Alterations in Behavior in AdultOffspring Mice FollowingMaternal InflammationDuring Pregnancy

Golan Hava

Lev Vered

Mazar YaelDepartment of Developmental Molecular

Genetics and ZlotowskiCenter for Neuroscience

Ben-Gurion University of the NegevBeer-Sheva, Israel

E-mail: havag@bgumail.bgu.ac.il

Hallak MordechaiDepartment of Obstetrics and Gynecology

Faculty of Health Sciences, Ben-GurionUniversity of the Negev, Beer-Sheva, Israel

Huleihel MahoudDepartment of Microbiology and

Immunology and Cancer Research CenterBen-Gurion University of the Negev

Beer-Sheva, Israel

ABSTRACT: Maternal intrauterine inflammation during pregnancy poses a majorthreat of neurodevelopmental brain damage in offspring and may cause poorcognitive and perceptual outcomes. In mice, we have previously shown thatmaternal inflammation induced by lipopolysaccharide (LPS) at gestation day 17thincreased the levels of the pro-inflammatory cytokine IL-6 in the fetal brain. In thisstudy, we used the same system and examined the effect of short, systemic maternalinflammation on anxiety and social behavior of the offspring. Adult offspring fromthe maternal inflammation group showed increased anxiety, as indicated by theelevated plus maze. Social interaction among offspring from the test groups wasexamined when two unfamiliar mice from different litters were introduced into anew home-cage. Offspring from the maternal inflammation group showed reducedactivity, indicating increased fear. In addition, offspring from the maternalinflammation group were less aggressive towards their cagemates and they spenta significantly longer time trimming the whiskers of their cagemates during the first30 min of their interaction, compared to offspring from the control group. Our datasuggest that short systemic maternal inflammation have long-lasting consequenceson the adult mouse stress and social behavior. � 2006 Wiley Periodicals, Inc.Dev Psychobiol 48: 162–168, 2006.

Keywords: cytokines; neurogenesis; social interactions; anxiety; prenatal stress;mouse

INTRODUCTION

Intrauterine inflammation during pregnancy increases the

risk of preterm delivery, neurodevelopmental brain

damage, neurological disorders, and mental retardation

in the offspring (Dammann, Kuban, & Leviton, 2002;

Saliba & Henrot, 2001). Intrauterine infection may affect

the immature brain by the induction of proinflammatory

cytokines.

The proinflammatory cytokines, tumor necrosis factor-

alpha (TNFa), interleukin-1 (IL-1) and IL-6 were

associated with intrauterine infection, preterm delivery,

neonatal infections, and neonatal brain damage. In the

brain, these proteins are expressed in both glial cells and

neurons. In addition to their function in the immune

response, these cytokines modulate neuron development

and function. TNFa is involved in the regulation of neurite

growth (Golan, Levav, Mendelsohn, & Huleihel, 2004;

Neumann et al., 2002), affects neuronal survival (Barker,

Middleton, Davey, & Davies, 2001; Yang, Lindholm,

Konishi, Li, & Shen, 2001), and regulates AMPA receptor

expression (Beattie et al., 2002). IL-1b and TNFaare involved in the regulation of synaptic plasticity

(Butler, O’Connor, & Moynagh, 2004; Cunningham,

Murray, O’Neill, Lynch, & O’Connor, 1996; Schneider

et al., 1998; Tancredi et al., 1992). In the animal model,

administration of lipopolysaccharide (LPS) to pregnant

rats increased the expression of TNFa and IL-1b mRNA

in a fetal brain in a dose-dependent manner. In addition,

Received 20 May 2005; Accepted 26 August 2005Correspondence to: Golan HavaPublished online in Wiley InterScience

(www.interscience.wiley.com). DOI 10.1002/dev.20116

� 2006 Wiley Periodicals, Inc.

decreased myelin basic protein expression and enhanced

expression of the astrocytes marker; glial fibrillary acidic

protein, was observed a week after application; (Cai,

Pan, Pang, Evans, & Rhodes, 2000). LPS administration

to pregnant rats upregulated the mRNA expression of the

stress-related peptide, corticotrophin-releasing hormone

(CRH) in the fetal brain (Gayle et al., 2004), suggesting

the possibility of inducing a fetal stress response.

