maternal exposure to lipopolysaccharide leads to transient motor dysfunction in neonatal rats

E-Mail [email protected]

Dev Neurosci DOI: 10.1159/000346579

Maternal Exposure to Lipopolysaccharide Leads to Transient Motor Dysfunction in Neonatal Rats

Catherine I. Rousset a, f Jinane Kassem a Arnaud Aubert c Deborah Planchenault b Pierre Gressens e, f Sylvie Chalon a Catherine Belzung a, d Elie Saliba a, b

a INSERM U930 and b Réanimation pédiatrique et Néonatologie, CHRU Tours, and c PAV EA2114 and d Departement des Sciences du Comportement, Université François Rabelais de Tours, Tours , and e INSERM U676, Faculté de Médecine, Université Diderot, and PremUP, Paris , France; f Centre for the Developing Brain, Department of Perinatal Imaging and Health, Division of Imaging & Biomedical Engineering, The Rayne Institute, King’s College London, St Thomas’ Hospital, London , UK

tion performances compared to vehicle. No myelination dif-ferences have been observed in the brains at adulthood. Ma-ternal LPS administration results in delayed motor develop-ment even though these alterations fade to reach control level by 5 weeks. Motor impairments observed at the early stage in this study could be linked to previously reported hypomyelination of the white matter induced by maternal LPS challenge in the neonates. Finally, the normal myelina-tion shown here at adulthood may explain the functional re-covery of the animals and suggest either a potential remye-lination of the brain or a delayed myelination in LPS pups.

Copyright © 2013 S. Karger AG, Basel

Introduction

White matter injury (WMI) is a severe neuropatholo-gy characterized by damage of the periventricular white matter combined with more diffuse injuries [1] . It is one of the most important lesions that occur in the immature brain and result in lifelong disorders of movement, pos-ture and cognition like cerebral palsy (CP) [2, 3] . The physiopathology of WMI is complex and maternal infec-tion is known as an important factor involved in its de-velopment [4, 5] . The maternal immune response follow-

Key Words

Motor development · Lipopolysaccharide · Hypomyelination · White matter disease

Abstract

Epidemiological and experimental data implicate maternal infection and inflammation in the etiology of brain white matter injury, which may lead to cerebral palsy in preterm newborns. Our aim was to investigate motor development of the offspring after maternal administration of lipopolysac-charide (LPS). Wistar rats were intraperitoneally injected with Escherichia coli LPS or saline on gestational days 19 and 20. From birth to 3 weeks, pups were tested for neurobehavioral development, neurological signs and reflexes. From 3 to 6 weeks, motor coordination was investigated. At 4 months, animals were tested for locomotion. Brain myelination was assessed by myelin basic protein immunohistochemistry. Days of appearance of several neurological reflexes were sig-nificantly delayed, and neonate LPS pups displayed retarded performance in righting, gait and negative geotaxis. At the juvenile stage, LPS animals showed important impairment in coordination. However, although the LPS group performed worse in most tests, they reached vehicle levels by 5 weeks. At 4 months, LPS animals did not show variations in locomo-

Received: August 23, 2012 Accepted after revision: December 12, 2012 Published online: February 27, 2013

Dr. Catherine Rousset Centre for the Developing Brain, Perinatal Brain Injury Group Kings College London, The Rayne Institute, St Thomas’ Hospital London SE1 7EH (UK) E-Mail catherine.rousset @ kcl.ac.uk

© 2013 S. Karger AG, Basel0378–5866/13/0000–0000$38.00/0

www.karger.com/dne

Rousset/Kassem/Aubert/Planchenault/Gressens/Chalon/Belzung/Saliba

Dev NeurosciDOI: 10.1159/000346579

2

ing infection may become deleterious for the fetus. Exces-sive secretion of inflammatory molecules such as cytokines and chemokines is cytotoxic for the fetus and may induce fetal inflammatory response syndrome [6–9] which con-tributes to neonatal brain injury and later developmental disability [8] . Because of an improved risk of white matter lesions after maternal infection, preterm newborns may have a high risk of developing neuromotor dysfunctions [1, 10–12] . White matter lesions are responsible for the majority of motor disorders of prematurity. If cognitive sequelae are very common in preterm infants, diffuse or cystic brain lesions can also lead to major motor deficits like spastic diplegia, visuomotor apraxia and CP [2] .

Several animal models have been used to understand the mechanisms underlying perinatal brain injury and the assessment of behavioral consequences is a way to test the relevance of a model compared to the clinical pheno-type. In models of hypoxia-ischemia, unilateral ligature of the carotid artery in postnatal day (P) 7 rats followed by hypoxia induced delayed appearance and develop-ment of neurological reflexes, as well as motor and mem-ory disorders by up to 5 weeks [13, 14] . Maternal hypox-ia during pregnancy has been shown to delay the develop-ment of motor reflexes in newborn mice [15] . In rabbits, prenatal hypoxia induced hypertonia with motor disor-ders (e.g. involving spontaneous locomotion, motor re-flexes and coordination of sucking and swallowing) [16] .

In inflammatory animal models, postnatal administra-tion of lipopolysaccharide (LPS) to newborn rats induces memory impairments [17] , as well as, surprisingly, accel-erated neurodevelopment and hyperactivity [18] . In a schizophrenia model, maternal administration of LPS at 18 and 19 days of gestation increases motor disorders and reactivity induced by amphetamine injections in adult-hood [19] . Two studies from the same group demonstrate opposite results depending on whether they induced acute (showing no behavioral phenotype) [20] or chronic (leading to developmental sequelae) [21] inflammation in the dam. Finally, maternal intrauterine endotoxin ad-ministration leads to early motor deficits in the newborn rabbit, resulting in a phenotype that resembles those found in cystic periventricular leukomalacia (PVL) and CP [22, 23] .

