Exercise during pregnancy increases hippocampal brain-derived

neurotrophic factor mRNA expression and spatial learning in

neonatal rat pup

Panaree Parnpiansila, Nuanchan Jutapakdeegula, Thyon Chentanezb, Naiphinich Kotchabhakdia,*

aNeuro-Behavioral Biology Center, Institute of Science and Technology for Research and Development, Mahidol University, Salaya Campus, Nakornpathom,

73170, ThailandbDepartment of Physiology, Faculty of Science, Mahidol University, Bangkok, 10400, Thailand

Received 27 June 2003; received in revised form 8 August 2003; accepted 14 August 2003

Abstract

Physical activities for a few days can increase brain-derived neurotrophic factor (BDNF) mRNA in rat hippocampus. To investigate the

influence of maternal exercise during pregnancy on rat pup hippocampal BDNF mRNA, we studied its expression by a semi-quantitative RT-

PCR method after young pregnant rats were exercised on a motor driven treadmill. Pups of exercised mothers had significantly increased

hippocampal BDNF mRNA expression compared to the control rat pups at birth (on postnatal day 0) (P , 0:001). In contrast, hippocampal

BDNF mRNA expression in pups of exercised mothers decreased significantly on postnatal day 28 (P , 0:002). Spatial learning of rat pups

was examined by multiple T maze training for 7 consecutive days between postnatal days 40 and 47. Pups of exercised mothers showed a

significant increase in spatial learning ability as demonstrated by significant decreases in total time from starting to target and total number of

errors as compared to age-matched control pups during the first 4 days of 7 consecutive days on multiple T maze training (P , 0:05). Thus,

physical exercise during gestation in pregnant mothers can increase hippocampal BDNF mRNA expression of postnatal pups and result in an

improvement in spatial learning in pups from exercised dams.

q 2003 Published by Elsevier Ireland Ltd.

Keywords: Brain-derived neurotrophic factor mRNA; Prenatal exercise; Spatial learning; Semiquantitative RT-PCR; T maze; Hippocampus

Moderate aerobic exercise during pregnancy is thought to

benefit the mother by maintaining her aerobic fitness during

pregnancy [19], and reducing pregnancy-associated dis-

comforts [8]. Exercise is presumed to have a positive

influence on fetal health as well, though such effects have

not been documented.

Physical exercise can also improve the performance of

experimental animals in tests of spatial learning [4,5] and

passive avoidance memory [14]. An increase in brain-

derived neurotrophic factor (BDNF) was postulated to play a

role in these processes [7]. The expression of BDNF mRNA

in rat brain is increased by both physical activity and by tasks

associated with learning and memory. In rat hippocampus,

BDNF mRNA increased in an age-dependent manner while

the pattern for its protein levels was mixed [2]. Further

increases in BDNF mRNA can be stimulated by running [13],

or by the radial arm maze test [10]. However, it is unknown

whether exercise in a pregnant mother can have effects on the

learning and memory of her offspring. This study was

performed to determine the effects of maternal exercise

during the gestation period on the postnatal expression of

BDNF in the hippocampus of rat pups. Furthermore, we

explored whether maternal exercise has effects on multiple T

maze, a test of spatial learning in these pups.

Eight-week-old pregnant Sprague–Dawley rats were

purchased from The National Laboratory Animal Center,

Mahidol University, Salaya. Gestational day was timed

from the appearance of a vaginal plug after mating. Rats

were housed individually in stainless steel cages. The rat

room was maintained on a reversed light/dark cycle (lights

off 08:00 h; lights on 20:00 h). The pregnant rats were

equally divided into control (n ¼ 20) and exercised groups

(n ¼ 20). During gestation exercised pregnant rats were

exercised by running on a horizontal motorized treadmill at

20 m/min for 30 min/day (approximately equal to 70% VO2

0304-3940/03/$ - see front matter q 2003 Published by Elsevier Ireland Ltd.

doi:10.1016/S0304-3940(03)01008-5

Neuroscience Letters 352 (2003) 45–48

www.elsevier.com/locate/neulet

* Corresponding author. Tel.: þ66-2-4419321; fax: þ66-2-4419743.