Activation of the stress response during pregnancy was

shown to have long-term consequences on the response of

adult offspring to stressful situations, as demonstrated in

rodents, primates, and humans (Breivik et al., 2002;

Clarke & Schneider, 1993; Kofman, 2002; Saliba &

Henrot, 2001).

In our previous study, we demonstrated that a single

intraperitoneal (i.p) administration of LPS (0.12 mg/g) to

pregnant mice on gestation day 17 (E17), produced

significantly increased IL-6 levels in the maternal spleen

and the fetal brain, (3 hr after the injection), compared to

the controls treated with saline (Golan, Lev, Hallak,

Sorokin, & Huleihel, 2005). Here, we examine the long-

term effects of this treatment on stress-related behavior in

the adult offspring.

MATERIALS AND METHODS

Study Design

Jackson Black C-57 mice were used. Pregnant mice were

treated on gestation day 17 (17 days after the day of

mating). Mice were randomly assigned to one of two

groups: (1) saline injections (i.p)—control group (n¼ 12),

(2) E. Coli LPS injection (.12 mg/g mouse/100 mL, i.p)—

test group (n¼ 12). This LPS dose was shown to induce

maternal inflammation without inducing preterm deliv-

ery. Measurements of IL-6 in maternal spleen indicate

maternal inflammatory response, which last up to 6 hr

after the LPS injection (data not shown).

The mouse colony was kept on a 12:12 hr light/dark

schedule; food and water were provided ad libitum. At the

age of 3 weeks, offspring were separated from their

mothers and housed in isolation, in separate home cages:

20 cm wide, 35 cm long, 15 cm high, until the age of

8 months, when they were given behavioral tests. Two

offspring from each mother, male and female, were

included in the elevated plus maze experiment and a

single male offspring from each litter was included in the

social-interaction study. All the procedures were per-

formed according to the ‘‘Principles of laboratory care’’

(NIH publication no. 86–23, revised 1985) as well as the

guidelines from the Israeli Council on Animal Care and

were approved by the Ben-Gurion University of the Negev

Animal Care and Use Committee.

Behavior Examination

After a week of daily handling (5 min a day) by the

researcher blind to experimental groups, the following

behavioral examinations were performed:

Elevated Plus Maze. A plus maze with 40 cm long, 10 cm

width, and 15 cm high arms was elevated 50 cm above

the floor and used to measure the time (in seconds) spent

in the open versus the closed arms and the number of

crossing between the closed and open arms during 5 min

of examination (Lister, 1987).

Social Interaction. Social interactions between unfa-

miliar offspring from the same treatment group were

examined in four pairs (each pair in a separate cage). The

mice were housed in separate cages after weaning and

until they were observed. Three episodes of 30 min each

were recorded and analyzed at three time-periods

(relative to the time when the mice were introduced)

0, 24 hr, and 7 days, all during the dark cycle.

The observations began the moment that two offspring

were introduced simultaneously into a new cage (day 1,

time 0–5). Mice behavior was videotaped for later

analysis. Each episode was analyzed for various home

cage behaviors (Lijam et al., 1997). The duration (in

seconds) of playing, resting huddled, whisker trimming,

fighting, and the number of times that a mouse sniffed the

anogenitals of its cagemate, or attacked its cagemate

from the rear, or dug tunnels. In addition, we also

measured a general parameter: general activity—the

number of times mice crossed a line along the midline of

the cage, this measure is affected by both: locomotion

and the interactions between the cagemates. In four cases

a mouse was severely wounded by its cagemate between

the second and third observation periods. In these cases,

the adversarial mice were separated and were not studied

during the last time period (the 7th day following their

first interaction).

Statistical Analysis

We used SPSS software for our statistical analyses. The

results of the elevated plus maze were analyzed by means

of the 2-way ANOVA for the effect of treatment and sex.

Changes in the behavior of the offspring during the social

interactions observed were evaluated for changes, each

parameter both within the group and between the groups

(2-way ANOVA for repeated measurements).

