements, apraxia and ataxia of gait, spasticity and manifestations of extrapyramidal dysfunction (2). Girls with RTT syndrome also display a range of very distinct behavioral manifestations. For instance, hand stereotypies—repetitive, involuntary stereotyped movements such as hand-wringing, clapping or waving—replace the purposeful use of the upper extremities. Normal breathing is replaced during the waking hours by hyperventilation and breath holding spells (2). Sleep/wake patterns also become disrupted and the patients show decreased amounts of rapid-eye-movement sleep (3,4). Among the behavioral features of RTT, symptoms of autism deserve special attention. RTT and autism have long been recognized to have common clinical manifestations (2,5), to the point that before recognition of RTT as a distinct disorder, Rett syndrome patients were regarded as autistic (6–9). According to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), RTT is one of the five sub-types of pervasive developmental disorder, together with classic autism and Asperger disorder (10). Great interest exists in the elucidation of the causes and pathogenesis of autism. A 556% increase in pediatric prevalence of autism has been reported between 1991 and 1997 (11), to a prevalence higher than spina bifida, cancer or Down syndrome (12). Although autism is recognized to be the common endpoint of neurological dysfunction of varying etiologies, common disease mechanisms may underlie the phenotypes shared by RTT and autism, and advances in our understanding of RTT may also shed some light into the pathogenesis of autism.

In addition to classic RTT, mutations in the MECP2 gene have been identified in a wide spectrum of neurological phenotypes. MECP2 mutations have been found in females with infantile autism, mild intellectual impairment and preserved speech RTT variants, and also in normal individuals (13–16). MECP2 mutations have also been found in several males with phenotypes including fatal neonatal encephalopathy, a constellation of severe mental retardation, seizures, tremor, ataxia and psychiatric manifestations (13). A major determinant of phenotype severity in females is the pattern of X-chromosome inactivation (generally skewed favoring the X-chromosome with the normal allele in mildly affected individuals and random in patients with classic presentation) (17). However, the location and type (missense versus non-sense) of the mutation also appear to play a role, especially in males, due to the presence of a single X-chromosome (17). Males with XXY karyotype who, like females, are mosaic for the normal and mutant X-chromosome, have disease manifestations very similar to classic RTT females (18). These findings suggest the potential for shared molecular mechanisms between male and female patients. This also implies that the study of affected males in human and animal models may facilitate the unraveling of RTT pathogenesis by eliminating variables found in females due to unpredictable patterns of X-chromosome inactivation. We have modeled RTT in the mouse by generating animals that express a truncating mutation of Mecp2 (Mecp2³⁰⁸), a type of mutation commonly found in classic RTT (19). In a mixed 129/SvEvxC57BL/6 genetic background, this mutation produces a neurological phenotype in both male and female mice (Mecp2^308/Y and Mecp2^308/X) that very closely recapitulates the motor dysfunction of classic human RTT. For example, after a period of normal development of several weeks, Mecp2^308/Y mice develop progressive motor symptoms including tremor, ataxia, stereotypic forepaw movements, hypoactivity, kyphosis, breathing abnormalities and seizures. Mecp2³⁰⁸ heterozygous females have unbalanced patterns of X-chromosome inactivation (favoring the expression of the wild-type allele) and exhibit a high degree of phenotypic variability beyond what is observed in human patients with similar mutations (20). Social interaction studies were performed in these mixed background animals at 6 months of age, a time at which Mecp2^308/Y mice are clearly symptomatic, and no social phenotype was detected in the mutants. On the other hand, the data suggested that wild-type social partners actively avoided the interactions with Mecp2^308/Y mice, a result that may have been related to the prominent physical and neurological abnormalities of mutant animals (19).

Given the prominence of disturbances of social interactions in RTT and autism, and the importance of an understanding of these deficits, we focused our attention on social interactions in Mecp2^308/Y mice. We hypothesized that mutant mice display social interaction abnormalities suggestive of behavioral deficits shown by individuals with RTT and infantile autism, and that these disturbances are manifested prior to development of a severe motor phenotype. Most behavioral tests in experimental animals depend on the measurement of motor output and may be influenced by underlying diurnal activity or motor dysfunction. Furthermore, controlling for these variables is essential in an animal model of RTT as patients are known to have both sleep/wake disturbances and motor deficits. To this end, we first performed a detailed analysis of motor skills and diurnal behavior in Mecp2^308/Y mice on a pure 129/SvEv genetic background. Using different tests, we then assessed social interactions prior to the development of motor dysfunction.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Ten-week-old Mecp2^308/Y mice display no locomotion deficits, abnormal anxiety responses, ataxia or grooming
Prior to tests of social interaction in Mecp2^308/Y mice, we first sought to determine whether mutant mice of pure 129/SvEv background have motor deficits that may impact social behavior. We tested 9–10-week-old mice using paradigms similar to those utilized on mixed background animals (19). In the open-field test, there were no significant differences between genotypes in distance traveled, time spent moving, movement speed, rearing and distribution of activity within the field (Fig. 1A–E). Using the accelerating rotarod apparatus, we did not detect any deficiency in the performance of Mecp2^308/Y mice (Fig. 1F). To further probe deficit of motor coordination, we performed the dowel, vertical pole and wire suspension tests (21). No differences between genotypes were noted in the dowel and vertical pole tests (Fig. 1G and H). In the wire test, however, wild-type mice could generally suspend themselves for the maximum time of 60 s (Fig. 1I), but Mecp2^308/Y mice tended to drop early (P<0.01). Although the majority of mutant mice displayed forepaw stereotypical movements and half of them showed tremor (Fig. 1J and K), self-grooming, a paradigm of stereotypical mouse behavior dependent on forepaw use, was not altered in Mecp2^308/Y mice sprayed with a light water mist (Fig. 1L). Furthermore, automated measurements in the open field and direct observations in a novel cage confirmed the absence of grooming abnormalities (data not shown). The lack of measurable deficits in most of these tasks was not due to poor sensitivity of the tests or absence of phenotypic differences between wild-type and mutant mice on a pure 129/SvEv genetic background. For all the coordination tasks, impairment of motor skills in mutant mice could be detected at more advanced ages (data not shown). Overall, these data indicate that 10-week-old Mecp2^308/Y mice have normal locomotor activity and anxiety-related responses in a novel environment, and do not show ataxia or gait abnormalities in three separate coordination tests and do not have impairments of grooming. This suggests that 10-week-old Mecp2^308/Y mice may be reliably used in social interaction studies detailed subsequently.

View larger version (31K): [in this window] [in a new window]   Figure 1. Mecp2 mutant mice are not impaired in the open-field paradigm and display no significant deficits of motor coordination or grooming at 10 weeks of age. (A–C) In the open-field test for locomotor activity, (A) Mecp2308/Y mice walked the same distance, (B) spent an equal amount of time moving and (C) traveled at a speed similar to wild-type mice. (D) Mutant mice showed similar levels of vertical activity as wild-type mice. (E) Wild-type and Mecp2 mutant mice displayed similar increases in activity in the center of the open-field arena over time. (F) Performance on an accelerating rotating rod (rotarod) apparatus was unaffected. (G) Mutant mice remained on wooden dowels for the maximum time of 2 min performing as well as wild-type mice and (H) were not impaired in their ability to hold onto a pole that transitions from a horizontal to a vertical position. (I) Mecp2308/Y mice were unable to hang onto a thin wire by their forepaws as long as wild-type mice. (J) Stereotypical forepaw movements were present in 90% Mecp2308/Y mice at 10 weeks of age. (K) Body tremor was present in 50% of mutant mice. (L) Grooming of the head and body was similar in mice of both genotypes during 10 min of observation after spraying the animal's fur with a light water mist. Values in panels (A–I) and (L) represent mean±SEM. The asterisks indicate significant genotype differences (*P<0.01; **P<0.001).