E-mail address: scnkc@mahidol.ac.th (N. Kotchabhakdi).

max) [16], 5 consecutive days/week. One-third of the total

number of pups (n ¼ 9) from each group were sacrificed on

postnatal days 0, 14 and 28, respectively, for measurement

of hippocampal BDNF. The remaining pups (n ¼ 15) of

each group were trained in a multiple T maze for 3

consecutive days and then tested on the maze daily on

postnatal days 40–47. These animals were sacrificed on

postnatal day 47, and their two sides of hippocampi were

collected for BDNF mRNA measurement.

Total RNA was isolated from rat pup hippocampus using

guanidine thiocyanate [1]. The b-actin sense primer is 50-

CCCAGAGCAAGAGAGGCATC-30 and the antisense

primer is 50-CTCAGGAGGAGCAATGATCT-30 (GenBank

accession number: NM031144). In the RT-PCR protocol 25

pmol of each b-actin primer, 200 mM dNTPs, 1.5 mM

MgCl2 and 1 unit of Taq DNA polymerase were used. For

BDNF, 2 mg total RNA was preheated with 40 pmol oligo

(dT) primer at 70 8C for 10 min in the RT reaction. The

mRNA was reversed transcribed into single stranded cDNA

using AMV reverse transcriptase (Pacific Science Com-

pany, Finland). Ten units of AMV reverse transcriptase was

activated at 42 8C for 60 min. The Oligo (dT) primer

sequence was obtained from Singapore (GENSET Singa-

pore Biotech. Pte Ltd, Singapore). BDNF primers obtained

from GenBank accession number M61175 were used [17].

To amplify BDNF 25 pmol of each sense and antisense

BDNF primer, 200 mM dNTPs, 2 mM MgCl2 and 1.5 units

of Taq DNA polymerase were used. PCR amplification of

BDNF and b-actin was carried out using the Gene Amp

PCR system (Perkin Elmer 9700). Initial denaturation of

BDNF mRNA was performed for 5 min. Each PCR cycle of

BDNF consisted of denaturation at 94 8C for 30 s, annealing

at 58 8C for 30 s, and extension at 72 8C for 45 s with final

extension at 72 8C for 7 min. For b-actin, AMV reverse

transcriptase was activated at 42 8C for 30 min. Then

denaturation was performed at 94 8C for 30 s, annealing at

58 8C for 30 s, and extension at 72 8C for 30 s. PCR products

of BDNF and b-actin were separated by electrophoresis on a

1.5% agarose gel and recorded by a gel documentation

system (Gel Doc, Bio-Rad). The density of the BDNF band

was analyzed with the Scion image software program

(http://www.scioncorp.com). The amount of PCR products

of hippocampal BDNF is normalized to that of PCR

products of b-actin in each sample.

Finally, RT-PCR of BDNF and b-actin conditions were

used for optimized semiquantitative RT-PCR conditions

using 2 and 0.25 mg of total RNA, respectively. The cycle

number for BDNF and b-actin amplification was varied

from 15 to 44 cycles to optimize condition. Thirty-four and

24 cycles were chosen for optimal conditions of BDNF and

b-actin amplification, respectively.

A time course of the BDNF and b-actin RT-PCR products

from rat pup hippocampus on postnatal days 0, 14 and 28 is

shown in Fig. 1. Densitometry scans of BDNF mRNA

normalized to b-actin mRNA are shown in Fig. 2. The

relative amount of hippocampal BDNF mRNA in the

exercised group at postnatal day 0 was significantly greater

than control pups (P , 0:001). At postnatal day 14

hippocampal BDNF mRNA increased from baseline in

both exercised and control groups, but there were no

significant differences between the two groups. At postnatal

day 28, the hippocampal BDNF mRNA level in the exercised

group was significantly lower than in the control group

(P , 0:01). At postnatal day 47 after 7 consecutive days of

training on multiple T maze, the BDNF mRNA expression

between the two groups showed no significant difference.

