Contents lists available at SciVerse ScienceDirect
Experimental Neurology
j ourna l homepage: www.e lsev ie r .com/ locate /yexnr
Adolescent hyperactivity and impaired coordination after neonatal hyperoxia
Thomas Schmitz a,⁎, Stefanie Endesfelder a, Marie-Christine Reinert b, Florian Klinker b, Susanne Müller c,Christoph Bührer a, David Liebetanz b
a Department of Neonatology, Charité University Medical Center, 13353 Berlin, Germanyb Department of Clinical Neurophysiology, Georg-August-University, 37099 Göttingen, Germanyc Berlin Center for Stroke Research, Charité University Medical Center, 10117 Berlin, Germany
⁎ Corresponding author at: Klinik für NeonatologieBerlin, Augustenburger Platz 1, D-13353 Berlin. Fax: +
In preterm infants, the risk to develop attention-deficit/hyperactivity disorder is 3 to 4-fold higher than interm infants. Moreover, preterm infants exhibit deficits in motor coordination and balance. Based on clinicaldata, higher oxygen levels in preterm infants lead to worse neurological outcome, and experimental hyper-oxia causes wide-ranging cerebral changes in neonatal rodents. We hypothesize that hyperoxia in the imma-ture brain may affect motor activity in preterm infants.We subjected newborn mice from P6 to P8 to 48 h of hyperoxia (80% O2) and tested motor activity in runningwheels starting at adolescent age P30. Subsequently, from P44 to P53, regular wheels were replaced by com-plex wheels with variable crossbar positions to assess motor coordination deficits. MRI with diffusion tensorimaging was performed in the corpus callosum to determine white matter diffusivity in mice after hyperoxiaat ages P30 and P53 in comparison to control animals.Adolescent mice after neonatal hyperoxia revealed significantly higher values for maximum velocity andmean velocity in regular wheels than controls (Pb0.05). In the complex running wheels, however, maximumvelocity was decreased in animals after hyperoxia, as compared to controls (Pb0.05). Decreased fractionalanisotropy and increased radial diffusion coefficient were observed in the corpus callosum of P30 and P53mice after neonatal hyperoxia compared to control mice.Hyperoxia in the immature brain causes hyperactivity, motor coordination deficits, and impaired whitematter diffusivity in adolescent and young adult mice.
Attention-deficit/hyperactivity disorder (ADHD) characterized byexcessive inattention, hyperactivity, and impulsivity, either alone orin combination (APA, 2000) is diagnosed in 8–12% of school age chil-dren born at term. Children born preterm even have a 3 to 4-foldhigher risk to develop ADHD than term infants (Aarnoudse-Moenset al., 2009; Biederman and Faraone, 2005; Delobel-Ayoub et al.,2009; Johnson et al., 2011). Apart from genetic linkage for ADHDfound in twin studies and in genome-wide linkage analysis studies(Sharp et al., 2009), environmental factors have also been associatedwith ADHD, often implicating adverse events during prenatal or earlypostnatal life, such as maternal alcohol use, asphyxia, and low birth-weight (Biederman and Faraone, 2005; Neuman et al., 2007; Pinedaet al., 2007; Saigal et al., 2003). In MRI studies, hyperactivity andADHD have been correlated to white matter pathologies both interm and preterm children (Qiu et al., 2011; Skranes et al., 2007).
Even though the rates of overt cerebral palsy are greatly increasedin preterm infants in comparison to term infants, modern medicalcare has recently caused a reduction of the incidence to less than 4per 100 liveborn very low birth weight infants (Platt et al., 2007). Asimilar decline has been observed for the hallmark lesions of overtcerebral palsy, cystic periventricular leukoamalacia. In contrast,there is a high prevalence of minor deficits of motor coordination aswell as of behavioral diseases such as ADHD persisting throughoutchildhood and adolescence of former very low birth weight infants(de Kieviet et al., 2009). Both locomotor deficits and ADHD are asso-ciated with diffuse periventricular white matter damage (PWMD)which is characterized by hypomyelination and requires magneticresonance imaging (MRI) and diffusion tensor imaging (DTI) for diag-nosis (Cheong et al., 2009; Constable et al., 2008; Counsell et al.,2008). While infection and ischemia are being thought to underliecystic periventricular leukomalacia, some observational data haverather implicated hyperoxia in the pathogenesis of PWMD (Collinset al., 2001; Deulofeut et al., 2007).