 
Abnormalities of diurnal control of motor behavior in Mecp2^308/Y mice
To further assess the motor function of Mecp2^308/Y mice and test the hypothesis that these animals may have subtle abnormalities of home-cage behavior or deficits of sleep/wake cycle, we studied motor activity in the home cage during a 12:12 light:dark cycle. As shown in Figure 2A, Mecp2^308/Y mice displayed relative hypoactivity in the dark phase (P<0.05) and hyperactivity in the light phase (P<0.005) when compared with wild-type littermates, although activity over a 24 h period did not differ between genotypes. The reduction in activity in the dark phase was mainly due to a decreased locomotion (P<0.005) (Fig. 2B), whereas the relative hyperactivity in the light phase was secondary to an increase of fine movements (P<0.005) (Fig. 2C). The spatial distribution of activity within the cage was similar between mutant and wild-type mice (data not shown). Other home-cage parameters, such as the number and distribution of fecal pellets (data not shown) and the position of the nest (discussed subsequently), did not differ between genotypes. Given the diurnal motor abnormalities uncovered in Mecp2^308/Y mice and the function of Mecp2 in transcriptional repression, we sought to determine circadian responses by testing motor activity in constant darkness after entrainment to a light:dark cycle. Two separate groups of mice were tested first under light:dark conditions and next during dark:dark cycles. The first group of animals was tested using the infrared beam system in their home cage. The second group was studied using a wheel running paradigm. In either of these experiments, there were no detectable differences in the circadian activity of wild-type and mutant mice after transition to constant darkness. In the first experiment, both genotypes showed similar degrees of delay of peak spontaneous activity under continuous dark conditions (Fig. 2D–G). In the running wheel paradigm, free-running period was similar in both genotypes, 24.03±0.07 h in wild-type and 23.78±0.22 h in mutants (mean±SEM; P>0.05). The known effects of wheel running on circadian activity were the most likely source of differences between the two experiments (22,23). To assess entrainment to light, activity onset was tested after returning the animals to a 12:12 h light:dark cycle for 3 days (infrared beam experiment) or following a single light pulse (wheel running experiment). No significant genotype differences were detected (data not shown).

View larger version (24K): [in this window] [in a new window]   Figure 2. Abnormalities of diurnal motor activity in Mecp2308/Y mice. (A–E) Motor activity was measured as number of beam interruptions per hour. The activity data are displayed for the dark and light phases of the 24 h cycle (see bar above each graph). (A) Mecp2308/Y mice displayed a decrease in total activity in the dark phase and hyperactivity in the light phase when compared with that in the wild-type littermates. The reduction in dark phase total activity was mainly due to decreased locomotor activity (B), whereas the relative light phase hyperactivity was secondary to an increase in fine movements (C). (D and E) Analysis of locomotor activity at 4 h interval during 12 : 12 h light : dark cycle (D) and after 2 weeks of constant darkness (E). Mecp2308/Y mice and wild-type littermates showed similar delays of activity. (F and G) Wheel running during 12 : 12 h light : dark cycle (LD) and after exposure to constant darkness (DD) in representative examples of a wild-type (F) and a Mecp2308/Y mouse (G). Values in panels (A–E) represent mean±SEM. The asterisks indicate significant genotype differences (*P<0.05; **P<0.005).

 
Nest building and utilization are impaired in Mecp2^308/Y mice
As a measure of home-cage activity related to social behavior, we studied nesting behavior. During the 3 weeks of analysis of motor activity in the home cage, we monitored the following: (1) the ability of mice to build the nest; (2) the location of the nest within the cage; and (3) the position of the mouse during periods of rest with respect to the position of the nest. As shown in Figure 3A and B, both height and quality of the nest, measured 24 h after introduction of nesting material into the cage, were reduced in Mecp2^308/Y mice. Examples in Figure 3C and D illustrate how a representative wild-type nest had the shape of a cocoon with well assembled walls, whereas a mutant nest consisted of largely untouched nesting material. The difference between genotypes did not appear to be related to slower building of the nest as it persisted even for up to 1 week (data not shown). Utilization of the nest also differed between the genotypes, whereas wild-type mice were found resting in the nest in 100% of the observations, and Mecp2^308/Y mice were observed inactive away from the nesting material 12.5% of the time (Fig. 3E). We next sought to determine whether the difference in nest building was due to impairment in the physical manipulation of nesting material or whether mutant mice spent less time building the nest. Mice of both genotypes were given two types of nesting material: normal pressed cotton which requires shredding or a paper towel (Kimwipes). This experiment showed that after both 30 min and 24 h, mutant mice showed significantly reduced nest height with both types of nesting material (Fig. 3F and G). Similar results were obtained assessing nest quality (results not shown). Finally, behavioral scoring demonstrated that Mecp2^308/Y mice spent significantly less time building the nest when given pressed cotton nestlets (Fig. 3H). After introduction of new nesting material into the cage, wild-type mice spent on average 11 out of 30 min building the nest, whereas Mecp2^308/Y mice spent only 2 min actively manipulating the same material.

View larger version (31K): [in this window] [in a new window]   Figure 3. Deficits of nesting behavior in Mecp2308/Y mice (A and B). The height (A) and quality score (B) of the nest measured 24 h after introduction of nesting material into the cage was reduced in Mecp2308/Y mice. (C and D) Examples of nests built by a wild-type mouse (C) and a Mecp2308 mutant (D) 24 h after introduction of nesting material into the cage. (E) Decreased use of the nest in Mecp2308/Y mice during periods of rest/sleep. (F and G) Nest building measured 0.5 and 24 h after placement in the cage of pressed cotton nesting material (F) or a Kimwipes paper towel (G). (H) Mecp2308/Y mice showed decreased interaction with nesting material compared with wild-type mice during 30 min of continuous behavioral observation. Black bars/lines, wild-type; gray bars/lines, Mecp2308/Y. Values in panels (A, B, F–H) represent mean±SEM. The asterisks indicate significant genotype differences (*P<0.01; **P<0.001).

 
Social interactions (I): Mecp2^308/Y mice show no abnormalities in the resident–intruder paradigm
In the wild, Mus musculus is a social species that establishes group territories (24). In this setting, mouse colonies are defended by a dominant male that attacks intruders that enter the territory. In the laboratory environment, the dominant male monitors and defends the entire cage space, occasionally attacking other males. The resident–intruder test measures territorial behavior in male rodents and is often used to study aggression. Isolation and repeated testing increase the likelihood of territorial attacks from the resident test mouse in his home cage. In addition to aggression, two mice that have not previously met display other types of social behaviors including following, sniffing, grooming and physical contacts that may lead to fighting. We used the resident–intruder paradigm to study the encounters between single-housed mice (Mecp2^308/Y and wild-type littermates) and socially housed wild-type controls. Each test mouse was tested for three consecutive days, each time with a different wild-type animal. Analysis of the time spent in each behavioral class across all 3 days showed a complicated interaction between test day, behavioral class and resident vs. intruder (P<0.05). In order to study this interaction, the analysis was run for residents and intruders separately. Although the behavior of both residents and intruders was significantly influenced by time (P<0.05), no significant genotype differences were detected. Overall, the residents showed higher social activity than intruders without excessive levels of aggression. Figure 4A–F shows the effect of repeated testing for each behavioral class. Although mild effects were found with repeated testing, none of the post hoc comparisons revealed significant changes. None of the tests had to be terminated due to excessive aggression.

View larger version (36K): [in this window] [in a new window]   Figure 4. No abnormalities of social interactions in the resident–intruder paradigm. (A–F) Wild-type and Mecp2308/Y mice were tested as residents in a resident–intruder paradigm on three consecutive days. Although there were obvious differences between resident and intruder mice, Mecp2308/Y and wild-type mice showed no difference in behavioral class or between test days. (A) Non-social behavior; (B) active social behavior; (C) passive social behavior; (D) passive aggressive behavior; (E) active aggressive behavior; and (F) active defensive behavior. Values represent mean±SEM.

 
Social interactions (II): Mecp2^308/Y mice are impaired in a test of social interaction without physical contact
As an alternative paradigm to measure the interaction between mice, we devised a partition test in which the interaction between two mice could be studied without direct physical contact and for a longer period of time. This test was employed as a tool to estimate the behavioral reactivity of wild-type and Mecp2^308/Y mice to a wild-type conspecific placed behind a transparent perforated barrier (Fig. 5A). Measurement of the position of wild-type and Mecp2^308/Y mice placed in the large cage compartment (Fig. 5B) showed that mutant mice spent significantly less time in the vicinity of the partition (P<0.05) and more time on the far side of the cage (P<0.001). A planned analysis of data from the first and second 30 min of the experiment showed a trend for an interaction between time and differences in the distribution of activity between wild-type and mutant mice (P=0.091). For the position close to the partition, the relative difference between wild-type and Mecp2^308/Y mice increased from 10 to 28% between the first and second half-hour (Fig. 5C and D). This analysis did not evaluate whether wild-type and mutant mice were simply resting near the partition or interacting with animals in the small cage compartments. As a measure of active interaction, we also determined whether mice close to the barrier were in the vicinity of the stimulus mice (<1xbody length) or near the barrier, but not close to the opponent (>1xbody length). Interestingly, both genotypes spent similar amounts of time near the partition at more than one body length from the stimulus mouse (Fig. 5E, ‘non-interacting’ mice). On the other hand, mutant mice showed a trend to be less often in close vicinity with the mice across the barrier (Fig. 5E, ‘actively interacting’ mice; P=0.058). To more directly assess the activity of test mice, we also used an automated system to study non-ambulatory movements (‘fine movements’) before and after introduction of stimulus mice in the small cage compartments. These measurements evaluated motor activity restricted to small cage areas and were not influenced by animals resting in a stationary position. As shown in Figure 5F, before introduction of stimulus animals, wild-type and Mecp2^308/Y mice had similar levels and distribution of fine movements. After controls were placed across the barrier, mice of both genotypes responded to the presence of stimulus animals with a marked change in the level and distribution of activity. Interestingly, when compared with wild-type, mutant mice showed a 6% reduction of total fine movements (P<0.05) and a difference in the distribution of fine movements within the cage compared with wild-type (P<0.005 for the interaction between genotype and cage position) (Fig. 5G). For the position closest to the barrier, fine movements were 50% lower in Mecp2^308/Y mice compared with wild-type animals (P<0.01). In a planned analysis of the effect of time, we compared the distribution of activity during the light and dark cycles. Differences in the distribution of fine movements between genotypes were not dependent upon the light cycle. Overall, these data indicate that Mecp2^308/Y mice have deficits in the interaction with unfamiliar control males in a partition test.