On the multiple T maze, pups from the exercised mothers

showed significant decreases in both total working time

(Fig. 3A) and total number of errors (Fig. 3B) from starting

to target during the first 4 days of training as compared to

the age-matched control pups (P , 0:05).

This study was designed to investigate the pattern of

hippocampal BDNF mRNA expression during postnatal

development in pups from exercised pregnant mothers

compared with sedentary pups. Finally, this study demon-

strates that treadmill exercise during pregnancy can cause a

significant increase in hippocampal BDNF mRNA of pups

that is measurable postnatally. The mechanisms underlying

this regulation, however, are currently undefined. BDNF

behaved as an immediate early gene and Ca2þ influx

increased its transcription through Ca2þ/cAMP within the

regulatory region of the BDNF gene [3]. It is well

established that BDNF promotes the phosphorylation of

synapsin I via activation of TrkB receptors in the

presynaptic terminals, resulting in neurotransmitter release

[9]. Exercise upregulated the MAP-K, PKC and CaM-K

signaling cascades of TrkB receptor stimulation [15].

Additionally, exercise elevates the expression of the

transcription factor CREB, as detected using microarray

and Taqman RT-PCR. The glutamatergic system seems to

be the most important in stimulating and the GABAergic

system in inhibiting BDNF mRNA levels in hippocampal

neurons [11]. Moreover, exercise may elevate levels of

BDNF by reducing GABA function. Maternal exercise may

be an important factor for the induction of BDNF gene

expression in hippocampal neurons by calcium influx via

NMDA receptors [7]. Zafra et al. [20] reported that basic

expression of BDNF mRNA is regulated by the balance

between the glutamate and GABA neurotransmitter sys-

tems. Accordingly, bicucullin, a GABAA receptor antagon-

ist, increased and muscimol, a GABAA receptor agonist,

decreased BDNF mRNA levels in vivo and in vitro. MK-

801, a noncompetitive NMDA receptor antagonist, blocked

Fig. 1. Gel electrophoresis of rat b-actin and hippocampal BDNF mRNA in

control (CTL) and pups from exercised mothers (EXS) at postnatal days (P)

0, 14 and 28, respectively.

P. Parnpiansil et al. / Neuroscience Letters 352 (2003) 45–4846

the increase in BDNF mRNA produced by bicucullin [20],

suggesting that the effect of glutamate on the basal

expression of BDNF mRNA is mediated through NMDA

rather than non-NMDA receptors [21]. Further studies have

demonstrated that MK-801 reduced the basal levels of

BDNF mRNA in the rat hippocampus in vitro and in vivo

[20]. On the other hand, MK-801, which specifically blocks

NMDA receptors, was ineffective in blocking the increase

in BDNF mRNA in hippocampal neurons. Zafra et al. [20]

found that the NMDA receptor antagonists, MK-801 and

ketamine could block the increase in BDNF and NGF

mRNA in vivo. However, neither MK-801 nor ketamine

inhibited the kainic acid mediated increase in hippocampal

BDNF mRNA, although both drugs effectively suppressed

seizures resulting from NMDA receptor activation. It has

been proposed that the placenta and amniotic fluid may also

be other sources of neurotrophic factors for the developing

fetus [6,18]. In addition, BDNF and NGF are expressed in

human amniotic epithelial tissue [18]. The sources of BDNF

and NGF in the amniotic fluid are unknown, though it is

likely that placenta amniotic epithelial tissue is one source.

There is evidence that some maternally derived growth

factors cross the placenta and are active in the fetus [6].