RESULTS

The elevated plus maze was used to evaluate the possible

effect of maternal inflammation during pregnancy on

Prenatal Maternal Inflammation Reduced Anxiety 163

anxiety-related behavior in the offspring. Offspring

behavior in the elevated plus maze showed that maternal

inflammation during pregnancy significantly increased

the preference of the offspring to stay in the close

arms, indicated by the ratio of the time spent in the close

arms versus the time spent in the open arms. This ratio was

2.09� .14 for offspring from the maternal inflammation

group, compared to 1.63� .18 in offspring from the

control group (n¼ 24 and 21, respectively, p¼ .03,

F¼ 4.6 between groups, no significant effect of sex was

detected p¼ .38, F¼ .76, Fig. 1A). No significant

differences were observed in the number of entries into

the close arms. Animals from the control group entered

30� 2.4 times, while the test group entered 33.17� 1.7

times during the 5 min of the test, as demonstrated in

Figure 1B.

Figure 2A demonstrates that during the 30 min of

observation on the first day, offspring of the control

group did not play with each other at all. Although not

significant (p¼ .68, F¼ .17) offspring of the maternal

inflammation group did playing for a short time during

the initial day. The duration of positive interactions, such

as playing, was increased on the second day of paired

housing and during the last observation of offspring

from the maternal inflammation group. Huddling

was observed only at the end of the last session in

both groups, no significant differences were observed

(p¼ .86, F¼ .03, Fig. 2B). In contrast, the anogenital

sniffing frequency was higher in the beginning and

declined over time. In both groups, a significant

reduction in the number of anogenital sniffing interac-

tions was observed within the first day (p< .03, control

and p< .004, LPS). Overall, the number of anogenital

sniffing events showed a similar trend in offspring from

both groups (p< .17, F¼ 2.04). An equal tendency was

FIGURE 1 Effect of maternal inflammation mice anxiety in elevated plus maze. (A) The ratio

between the time mice spent in the close arms versus the time mice spent in the open arms is

presented. (B) The number of entries into the close arms. n(control)¼ 21, and n(maternal

inflammation )¼ 24, � p¼ .03, F¼ 4.6 between groups, no significant effect of sex was detected

p¼ .38, F¼ .76, p¼ .03,F¼ 4.6 between groups, no significant effect of sex was detected p¼ .38,

F¼ .76, possibly due to small number of female subjects.

FIGURE 2 Effect of maternal inflammation on social interac-

tions. (A) Playing, p¼ .68, F¼ .17; (B) huddling, p¼ .86,

F¼ .03; (C) nose to back, p¼ .17, F¼ 2.04. (Group X episode:

control, p¼ .29; LPS, p< .001). Analyses of the different

parameters are presented for 5-min observations: during the first

0–30 min of paired housing on the first day and during the second

and last observation periods at 24 hr and 7 days later.

164 Hava et al.

observed when the general activity of the mice was

compared, as depicted in Figure 3A. A similar significant

reduction over time in general activity was observed in

both groups, (Group X episode: control, p¼ .059; LPS,

p¼ .01; and between groups p<.58, F¼ .32). The time

offspring spent digging was significantly higher in the

control group on the first day (p< .02). This tendency

was reversed at the beginning of the second observation

day. An increase of up to 144.9� 88 s was observed in

offspring from the maternal inflammation group, com-

pared to 52� 35 s in the controls (p< .016, t-test). Later

on the same day and on the 7th day, the offspring of both

groups behaved similarly (Fig. 3B). The time mice spent

trimming whiskers was significantly longer for the

offspring of the maternal inflammation group on the first

day of observation (p¼ .0001, F¼ 28.5 between groups;

Group X episode: control, p¼ .68; LPS, p¼ .023;

Fig. 3C): about 20%–40% of the time (47.6� 44.2 s in

the first 5-min episode), compared to the control group,

which spent about 6% of the time (12.67� 5.3 s in the

first 5-min episode) doing it. On the 2nd and 7th days,

mice from both groups spent a similar amount of time on

this type of behavior. Also, the time offspring from the

control group spent trimming whiskers did not change

during the week (p> .1), compared to the offspring from

the maternal inflammation group that adapted during the

week (p< .02). In general, the frequency of interactions

and the amount of activity or neutral interaction was

reduced over time during the week in both groups, and

the only significant difference found between the groups

was in the time spent trimming whiskers.

Aggressive behavior, such as chasing, frequently

evolved following anogenital sniffing. In such cases,

chasing developed into an attack from the rear, followed

by fighting. Chasing was observed on the first day, but

diminished significantly during the first day in offspring

of both groups as depicted in Figure 4A (Group X

episode: control, p¼ .008; LPS, p< .017). No significant

difference was observed between the groups for this

parameter (p> .1, F¼ .2). A high frequency of attacks

from the rear was observed on the first day in both groups.