View larger version (28K): [in this window] [in a new window]   Figure 5. Mecp2308/Y mice are impaired in a test of social interaction without physical contact. (A) Example of a test cage containing two mice separated by a clear acrylic barrier with holes. The position of wild-type and Mecp2308/Y mice placed in the large cage area was manually scored once a minute for 1 h by dividing this compartment in three equal parts (beams 1+2, 3+4 and 5+6). Motor activity of mice in the large compartment was followed by interruption of infrared beams positioned as depicted in the photograph. (B) Mecp2308/Y mice spent less time in close proximity to the mouse in the small compartment, and more time away from it. (C and D) Differences in the distribution of wild-type and Mecp2308/Y mice within the cage increase between the first half-hour (C) and the second half-hour (D) of observation. (E) Differences at the position closest to the partition are mostly related to a decrease of the time mutant mice spent within less than one body length from the opponent (P=0.058). (F–I) Wild-type and Mecp2308/Y mice show no significant differences in the level or the spatial distribution of fine movements in the cage before introduction of the stimulus mouse on the opposite side of the partition (F). After introduction of the stimulus mouse, Mecp2308/Y mice show a significant decrease in total fine movements and a change of their distribution within the cage (G). The differences between genotypes are not dependent on the phase of the light cycle (H and I). Values represent mean±SEM. The asterisks indicate significant genotype differences (*P<0.05; **P<0.001).

 
Social interactions (III): the tube test of social dominance reveals a subtle phenotype in Mecp2^308/Y mice
As an additional method to evaluate social interactions, we next used the tube test, a paradigm used to evaluate social dominance (25). In this experiment, two mice were positioned at the opposite end of an acrylic tube and released to meet inside it. After a period of time during which the animals explored each other, one mouse backed out of the tube ending the test. Single-housed wild-type and Mecp2^308/Y mice were first tested against socially housed wild-type mice and then against each other. When single-housed wild-type and Mecp2^308/Y mice were paired with group-housed controls, no differences between genotypes were observed in the retreats from the opponent (Fig. 6A), distance traveled inside the tube (Fig. 6B) or time to retreat from the opponent (Fig. 6C). There were also no differences in the retreat from the opponent when testing single-housed mice of both genotypes against each other (Fig. 6D). Interestingly, genotype differences were detected in single-housed mice in the maximum distance covered inside the tube before reaching the opponent and time to retreat. Mecp2^308/Y mice traveled a shorter distance inside the tube when they were paired with wild-type single-housed mice (Fig. 6E, P<0.001). Furthermore, although the interaction between wild-type/wild-type or mutant/wild-type pairs did not show differences between genotypes, when mutant mice were paired with each other, there was a significant increase in test duration (Fig. 6F, P<0.01). To control for motor impairments or poor motivation, we tested the ability of naïve wild-type and Mecp2^308/Y mice to walk across the tube in the absence or presence of familiar food at its opposite end. In either setting, there were no significant differences between genotypes (data not shown). These results show that Mecp2^308/Y mice have subtle impairments in the interaction with unfamiliar single-housed controls in the tube test.

View larger version (25K): [in this window] [in a new window]   Figure 6. Abnormal social interactions in the tube test. Single-housed wild-type and Mecp2308/Y mice were first tested against socially housed wild-type animals (A–C) and then against each other (D–F). (A–C) No significant differences between single-housed wild-type and Mecp2308/Y mice in the interaction with socially housed controls for retreats (A), distance traveled (B) and test duration (C). (D) No difference in the dominance of single-housed wild-type and Mecp2308/Y mice. (E) Mecp2308/Y mice travel a shorter distance when encountering a single-housed wild-type on the opposite side of the tube. (F) Mecp2308/Y mice take longer to be defeated or to defeat a single-housed opponent of the same genotype. Values represent mean±SEM. The asterisks indicate significant genotype differences (*P<0.01; **P<0.001).

 
Social interactions (IV): Mecp2^308/Y mice have impairments of social interaction, but no deficit of social recognition or olfaction
The social interaction experiments described earlier were performed pairing subjects that had never encountered one another. Furthermore, behavioral testing was started immediately after introduction of the unfamiliar mouse to the test subject. We sought to determine whether the social interaction abnormalities seen in Mecp2^308/Y mice persisted after a period of habituation to the partner/opponent or whether the ability of recognizing familiar from unfamiliar mice was impaired. We measured the reactivity of wild-type and Mecp2^308/Y mice to familiar or unfamiliar adult mice in the neighboring compartment of a common cage divided in halves by a perforated transparent barrier. After 18 h of exposure to a conspecific across the partition, Mecp2^308/Y mice spent significantly less time in direct contact with the barrier (with whiskers, nose or paws) than wild-type littermates (Fig. 7A, familiar adult, P<0.01). Replacement of the familiar adult mouse with a different subject showed a significant increase in the amount of time spent in contact with the partition for both mutant and wild-type mice. Both wild-type and Mecp2^308/Y mice spent approximately twice as much time as in contact with the partition (Fig. 7A, unfamiliar adult, P<0.01). After reintroduction of the familiar mouse into the cage, animals of both genotypes showed a return to baseline levels (Fig. 7A, familiar adult). Interestingly, the level of social interaction of Mecp2^308/Y mice with both familiar and unfamiliar mice was significantly lower than wild-type littermates. These data demonstrate that Mecp2^308/Y mice display social interaction abnormalities even after a prolonged period of exposure to an adult conspecific, but importantly their ability to discriminate between familiar and unfamiliar individuals is not impaired. These results also suggest that sensory modalities required for social recognition are not impaired in Mecp2^308/Y mice. MeCP2 is highly expressed in the olfactory bulb and defects in the olfactory neuronal maturation have been described in Mecp2 null mice and RTT patients. To formally exclude the presence of olfactory deficits in Mecp2^308/Y mice, we measured the amount of time required by wild-type and mutant mice to find food hidden under the cage bedding. Animals of both genotypes took similar amounts of time in this task (Fig. 7B).

View larger version (20K): [in this window] [in a new window]   Figure 7. Abnormal social interactions with familiar and unfamiliar mice, intact social recognition, and absence of olfactory deficits or abnormalities in the exploration of inanimate unfamiliar stimuli. (A) Wild-type and Mecp2308/Y mice were tested against wild-type animals in a standard housing cage divided in half by a clear perforated plastic partition. Mecp2308/Y mice spent less time in direct contact with the partition with forepaws, nose or whiskers in the presence of a familiar adult male mouse (a subject housed across the barrier for the previous 18 h; left and rightmost columns) and in the presence of an unfamiliar adult male (center columns). (B) Mecp2308/Y mice required similar amounts of time as wild-type littermates to find food hidden under the cage bedding. (C–E) Wild-type and Mecp2308/Y mice were tested against three types of novel stimuli in a standard housing cage. Mecp2308/Y mice spent less time interacting with unfamiliar naïve juvenile males (C), but explored the nesting material of unfamiliar males (D) and an unfamiliar plastic object (E) as long as wild-type littermates. Values represent mean±SEM. The asterisks indicate significant genotype differences (*P<0.01 in the comparison of wild-type versus mutant; +P<0.01 in the comparison of familiar adult versus unfamiliar adult interaction; **P<0.001).