Maternal inflammatory challenge brings quite differ-ent results depending on the model, animal strain, timing and dose of the infectious reagent, and to our knowledge, very few studies focused on very late infection in preg-nancy, which is of even more relevance for WMI consid-ering that oligodendrogenesis is initiated at that stage in rodents [24] . The behavioral assessment is very often

made during the early life of the animals; however, in clin-ic, PVL and CP diagnosis is frequently made during child-hood and not infancy. The purpose of this study was to determine, in our previously described maternal infec-tion model [25] , the motor consequences of maternal LPS administration on offspring from birth to adulthood. This model utilized a dose and a timing of administration of LPS leading to subtle and diffuse brain lesions of hypo-myelination which allowed us to follow the animal until late in life.

Materials and Methods

Animals and Drugs Experiments were carried out in compliance with appropriate

European Community Commission directive guidelines (86/609/EEC). Animals were housed in a temperature-controlled room (22 ± 1 ° C) under light/dark reverse 12-hour cycles (between 8: 00 p.m. and 8: 00 a.m.) and had access to food and water ad libitum. All animals (mothers and pups) were weighted at 9 a.m. every morn-ing as an indicator of well-being and normal growth. Timed-preg-nant Wistar rats used in this study were purchased from CERJ (Le Genest, France). Dams were intraperitoneally injected with saline (vehicle group) or 300 μg/kg LPS ( Escherichia coli , serotype 055:B5; Sigma; LPS group) at 11 a.m. on gestational days 19 and 20 [25] . The progeny was sex typed from P0, and all experiments have been conducted on 1 male and 1 female pup from each dam. A total of 11 litters was used, 6 for the vehicle and 5 for the LPS group. In a first group, 2 pups from each dam (1 male and 1 female) were used for the study of neurobehavioral development between P1 and P21. In a second group, 2 pups (1 male and 1 female) per dam were used for motor coordination tests between 2 and 6 weeks of age. The rest of animals were used at 4 months for the locomotion ac-tivity test and killed. Animals used for one test were not used for another test.

For the examination of neurobehavioral development and the motor coordination test, we used the Fox battery test [26] rear-ranged by Lubics et al. [13] .

Examination of Neurobehavioral Development Examinations were started at P1 and were carried out daily be-

tween 9 and 12 a.m. until P21. Inspections were made for matura-tion of physical characteristics such as eye opening, incisor erup-tion and ear unfolding [27, 28] . Weight was also recorded every day. Pups were tested for neurological signs and reflexes outlined in what follows [27–31] . (1) Sensory reflexes: the ear and the eyelid were gently touched with a cotton swab and the first day of the ear twitch reflex and the contraction of the eyelid were recorded. (2) Limb grasp: the fore- and hindlimbs were touched with a thin rod, and the first day of grasping onto the rod was recorded. (3) Limb placing: the back of the forepaw and the hindpaw was touched with the edge of the bench while the animal was suspended, and the first day of lifting and placing the paws on the table was noted. (4) Au-ditory startle: the first day of the startle response to a clapping sound was observed. (5) Negative geotaxis: animals were placed head down on an inclined board. The hindlimbs of the pups were

Maternal LPS Delays Rat Motor Development


3

placed in the middle of the board. The day they began to turn around and climb up the board with their forelimbs reaching the upper rim was observed; the time (in seconds) to reach the upper end of the board was recorded daily. The test was considered neg-ative if the animal did not succeed within 30 s. (6) Righting reflex: rats were placed in supine position and the time (in seconds) to turn over and come back to prone position and place all 4 paws in contact with the surface was recorded. (7) Gait: each animal was placed in the center of a white paper circle of 14 cm in diameter. The first day the pup began to walk off the circle with both fore-limbs was observed, and the time (in seconds) was recorded daily. If the animal did not leave the circle within 30 s, the test was con-sidered negative. For righting and gait, an animal is considered to be efficient for a task when it completes it in less than 2 s on at least 3 consecutive days.

Motor Coordination Tests Pups were tested for motor coordination twice a week between

2 and 6 weeks of age. (1) Rope suspension test: animals were sus-pended by both their forepaws on a 4-mm diameter iron rope, hung horizontally 60 cm over a foam pad. The time the animals stayed suspended was recorded (maximum 30 s until 3 weeks and 1 min afterwards). (2) Rotarod test: animals were tested on a tread-mill of 4 cm diameter, attached to a rotating motor. The test was performed at a speed of 10 rpm. Animals were placed on the rotat-ing drum and the time the animal could stay on the rotarod was measured (maximum 2 min). This test was only performed start-ing at 3 weeks of age, as younger pups cannot walk efficiently enough to do the test. (3) Grid walking and foot fault test: rats were placed on a stainless steel grid floor (20 × 40 cm with a mesh size of 4 cm 2 ) elevated 1 m above the floor. During the 2-min observa-tion period, the total number of steps was counted, as well as the number of foot faults (when the animal misplaced a limb outside the grid). (4) Inclined board test: animals were placed on a wooden board which was gradually elevated by 5°. The maximum angle at which the animals could maintain position on the inclined board for 5 s was recorded. (5) Walking initiation: the rats were placed on a horizontal surface in the center of two concentric circles (10 and 45 cm diameter); the time (in seconds) to walk off the circles with both forelimbs was noted. (6) Traversing a square bridge: rats were placed on a bridge (1 × 40 × 1 cm, elevated 40 cm above the floor), with two platforms on both ends. The duration the rat stayed on the bridge was measured to a maximum of 120 s. When a rat escaped to one of the platforms, the duration was noted and we considered the test positive.