Exercise upregulates maternal hippocampal neuron activity

via glutamatergic neurons resulting in a significant increase

in hippocampal BDNF mRNA. Maternal BDNF gene

expression may mediate the signal transduction pathways

via release of glutamate, depolarization and Ca2þ dependent

BDNF transcription to cross the placenta via an autocrine or

paracrine mechanism and indirectly activate fetal BDNF

gene expression in hippocampal neurons. Therefore,

hippocampal BDNF mRNA of pups from exercised dams

increases immediately after birth via still unknown

mechanisms.

During neural development, hippocampal BDNF mRNA

expression reached the peak level at postnatal day 14 while

BDNF protein reached its highest level between postnatal

days 14 and 21 [2]. The hippocampus plays a central role in

spatial learning and memory [12]. Spatial learning is

associated with an increase in BDNF mRNA levels in the

hippocampus within 15 and 30 min after radial arm maze

test [10]. The induction of BDNF mRNA is relatively rapid,

detectable within 30 min and peaking at about 3 h [21]. In

this study we collected pup brain 24 h after T maze training

for quantitative measurement of hippocampal BDNF

mRNA. Thus, we could not detect an increased hippocam-

pal BDNF mRNA level at P47; it may return to the level in

control pups at 24 h after T maze training. With respect to

total working time and total number of errors from starting

to target, the maze shows a significant difference between

two groups of pups at the first 4 days of 7 consecutive days

of training. The behavioral differences at the first 4 days in

our study result from an elevation of hippocampal BDNF

mRNA expression after T maze training.

In conclusion, these results indicate that maternal

exercise by treadmill running at submaximal intensity

during pregnancy can upregulate rat pup hippocampal

BDNF mRNA expression at birth associated with an

increase in spatial learning. Increased BDNF mRNA

expression in the pups by physical exercise in pregnant

Fig. 2. Alterations of relative hippocampal BDNF mRNA expression in the

control and pups from exercised mothers at different postnatal ages: P0, P14

and P28, respectively. Each bargraph represents three sessions of RNA

determinations prepared totally from both sides of nine hippocampi for

each postnatal group. BDNF mRNA data are expressed as a BDNF/actin

ratio. Each value represents the mean ^ SEM of three experiments:

***P , 0:001, **P , 0:01, NS means no significant difference.

Fig. 3. (A,B) Bargraphs show the total working time from starting to target

and total number of errors in T maze during 7 consecutive days of T maze

training respectively. * means a significant difference between control and

pups from exercised mothers at P , 0:05.

P. Parnpiansil et al. / Neuroscience Letters 352 (2003) 45–48 47

mothers may have beneficial effects on spatial learning and

other intellectual development. Future studies will explore

whether maternal exercise during pregnancy and lactation

can maintain the neonatal elevation of hippocampal BDNF

mRNA expression, and spatial learning from exercised

mothers.

References

[1] P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by

acid guanidinium thiocyanate-phenol-chloroform extraction, Anal.

Biochem. 162 (1987) 156–159.

[2] K.P. Das, S.L. Chao, L.D. White, W.T. Haines, G.J. Harry, H.A.

Tilson, S. Barone Jr., Differential patterns of nerve growth factor,

brain-derived neurotrophic factor and neurotrophin-3 mRNA and

protein levels in developing regions of rat brain, Neuroscience 103

(2001) 739–761.

[3] S. Finkbeiner, Calcium regulation of the brain-derived neurotrophic

factor gene, Cell. Mol. Life Sci. 57 (2000) 394–401.

[4] D.E. Fordyce, R.P. Farrar, Effect of physical activity on hippocampal

high affinity choline uptake and muscarinic binding: a comparison

between young and old F344 rats, Brain Res. 541 (1991) 57–62.

[5] D.E. Fordyce, J.M. Wehner, Physical activity enhances spatial

learning performance with an associated alteration in hippocampal

protein kinase C activity in C57BL/6 and DBA/2 mice, Brain Res. 619

(1993) 111–119.

[6] J.H. Gilmore, L.F. Jarskog, S. Vadlamudi, Maternal infection

regulates BDNF and NGF expression in fetal and neonatal brain

and maternal-fetal unit of the rat, J. Neuroimmunol. 138 (2003)

49–55.