 
Social interactions (V): Mecp2^308/Y mice are impaired in the interaction with unfamiliar juveniles, but show no olfactory deficits or abnormalities in the interaction with an unfamiliar inanimate object
In order to determine whether the social interactions deficits displayed by Mecp2^308/Y mice extended to the interaction with inanimate objects, we compared the time wild-type and mutant animals spent exploring three different types of stimuli: unfamiliar juvenile males; the nesting material used by unfamiliar mice (a familiar object with a novel odor); and a small clear plastic tube (an unfamiliar object). Juvenile males were used instead of adults to exclude the effect of mutual aggression. All stimuli were presented to the mice in standard cages in the absence of a partition to allow for maximal interaction. As shown in Figure 7C, Mecp2^308/Y mice spent approximately half as much time as wild-type animals exploring unfamiliar juvenile males placed in their cages (P=0.001). On the other hand, mice of both genotypes responded similarly to the presence of nesting material used by unfamiliar mice or the novel plastic object. These data demonstrate that the social deficits of mutant mice extend to the interaction with juveniles. More importantly, the results demonstrate that although mutant mice are impaired in the interaction with a conspecific, they behave normally in the presence of inanimate novel stimuli. Interestingly, the data indicate that the simple presence of the odor of unfamiliar mice is not sufficient to elicit abnormalities of exploratory behavior.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Abnormalities of diurnal activity in Mecp2^308/Y mice
The present study shows that Mecp2^308/Y mice have abnormalities in the diurnal control of motor activity that result in a marked reduction of the physiological difference in activity between dark and light phases. Mutant mice display hypoactivity in the dark phase and hyperactivity in the light phase. The relative decrease of activity in the dark phase is largely due to reduced locomotion, whereas the light-phase hyperactivity is mainly caused by an increase of fine movements. The amplitude of the difference in activity between light and dark phases is severely reduced in Mecp2^308/Y mice: where in wild-type mice, total activity was 146% higher in the dark than in the light, this increase was only 33% in mutants. Locomotor activity and fine movements are similarly altered, although to different degrees. Ambulation was nearly equal in the dark and light phases in Mecp2^308/Y mice (only 8% difference compared with 91% change in wild-type). Fine movements were only 57% higher for the mutants in the dark phase compared with that in the light phase, whereas in wild-type mice this difference was 243%. Interestingly, despite these large shifts, the level of activity across the entire 24 h period was not affected. This suggests that the Mecp2³⁰⁸ mutation does not result in a generalized reduction of motor function in adult male mice of pure 129/SvEv background. In agreement with this finding, experiments directed at assessing the motor skills of 10-week-old Mecp2^308/Y mice using a variety of tests did not reveal significant impairments in exploratory behavior, locomotion or coordination. In addition, the changes of activity during the light and dark phases do not appear to be related to dysfunction of the master circadian clock residing in the suprachiasmatic nucleus (SCN) of the anterior hypothalamus. In fact, behavioral assessment of the activity of the SCN or its entrainment to light did not show significant differences between wild-type and Mecp2^308/Y mice. It is possible that the coupling between the SCN and neuronal systems controlling motor output may be impaired in RTT. Although the mechanism of these diurnal abnormalities remains to be determined, it is interesting to note a parallel with disturbances of activity reported in RTT patients. Affected girls have abnormalities of sleep/wake patterns with poor quality of sleep, significantly less nighttime sleep and more daytime sleep (3). Sleep studies have also shown a decrease of rapid-eye-movement sleep (4). Mood changes, including anxiety, have been reported in RTT and some of these manifestations may also have an impact on sleep. For instance, patients have been described with inconsolable crying and screaming at nighttime. In light of these observations, it is conceivable that the increased fine movements of Mecp2^308/Y mice during the light phase may be related to disturbances of sleep, and that the decreased ambulatory activity in these mice during the dark cycle may be functionally equivalent to the patient's increase of daytime sleepiness. Although sleep disturbances are not diagnostic of RTT, they point to the dysfunction of the neurological substrate controlling circadian behavior and/or sleep. Further studies directed at understanding the basis of these phenotypes in Mecp2^308/Y mice may provide a key to understand sleep/wake abnormalities in RTT patients and perhaps provide some measures for control of these symptoms.

Nesting, a home-cage phenotype related to social behavior, is impaired in Mecp2^308/Y mice
Mice of both sexes build a nest when provided with suitable material and are typically found lying in it during the daytime. In the wild, the nest provides shelter, camouflage from predators and conservation of body heat, and is an important component of fitness (26). Parents and offspring share the nest and both parents retrieve pups into the nest. We have shown that male Mecp2³⁰⁸ mutant mice have significant impairment of nesting behavior in the home cage. Although several mouse models have deficits of nesting behavior, this phenotype is often seen in association with considerable global motor impairments (27,28). In contrast, Mecp2^308/Y mice did not display significant motor deficits, suggesting that their nest building impairments may not simply result from general loss of motor skills. For instance, use of a nesting material that eliminated the need for extensive shredding and manipulation did not improve the performance of Mecp2^308/Y mice. Secondly, even though the height and quality of the nest increased in both genotypes between 30 min and 24 h from introduction of the nesting material in the cage, mutant mice were unable to achieve the complexity of the nest of wild-type mice. Nesting was also decreased in Mecp2^308/X heterozygous females in the absence of measurable motor impairments and not only in the generally more severely affected males (Paolo Moretti and Huda Y. Zoghbi, personal communication). Finally, Mecp2^308/Y mice spent significantly less time actively interacting with the nesting material suggesting decreased interest in building the nest. In observations conducted over several days, mutant mice were found outside the nest during periods of inactivity indicating that nest utilization, and not only its construction, were affected and suggesting the existence of non-motor deficits in nesting behavior. Overall, these data indicate the presence of behavioral phenotypes that go beyond the impairment in execution of a motor task. Mecp2^308/Y mice have stereotypical movements of the forepaws when held by the tail and are impaired on wire suspension, a test largely dependent on forepaw use. Although we cannot exclude that subtle motor dysfunction affecting predominantly the forepaws may disrupt execution of complex motor tasks requiring their use, not all forepaw-mediated behaviors are impaired in mutant mice. Self-grooming, a complex stereotyped behavior involving the forepaws, was not affected in the mutants. This suggests that simple forepaw stereotypies do not interfere with all forepaw movements in Mecp2^308/Y mice, and may not affect complex stereotyped and non-stereotyped behaviors such as grooming and nesting. It is tempting to speculate that the severe deficits in the interaction with nesting material found in mutant mice may correspond to the apraxia of hand use seen in RTT patients. Although speculative, this hypothesis is strengthened by the presence of many obvious phenotypic similarities between Mecp2^308/Y mice and RTT patients, including the resemblance of forepaw stereotypies with hand-wringing (19).

Interestingly, changes in nesting have been associated with abnormalities of social interactions in other mouse models. For example, mice deficient for Dvl1, a mouse homolog of the Drosophila segment polarity gene Dishevelled, exhibit deficits in nest building and reduced social interactions (29,30). Male mice selected for high nest-building behavior are more aggressive than either mice selected for low nest-building behavior or randomly bred control lines (31). Conversely, male mice selected for short attack latency show more nest-building behavior than those selected for long attack latency. It remains to be determined whether the neurological substrate regulating social interactions is also responsible for controlling nesting behavior.

Mecp2^308/Y mice display test-dependent impairments of social behavior
Using five different behavioral tests to study social interactions, we have shown that Mecp2^308/Y mice on a pure 129/SvEv genetic background exhibit paradigm-dependent deficits of social interactions. In the first test of social interactions without physical contact, measures of position showed that mutant mice spent significantly less time in close proximity to the mouse on the opposite side of the partition suggesting decreased interaction. This difference was accounted for by a reciprocal increase in the time spent in the cage area farthest away from the opponent. Measures of motor activity demonstrated that the presence of an opponent across the partition was associated with mild reduction of total fine movements in the mutants and redistribution of activity across the cage when compared with wild-type mice. These abnormalities cannot be attributed to global motor deficits and are not accounted for by differences in diurnal motor activity as under single-housed conditions, Mecp2^308/Y mice did not show similar changes. Analogous results were obtained with independent groups of mice in two additional paradigms: interaction with adult males in different partition test and exploration of male juveniles in a common cage space. Overall, these deficits may not be attributed to difference in aggression or dominance in Mecp2^308/Y mice. In fact, direct contact was prevented by the barrier in two of the tests and no significant differences in aggression or dominance were seen in the resident–intruder test and the tube test. No aggression was present in the interaction with juveniles. Possible explanations are reduced interest and initiative in Mecp2^308/Y mice. This hypothesis is supported by results of the tube test. No dominance differences were found in encounters with socially housed or single-housed mice, as seen in the equal distribution of retreats between genotypes. On the other hand, mutant mice remained close to the entrance of the tube when facing a single-housed unfamiliar wild-type mouse. Pairing of two single-housed mutants also increased the duration of the test suggesting that the social impairments of Mecp2^308/Y mice may be more obvious in the absence of a wild-type mouse which can take the initiative in the interaction. An additional possibility is that the behavior of Mecp2^308/Y mice during social interactions may be affected by increased anxiety in a social encounter with an adult male. In the first partition test, mutants spent more time than wild-type as far away as possible from the opponent across the barrier and showed less activity close to the partition. It is conceivable that the close setting of the resident–intruder test may mask the effect of social anxiety by not allowing sufficient physical separation between the subjects. Interestingly, the social interaction deficits of Mecp2^308/Y mice do not appear to extend to the recognition of familiar from unfamiliar mice or the interaction with novel olfactory stimuli or inanimate objects. These findings, together with the absence of olfactory deficits in Mecp2^308/Y mice, suggest that the impairment of social interactions with unfamiliar or familiar mice cannot be attributed to global sensory or cognitive deficits.