Cyclotron Test Animals were tested for locomotion behavior at 4 months. The

apparatus consisted of a 60-cm diameter walking ring of 6 cm width with 50-cm walls around. The system recorded the number of times the animals walked past 1 of 4 electrical captors evenly placed on the ring. Each animal had a 15-min session in the appa-ratus every day for 5 days. The scores at each electrical captor were recorded at the end of each session and added to each other.

Maternal Care Litters were filmed at P3, 4, 7 and 8 for observation of maternal

care to the pups. Principal components of maternal behavior were observed, i.e. nursing and crouching, licking, and nesting [32] . Each behavioral item was counted according to a scanning method

applied on 20-min videos: every minute, films were scanned in a slow motion and occurrences of focal behavioral items were marked.

Myelin Basic Protein Immunohistochemistry After they had been used for functional tests, animals were

sacrificed. Brains were quickly dissected on ice, frozen in cooled isopentane and stored at –80 ° C. Coronal sections of 16-μm thick-ness were prepared at –20 ° C in a cryostat microtome (Jung CM3000, Leica, Wetzlar, Germany), mounted on slides and stored at –80 ° C until use. Sections were taken from brain regions corresponding to plates 12 and 30 of the atlas by Paxinos and Watson [33] . Sections were fixed for 3 min in methanol, for 4 min in acetone and dried for 20 min. After rinsing twice with PBS (w/o calcium and magnesium 1×, Gibco, gelatin 2% and Triton 0.25%), brain sections were incubated with a monoclonal mouse anti-myelin basic protein (MBP) to detect myelin (Chemicon) diluted at 1; 1,000 in PBS overnight. Sections were rinsed 4 times in PBS and incubated with biotinylated anti-mouse IgG diluted at 1; 400 in PBS for 90 min. After 3 washes, sections were incu-bated with streptavidin-biotin-peroxidase complex (Amersham Biosciences) diluted at 1; 400 in PBS for 90 min. Sections were rinsed once in PBS and once in Tris 50 m M . The reaction was visualized with 3,3 ′ -diaminobenzidine (Sigma) and mounted with counterstaining. MBP quantification was performed by a researcher unaware of the experimental groups. The optical den-sity of MBP-stained fibers was measured in the external capsule of coronal sections. At least 4 sections per brain of each animal were examined. Optical density was measured at ×200 magnifica-tion using a computerized image analysis system (Mercator; Ex-plora Nova, La Rochelle, France), which reads optical density as gray levels. Nonspecific background densities were measured at each brain level in a region devoid of MBP immunostaining and were subtracted [34] .

Statistics Statistical analyses were conducted with Prism 4.0 (GraphPad).

Data are presented as mean ± SEM. They were analyzed with a 2-way ANOVA test followed by the Bonferroni post hoc test or with a Mann-Whitney U test. The significance level was set at p < 0.05.

Results

Physiological Parameters Mothers were weighted at 9 a.m. every morning (injec-

tions at 11 a.m. on gestational days 19 and 20). We ob-served a significant decrease of dams’ weight after the first and the second LPS injections (comparing the weight of each animal 2 h before the injection vs. the following morning). The LPS mothers lost weight whereas vehicle animals gained weight (injection 1: LPS group, n = 7, –2.29 ± 7.09 g vs. vehicle group, n = 6, +13.67 ± 1.89 g, U = 7.5, p = 0.05; injection 2: LPS group, n = 7, –2.57 ± 4.27 g vs. vehicle group, n = 6, +7.17 ± 0.98 g, U = 3.5, p < 0.01).



4

We also noticed a significant hypothermia in LPS females after the first injection compared to the vehicle group (LPS group, n = 8, –0.8 ± 0.3 ° C vs. vehicle group, n = 6, +0.2 ± 0.1 ° C, U = 6.0, p < 0.05), when we compared the temperature of each animal 1 h after the LPS challenge with the temperature taken the same morning after the weighting. This effect was transient and confirms previ-ously reported data [35] . We did not observe any differ-ence in the delivery day (LPS group, n = 5, 21.30 ± 0.20 days vs. vehicle group, n = 6, 21.50 ± 0 days, U = 12, p = 0.66), nor in the number of living newborns per litter (LPS group, n = 5, 10 ± 1 pups vs. vehicle group, n = 6, 12 ± 1 pups, U = 9, p = 0.31).

Although there was a significant difference in weight between males and females, we did not observe an LPS effect on offspring’s weight at any stage (data not shown). The general state of the animals was good; the behavioral observations made are compatible with animal survey. Overall, we did not observe any differences between males and females except for the weight; consequently, we pooled the results of both sexes together.

Neurological Reflexes Among the physical characteristics studied, none of

them was delayed in LPS animals compared to vehicle animals ( table 1 ). However, several reflexes such as effi-

ciency in righting to prone position and efficiency in gait appeared significantly later (U = 19.5 and U = 19, respec-tively, p < 0.01) compared to vehicle pups ( table 1 ).

LPS pups performed worse in certain tasks compared to the vehicle group. If the animals are significantly im-proving over time, righting to prone position took longer for the LPS pups (2-way ANOVA, time p < 0.0001, treat-ment p < 0.001, interaction n.s.) until they reached the same time as the vehicle group at around P10 ( fig. 1 a). The time taken to perform the gait test was significantly longer in the LPS group (2-way ANOVA, time p < 0.0001, treatment p < 0.01, interaction n.s.) until P19 after which they reached the vehicle level ( fig. 1 b). The results of the negative geotaxis tests were less straightforward; never-theless, LPS animals took longer to complete the task for both the 20° (although not significant, there is a clear trend: 2-way ANOVA, time p < 0.0001, treatment p = 0.087, interaction n.s.) and 40° angled boards (2-way ANOVA, time p < 0.0001, treatment p = 0.001, interac-tion n.s.) ( fig. 1 c, d).