The number of attacks was significantly reduced in

offspring of both the control and the maternal inflamma-

tion groups (Group X episode: control, p¼ .004; LPS,

p< .003). Like the time spent chasing, the frequency of

such observed attacks was equal in both groups (Fig. 4B.

p¼ .8, F¼ .05). The time that offspring spent fighting

was, in general, higher on the first day. However, no clear

tendency was observed over time. Offspring from the

control group spent more time fighting, as compared to

offspring from the maternal inflammation group. How-

ever, differences in this parameter did not reach

significance (Fig. 4C. p< .1, F¼ 2.7). Although aggres-

sive behavior diminished significantly during the first

day of observation, in three out of five cages in the control

group, one mouse was severely wounded between the

2nd and the 7th day of paired housing. In contrast, only

one mouse (out of four pairs) was wounded in the

maternal inflammation group.

DISCUSSION

Our main findings in the present study demonstrate that 8-

month-old mice, that had experienced an inflammatory

response during the last stages of their embryonic

development, E17, had a distinct response to stressful

situations, mainly confirmed by the longer time spent in

FIGURE 3 Effect of maternal inflammation on social

interactions. (A) Activity, p¼ .5, F¼ .3 (Group X episode:

control, p¼ .059; LPS, p¼ .01); (B) digging, p¼ .6,F¼ .21; (C)

whisker trimming, p¼ .0001, F¼ 28.5 (Group X episode:

control, p¼ .68; LPS, p¼ .023). Analyses of the different

parameters are presented for 5-min observations during the first

0–30 min of paired housing on the first day and for the second

and last observation periods at 24 hr and 7 days later.

Prenatal Maternal Inflammation Reduced Anxiety 165

the close arms of the elevated plus maze and a tendency of

reduced aggressiveness towards their cagemates.

Hyperanxiety in the elevated plus maze was observed

in adult rodents previously exposed to prenatal stress (PS).

This response was induced by different type of stressors,

such as immobilization (Takahashi, Turner, & Kalin,

1992), cold-water immersion (Velazquez-Moctezuma,

Dominguez Salazar, & Cruz Rueda, 1993), noise and

flashing lights (Fride & Weinstock, 1988), and additional

paradigms (Weinstock, 1997; Weinstock, Fride, &

Hertzberg, 1988). It is well documented that the stress

response was associated with modifications in the

hypothalamus-pituitary-adrenal (HPA) axis. Alterations

of the maternal HPA axis increased plasma levels of CRH,

ACTH, and cortisol. The developing HPA axis undergoes

extensive changes during gestation days 15–21 when it

begins to release ACTH and corticosteron in response to

the stress (Boudouresque et al., 1988). At this stage, the

embryonic HPA axis is highly sensitive to maternal stress

and the subsequent hormones released, which may cause

long-lasting alteration (Marchlewska-Koj, Kapusta, &

Kruczek, 2003). Activation of the HPA is achieved also by

the innate immune response. The inflammatory response

involves the induction of proinflammatory cytokines,

which activate the HPA axis to release glucocorticoide

(GC), acting to suppress cytokines levels. Increased fetal

brain CRH was shown in rats following similar maternal

treatment (.1 mg/g LPS at E18, Gayle et al., 2004). We

have already established that the protocol of maternal

inflammation used in the present study indeed elevates the

proinflammatory cytokine IL-6 in the fetal brain (Golan

et al., 2005) and thus, may cause the induction of CRH in

the fetal brain. Therefore, the similarity between the

behaviors of the adult offspring in the elevated plus maze

reported here and in studies of PS is to be expected.

Other stressful situations for mice are a novel

environment and the presence of unfamiliar cagemates.

In such situations, the offspring from the maternal

inflammation group showed reduced aggression towards

their cagemates. This is in accordance with previous

reports, which examined the long-term modification of the

behavior of adult offspring after experiencing PS. In

addition, reduced social interaction in adulthood follow-

ing PS was reported in humans, primates, and rodents

(see: Kofman, 2002; Weinstock, 2001). Similar beha-

vioral consequences were also observed in C57BL/6 mice

after maternal separation during the first postnatal week

(postnatal day 5), (Romeo et al., 2003) this is the rodent’s

equivalent of the third trimester in human embryonic

development.

A remarkable effect of maternal inflammation,

observed during the first day of social interaction, was

an increase in the time spent trimming whiskers; this type

of behavior is usually associated with social dominance

(Lijam et al., 1997). Mice from the control group spent a

constant amount of time in whisker trimming throughout

the entire test period. In contrast, in the maternal

inflammation group, a significantly longer time for

whisker trimming was observed only on the first day

(Fig. 3C). It is possible to suggest that feelings of anxiety

and insecurity in mice from this group caused the

extended amount of time required for the establishment

of a hierarchy between the cagemates. The fact that on the

2nd and 7th days of observation the time spent on whisker

FIGURE 4 Effect of maternal inflammation on social

interactions. (A) Chasing, p¼ .6, F¼ .2, (Group X episode:

control, p¼ .008; LPS, p< .017); B) Attacking from the rear

p¼ .8, F¼ .05, (Group X episode: control, p¼ .004; LPS,

p< .003); (C) fighting, p¼ .12, F¼ 2.7 (Group X episode:

control, p¼ .17; for LPS p cannot be calculated). Analyses of

the different parameters are presented for 5-min observations

during the first 0–30 min of paired housing on the first day and

the second and the last observation periods at 24 hr and 7 days

later.

166 Hava et al.

trimming was similar for both groups may suggest that

offspring of the maternal inflammation group require

more time to adapt to their new conditions, compared to

the offspring from the control groups.

In conclusion, maternal administration of LPS caused

long-term changes in the responses of the adult

offspring, mainly increased anxiety-related behavior

and an extension of the time required for their

adaptation to a new social situation. The observed effects

of maternal inflammation during pregnancy may be

explained by the activation of the maternal and/or fetal

stress response.

REFERENCES

Barker, V., Middleton, G., Davey, F., & Davies, A. (2001).

TNFa contribute to the death of NGF-dependent neu-

rons during development. Nature Neuroscience, 4, 1194–

1198.

Beattie, E., Stellwagen, D., Morishita, W., Bresnahan, J., Ha,

B., Von Zastrtrow, M., Beattie, M., & Malenka, R. (2002).

Control of synaptic strength by glial TNFa. Science, 295,

2282–2285.

Boudouresque, F., Guillaume, V., Grino, M., Strbak, V.,

Chautard, T., Conte-Devolx, B., & Oliver, C. (1988).

Maturation of the pituitary-adrenal function in rat fetuses.

Neuroendocrinology, 48, 417–422.

Breivik, T., Stephan, M., Brabant, G. E., Straub, R. H.,

Pabst, R., & Von-Horsten, S. (2002). Postnatal lipopolysac-

charide-induced illness predisposes to periodontal disease

in adulthood. Brain Behavior and Immunity, 16, 421–

438.

Butler, M. P., O’Connor, J. J., & Moynagh, P. N. (2004).

Dissection of tumor-necrosis factor-a inhibition of

long-term potentiation (LTP) reveals A p38 mitogen-

activated protein kinase-dependent mechanism which maps

to early-but not late-phase LTP. Neuroscience, 124, 319–

326.

Cai, Z., Pan, Z. L., Pang, Y., Evans, O. B., & Rhodes, P. G.

(2000). Cytokine induction in fetal rat brains and brain injury

in neonatal rats after maternal lipopolysaccharide adminis-

tration. Pediatric Research, 47, 64–72.

Clarke, A. S., & Schneider, M. L. (1993). Prenatal stress has

long-term effects on behavioral responses to stress in

juvenile rhesus monkeys. Devlopmental Psychobiology,

26(5), 293–304.

Cunningham, A. J., Murray, C. A., O’Neill, L. A. J., Lynch, M.

A., & O’Connor, J. J. (1996). IL-1beta and TNF inhibit LTP

in the rat dentate gyrus in vitro. Neuroscience Letter, 203,

17–20.

Dammann, O., Kuban, K. C. K., & Leviton, A. (2002).

Perinatal infection, fetal inflammatory response, white

matter damage and cognitive limitation in children born

preterm. Mental Retardation and Developmental Disabilities,

8, 46–50.

Fride, E., & Weinstock, M. (1988). Prenatal stress in-

creases anxiety related behavior and alters cerebral

lateralization of dopamine activity. Life Science, 42,

1059–1065.

Gayle, D. A., Beloosesky, R., Desai, M., Amidi, F., Nunez, S.

E., & Ross, M. G. (2004). Maternal lipopolysaccharide

induces cytokines in the amniotic fluid and corticotropin

releasing hormone in the fetal rat brain. American Journal

of Physiology, Regulatory, Integrative and Comparitive

Physiology, 286, R1024–9. Epub 2004 Feb 26.

Golan, H., Levav, T., Mendelsohn, A., & Huleihel, M. (2004).

TNF alpha involvement in hippocampus development and

function. Cerebral Cortex, 14, 97–105.

Golan, H., Lev, V., Hallak, M., Sorokin, Y., & Huleihel, M.

(2005). Specific neurodevelopmental damage in offspring

following maternal inflammation during pregnancy. Neuro-

pharmacology, 48, 903–917.

Kofman, O. (2002). The role of prenatal stress in the etiology of

developmental behavioral disorders. Neuroscience and

Biobehavioral Reviews, 26, 457–470.

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., & Wynsham-Boris, A. (1997). Social

interaction and sensorymotor gating abnormalities in mice

lacking dvl1. Cell, 90, 895–905.

Lister, R. G. (1987). The use of a plus maze to measure anxiety

in the mouse. Psychopharmacology, 92, 180–185.

Marchlewska-Koj, A., Kapusta, J., & Kruczek, M. (2003).

Prenatal stress modifies behavior in offspring of bank voles

(clethrionomys glareolus). Physiology and Behavior, 79,

671–678.

Neumann, H., Schweigreiter, R., Yamashita, T., Rosenkranz,

K., Wekerle, H., & Barde, Y. (2002). Tumor necrosis

factor inhibits neurite outgrowth and branching of hippo-

campal neurons by rho-dependent mechanism. Journal of

Neuroscience, 22, 854–862.

Romeo, R. D., Mueller, A., Sisti, H. M., Ogawa, S., McEwen,

B. S., & Brake, W. G. (2003). Anxiety and fear behaviors

in adult male and female C57BL/6 mice are modulated

by maternal separation. Hormones and Behavior, 43, 561–

567.

Saliba, E., & Henrot, A. (2001). Inflammatory mediators and

neonatal brain damage. Biology of the Neonate, 79, 224–

227.

Schneider, H., Pitossi, F., Balschun, D., Wagner, A., del Roy,

A., & Besedovsky, H. O. (1998). A neuromodulatory role of

interleukin-1beta in the hippocampus. PNAS, 95, 7778–

7783.

Takahashi, L. K., Turner, J. G., & Kalin, N. H. (1992). Prenatal

stress alters brain catecholaminergic activity and potentiates

stress-induced behavior in adult rats. Brain Research, 574,

131–137.

Tancredi, V., D’Arcangelo, G., Grassi, F., Tarroni, P., Palmieri,

G., Santoni, A., & Eusebi, F. (1992). Tumor necrosis factor

alters synaptic transmission in rat hippoacmpal slice.

Neuroscience Letter, 146, 176–178.

Velazquez-Moctezuma, J., Dominguez Salazar, E., & Cruz

Rueda, M. L. (1993). The effect of prenatal stress on adult

Prenatal Maternal Inflammation Reduced Anxiety 167

sexual behavior in rats depends on the nature of the stressor.

Physiology and Behavior, 53, 443–448.

Weinstock, M. (1997). Does prenatal stress impair coping

and hypothalamic-pituitary-adrenal axis? Neuroscience and

Biobehavioral Reviews, 21, 1–10.

Weinstock, M. (2001). Alterations induced by gestational stress

in brain morphology and behavior of the offspring. Progress

in Neurobiology, 65, 427–451.

Weinstock, M., Fride, E., & Hertzberg, R. (1988). Prenatal

stress effects on functional development of the offspring.

Progress in Brain Research, 73, 319–331.

Yang, L., Lindholm, K., Konishi, Y., Li, R., & Shen, Y. (2001).

Target depletion of distinct tumor necrosis factor receptor

subtype reveals hippocampal neuron death and survival

through different signal transduction pathways. Journal of

Neuroscience, 22, 3025–3032.

168 Hava et al.