In previous studies, using the resident–intruder paradigm and the tube test, Mecp2^308/Y mice of mixed 129/SvEvxC57BL/6 genetic background did not show significant social interaction deficits when compared with wild-type mice. In contrast, these experiments suggested that wild-type animals actively avoided Mecp2^308/Y mice (19). For instance, wild-type mice retreated from mutant mice in most encounters in the tube test. In the resident–intruder test, intruders spent less time interacting with Mecp2^308/Y mice when compared with interactions with wild-type residents. There are several explanations for these differences. First, Shahbazian et al. studied 6-month-old symptomatic mutant mice which displayed severe motor dysfunction and fur abnormalities. It is possible that these overt manifestations of disease influence experimental outcomes in social interactions. Secondly, behavioral responses are known to vary among animals of different genetic backgrounds (32). Thirdly, in the present study, animals had multiple encounters in various experimental settings.

Prior to the identification of RTT as a distinct disorder, patients with Rett syndrome were diagnosed as autistic (6–9), based on the overlap in clinical manifestations. Especially in the early stages of the disorder, RTT girls show significant phenotypic overlap with children with infantile autism. For example, language and social skills are commonly lost in autism and a period of regression is seen in some individuals (33). Recent studies have shown that features of autism are more common in RTT girls than in children with severe to profound mental retardation, with more autistic-relating deficits and less anti-social behavior (34). Girls with RTT are also indistinguishable from autistic children in the overall severity of autistic-relating manifestations. Although genetic disorders, such as Fragile X syndrome, Angelman syndrome, Tuberous Sclerosis and Down syndrome, are all associated with autistic features (35–38), RTT has perhaps the closest clinical overlap with autistic spectrum disorders. Recent molecular data support the similarities between RTT and autism found at the clinical level. Although mutations in the coding region of MECP2 occur at low frequency in patients with a diagnosis of autism (14–16), a recent study described significant differences in MeCP2 expression in brain samples from several neurodevelopmental disorders including autism, pervasive developmental disorder, Prader-Willi and Angelman syndromes (39). These results suggest that the regulation of MeCP2 expression is defective in autistic spectrum disorders.

There are currently no known genetic causes or accepted animal models of autism. To our knowledge, this is the first detailed study of social behavior in an animal model of a genetic disorders associated with autistic manifestations. The identification of social interaction deficits in a mouse model of RTT allows the dissection of the cellular and molecular determinants of social behavior in these animals. We propose that the elucidation of the causes of social interaction deficits of Mecp2^308/Y mice may prove useful in the study of autistic features in RTT and perhaps serve as a model for the pathogenesis of autism. In addition to autism, abnormal social interactions are found in several human psychiatric disorders and Mecp2^308/Y mice may provide a genetic model to study factors that influence abnormal social interactions. For instance, drugs may be screened for their influence on the abnormal social interactions observed in these mice. Future studies will provide an opportunity to dissect the molecular basis of these defects.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Subjects
The experimental subjects used in all experiments were pure 129/SvEv wild-type and Mecp2^308/Y littermates that were bred in our animal facility from crosses of heterozygous Mecp2^308/X mutant females with wild-type males. All testing was done by observers blind to the genotype of the subjects. First group of wild-type and Mecp2^308/Y mice (n=23 and 31, respectively, age 10 weeks) was used to study open-field activity, and a second group (n=16 and 18, respectively, age 10 weeks) was tested in experiments on motor skills (e.g. dowel test). These mice were housed in groups of 4–5 animals per cage for the duration of all experiments. Home-cage behavior and social interaction studies (tests labeled I through III in Results) were performed in a third group of wild-type and Mecp2^308/Y mice (n=12 per genotype). At the beginning of the study, mice in the latter group were 8-week-old. Body weight was 21±0.8 g in wild-type and 23±0.7 g in Mecp2^308/Y mutants (mean±SEM). The animals were single housed in polycarbonate cages with water and food available ad libitum. Partners for social interaction tests (n=24; 129S6/SvEvTac strain from Taconic, Albany, NY, USA) were 5-week-old at the beginning of the study and were housed in groups of 4–5 mice per cage. Additional social interaction (tests IV and V) and olfaction studies were performed in two groups of wild-type and Mecp2^308/Y mice (n=15 per genotype each group). Two additional groups of wild-type and Mecp2^308/Y pure 129/SvEv mice were used for further circadian rhythm experiments (i.e. light–dark exposure and wheel running; n=10 and 6 per genotype, respectively). The experimental rooms were kept on a 12:12 light:dark cycle with lights on from 7:00 AM to 7:00 PM. Ambient temperature was maintained at 21°C. Mice were genotyped by PCR using DNA prepared from tail biopsies.

Open-field test
Mice were placed in the center of an open-field arena (40x40x30 cm³). Activity was quantified by a computer-operated Digiscan optical animal activity system (RXYZCM, Accuscan). Each test session lasted 30 min, and data were analyzed as three blocks of 10 min each. Analysis of data was performed using a two-way ANOVA (genotypex10 min interval) with repeated measures, using genotype as a between-subject and time as a within-subject factors.

Rotating rod test
Mice were placed on an accelerating rotarod apparatus (Ugo Basile) for 16 trials (four trials a day on four consecutive days) with a 30–60 min rest interval between trials. Each trial lasted for a maximum of 10 min, during which the rod accelerated linearly from 4 to 40 rpm for the first 5 min and stayed at a stable speed for another 5 min. The amount of time for each mouse to fall from the rod was recorded for each trial. Data were analyzed using a two-way ANOVA (genotypextrial) with repeated measures, using genotype as a between-subject and trial as a within-subject factors.

Dowel test
Mice were placed in the center of a horizontal dowel, and the time they remained on the rod was recorded. If mice walked across and off the dowel, they were placed back onto the center of the dowel. Trials lasted for a maximum of 2 min. Mice were first tested on a 0.9 cm diameter dowel and 2 days later on a 0.7 cm diameter dowel. Data were analyzed using a Mann–Whitney U-test.

Vertical pole test
Mice were placed onto one end of a horizontal pole covered with cloth tape (1.9 cm diameter and 43 cm length). The pole was then slowly shifted to a vertical position, and the time the mice remained on the pole was scored as previously described. Data were analyzed using a Mann–Whitney U-test.

Suspended wire test
Mice were suspended by their forepaws on a 2 mm wire, and the length of time they remained on the wire was recorded. The maximum length of the test was 1 min. Data were analyzed using a Mann–Whitney U-test.

Stereotypic forepaw movements
Mice were suspended by the tail for 10 s at 40 cm above the cage bedding. Presence or absence of stereotypic forepaw movements was scored and analyzed with Fisher's exact test.

Body tremor
Mice were held on the hand of the examiner for 10 s and the presence or absence of body tremor was recorded and analyzed with Fisher's exact test.

Grooming induced by water mist
Single-housed mice were sprayed with a light water mist on the back and neck area and their behavior was videotaped for 10 min. The videotape was subsequently observed by an examiner blind to the genotype of the mice. The amount of time spent grooming the head or the rest of the body was recorded. Data were analyzed using a two-way ANOVA (genotypexbody area) with repeated measures, using genotype as a between-subject factor and body area as a within-subject factor.

Wheel running
Mice were housed individually in cages equipped with running wheels. The cages were placed within light-tight, ventilated environmental compartments, where except as noted, the animals were exposed to a 12:12 h light:dark cycle. Magnetic reed switches mounted near the running wheel detected movement of a magnet mounted on the wheel. Switch closures were recorded and plotted by a computer-based system (VitalView 4.0, and ActiView 1.3 software MiniMitter, Bend, OR, USA). To assess the presence of rhythmicity during various phases of the experiment, actograms were scored by an observer blind to the genotype of the animals. The behavior of each animal in constant darkness and after the light pulse was compared with the behavior of the same mouse during light:dark cycles. Free-running period was estimated from actograms without knowledge of the genotype of the animals. The period was determined from the slope of a hand-drawn line through activity onsets of each animal housed in constant darkness (40). The first 4 days after discontinuation of the lighting cycle were excluded from the analysis to allow stabilization to constant conditions. Data were analyzed using ANOVA.