Motor Coordination and Strength In the grid walking and foot fault test, there are no

differences between the 2 groups for the total number of steps throughout the observation period ( fig. 2 a). How-ever, the LPS animals made significantly more mistakes than the vehicle animals (2-way ANOVA, time p < 0.0001, treatment p < 0.001, interaction p < 0.01; fig. 2 b). Interestingly, no significant differences have been ob-served between the 2 groups in the other tests performed ( fig. 3 ). For the bridge test, the inclined board test and the rope suspension test, the results between the 2 groups were very similar ( fig. 3 a–c). Even though we ob-served more differences, the rotarod and the gait tests did not reveal significant differences between LPS and vehicle animals throughout the experimental period ( fig. 3 d–f).

Locomotion In the cyclotron test, when measuring the number of

times each animal walks past one of the captor, no sig-nificant difference in locomotion was observed between the 2 groups over the 5 days even though we observed that the LPS animals (n = 16) constantly made slightly lower scores than the vehicle group (n = 24) (day 1: LPS 204.3 ± 6.4 vs. vehicle 212.2 ± 6.3; day 2: LPS 195.4 ± 8.4 vs. ve-hicle 199.4 ± 6.1; day 3: LPS 197.4 ± 8.3 vs. vehicle 204.6 ± 7.3; day 4: LPS 189.6 ± 8 vs. vehicle 202.6 ± 7.6; day 5: LPS 190.7 ± 6.5 vs. vehicle 202.5 ± 7.9).

Table 1. Average days ± SEM of appearance of physical and neuro logical signs in control and LPS animals

Sign Day of appearance Mann-WhitneyU test

controls LPS group

Eye opening 15.2±0.2 15.3±0.2 0.76Incisor eruption 7.6±0.2 8.0±0.3 0.26Ear unfolding 15.0±0.0 15.3±0.2 0.34Ear twitch reflex 17.3±0.2 17.5±0.3 0.71Eyelid reflex 4.5±0.4 5.6±0.4 0.09Forelimb placing 15.3±0.6 14.8±0.4 0.51Hindlinb placing 20.3±0.3 20.9±0.4 0.32Forelimb grasp 12.7±0.3 12.6±0.4 0.89Hindlimb grasp 16.5±0.2 16.9±0.2 0.22Auditory startle 12.1±0.1 12.2±0.1 0.65Righting prone position 1.3±0.1 1.7±0.2 0.14Righting prone position

efficiencya 5.6±0.4 8.2±0.7 0.008*Gait 11.7±0.6 11.4±0.7 0.69Gait efficiencya 17.3±0.6 19.8±0.6 0.007*

* p < 0.01, Mann-Whitney U test.a An animal is considered to be efficient for a task when it

completes it in less than 2 s on at least 3 consecutive days.



5

**

*

**

Righ

ting

(s)

Postnatal day

0P1 P2 P3 P4 P5 P6 P7 P8 P9 P1

0P1

1P1

2P1

3P1

4

5

10

15

20

25

Gait

test

(s)

Postnatal day

0

P13

P14

P15

P16

P17

P18

P19

P20

P21

10

20

30LPSVehicle

Neg

ativ

e ge

otax

is 40

° (s)

Postnatal day

0

P11

P12

P13

P14

P15

P16

P17

P18

P19

P20

P21

10

20

30

Neg

ativ

e ge

otax

is 20

° (s)

Postnatal day

0

P11

P10P9 P1

2P1

3P1

4P1

5P1

6P1

7P1

8P1

9P2

0P2

1

10

15

5

20

25

30

35

a b

dc

Fig. 1. Motor impairment in pups after maternal LPS challenge during pregnancy. Daily performance of motor tests and reflexes at the neonatal stage until 3 weeks. Righting ( a ), gait ( b ), negative geotaxis 20° ( c ) and negative geotaxis 40° ( d ) tests. Vehicle group,

n = 12; LPS group, n = 10. Data are presented as means ± SEM and analyzed with 2-way ANOVA followed by the Bonferroni post hoc test. * p < 0.05; * * p < 0.01.

0P20 P23 P27 P30

20

40

60

80

100

Num

ber o

f ste

ps

LPSVehicle

Postnatal daya

***

0P20 P23 P27 P30

1

2

3

4

5

6

7

8

Num

ber o

f foo

t fau

lts

LPSVehicle

Postnatal dayb

Fig. 2. Motor coordination deficit in prenatally challenged juvenile rats, performances in the grid walking test. Vehicle group, n = 12; LPS group, n = 10. Number of total steps ( a ), and number of foot

faults ( b ). Data are presented as means ± SEM and analyzed with 2-way ANOVA followed by the Bonferroni post hoc test. * * * p < 0.001.



6

Maternal Behavior We found that LPS mothers (n = 5) overall expressed

more maternal behaviors (i.e. the sum of the occurrences of every maternal item measured) than vehicle mothers (n = 6) even though it is not statistically significant (P3: LPS 11.2 ± 1.5 vs. vehicle 6.5 ± 0.9; P4: LPS 13.8 ± 1.6 vs. vehicle 8.7 ± 1.1; P7: LPS 13.6 ± 1.6 vs. vehicle 11.2 ± 3;

P8: LPS 9.8 ± 2.1 vs. vehicle 8.1 ± 2.4). When we compared each maternal behavior assessed independently (i.e. lick-ing, nursing and crouching, nesting), we did not find sig-nificant differences between vehicle and experimental dams (data not shown).