While in utero, paO2 is maintained at low levels of 24–28 mm Hg(Hoffmann, 2002), preterm infants after birth experience a several-fold increase in arterial oxygen tension to 65–80 mm Hg, even with-out supplemental oxygen (Castillo et al., 2008). At the same time,
Fig. 1. Regular and complex running wheel for adolescent and adult mice.Mice afterneonatal hyperoxia and control mice had access to running wheels composed of regu-larly spaced crossbars for 2 weeks (P30–P43). Thereafter, running performance wasrecorded for additional ten days (P44–P53) on complex wheels with variable crossbarpositions.
Table 1Motor behavior in mice exposed to hyperoxia obtained in regular and complex runningwheels.
Running parameter Variables Degrees offreedom
F values P value
Conventional running wheelsVmax Time course 12 9.653 b0.001*
Time course×group 12 0.395 0.945Group 1 5.058 0.033*
Shown is the overview of values for Vmax = maximum velocity achieved within 24 h,Vmean = mean running velocity averaged for all running periods within 24 h, Distac= accumulative distance per 24 h, Nrun = numbers of runs performed per 24 h.
375T. Schmitz et al. / Experimental Neurology 235 (2012) 374–379
the immature brain has a markedly lower antioxidant capacity andimmature stages of oligodendroglial cells show a particular suscepti-bility to oxidative stress (Deng et al., 2003 , Gerstner et al., 2007). In amouse model for neonatal hyperoxia, 48 h exposure to 80% O2 fromP6 to P8 resulted in cell death, decreased proliferation, and delayedmaturation of oligodendroglial precursor cells followed by hypomye-lination of the white matter (Gerstner et al., 2008; Schmitz et al.,2011). However, behavioral changes caused by exposure of theimmature brain to hyperoxia have not been investigated yet.
In the present study, a running wheel technique established foranalysis of running activity and motor learning behavior in mice(Dowling et al., 2011; Hibbits et al., 2009; Liebetanz and Merkler,2006; Liebetanz et al., 2007) was used to analyze motor behaviorafter neonatal exposure to hyperoxia in comparison to matchedcontrol mice. In addition, diffusivity in the corpus callosum was mea-sured by MRI with DTI. The results of this study implicate unphysiolo-gical increase of oxygen tension in the immature brain to causehyperactivity and impaired motor learning in association with per-turbed white matter diffusivity during later development.
Methods
All animal experiments were performed in accordance with inter-national guidelines for good laboratory practice and were approvedby the animal welfare committees of both study centers in Berlinand Göttingen.
Hyperoxia exposure
Six-day-old (P6) C57B/6J wild-type mice were subjected to 48 hhyperoxia till P8. Litters were divided into hyperoxia and controlgroups. Pups exposed to hyperoxia were placed along with theirmothers, in a chamber containing 80% O2. This procedure of exposureto hyperoxia has been demonstrated elsewhere to cause a 3 to 4-foldincrease of oxygen tension in the blood of newborn mice (Schmitzet al., 2011). The control pups of each litter were kept in room-airwith a second lactating mother. The mothers of the two groupswere replaced after 24 h to prevent oxygen-induced acute lung injury(Taglialatela et al., 1998). During recovery in room air, all pupsexposed to hyperoxia were reunited with their biological motheruntil the age of P30 when MRI or running wheel tests were started.The pups appeared normal and did not suffer weight loss or othermorbidities.
Voluntary wheel running behavior
Investigators performing the running wheel experiments wereblinded for group assignment of mice to control versus hyperoxia.From P30, all mice were separated into single cages and had continu-ous and individual access to a running wheel (Fig. 1) placed insideeach cage. The axis of each running wheel was connected to a rotationsensor with a resolution of 16 counts/revolution, one revolution cor-responding to 35.5 cm. Using a customized recording device and soft-ware (Boenig and Kallenbach oHG), running wheel revolutions wererecorded continuously at a sampling rate of 1/0.48 s. Based on theseaccumulative data, several parameters were calculated using a cus-tom designed MatLab program (The MathWorks). To assess wheelrunning activity as a possible correlate for hyperactivity as one symp-tom of ADHD and also to determine physical fitness and endurance ofthe mouse throughout the experiment, several parameters weremeasured (Table 1), i.e. maximum velocity (Vmax), mean velocity(Vmean), number of runs per 24 h (Nrun), accumulative distanceper 24 h (Distac). Vmax in particular has been demonstrated to pro-vide high intraindividual stability (Liebetanz et al., 2007).
At a time when all mice had reached a stable plateau on the con-ventional running wheels, the wheels were replaced by complex
running wheels with irregularly spaced cross bars on day 13 of thebehavioral experiment, where the wheel running activity was moni-tored for further 10 days.
The complex design of the latter wheels (Fig. 1) was chosen tomake wheel running more difficult in such a way that any improve-ment of running performance on these wheels requires motor learn-ing and coordinative skills' functions on a supraspinal level (Liebetanzand Merkler, 2006).
Magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI)
Investigators performing the MRI measurements were blinded forgroup assignment of mice to control versus hyperoxia. MRI was per-formed using a 7 Tesla rodent scanner (Pharmascan 70⁄16, BrukerBioSpin, Ettlingen, Germany) with a 16 cm horizontal bore magnetand a 9 cm (inner diameter) shielded gradient with an H-
376 T. Schmitz et al. / Experimental Neurology 235 (2012) 374–379
resonance-frequency of 300 MHz and a maximum gradient strengthof 300 mT/m. For imaging a 1H-Phased-Array surface coil for mousehead and a
1H-RF-Volumeresonator (72 mm) for transmission were
used. Data acquisition and image processing were carried out withthe Bruker software Paravision 4.0. During the examinations micewere anaesthetized with 2.0–1.0% isoflurane (Forene, Abbot, Wiesba-den, Germany) delivered in a O2/N2O mixture under constant ventila-tion monitoring (Small Animal Monitoring & Gating System, SAInstruments, Stony Brook, New York, USA). To ensure physiologicalbody temperature animals were placed on a heated circulatingwater blanket. To localize the corpus callosum we first used a T2-weighted 2D turbo spin-echo sequence (TR/TE=4200/36 ms, rarefactor 8, 4 averages) with a field of view (FOV) of 2.56×2.56 cmand a matrix size of 256×256 with 20 slices at 0.5 mm to cover thewhole brain. For following DTI, a 1 mm slice with a FOV of2.60×2.60 cm and a matrix size of 128×128 was placed over the cor-pus callosum. Imaging parameters are: DTI Epi with Spin Echo prepa-ration and 16 diffusion directions, b-value=1000 s/mm2, diffusionduration δ=4ms, diffusion separation Δ=20 ms, TR/TE=3000/34 ms. If there were movement artefacts, the whole data set of the an-imal was excluded from analysis. On a pixel by pixel basis, fractionalanisotropy and apparent diffusion coefficient (ADC) for radial (longi-tudinal) and axial (perpendicular) diffusivity were calculated for thecorpus callosum using Bruker software Paravision 4.0. One value peranimal was obtained by averaging the measurements of nine separatepixels along the radial directory of corpus callosum fibers. Direction-ally encoded color (DEC) maps were used to represent anisotropy inthree directional manners in coronal images; red for lateral–medial,blue for anterior–posterior, green for in–out (Pajevic and Pierpaoli,1999). Given the axonal directory in the corpus callosum, the domi-nant direction for fractional anisotropy in this region is medial–lateral(red).
Statistics
All data of running wheel experiments are expressed as mean±S.E.M. The effect of hyperoxia on the wheel running performanceas assessed by the different running parameters was tested by 2-factorial repeated measurement ANOVAs for each running parameter.Separate ANOVAs were performed for the phase on the conventionalrunning wheels (days 1–13) and for the phase on the complexrunning wheels (days 14–23). Group (normoxia versus hyperoxia)and time course were defined as independent variables whereas thecurrent running parameter served as the dependent variable. Datafrom complex wheels were normalized to the individual plateauvalues on the conventional running wheel.
Post hoc tests (t-test, one-tailed) were performed in order tocompare the values of both groups separately for each time point, ifsignificant effects were indicated by the ANOVAs. A P level of b0.05was considered significant in all tests. Statistics were performedwith SPPS™ 11.0 software (Chicago, IL, USA).
Values of FA, radial and axial diffusivity data obtained by MRI withDTI showed nonnormality in the Kolmogorov–Smirnov test. Resultsof diffusivity are therefore given as median and interquartile range.GraphPadPrism™ software (San Diego, CA, USA) was used forMann–Whitney-U test; a P valueb0.05 was considered significant.
Results
Motor behavior after neonatal hyperoxia: wheel running
The wheel running experiment was started with 15 control miceand 15 mice exposed to hyperoxia. During its course, one animalhad to be excluded because of unexpected death at day 13 of the run-ning experiment. A subsequent autopsy of this animal revealed anabnormal reduced heart size pointing to cardiac failure as possible
cause of death. Three mice were excluded for the time on the complexrunning wheel: two of them caught their neck in the running wheeland needed to recover for the rest of the experiment, the third onehad stopped wheel running for unknown reasons. Finally, therewere data available from 29 mice (n=14 hyperoxia; n=15 nor-moxia) for analysis of the regular running wheels, and from 26 micefor the complex wheels (n=13 each group).