Home-cage activity
Mice were single-housed and tested in their home cage (28.5x17.5x12 cm³). Home-cage activity was determined using the Photobeam Activity System (San Diego Instruments), a computerized system that measures the frequency of photobeam breaks along the side of the cage. The software discriminates between locomotor activity, fine movements and total activity (the sum of locomotor activity and fine movements). Locomotion was defined as interruption of two adjacent beams, whereas fine movements were defined as breaking the same beam twice in a row. A 1 s refractory period was introduced to avoid counting high frequency repetitive behavior such as scratching or tail flipping. Data were collected at 1 min interval. In the first experiment, activity was recorded during four separate 2 day sessions, each beginning at 7:00 PM. All recordings were spaced 2 days apart from one another and occurred during the 3 weeks of single housing (age 8–11 weeks). For analysis, the activity measures were collapsed into 12 h bins to examine the distribution of activity during the light and dark periods. Data analysis was performed using a two-way ANOVA (genotypex12 h interval) with repeated measures, using genotype as a between-subject and time interval as a within-subject factors. In a second experiment (designed to test exposure to constant darkness), activity was recorded in a separate group of mice (n=10 per genotype). Here, 5 days of 12:12 h light:dark cycle were followed by 15 days of constant darkness at the end of which mice were re-exposed to the original 12:12 h light:dark cycle for 12 days. An infrared light was present at all times to allow daily monitoring of the animals. Activity measures collected during the 12:12 h light:dark cycle before exposure to constant darkness, during the last 2 days of dark:dark cycles and during the last 2 days after re-exposure to a 12:12 h light:dark cycle were collapsed into 4 h interval to examine the distribution of activity during the 24 h. Analysis of the data was performed using a three-way ANOVA (genotypex4 h intervalxlight phase) with repeated measures, using genotype as a between-subjects factor and time interval and light phase as within-subject factors.

Nesting behavior
Nest building and utilization were assessed during 3 weeks of single housing between age 8 and 11 weeks prior to the social interaction tests. Nesting material, a 5x5 cm² piece of cotton (Ancare, Bellmore, NY, USA), was introduced in the cage four times for 2 days during 3 weeks. After 24 h, the position of the mouse compared with the location of the nest and the quality of the nest were recorded. After 48 h, the same two measures were recorded, together with the height of the nest. Nest quality was measured using the following scale: (0) nesting material unmodified; (1) flat nest with partially shredded nesting material; (2) shallow nest with shredded material, but lacking fully formed walls; (3) nest with well developed walls; and (4) nest in a shape of a cocoon with partial or complete roof. The four 2 days period were averaged for the analysis. Nest height was analyzed with t-test, and nest quality and nest utilization with Fisher's exact test. Additional nest building studies were performed in the same mice at 17 weeks of age. Here, the behavior of the mice was scored during the initial 30 min of exposure to the nesting material (starting from a cotton nestlet wafer), and data were collected as three 10 min interval. In addition, nest height and quality were measured 30 min and 24 h after introduction to two different types of nesting material: a block of pressed cotton and a Kimwipe paper towel. Data were analyzed using a two-way ANOVA (genotypextime) with repeated measures. Significant interactions were analyzed by post hoc T-tests.

Resident–intruder test
After 3 weeks of single housing, resident–intruder tests were performed on three consecutive days. On each day, the single-housed residents were exposed for 10 min to an unfamiliar intruder (socially housed 129S6/SvEv). Behavior was recorded on digital video and the tapes were scored afterwards using a detailed ethogram in the Observer^® software (Noldus Information Technologies, Wageningen, The Netherlands). The ethogram contained a total of 26 behaviors grouped into six behavioral classes: non-social exploratory behaviors (attention, sniffing, walking, exploring, rearing, sifting, self-grooming), active social behavior (approaching, social sniffing, genital sniffing, social pushing, following, mounting); active aggression (lateral threatening, tail rattling, hanging over, biting, chasing, clinching); passive social behavior (being groomed, being mounted); active defense (walking away, running away, defensive kicking); and passive defense (freezing, in submission). At any time, only one behavior was recorded as a state, which means that both duration and frequency are computed. The tapes were scored and analyzed for the resident and intruder independently. Total scores for the six behavioral classes for the tests performed on three consecutive days were analyzed with repeated measures of ANOVA, using genotype as a between-subject and day as a within-subject factors. When appropriate, the ANOVA was followed by post hoc T-tests.

Social interaction test without physical contact (I)
An experimental cage (48x26x15 cm³) was divided into two parts by a transparent perforated barrier separating a larger (26x38 cm²) from a smaller compartment (26x10 cm²). Two days after completion of the resident–intruder test, each single-housed test subject was placed in the large compartment at 5:00 PM and left undisturbed until 9:00 AM the following day. Then, an unfamiliar individual from the socially housed mice (129S6/SvEvTac) was placed in the small compartment behind the barrier. The position of the test mouse in the large cage compartment was recorded every minute for 1 h. The large cage compartment was divided by two imaginary lines in three equal sections, i.e. close to, intermediate and far away from the partition. When close to the partition, the examiner scored whether the test mouse was within one body length from the stimulus mouse on the opposite side of the barrier. In a subset of mice (n=6 for each genotype), in addition to manual scoring, the Photobeam Activity System was used simultaneously to record the activity of test subjects in the larger cage compartment. Locomotor activity and fine movements were recorded automatically and used to determine the distribution of motor activity at various distances from the partition. The data were analyzed for the periods before and after introduction of the stimulus mouse. Beam interruptions after introduction of the stimulus mouse were also analyzed separately for the light and dark phases of the 24 h cycle. Data were analyzed using a two-way ANOVA with repeated measures using genotype as a between subject and cage position as a within subject factors. Significant interactions were then analyzed by T-tests.

Tube test
The tube test was used as an additional measure of social behavior (25). Here, two mice were placed into opposite ends of a clear acrylic cylindrical tube (3.5 cm in diameter and 33 cm in length), and the following parameters were recorded: which mouse retreated backwards, the distance each mouse moved forward inside the tube until the heads met (measured in increments of one-third of the total length of the tube) and the latency to retreat (i.e. end of test). The percentage of retreats, distance moved and retreat latency were calculated from the total number of encounters per test subject. This assay was performed in two ways. First, single-housed wild-type and Mecp2^308/Y mice were tested against unfamiliar group-housed mice (four trials per day for two consecutive days). Secondly, all 24 single-housed mice were tested against each other (23 trials over four consecutive days). In the first experiment, data were analyzed using T-tests. In the second experiment, retreat and distance traveled data were analyzed by T-tests. Data on test duration were analyzed using a two-way ANOVA (genotype of test mousexgenotype of opponent mouse) with repeated measures.

Partition test in standard housing cage
Pure 129/SvEv wild-type and Mecp2^308/Y littermates of 30 weeks of age were caged individually for 3 days to reduce group effects. Pure C57BL/6J wild-type male mice of 15 weeks of age were used as stimulus mice. One 129/SvEv and one C57BL/6J mice were placed in pairs into separate halves of an experimental cage (28.5x17.5x12 cm³) separated by a perforated transparent partition that permitted animals to see, hear and perceive the smell of a neighbor, but prevented direct physical contact. Typically, mice move near the partition where they smell, touch, clutch, hang on, put their noses into or gnaw the holes. The mice were left undisturbed in these housing conditions for 18 h. Testing was started at 10:00 AM the following day. The time spent by test subjects in direct contact with the partition (with whiskers, nose or paws) was measured using a hand-held computer (Psion Workabout mx, Psion Teklogix, Erlanger, KY, USA) and the software The Observer® (Noldus Information Technologies, Leesburg, VA, USA). Test subjects remained in their own compartment and were observed for three consecutive periods of 5 min each. During the first period, the caging arrangements of the previous 18 h were not changed (familiar adult). During the second period, the stimulus mouse was replaced by an animal previously housed in a different cage under identical circumstances (unfamiliar adult). During the third period, the original stimulus mouse was returned to its compartment (familiar adult). Data were analyzed using a two-way ANOVA (genotypextime) with repeated measures. Significant interactions were analyzed by post hoc T-tests.

Olfaction test
Mice were habituated to the flavor of a novel food (Kellogg's blueberry cereal bar) for 3 days prior to testing. On the fourth day, a 1 cm³ piece of blueberry bar was buried under 2 cm of bedding in a clean cage. The mice were placed in the cage, and the time required to find the food was measured. Data were analyzed by T-test.

Interaction with unfamiliar juveniles, nesting material used by unfamiliar males and a novel object
Pure 129/SvEv wild-type and Mecp2^308/Y littermates of 30 weeks of age were placed into individual cages immediately prior to the experimental session and allowed to habituate to the new environment for 15 min. The cages used were identical to those in which the animals were normally housed (28.5x17.5x12 cm³). A male juvenile (pure C57BL/6J wild-type of 3–4 weeks of age) was placed into the cage with an adult for a total of 2 min. The social investigation of the juvenile by the adult was observed by a trained observer who timed the duration of the investigation with a hand-held stopwatch. Behaviors that were scored as social investigation included the following: direct contact, sniffing and close following (within <1 cm). There were no aggressive encounters between animals. Naïve mice were used in all encounters. After 1 week, the same adult mice were habituated to individual cages for 15 min after which nesting material that had been used by unfamiliar mice was placed in the cage. The time spent investigating the nesting material was measured during a 2 min period. One week later, after a period of habituation to individual cages for 15 min, the same adult mice were exposed for 2 min to an unfamiliar object (plastic tube 1.5 ml). The time spent investigating the tube was measured using a hand-held stopwatch. Data for these three experiments were analyzed by T-test.