[7] F. Gomez-Pinilla, V. So, J.P. Kesslak, Spatial learning induces

neurotrophin receptor and synapsin I in the hippocampus, Brain Res.

904 (2001) 13–19.

[8] A.E. Heffernan, Exercise and pregnancy in primary care. Nurse Pract.,

25 (2000) 42, 49, 53–56, passim.

[9] J.N. Jovanovic, A.J. Czernik, A.A. Fienberg, P. Greengard, T.S. Sihra,

Synapsins as mediators of BDNF-enhanced neurotransmitter release,

Nat. Neurosci. 3 (2000) 323–329.

[10] M. Mizuno, K. Yamada, A. Olariu, H. Nawa, T. Nabeshima,

Involvement of brain-derived neurotrophic factor in spatial memory

formation and maintenance in a radial arm maze test in rats,

J. Neurosci. 20 (2000) 7116–7121.

[11] R. Molteni, Z. Ying, F. Gomez-Pinilla, Differential effects of acute

and chronic exercise on plasticity-related genes in the rat hippo-

campus revealed by microarray, Eur. J. Neurosci. 16 (2002)

1107–1116.

[12] R.G. Morris, P. Garrud, J.N. Rawlins, J. O’Keefe, Place navigation

impaired in rats with hippocampal lesions, Nature 297 (1982)

681–683.

[13] S.A. Neeper, F. Gomez-Pinilla, J. Choi, C.W. Cotman, Physical

activity increases mRNA for brain-derived neurotrophic factor and

nerve growth factor in rat brain, Brain Res. 726 (1996) 49–56.

[14] T. Samorajski, C. Delaney, L. Durham, J.M. Ordy, J.A. Johnson, W.P.

Dunlap, Effect of exercise on longevity, body weight, locomotor

performance, and passive-avoidance memory of C57BL/6J mice,

Neurobiol. Aging 6 (1985) 17–24.

[15] R.A. Segal, M.E. Greenberg, Intracellular signaling pathways

activated by neurotrophic factors, Annu. Rev. Neurosci. 19 (1996)

463–489.

[16] R.E. Shepherd, P.D. Gollnick, Oxygen uptake of rats at different work

intensities, Pflug. Arch. 362 (1976) 219–222.

[17] W. Tokuyama, T. Hashimoto, Y.X. Li, H. Okuno, Y. Miyashita,

Highest trkB mRNA expression in the entorhinal cortex among

hippocampal subregions in the adult rat: contrasting pattern with

BDNF mRNA expression, Brain Res. Mol. Brain Res. 62 (1998)

206–215.

[18] S. Uchida, Y. Inanaga, M. Kobayashi, S. Hurukawa, M. Araie, N.

Sakuragawa, Neurotrophic function of conditioned medium from

human amniotic epithelial cells, J. Neurosci. Res. 62 (2000) 585–590.

[19] A.M. Uzendoski, R.W. Latin, K.E. Berg, S. Moshier, Physiological

responses to aerobic exercise during pregnancy and post-partum,

J. Sports Med. Phys. Fitness 30 (1990) 77–82.

[20] F. Zafra, E. Castren, H. Thoenen, D. Lindholm, Interplay between

glutamate and gamma-aminobutyric acid transmitter systems in the

physiological regulation of brain-derived neurotrophic factor and

nerve growth factor synthesis in hippocampal neurons, Proc. Natl.

Acad. Sci. USA 88 (1991) 10037–10041.

[21] F. Zafra, B. Hengerer, J. Leibrock, H. Thoenen, D. Lindholm, Activity

dependent regulation of BDNF and NGF mRNAs in the rat

hippocampus is mediated by non-NMDA glutamate receptors, Eur.

Mol. Biol. Org. J. 9 (1990) 3545–3550.

P. Parnpiansil et al. / Neuroscience Letters 352 (2003) 45–4848