Training phaseThe ANOVAs (Table) for the phase on the conventional running
wheels revealed a significant effect of time in all measured parame-ters except for Nrun, indicating that running performance in termsof velocity (Vmax and Vmean), and distance (Distac) increased dur-ing training period. In addition, the factor group (normoxia versushyperoxia) was significant in the ANOVAs for the velocity parametersVmax and Vmean (Table). However, interactions of time and groupwere not significant in the ANOVAs of any parameter. This indicatesthat both groups had comparable training curves on the regular run-ning wheel and corresponds to the fact that all mice improved theirrunning performance during this period as assessed by Distac,Vmax, and Vmean during the first week and then reached a plateaulevel in the second week (Fig. 2).
However, the significant effect of the factor group for the velocityparameters indicates that neonatal exposure to hyperoxia specificallyinfluenced these parameters. According to the post hoc tests, animalsafter hyperoxia reached higher values for Vmax and Vmean at severaldays of the training phase (Fig. 2). Hence, on conventional runningwheels, mice after neonatal hyperoxia ran faster than control mice.Mice after hyperoxia also had higher values in Distac, but this differ-ence did not reach significance on statistical analysis.
Learning phaseAfter replacement of conventional by complex running wheels,
mice of both groups dramatically dropped in all running parameters.Thereafter, mice improved their motor performance in the complexwheels. The ANOVAs for the phase during which the mice had accessto the complex running wheels (days 14–24) revealed a main effectof time course with regards to Vmax, Vmean, Distac and Nrun(Table). This indicates that all mice improved their running abilityafter the overall drop during the initial days on the complex runningwheels.
The factor group was significant in the ANOVA for the parameterVmax (Table), while the interactions of time and group were not sig-nificant for any parameter. This indicates that postnatal hyperoxiaalso had an effect on the learning curve on the complex wheels.There was a trend in mice after hyperoxia for lower Distac, but statis-tical analysis did not reveal significance (Table).
Post hoc test showed that mice after hyperoxia reached a signifi-cantly lower maximum velocity (Vmax, Fig. 2) indicating a decreasedcapacity to compensate the motor challenge encompassed by thecomplex running wheels.
Hyperoxia changes diffusivity in the corpus callosum
In order to determine whether hyperactivity and impaired motorcoordination in mice after neonatal exposure to hyperoxia correlateto changes of white matter diffusion characteristics, we measuredFA as well as axial and radial diffusivity in the corpus callosum atP30 and at P53. Therefore, we performed T2-weighted magnetic res-onance imaging (MRI) sequences to select the coronal slice for DTImeasurement and generated directionally encoded color (DEC)maps to represent anisotropy in the lateral-medial directional man-ners of the corpus callosum. In the anisotropy measurements, FAwas significantly decreased in the corpus callosum of P30 animalsthat had been exposed to hyperoxia from P6 to P8 (median=0.377;25–75% interval 0.357–0.423) compared to controls always kept in
Max
imum
vel
ocity
in
met
ers/
min
ute
(Vm
ax)
Mea
n ve
loci
ty in
met
ers/
min
ute
(Vm
ean)
A) Conventional running wheels B) Complex running wheels
5
10
15
20
1 2 3 4 5 6 7 8 9 10 1 1 12 130
20
40
60
80
100
14 15 16 17 18 19 20 21 22 23 24
0
20
40
60
80
100
120
14 15 16 17 18 19 20 21 22 23 24
Max
imum
vel
ocity
in %
(V
max
)M
ean
velo
city
in %
(V
mea
n)A
ccum
ulat
ive
dist
ance
in %
(D
ista
c)
* **
* * * * *
* *
45
50
55
60
65
1 2 3 4 5 6 7 8 9 10 1 1 12 1355
65
75
85
95
14 15 16 17 18 19 20 21 22 23 24
Normoxia Hyperoxia
* * * * * *
* * * *
*
Time in days
* ** * * * * ** *
Acc
umul
ativ
e di
stan
cein
10x
3 m
eter
s (D
ista
c)
0
2
4
6
8
10
12
1 2 3 4 5 6 7 8 9 10 1 1 12 13
Fig. 2. Wheel running performance in mice is affected by neonatal hyperoxia.Mice after hyperoxia are shown by circles, control mice by triangles. Left panels: Training phase onconventional running wheels (days 1–14; P30–P43). Right panels: Learning phase on complex running wheels (days 15–24; P44–P53). If ANOVAs showed significant differencesfor the factor group, post hoc tests were applied (* = Pb0.05). All data represent 24 h time blocks and are shown as mean±S.E.M.
377T. Schmitz et al. / Experimental Neurology 235 (2012) 374–379
room air (median=0.480; 0.447–0.506; n=10; Mann–Whitney ranktest, Pb0.01) (Fig. 3). This significant decrease of FA in the corpus cal-losum was still present at P53 (median=0.490; 25–75% interval0.462–0.530; control median=0.573; 0.523–0.605; n=10, Pb0.01)(Fig. 2).
The apparent diffusion coefficient (ADC) for radial diffusivity wasincreased after hyperoxia both at P30 (median=0.769×10−3 mm2/s;25–75% interval 0.715–0.805) and at P53 (median=0.610×10−3 mm2/s; 0.583–0.638) in comparison to age matched controls(P30 median=0.627×10−3 mm2/s; 25–75% interval 0.613–0.690;P53 median=0.510×10−3 mm2/s; 0,470–0,523) (Mann–Whitneyrank test: both ages Pb0.001, n=10) (Fig. 3). Axial diffusivity wassimilar in both groups, with a median ADC of 1.154×10−3 mm2/sin controls and 1.170×10−3 mm2/s after hyperoxia at P30, and1.180×10−3 mm2/s and 1.168×10−3 mm2/s, respectively, at P53(data not shown).
Hence, based on these data, neonatal hyperoxia leads to long termimpairment of white matter diffusivity.
Discussion
In our study in mice, 48 h exposure from P6 to P8 to 80% oxygencaused hyperactivity at young adult ages. In a running wheel systemestablished for motor learning analysis, mice at P30 after hyperoxiaexhibited higher velocity of wheel running than matched control an-imals always kept in normal room air. However, replacement of the
regular wheel by a complex wheel with irregularly spaced crossbarsrevealed impaired capacity of mice after hyperoxia to compensatethe complex motor challenge. Moreover, the changes of motor behav-ior correlated to perturbed white matter diffusivity in the corpus cal-losum determined by MRI with diffusion tensor imaging (DTI). Thiscombination is reminiscent of the phenotype of a substantial propor-tion of children born very preterm (de Kieviet et al., 2009; Johnsonet al., 2011)
In children and adolescents born preterm, PWMD and globalwhite matter reduction are associated with poor academic perfor-mance, behavioral difficulties, and subtle motor impairment(Northam et al., 2011; Shah et al., 2008). PWMD most likely occursdue to the selective vulnerability of late oligodendroglial progenitorcells (OPC) which die or display arrested maturation in response tonoxious stimuli. These cellular pathologies perturb development ofother brain white matter structures such as the corpus callosum. Fail-ure of normal myelination has been established as a hallmark ofPWMD by diffusion tensor imaging studies (Counsell et al., 2008),although there are conflicting data for white matter tracts otherthan the corpus callosum (Bonifacio et al., 2010). While previousmodels have invoked hypoxia–ischemia and inflammation as culpritsof oligodendroglial degeneration and maturational arrest, recent ani-mal models have shown that hyperoxia in neonatal rodents (80% O2
for 24–48 h) causes apoptosis, decreased proliferation, and delayedmaturation of OPC and subsequent hypomyelination (Schmitz et al.,2011). The results presented here demonstrate that these early
Fig. 3. Long term of diffusivity in the corpus callosum after neonatal hyperoxia.(A,B) Coronal black-and-white images of T2-weighted MRI in P30 and P53 previously exposed to 48 hhyperoxia from P6 to P8 in comparison to age matched control mice always kept in room air. At the colored images, fractional anisotropy (FA) in medial-lateral directed fibers of thecorpus callosum (CC) is shown in red by directionally encoded color (DEC) maps. The region of CC selected for diffusion tensor imaging (DTI) measurements is indicated by whiteboxes, dotted lines are pointing to enlarged images of the CC. (A–C) FA in the CC is significantly decreased at P30 and P53 in mice after neonatal exposure to hyperoxia in compar-ison to controls. (D) Apparent diffusion coefficient (ADC) for radial diffusivity is significantly increased in the CC both at P30 and at P53 in mice that were previously exposed tohyperoxia from P6 to P8. Notably, within control animals, there is a developmentally regulated increase of FA and a decrease of radial diffusivity from P30 to P53, which is in agree-ment with continuing white matter development at these ages (Chahboune et al., 2007). Data are shown as boxes-and-whiskers diagrams (n=10 animals per group, using aMann–Whitney rank test *Pb0.01 and **Pb0.001).
378 T. Schmitz et al. / Experimental Neurology 235 (2012) 374–379
histological changes in response to hyperoxia translate into long termimpaired white matter integrity as revealed by decreased FA and in-creased radial diffusivity in DTI, and into altered brain function asdemonstrated by hyperactivity and motor coordination dysfunctionin the running-wheel setting of grown-up animals.
In former preterm infants at adolescent ages, decreased FA in thecorpus callosum has been correlated to ADHD symptoms, pointingto impaired white matter integrity as a factor for ADHD in preterm in-fants (Johnson et al., 2010; Skranes et al., 2007). In our DTI analysis,callosal FA in mice after neonatal hyperoxia was decreased both atage P30 and P53 as a sign of microstructural white matter changeslong term. The increase in radial diffusivity in the corpus callosumof these mice after hyperoxia are likely to be caused by hypomyelina-tion that has been demonstrated before to be caused by exposure tohyperoxia in newborn mice and rats through the analysis of myelinbasic protein expression (Gerstner et al., 2008; Schmitz et al., 2011).In contrast, axial diffusivity was equal in both controls and miceafter hyperoxia, which has often been interpreted as absence of axo-nal damage (Song et al., 2002). Hence, our data on hyperactivity inthe hyperoxia model are in concordance both with DTI characteristics
found in the corpus callosum of VLBW preterm infants and in patientswith ADHD both born preterm and term (Nagel et al., 2011; Skraneset al., 2007).
A broad array of factors has been associated with ADHD, both ofgenetical and environmental nature (Biederman and Faraone, 2005).Mouse models of genetic etiology of ADHD have been created involv-ing G-coupled receptor kinase-interacting protein-1 mutations (Wonet al., 2011), X-monosomy (Davies et al., 2007), and Cdk5-activatingcofactor p35 lack (Drerup et al., 2010). Adverse environmental eventsduring early brain development have also been associated withADHD, however, these factors do not explain the increased riskin preterm infants to develop ADHD compared to term infants(Biederman and Faraone, 2005). Environmental models of ADHDavailable so far are based on brain insults unrelated to brain injuryin preterm infants, i.e. prenatal exposure to alcohol, cerebrallesion by 5-hydroxydopamine injection, exposure to heavy metals(Kostrzewa et al., 2008). Moreover, there is no experimental modelavailable to mimic ADHD as a result of perturbed brain developmentin preterm infants, and causes of ADHD in the immature brain remainunclear.
379T. Schmitz et al. / Experimental Neurology 235 (2012) 374–379
In the rodent hyperoxia model, the toxic stimulus is in agreementto environmental risks in preterm infants, and injury occurs througholigodendroglial deficits and hypomyelination (Schmitz et al., 2011)both of which represent important features of white matter damagein the immature brain. The present study suggests that hyperoxiain the immature brain may be an important factor for the increasedADHD incidence in preterm infants. Furthermore, the hyperoxiamodel utilized provides a novel scientific tool for further researchon environmental etiology of ADHD.
Acknowledgments
Marie-Christine Reinert was supported by the Jacob Henle Pro-gram for Experimental Medicine of the Medical Faculty, Universityof Göttingen.
References
Aarnoudse-Moens, C.S., Weisglas-Kuperus, N., van Goudoever, J.B., Oosterlaan, J., 2009.Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birthweight children. Pediatrics 124, 171–728.
APA, 2000. Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR). Amer-ican Psychiatric Association, Washington, DC.
Bonifacio, S.L., Glass, H.C., Chau, V., Berman, J.I., Xu, D., Brant, R., Barkovich, A.J., Poskitt,K.J., Miller, S.P., Ferriero, D.M., 2010. Extreme premature birth is not associatedwith impaired development of brain microstructure. J. Pediatr. 157, 729–732 e721.
Castillo, A., Sola, A., Baquero, H., Neira, F., Alvis, R., Deulofeut, R., Critz, A., 2008. Pulseoxygen saturation levels and arterial oxygen tension values in newborns receivingoxygen therapy in the neonatal intensive care unit: is 85% to 93% an acceptablerange? Pediatrics 121, 882–889.
Chahboune, H., Ment, L.R., Stewart, W.B., Ma, X., Rothman, D.L., Hyder, F., 2007. Neuro-development of C57B/L6 mouse brain assessed by in vivo diffusion tensor imaging.NMR Biomed. 20, 375–382.
Cheong, J.L.Y., Thompson, D.K., Wang, H.X., 2009. Abnormal white matter signal on MRimaging is related to abnormal tissue microstructure. AJNR Am. J. Neuroradiol. 30,623–628.
Collins, M.P., Lorenz, J.M., Jetton, J.R., Paneth, N., 2001. Hypocapnia and otherventilation-related risk factors for cerebral palsy in low birth weight infants.Pediatr. Res. 50, 712–719.
Constable, R.T., Ment, L.R., Vohr, B.R., Kesler, S.R., Fulbright, R.K., Lacadie, C., Delancy, S.,Katz, K.H., Schneider, K.C., Schafer, R.J., Makuch, R.W., Reiss, A.R., 2008. Premature-ly born children demonstrate white matter microstructural differences at 12 yearsof age, relative to term control subjects: an investigation of group and gendereffects. Pediatrics 121, 306–316.
Counsell, S.J., Edwards, A.D., Chew, A.T., Anjari, M., Dyet, L.E., Srinivasan, L., Boardman,J.P., Allsop, J.M., Hajnal, J.V., Rutherford, M.A., Cowan, F.M., 2008. Specific relationsbetween neurodevelopmental abilities and white matter microstructure in chil-dren born preterm. Brain 131, 3201–3208.
Davies, W., Humby, T., Isles, A.R., Burgoyne, P.S., Wilkinson, L.S., 2007. X-monosomy ef-fects on visuospatial attention in mice: a candidate gene and implications for Turn-er syndrome and attention deficit hyperactivity disorder. Biol. Psychiatry 61,1351–1360.
de Kieviet, J.F., Piek, J.P., Aarnoudse-Moens, C.S., Oosterlaan, J., 2009. Motor develop-ment in very preterm and very low-birth-weight children from birth to adoles-cence: a meta-analysis. JAMA 302 (20), 2235–2242 Nov 25, 302, 2235–2242.
Delobel-Ayoub, M., Arnaud, C., White-Koning, M., Casper, C., Pierrat, V., Garel, M.,Burguet, A., Roze, J.C., Matis, J., Picaud, J.C., Kaminski, M., B L., EPIPAGE StudyGroup, 2009. Behavioral problems and cognitive performance at 5 years of ageafter very preterm birth: the EPIPAGE Study. Pediatrics 123, 1485–1492.
Deng, W., Rosenberg, P.A., Volpe, J.J., Jensen, F.E., 2003. Calcium-permeable AMPA/kainate receptors mediate toxicity and preconditioning by oxygen-glucose depri-vation in oligodendrocyte precursors. Proc. Natl. Acad. Sci. U. S. A. 100, 6801–6806.
Deulofeut, R., Dudell, G., AT, S., 2007. Treatment-by-gender effect when aiming to avoidhyperoxia in preterm infants in the NICU. Acta Paediatr. 96, 990–994.
Dowling, P., Klinker, F., Stadelmann, C., Hasan, K., Paulus, W., Liebetanz, D., 2011. Dopa-mine D3 receptor specifically modulates motor and sensory symptoms in iron-deficient mice. J. Neurosci. 31, 70–77.
Drerup, J.M., Hayashi, K., Cui, H., Mettlach, G.L., Long, M.A., Marvin, M., Sun, X.,Goldberg, M.S., Lutter, M., Bibb, J.A., 2010. Attention-deficit/hyperactivity pheno-
type in mice lacking the cyclin-dependent kinase 5 cofactor p35. Biol. Psychiatry68, 1163–1171.
Gerstner, B., Sifringer, M., Dzietko, M., Schüller, A., Lee, J., Simons, S., Obladen, M., Volpe,J.J., Rosenberg, P.A., Felderhoff-Mueser, U., 2007. Estradiol attenuates hyperoxia-induced cell death in the developing white matter. Ann. Neurol. 61, 562–573.
Gerstner, B., DeSilva, T.M., Genz, K., Armstrong, A., Brehmer, F., Neve, R.L., Felderhoff-Mueser, U., Volpe, J.J., Rosenberg, P.A., 2008. Hyperoxia causes maturation-dependent cell death in the developing white matter. J. Neurosci. 28, 1236–1245.
Hibbits, N., Pannu, R., Wu, T.J., Armstrong, R.C., 2009. Cuprizone demyelination of thecorpus callosum in mice correlates with altered social interaction and impairedbilateral sensorimotor coordination. ASN Neuro 1. doi:10.1042/AN20090032 pii:e00013.
Hoffmann, J.I.E., 2002. Basis science: the circulatory system. August In: Rudolph, C.D.,AM, R, Hostetter, M.K., Lister, G., Siegel, N.J. (Eds.), Rudolph's Pediatrics. Mcgraw-Hill Professional, p. 1750.
Johnson, S., Hollis, C., Kochhar, P., Hennessy, E., Wolke, D., Marlow, N., 2010. Psychiatricdisorders in extremely preterm children: longitudinal finding at age 11 years in theEPICure study. J. Am. Acad. Child Adolesc. Psychiatry 49, 453–463.
Johnson, S., Wolke, D., Hennessy, E., Marlow, N., 2011. Educational outcomes inextremely preterm children: neuropsychological correlates and predictors ofattainment. Dev. Neuropsychol. 36, 74–95.
Kostrzewa, R.M., Kostrzewa, J.P., Kostrzewa, R.A., Nowak, P., Brus, R., 2008. Pharmaco-logical models of ADHD. J. Neural. Transm. 115, 287–298.
Liebetanz, D., Merkler, D., 2006. Effects of commissural de- and remyelination on motorskill behaviour in the cuprizone mouse model of multiple sclerosis. Exp. Neurol.202, 217–224.
Liebetanz, D., Baier, P.C., Paulus, W., Meuer, K., Bähr, M., Weishaupt, J.H., 2007. A highlysensitive automated complex running wheel test to detect latent motor deficits inthe mouse MPTP model of Parkinson's disease. Exp. Neurol. 502, 207–213.
Nagel, B.J., Bathula, D., Herting, M., Schmitt, C., Kroenke, C.D., Fair, D., Nigg, J.T., 2011.Altered white matter microstructure in children with attention-deficit/hyperactivity disorder. J. Am. Acad. Child Adolesc. Psychiatry 60, 283–290.
Neuman, R.J., Lobos, E., Reich, W., Henderson, C.A., Sun, L.-W., Todd, R.D., 2007. Prenatalexposure and dopaminergic genotypes interact to cause a severe ADHD subtype.Biol. Psychiatry 61, 1320–1328.
Northam, G.B., Liégeois, F., Chong, W.K., Wyatt, J.S., Baldeweg, T., 2011. Total brainwhite matter is a major determinant of IQ in adolescents born preterm. Ann. Neu-rol. 69, 702–711. doi:10.1002/ana.22263.
Pajevic, S., Pierpaoli, C., 1999. Color schemes to represent the orientation of anisotropictissues from diffusion tensor data: application to white matter fiber tract mappingin the human brain. Magn. Reson. Med. 42, 526–540.
Pineda, D.A., Guillermo Palacio, L., Puerta, I.C., Merchán, V., Arango, C.P., Galvis, A.Y.,Gómez, M., Camilo Aguirre, D., Lopera, F., Arcos-Burgos, M., 2007. Environmentalinfluences that affect attention deficit/hyperactivity disorder. Eur. Child Adolesc.Psychiatry 16, 337–346.
Platt, M.J., Cans, C., Johnson, A., Surman, G., Topp, M., Torrioli, M.G., Krägeloh-Mann, I.,2007. Trends in cerebral palsy among infants of very low birthweight (b1500 g) orborn prematurely (b32 weeks) in 16 European centres: a database study. Lancet369, 43–50.
Qiu, M.G., Ye, Z., Li, Q.Y., Liu, G.J., Xie, B., Wang, J., 2011. Changes of brain structure andfunction in ADHD children. Brain Topogr. 24 (3–4), 243–252. [Epub 2010 Dec 30].
Saigal, S., Pinelli, J., Hoult, L., Kim, M.M., Boyle, M., 2003. Psychopathology and socialcompetencies of adolescents who were extremely low birth weight. Pediatrics111, 969–975.
Shah, D.K., Doyle, L.W., Anderson, P.J., Bear, M., Daley, A.J., Hunt, R.W., Inder, T.E., 2008.Adverse neurodevelopment in preterm infants with postnatal sepsis or necrotizingenterocolitis is mediated by white matter abnormalities on magnetic resonanceimaging at term. J. Pediatr. 153, 170–175.
Sharp, S.I., McQuillin, A., Gurling, H.M., 2009. Genetics of attention-deficit hyperactivitydisorder (ADHD). Neuropharmacology 57, 590–600.
Skranes, J., Vangberg, T.R., Kulseng, S., Indredavik, M.S., Evensen, K.A., Martinussen, M.,Dale, A.M., Haraldseth, O., Brubakk, A.M., 2007. Clinical findings and white matterabnormalities seen on diffusion tensor imaging in adolescents with very lowbirth weight. Brain 130, 654–666.
Taglialatela, G., Perez-Polo, J.R., Rassin, D.K., 1998. Induction of apoptosis in the CNSduring development by the combination of hyperoxia and inhibition of glutathionesynthesis. Free Radic. Biol. Med. 25, 936–942.
Won, H., Mah, W., Kim, E., Kim, J.W., Hahm, E.K., Kim, M.H., Cho, S., Kim, J., Jang, H., Cho,S.C., Kim, B.N., Shin, M.S., Seo, J., Jeong, J., Choi, S.Y., Kim, D., Kang, C., Kim, E., 2011.GIT1 is associated with ADHD in humans and ADHD-like behaviors in mice. Nat.Med. 17, 566–572.