	ACKNOWLEDGEMENTS

 
The authors wish to thank members of the Zoghbi laboratory for feedback provided during the course of this study. We also wish to thank Millan Patel and Gerard Karsenty for use of their wheel running equipment, Corinne Spencer for advice on the partition test and O'Brian Smith for his assistance with statistical analysis. We are grateful to Jonathan Levenson, Adriano Flora, Jeff Neul, Aaron Bowman and Noah Shroyer for their critical reading of the manuscript. This work was supported by an NIH postdoctoral fellowship to P.M. (NS043969-03), NIH grant P01 HD40301 and funds from Cure Autism Now to H.Y.Z. and a NIH grant to the Baylor College of Medicine Mental Retardation Research Center (HD24064). H.Y.Z. is an investigator with the Howard Hughes Medical Institute.

	FOOTNOTES

 
The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors.

Present address: Henry Wellcome Laboratories for Integrative Neuroscience and Endocrinology (LINE), Dorothy Hodgkin Building, University of Bristol, Whitson Street, Bristol BS1 3NY, UK.

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 

1. Amir, R.E., Van den Veyver, I.B., Wan, M., Tran, C.Q., Francke, U. and Zoghbi, H.Y. (1999) Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl-CpG-binding protein 2. Nat. Genet., 23, 185–188.[CrossRef][Web of Science][Medline]

2. Hagberg, B., Aicardi, J., Dias, K. and Ramos, O. (1983) A progressive syndrome of autism, dementia, ataxia, and loss of purposeful hand use in girls: Rett's syndrome: report of 35 cases. Ann. Neurol., 14, 471–479.[CrossRef][Web of Science][Medline]

3. Glaze, D.G., Frost, J.D., Jr, Zoghbi, H.Y. and Percy, A.K. (1987) Rett's syndrome: characterization of respiratory patterns and sleep. Ann. Neurol., 21, 377–382.[CrossRef][Web of Science][Medline]

4. Piazza, C.C., Fisher, W., Kiesewetter, K., Bowman, L. and Moser, H. (1990) Aberrant sleep patterns in children with the Rett syndrome. Brain Dev., 12, 488–493.[Web of Science][Medline]

5. Mount, R.H., Charman, T., Hastings, R.P., Reilly, S. and Cass, H. (2003) Features of autism in Rett syndrome and severe mental retardation. J. Autism Dev. Disord., 33, 435–442.[CrossRef][Web of Science][Medline]

6. Olsson, B. (1987) Autistic traits in the Rett syndrome. Brain Dev., 9, 491–498.[Web of Science][Medline]

7. Olsson, B. and Rett, A. (1987) Autism and Rett syndrome: behavioural investigations and differential diagnosis. Dev. Med. Child Neurol., 29, 429–441.[Web of Science][Medline]

8. Olsson, B. and Rett, A. (1985) Behavioral observations concerning differential diagnosis between the Rett syndrome and autism. Brain Dev., 7, 281–289.[Web of Science][Medline]

9. Percy, A.K., Zoghbi, H.Y., Lewis, K.R. and Jankovic, J. (1988) Rett syndrome: qualitative and quantitative differentiation from autism. J. Child Neurol., 3 (suppl.), S65–67.

10. American Psychiatric Association, Task Force on DSM-IV (2000) Diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR, 4th edn. Washington DC.

11. Stokstad, E. (2001) Development. New hints into the biological basis of autism. Science, 294, 34–37.[Free Full Text]

12. Filipek, P.A., Accardo, P.J., Baranek, G.T., Cook, E.H., Jr, Dawson, G., Gordon, B., Gravel, J.S., Johnson, C.P., Kallen, R.J., Levy, S.E. et al. (1999) The screening and diagnosis of autistic spectrum disorders. J. Autism Dev. Disord., 29, 439–484.[CrossRef][Web of Science][Medline]

13. Percy, A.K. (2002) Rett syndrome. Current status and new vistas. Neurol. Clin., 20, 1125–1141.[CrossRef][Web of Science][Medline]

14. Carney, R.M., Wolpert, C.M., Ravan, S.A., Shahbazian, M., Ashley-Koch, A., Cuccaro, M.L., Vance, J.M. and Pericak-Vance, M.A. (2003) Identification of MeCP2 mutations in a series of females with autistic disorder. Pediatr. Neurol., 28, 205–211.[Web of Science][Medline]

15. Lam, C.W., Yeung, W.L., Ko, C.H., Poon, P.M., Tong, S.F., Chan, K.Y., Lo, I.F., Chan, L.Y., Hui, J., Wong, V. et al. (2000) Spectrum of mutations in the MECP2 gene in patients with infantile autism and Rett syndrome. J. Med. Genet., 37, E41.

16. Beyer, K.S., Blasi, F., Bacchelli, E., Klauck, S.M., Maestrini, E. and Poustka, A. (2002) Mutation analysis of the coding sequence of the MECP2 gene in infantile autism. Hum. Genet., 111, 305–309.[CrossRef][Web of Science][Medline]

17. Amir, R.E., Van den Veyver, I.B., Schultz, R., Malicki, D.M., Tran, C.Q., Dahle, E.J., Philippi, A., Timar, L., Percy, A.K., Motil, K.J. et al. (2000) Influence of mutation type and X chromosome inactivation on Rett syndrome phenotypes. Ann. Neurol., 47, 670–679.[CrossRef][Web of Science][Medline]

18. Leonard, H., Silberstein, J., Falk, R., Houwink-Manville, I., Ellaway, C., Raffaele, L.S., Engerstrom, I.W. and Schanen, C. (2001) Occurrence of Rett syndrome in boys. J. Child Neurol., 16, 333–338.[Abstract/Free Full Text]

19. Shahbazian, M., Young, J., Yuva-Paylor, L., Spencer, C., Antalffy, B., Noebels, J., Armstrong, D., Paylor, R. and Zoghbi, H. (2002) Mice with truncated MeCP2 recapitulate many Rett syndrome features and display hyperacetylation of histone H3. Neuron, 35, 243–254.[CrossRef][Web of Science][Medline]

20. Young, J.I. and Zoghbi, H.Y. (2004) X-chromosome inactivation patterns are unbalanced and affect the phenotypic outcome in a mouse model of Rett syndrome. Am. J. Hum. Genet., 74, 511–520.[CrossRef][Web of Science][Medline]

21. Paylor, R., Nguyen, M., Crawley, J.N., Patrick, J., Beaudet, A. and Orr-Urtreger, A. (1998) Alpha7 nicotinic receptor subunits are not necessary for hippocampal-dependent learning or sensorimotor gating: a behavioral characterization of Acra7-deficient mice. Learn. Mem., 5, 302–316.[Abstract/Free Full Text]

22. Edgar, D.M., Kilduff, T.S., Martin, C.E. and Dement, W.C. (1991) Influence of running wheel activity on free-running sleep/wake and drinking circadian rhythms in mice. Physiol. Behav., 50, 373–378.[CrossRef][Medline]

23. Edgar, D.M., Martin, C.E. and Dement, W.C. (1991) Activity feedback to the mammalian circadian pacemaker: influence on observed measures of rhythm period length. J. Biol. Rhythms, 6, 185–199.[Abstract/Free Full Text]

24. Van Oortmerssen, G.A. (1971) Biological significance, genetics and evolutionary origin of variability in behaviour within and between inbred strains of mice (Mus musculus). A behaviour genetic study. Behaviour, 38, 1–92.[Medline]

25. Lindzey, G., Winston, H. and Manosevitz, M. (1961) Social dominance in inbred mouse strains. Naturwissenschaften, 191, 474–476.