Brid

ge te

st (s

)

Postnatal day

0P16 P20 P23 P27 P30 P34 P37 P41

50

100

150

200LPSVehicle

a

Max

imum

ang

le (s

)

Postnatal day

0P16 P20 P23 P27 P30 P34 P37 P41

10

20

30

40

50

b

Rota

rod

test

(s)

Postnatal day

0P20 P23 P27 P30 P34 P37 P41

20

40

60

100

80

120

dGa

it bi

g ci

rcle

(s)

Postnatal day

0P20 P23 P27 P30 P34 P37 P41

2

4

8

6

10

f

Gait

smal

l circ

le (s

)

Postnatal day

0P20 P23 P27 P30 P34 P37 P41

2

1

3

5

4

6

e

Susp

ensio

n (s

)

Postnatal day

0P16 P20 P23 P27 P30 P34 P37 P41

10

20

30

50

40

60

c

Fig. 3. No motor deficits or very subtle deficits in juvenile rats from LPS-treated mothers. Performances in tests for motor coordination and strength. Vehicle group, n = 12; LPS group, n = 10. Time on the bridge ( a ), maximum angle ( b ) time suspended on the rope ( c ), time on the rotarod ( d ) and time to walk out the small ( e ) and big ( f ) circles. Data are presented as means ± SEM and analyzed with 2-way ANOVA.



7

Histologic Examination and Myelination At 4 months, no histologic lesions were observed in the

LPS group (data not shown). Moreover, myelination was similar in both groups in the external capsule ( fig. 4 ).

Discussion

Our data show that maternal administration of E. coli LPS at 19 and 20 days of gestation induces motor disor-ders in offspring such as delayed control of different re-flexes and tasks (righting, gait, negative geotaxis) and de-creased performances compared with vehicle animals

from birth until 3 weeks. Based on previous reports, those deficits are most likely of cerebral origins. A common fea-ture of all the different prenatal inflammatory models published, regardless of the lesion size, is that the cerebral damage (astrogliosis, hypomyelination, cell death) is pre-dominantly localized in the white matter, i.e. internal and external capsules [36] . Our results (impairment of motor reflexes, lack of coordination and decrease of reaction) may be explained in part by cerebral lesion in the white matter. This structure governs motricity, and lesions or disorganization in the white matter may promote the oc-currence of motor sequelae. The white matter is com-posed of nerve tracts, myelinated axons, which make a

0

2

4

6

8

Densitometry

Left Right

VehicleLPS

a b

c d

Fig. 4. Normal myelination in adult brain after prenatal LPS challenge. MBP-labelled sections after immunohistochemistry on 4-month brains and densitometric analysis of MBP immunostaining in the external capsule. Vehicle group, n = 8; LPS group, n = 8. Normal myelination was observed in LPS ( b, d ) compared to vehicle brains ( a, c ) in the external capsule. Bar = 500 μm ( a , b ) and 100 μm ( c , d ).



8

link between the different parts of the brain and transmit nervous influx between neurons. White matter lesions in-duce disorders in nervous transmission paths between ce-rebral structures. The internal capsule is a link between the cortex and other cerebral structures like the thalamus and basal ganglia in both rodents and humans. In the lat-ter, these regions have been significantly correlated with general movement disorders [2, 37] in children who suf-fered perinatal brain injury. MRI studies showed that ab-normal intensity in the internal capsule may be predictive of neurodevelopmental disabilities. That kind of lesion, already observed in CP [38] , may provoke alterations of the control of the voluntary movements.

Another aspect to consider is axonopathy, provoked by an impaired axonal growth or a myelination defect in the sheath wrapping the axons. The nude axons will not be able to transmit the signal properly and will start to degen-erate. Even if our model was mainly designed for WMI, it is now recognized that white matter lesions, even if pre-dominant, are often associated with neuronal and axonal injury [1] . A recent paper of systemic inflammation in newborn mice showed that the myelin defect observed was associated with axonopathy (reduction of axon diameters and thickness of myelin sheath [39] ). Axonopathy can therefore not be excluded as a contributing factor in the behavioral defects we observe in this work, and investigat-ing axon thickness and myelin sheath by electron micros-copy in the motor cortex would be informative.

In our study, functional impairments disappear rapidly except for the motor coordination dysfunction which con-tinues until 5 weeks of age. At 4 months, there is no more locomotor dysfunction observable. These results suggest that motor disorders observed in the first weeks improve and confirm the results of two previous studies showing that initially observed motor disorders disappear with time [13, 21] . The functional deficiency observed at differ-ent tasks is not permanent. One hypothesis to explain the motor recovery might be a higher degree of brain plastic-ity. It has been reported that after a hemispherectomy in P14 rats, an aberrant ipsilateral corticospinal tract appears and grows to reach the spinal cord [40] . This new tract is not collateral to the normal crossed corticospinal tract [41, 42] . It seems that the optimal functional plasticity period corresponds to the stage of the development of corticospi-nal axons, i.e. the first 2 weeks of life. Another hypothesis could be that maternal behavior may play an important role in the improvement of performances by a stronger stimulation. Beyond changes in physiological parameters per se, LPS could modify maternal care to newborns. The maternal tendency to give more care to the LPS progeny

in our study may be caused either by a greater demand for care by the pups of LPS-treated mothers, although it has recently been shown that prenatally LPS-exposed pups had a significant decrease in the number and duration of ultrasonic vocalization (i.e. potent maternal behavior in-ducers) [43] , or the mother recovering from the LPS chal-lenge is more stressed and increases her care to the pups. Regardless, more maternal care and increased stimulation of the pups could explain the improvement in the tests later on. These two hypotheses of brain plasticity and ma-ternal care are probably intertwined as the animal envi-ronment may be a factor influencing brain plasticity.

The results of our previous studies [25] showing strong hypomyelination in LPS brains at P7 using the same mod-el, and the complementary results obtained here showing normal myelination at adulthood in the LPS animals seem to correlate very well with the functional deficit and recovery that we are observing and could explain at the cellular level the behavioral recovery. Once again, two hy-potheses can be evoked. We previously reported an in-crease in cell death in the subventricular white matter suggestive of oligodendrocyte death. Remyelination would then be explained by an increase in oligodendro-cyte production after the insult to renew the pool of cells. However, a new interesting recently exposed hypothesis suggests that the hypomyelination observed is not caused by death of the preoligodendrocytes but a delay in their maturation which stops after the insult and restarts later on. This has been shown in an organotypic culture mod-el of chronic WMI [44] and in humans [45, 46] . This lat-ter hypothesis could be of particular relevance to our model since the Wistar rat has the highest percentage and density of preoligodendrocytes relative to other rat strains [47] . To decipher what is happening to the preoligoden-drocytes (death and renewal or delayed maturation), fur-ther investigations such as BrdU staining are needed.

A critical issue in animal models of perinatal brain in-jury is to adapt the pertinent pathophysiology scenarios to their corresponding developmental window in order to induce neuropathological and behavioral characteristics reminiscent of the clinical phenotype. Using antenatal in-flammatory models, we and others were able to demon-strate in the treated pups the histopathologic changes as-sociated with diffuse white matter disease observed in the preterm infant [23, 25, 48] . These models are clinically rel-evant and demonstrate a standard histopathologic out-come. However, few studies were able to evaluate the long-term motor or cognitive outcome. Despite motor impair-ment and a slightly delayed neurological development during the first weeks, we did not observe discernible dif-



9

ferences in the motor outcome of adult animals. This find-ing is relevant as in clinic, the diagnosis of PVL and CP is typically made in childhood, and the outcome is not re-versible. This shows a major limitation of our model as the long-term phenotype does not mimic the clinical observa-tions. Nevertheless interesting and relevant to a certain extent, our model leads to subtle brain injury with tran-sient motor disorders. The early injury and behavioral dis-order we observed is compensated until reaching vehicle level in adults. It has been shown that adding to a unique rat paradigm of prenatal inflammation, a sensitizing event pertinent to the context of very early premature human newborns, such as hypoxia-ischemia, can induce more se-vere cerebral lesions associated with long-lasting motor deficits remindful of the human context [49] . More gener-ally, our results and those of others demonstrate one of the limitations of the use of animal models to mimic disease. The brains of rodents are more primitive, less complex than human brains and probably more plastic, which could explain the recovery. White matter lesions tend to disappear in rodents after an LPS challenge, when unfor-tunately white matter damage is persistent and permanent in preterm babies. Large animal models (piglet or sheep) provide better models experimentally and clinically due to the similarity and complexity of their brain compared to humans. However, there is currently no behavioral assess-ment available yet for those species, not even considering the logistic and cost of such experiments.

Finally, in the present study, our purpose was to inves-tigate motor impairment and we mainly observed coor-dination disorders. We were not able to study potential cognitive disorders in the offspring. However, motor dis-abilities observed in children with WMI such as PVL are often associated with important behavioral, attention and

cognitive deficiencies [2] . Recent papers have focused on the cognitive impairments related to LPS challenge. Short systemic maternal inflammation has been shown to cause long-lasting consequences on the adult mouse stress and social behavior [50, 51] . The hippocampus is a cerebral structure involved in cognition and memory. Lesions in this area may provoke cognitive impairment in children with WMI. In our model, hippocampal astrogliosis after maternal LPS administration [25] was observed; altered synaptic transmission in the hippocampus of juvenile off-spring rats [52] has also been reported. More in-depth investigation of cognitive behavior impairment in our model is needed.

In conclusion, our results show that maternal admin-istration of LPS has important but transient consequenc-es on motor behavior in the progeny, especially concern-ing coordination and gait. These alterations, despite a slight coordination deficit, differ from that observed in children with CP as, after the initial motor impairment, animals recover. We also observed that the hypomyelin-ation previously described at the neonatal stage, probably explaining the functional deficit we report here, is no lon-ger observable at the adult stage. This late normal my-elination may explain the functional recovery of the ani-mals and could suggest either a potential remyelination of the brain or a delayed myelination in LPS pups, that remains to be investigated.

Acknowledgements

We would like to thank Sylvie Bodard for excellent technical assistance and Philippe Bertrand for statistical advice. This work was supported by INSERM and Region Centre, France.

References

1 Khwaja O, Volpe JJ: Pathogenesis of cerebral white matter injury of prematurity. Arch Dis Child Fetal Neonatal Ed 2008; 93:F153–F161.

2 Volpe JJ: Brain injury in premature infants: a complex amalgam of destructive and devel-opmental disturbances. Lancet Neurol 2009; 8: 110–124.

3 Wu YW, Colford JM Jr: Chorioamnionitis as a risk factor for cerebral palsy: a meta-analy-sis. JAMA 2000; 284: 1417–1424.

4 Saliba E, Marret S: Cerebral white matter damage in the preterm infant: pathophysiol-ogy and risk factors. Semin Neonatol 2001; 6: 121–133.

5 Back SA, Rivkees SA: Emerging concepts in periventricular white matter injury. Semin Perinatol 2004; 28: 405–414.

6 Bell MJ, Hallenbeck JM, Gallo V: Determin-ing the fetal inflammatory response in an ex-perimental model of intrauterine inflamma-tion in rats. Pediatr Res 2004; 56: 541–546.

7 Dammann O, Leviton A: Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res 1997; 42: 1–8.

8 Dammann O, Leviton A: Role of the fetus in perinatal infection and neonatal brain dam-age. Curr Opin Pediatr 2000; 12: 99–104.

9 Plaisant F, Dommergues MA, Debillon T, Gressens P: Pathophysiology of periventricular leukomalacia. Arch Pediatr 2005; 12: 734–736.

10 Marret S, Marpeau L: Grand prematurity, risk of neuropsychic handicaps and neuroprotec-tion. J Gynécol Obstét Biol Reprod 2000; 29: 373–384.

11 Nelson KB: The epidemiology of cerebral pal-sy in term infants. Ment Retard Dev Disabil Res Rev 2002; 8: 146–150.

12 Nelson KB, Willoughby RE: Infection, in-flammation and the risk of cerebral palsy. Curr Opin Neurol 2000; 13: 133–139.

13 Lubics A, Reglodi D, Tamas A, Kiss P, Szalai M, Szalontay L, Lengvari I: Neurological re-flexes and early motor behavior in rats sub-jected to neonatal hypoxic-ischemic injury. Behav Brain Res 2005; 157: 157–165.



10

14 Wang LS, Zhou J, Shao XM, Tang XC: Huper-zine A attenuates cognitive deficits and brain injury in neonatal rats after hypoxia-isch-emia. Brain Res 2002; 949: 162–170.

15 Golan H, Kashtutsky I, Hallak M, Sorokin Y, Huleihel M: Maternal hypoxia during preg-nancy delays the development of motor re-flexes in newborn mice. Dev Neurosci 2004; 26: 24–29.

16 Derrick M, Luo NL, Bregman JC, Jilling T, Ji X, Fisher K, Gladson CL, Beardsley DJ, Mur-doch G, Back SA, Tan S: Preterm fetal hypox-ia-ischemia causes hypertonia and motor def-icits in the neonatal rabbit: a model for human cerebral palsy? J Neurosci 2004; 24: 24–34.

17 Bilbo SD, Levkoff LH, Mahoney JH, Watkins LR, Rudy JW, Maier SF: Neonatal infection induces memory impairments following an immune challenge in adulthood. Behav Neu-rosci 2005; 119: 293–301.

18 Roberson R, Woodard JE, Toso L, Abebe D, Poggi SH, Spong CY: Postnatal inflammatory rat model for cerebral palsy: too different from humans. Am J Obstet Gynecol 2006; 195: 1038–1044.

19 Fortier ME, Joober R, Luheshi GN, Boksa P: Maternal exposure to bacterial endotoxin during pregnancy enhances amphetamine-induced locomotion and startle responses in adult rat offspring. J Psychiatr Res 2004; 38: 335–345.

20 Poggi SH, Park J, Toso L, Abebe D, Roberson R, Woodard JE, Spong CY: No phenotype as-sociated with established lipopolysaccharide model for cerebral palsy. Am J Obstet Gyne-col 2005; 192: 727–733.

21 Toso L, Poggi S, Park J, Einat H, Roberson R, Dunlap V, Woodard J, Abebe D, Spong CY: Inflammatory-mediated model of cerebral palsy with developmental sequelae. Am J Ob-stet Gynecol 2005; 193: 933–941.

22 Kannan S, Saadani-Makki F, Balakrishnan B, Chakraborty P, Janisse J, Lu X, Muzik O, Romero R, Chugani DC: Magnitude of [(11)C]PK11195 binding is related to severity of motor deficits in a rabbit model of cerebral palsy induced by intrauterine endotoxin ex-posure. Dev Neurosci 2011; 33: 231–240.

23 Saadani-Makki F, Kannan S, Lu X, Janisse J, Dawe E, Edwin S, Romero R, Chugani D: In-trauterine administration of endotoxin leads to motor deficits in a rabbit model: a link be-tween prenatal infection and cerebral palsy. Am J Obstet Gynecol 2008; 199: 651, e651–e657.

24 Liu Y, Wu Y, Lee JC, Xue H, Pevny LH, Kapri-elian Z, Rao MS: Oligodendrocyte and astro-cyte development in rodents: an in situ and immunohistological analysis during embry-onic development. Glia 2002; 40: 25–43.

25 Rousset CI, Chalon S, Cantagrel S, Bodard S, Andres C, Gressens P, Saliba E: Maternal ex-posure to LPS induces hypomyelination in the internal capsule and programmed cell death in the deep gray matter in newborn rats. Pe-diatr Res 2006; 59: 428–433.

26 Fox WM: Reflex-ontogeny and behavioural development of the mouse. Anim Behav 1965; 13: 234–241.

27 Smart JL, Dobbing J: Vulnerability of devel-oping brain. 6. Relative effects of foetal and early postnatal undernutrition on reflex on-togeny and development of behaviour in the rat. Brain Res 1971; 33: 303–314.

28 Smart JL, Dobbing J: Vulnerability of devel-oping brain. 2. Effects of early nutritional de-privation on reflex ontogeny and develop-ment of behaviour in the rat. Brain Res 1971; 28: 85–95.

29 Altman J, Sudarshan K: Postnatal develop-ment of locomotion in the laboratory rat. Anim Behav 1975; 23: 896–920.

30 Dam K, Seidler FJ, Slotkin TA: Chlorpyrifos exposure during a critical neonatal period elic-its gender-selective deficits in the development of coordination skills and locomotor activity. Brain Res Dev Brain Res 2000; 121: 179–187.

31 Hill JM, Gozes I, Hill JL, Fridkin M, Brenneman DE: Vasoactive intestinal peptide antagonist retards the development of neonatal behaviors in the rat. Peptides 1991; 12: 187–192.

32 Rosenblatt JS: Prepartum and postpartum regulation of maternal behaviour in the rat. Ciba Found Symp 1975; 33: 17–37.

33 Paxinos G, Watson C: The Rat Brain in Ste-reotaxis Coordinates. New York, Academic Press, 1986.

34 Rousset CI, Kassem J, Olivier P, Chalon S, Gressens P, Saliba E: Antenatal bacterial en-dotoxin sensitizes the immature rat brain to postnatal excitotoxic injury. J Neuropathol Exp Neurol 2008; 67: 994–1000.

35 Bull DF, Brown R, King MG, Husband AJ: Modulation of body temperature through taste aversion conditioning. Physiol Behav 1991; 49: 1229–1233.

36 Wang X, Rousset CI, Hagberg H, Mallard C: Lipopolysaccharide-induced inflammation and perinatal brain injury. Semin Fetal Neo-natal Med 2006; 11: 343–353.

37 Ferrari F, Todeschini A, Guidotti I, Martinez-Biarge M, Roversi MF, Berardi A, Ranzi A, Cowan FM, Rutherford MA: General move-ments in full-term infants with perinatal as-phyxia are related to basal ganglia and tha-lamic lesions. J Pediatr 2011; 158: 904–911.

38 Stewart RM: Observations on the pathology of cerebral diplegia. [Abridged.]: President’s address. Proc R Soc Med 1942; 36: 25–32.

39 Favrais G, van de Looij Y, Fleiss B, Ramanant-soa N, Bonnin P, Stoltenburg-Didinger G, Lacaud A, Saliba E, Dammann O, Gallego J, Sizonenko S, Hagberg H, Lelievre V, Gressens P: Systemic inflammation disrupts the devel-opmental program of white matter. Ann Neu-rol 2011; 70: 550–565.

40 Hicks SP, D’Amato CJ: Motor-sensory cor-tex-corticospinal system and developing lo-comotion and placing in rats. Am J Anat 1975; 143: 1–42.

41 Huttenlocher PR, Bonnier C: Effects of changes in the periphery on development of the corticospinal motor system in the rat. Brain Res Dev Brain Res 1991; 60: 253–260.

42 Huttenlocher PR, Raichelson RM: Effects of neonatal hemispherectomy on location and number of corticospinal neurons in the rat. Brain Res Dev Brain Res 1989; 47: 59–69.

43 Baharnoori M, Bhardwaj SK, Srivastava LK: Neonatal behavioral changes in rats with ges-tational exposure to lipopolysaccharide: a prenatal infection model for developmental neuropsychiatric disorders. Schizophr Bull 2012; 38: 444–456.

44 Dean JM, Riddle A, Maire J, Hansen KD, Preston M, Barnes AP, Sherman LS, Back SA: An organotypic slice culture model of chron-ic white matter injury with maturation arrest of oligodendrocyte progenitors. Mol Neuro-degener 2011; 6: 46.

45 Billiards SS, Haynes RL, Folkerth RD, Boren-stein NS, Trachtenberg FL, Rowitch DH, Li-gon KL, Volpe JJ, Kinney HC: Myelin abnor-malities without oligodendrocyte loss in peri-ventricular leukomalacia. Brain Pathol 2008; 18: 153–163.

46 Buser JR, Maire J, Riddle A, Gong X, Nguyen T, Nelson K, Luo NL, Ren J, Struve J, Sherman LS, Miller SP, Chau V, Hendson G, Ballabh P, Grafe MR, Back SA: Arrested preoligoden-drocyte maturation contributes to myelina-tion failure in premature infants. Ann Neurol 2012; 71: 93–109.

47 Dean JM, Moravec MD, Grafe M, Abend N, Ren J, Gong X, Volpe JJ, Jensen FE, Hohimer AR, Back SA: Strain-specific differences in perinatal rodent oligodendrocyte lineage pro-gression and its correlation with human. Dev Neurosci 2011; 33: 251–260.

48 Mallard C, Welin AK, Peebles D, Hagberg H, Kjellmer I: White matter injury following sys-temic endotoxemia or asphyxia in the fetal sheep. Neurochem Res 2003; 28: 215–223.

49 Girard S, Kadhim H, Beaudet N, Sarret P, Se-bire G: Developmental motor deficits induced by combined fetal exposure to lipopolysac-charide and early neonatal hypoxia/ischemia: a novel animal model for cerebral palsy in very premature infants. Neuroscience 2009; 158: 673–682.

50 Golan HM, Lev V, Hallak M, Sorokin Y, Hu-leihel M: Specific neurodevelopmental dam-age in mice offspring following maternal in-flammation during pregnancy. Neurophar-macology 2005; 48: 903–917.

51 Hava G, Vered L, Yael M, Mordechai H, Ma-houd H: Alterations in behavior in adult off-spring mice following maternal inflammation during pregnancy. Dev Psychobiol 2006; 48: 162–168.

52 Lowe GC, Luheshi GN, Williams S: Maternal infection and fever during late gestation are associated with altered synaptic transmission in the hippocampus of juvenile offspring rats. Am J Physiol Regul Integr Comp Physiol 2008; 295:R1563–R1571.

maternal exposure to lipopolysaccharide leads to transient motor dysfunction in neonatal rats

Documents