26. Bult, A. and Lynch, C.B. (1997) Nesting and fitness: lifetime reproductive success in house mice bidirectionally selected for thermoregulatory nest-building behavior. Behav. Genet., 27, 231–240.[CrossRef][Web of Science][Medline]

27. Szczypka, M.S., Kwok, K., Brot, M.D., Marck, B.T., Matsumoto, A.M., Donahue, B.A. and Palmiter, R.D. (2001) Dopamine production in the caudate putamen restores feeding in dopamine-deficient mice. Neuron, 30, 819–828.[CrossRef][Web of Science][Medline]

28. Ballard, T.M., Pauly-Evers, M., Higgins, G.A., Ouagazzal, A.M., Mutel, V., Borroni, E., Kemp, J.A., Bluethmann, H. and Kew, J.N. (2002) Severe impairment of NMDA receptor function in mice carrying targeted point mutations in the glycine binding site results in drug-resistant nonhabituating hyperactivity. J. Neurosci., 22, 6713–6723.[Abstract/Free Full Text]

29. Lijam, N., Paylor, R., McDonald, M.P., Crawley, J.N., Deng, C.X., Herrup, K., Stevens, K.E., Maccaferri, G., McBain, C.J., Sussman, D.J et al. (1997) Social interaction and sensorimotor gating abnormalities in mice lacking Dvl1. Cell, 90, 895–905.[CrossRef][Web of Science][Medline]

30. Long, J.M., LaPorte, P., Paylor, R. and Wynshaw-Boris, A. (2004) Expanded characterization of the social interaction abnormalities in mice lacking Dvl1. Genes Brain Behav., 3, 51–62.[CrossRef][Web of Science][Medline]

31. Sluyter, F., Bult, A., Lynch, C.B., van Oortmerssen, G.A. and Koolhaas, J.M. (1995) A comparison between house mouse lines selected for attack latency or nest-building: evidence for a genetic basis of alternative behavioral strategies. Behav. Genet., 25, 247–252.[CrossRef][Web of Science][Medline]

32. Crawley, J.N. and Paylor, R. (1997) A proposed test battery and constellations of specific behavioral paradigms to investigate the behavioral phenotypes of transgenic and knockout mice. Horm. Behav., 31, 197–211.[CrossRef][Medline]

33. Fombonne, E. and Chakrabarti, S. (2001) No evidence for a new variant of measles–mumps–rubella-induced autism. Pediatrics, 108, E58.

34. Mount, R.H., Charman, T., Hastings, R.P., Reilly, S. and Cass, H. (2002) The Rett syndrome behaviour questionnaire (RSBQ): refining the behavioural phenotype of Rett syndrome. J. Child Psychol. Psychiatry, 43, 1099–1110.[CrossRef][Web of Science][Medline]

35. Kielinen, M., Rantala, H., Timonen, E., Linna, S.L. and Moilanen, I. (2004) Associated medical disorders and disabilities in children with autistic disorder: a population-based study. Autism, 8, 49–60.[Abstract/Free Full Text]

36. Asato, M.R. and Hardan, A.Y. (2004) Neuropsychiatric problems in tuberous sclerosis complex. J. Child Neurol., 19, 241–249.[Abstract/Free Full Text]

37. Trillingsgaard, A. and Ostergaard, J.R. (2004) Autism in angelman syndrome: an exploration of comorbidity. Autism, 8, 163–174.[Abstract/Free Full Text]

38. Peters, S.U., Goddard-Finegold, J., Beaudet, A.L., Madduri, N., Turcich, M. and Bacino, C.A. (2004) Cognitive and adaptive behavior profiles of children with Angelman syndrome. Am. J. Med. Genet., 128A, 110–113.

39. Samaco, R.C., Nagarajan, R.P., Braunschweig, D. and LaSalle, J.M. (2004) Multiple pathways regulate MeCP2 expression in normal brain development and exhibit defects in autism-spectrum disorders. Hum. Mol. Genet., 13, 629–639.[Abstract/Free Full Text]

40. Bae, K., Jin, X., Maywood, E.S., Hastings, M.H., Reppert, S.M. and Weaver, D.R. (2001) Differential functions of mPer1, mPer2, and mPer3 in the SCN circadian clock. Neuron, 30, 525–536.[CrossRef][Web of Science][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				M. Adachi, A. E. Autry, H. E. Covington III, and L. M. Monteggia MeCP2-Mediated Transcription Repression in the Basolateral Amygdala May Underlie Heightened Anxiety in a Mouse Model of Rett Syndrome J. Neurosci., April 1, 2009; 29(13): 4218 - 4227. [Abstract] [Full Text] [PDF]

				J. Molina, P. Carmona-Mora, J. Chrast, P. M. Krall, C. P. Canales, J. R. Lupski, A. Reymond, and K. Walz Abnormal social behaviors and altered gene expression rates in a mouse model for Potocki-Lupski syndrome Hum. Mol. Genet., August 15, 2008; 17(16): 2486 - 2495. [Abstract] [Full Text] [PDF]

				B. Kerr, M. Alvarez-Saavedra, M. A. Saez, A. Saona, and J. I. Young Defective body-weight regulation, motor control and abnormal social interactions in Mecp2 hypomorphic mice Hum. Mol. Genet., June 15, 2008; 17(12): 1707 - 1717. [Abstract] [Full Text] [PDF]

				R. C. Samaco, J. D. Fryer, J. Ren, S. Fyffe, H.-T. Chao, Y. Sun, J. J. Greer, H. Y. Zoghbi, and J. L. Neul A partial loss of function allele of Methyl-CpG-binding protein 2 predicts a human neurodevelopmental syndrome Hum. Mol. Genet., June 15, 2008; 17(12): 1718 - 1727. [Abstract] [Full Text] [PDF]

				S. Jamain, K. Radyushkin, K. Hammerschmidt, S. Granon, S. Boretius, F. Varoqueaux, N. Ramanantsoa, J. Gallego, A. Ronnenberg, D. Winter, et al. Reduced social interaction and ultrasonic communication in a mouse model of monogenic heritable autism PNAS, February 5, 2008; 105(5): 1710 - 1715. [Abstract] [Full Text] [PDF]

				M. Alvarez-Saavedra, M. A. Saez, D. Kang, H. Y. Zoghbi, and J. I. Young Cell-specific expression of wild-type MeCP2 in mouse models of Rett syndrome yields insight about pathogenesis Hum. Mol. Genet., October 1, 2007; 16(19): 2315 - 2325. [Abstract] [Full Text] [PDF]

				B. E. McGill, S. F. Bundle, M. B. Yaylaoglu, J. P. Carson, C. Thaller, and H. Y. Zoghbi From the Cover: Enhanced anxiety and stress-induced corticosterone release are associated with increased Crh expression in a mouse model of Rett syndrome PNAS, November 28, 2006; 103(48): 18267 - 18272. [Abstract] [Full Text] [PDF]

				I. Nishijima, T. Yamagata, C. M. Spencer, E. J. Weeber, O. Alekseyenko, J. D. Sweatt, M. Y. Momoi, M. Ito, D. L. Armstrong, D. L. Nelson, et al. Secretin receptor-deficient mice exhibit impaired synaptic plasticity and social behavior Hum. Mol. Genet., November 1, 2006; 15(21): 3241 - 3250. [Abstract] [Full Text] [PDF]

				G. J. Pelka, C. M. Watson, T. Radziewic, M. Hayward, H. Lahooti, J. Christodoulou, and P. P. L. Tam Mecp2 deficiency is associated with learning and cognitive deficits and altered gene activity in the hippocampal region of mice Brain, April 1, 2006; 129(4): 887 - 898. [Abstract] [Full Text] [PDF]

				P. Moretti, J. M. Levenson, F. Battaglia, R. Atkinson, R. Teague, B. Antalffy, D. Armstrong, O. Arancio, J. D. Sweatt, and H. Y. Zoghbi Learning and Memory and Synaptic Plasticity Are Impaired in a Mouse Model of Rett Syndrome J. Neurosci., January 4, 2006; 26(1): 319 - 327. [Abstract] [Full Text] [PDF]

				J. I. Young, E. P. Hong, J. C. Castle, J. Crespo-Barreto, A. B. Bowman, M. F. Rose, D. Kang, R. Richman, J. M. Johnson, S. Berget, et al. Inaugural Article: Regulation of RNA splicing by the methylation-dependent transcriptional repressor methyl-CpG binding protein 2 PNAS, December 6, 2005; 102(49): 17551 - 17558. [Abstract] [Full Text] [PDF]

				H. Y. Zoghbi MeCP2 Dysfunction in Humans and Mice J Child Neurol, September 1, 2005; 20(9): 736 - 740. [Abstract] [PDF]

				N. Kishi and J. D. Macklis Dissecting MECP2 Function in the Central Nervous System J Child Neurol, September 1, 2005; 20(9): 753 - 759. [Abstract] [PDF]

				H. Y. Zoghbi MeCP2 Dysfunction in Humans and Mice J Child Neurol, August 1, 2005; 20(8): 736 - 740. [Abstract] [PDF]

				N. Kishi and J. D. Macklis Dissecting MECP2 Function in the Central Nervous System J Child Neurol, August 1, 2005; 20(8): 753 - 759. [Abstract] [PDF]

				R. C. Samaco, A. Hogart, and J. M. LaSalle Epigenetic overlap in autism-spectrum neurodevelopmental disorders: MECP2 deficiency causes reduced expression of UBE3A and GABRB3 Hum. Mol. Genet., February 15, 2005; 14(4): 483 - 492. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 14/2/205    most recent ddi016v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (48) Request Permissions Google Scholar Articles by Moretti, P. Articles by Zoghbi, H. Y. Search for Related Content PubMed PubMed Citation Articles by Moretti, P. Articles by Zoghbi, H. Y. Social Bookmarking          What's this?

Online ISSN 1460-2083 - Print ISSN 0964-6906

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics