Maternal Folic Acid Supplementation During PregnancyImproves Neurobehavioral Development in Rat Offspring

Xinyan Wang1 & Wen Li1 & Shou Li1 & Jing Yan1& John X. Wilson2

& Guowei Huang1

Received: 8 February 2017 /Accepted: 6 April 2017 /Published online: 18 April 2017# Springer Science+Business Media New York 2017

Abstract Maternal folate status during pregnancy may influ-ence central nervous system (CNS) development in offspring.However, the recommended intakes of folic acid for women ofchildbearing age differ among countries and there is still noconsensus about whether folic acid should be supplementedcontinuously throughout pregnancy. We hypothesized thatfolic acid supplementation may be more beneficial for off-spring’s neurobehavioral development if prolonged through-out pregnancy instead of being limited to the periconceptionalperiod. In this study, three groups of the female rats were fedfolate-normal, folate-deficient, or folate-supplemented dietsthroughout pregnancy. In another group, the female rats werefed folate-supplemented diet from mating for 10 consecutivedays and then fed folate-normal diet for remainder days ofpregnancy. The results showed that maternal folate deficiencyincreased plasma homocysteine (Hcy) concentration in dams,delayed early sensory-motor reflex development, impairedspatial learning and memory ability, and caused ultrastructuraldamages in the hippocampus of offspring. Maternal folic acidsupplementation would be more effective on improving earlysensory-motor reflex development and spatial learning andmemory ability in offspring if prolonged throughout pregnan-cy instead of being limited to the periconceptional period. Inconclusion, prolonged maternal folic acid supplementation

throughout pregnancy would be more effective in neurobe-havioral development of offspring in rats.

Keywords Folate deficiency . Folic acid supplementation .

Periconceptional period . Pregnancy . Neurobehavioraldevelopment . Hippocampus

AbbreviationsCNS Central nervous systemHcy HomocysteineHHcy HyperhomocysteinemiaMWM Morris water mazeNTDs Neural tube defectsPND Postnatal day

Introduction

Folate is essential for nucleic acid synthesis, methylation re-actions, and cell division [1–3]. Inadequate dietary folate inpregnant women is a well-known risk factor for the occur-rence of the neural tube defects (NTDs) spina bifida and an-encephaly in their offspring. Besides, in rodent, maternal fo-late deficiency may induce increased homocysteine (Hcy)transfer to fetus [4], inhibit progenitor cells proliferation, in-crease apoptosis in fetal forebrain [5], and persistently impairmemory ability in offspring [6]. Similar impaired memory andincreased apoptosis along with hyperhomocysteinemia(HHcy) and disrupted choline metabolism were also foundin adult male mice with severe methylenetetrahydrofolate re-ductase (MTHFR) deficiency [7]. Maternal supplementationwith folic acid decreases NTDs incidence [8] and has somebeneficial effects on cognitive developments in offspring [9].

* Guowei Huanghuangguowei@tmu.edu.cn

1 Department of Nutrition and Food Science, School of Public Health,Tianjin Medical University, 22 Qixiangtai Road, Heping District,Tianjin 300070, China

2 Department of Exercise and Nutrition Sciences, School of PublicHealth and Health Professions, University at Buffalo, Buffalo, NY,USA

Mol Neurobiol (2018) 55:2676–2684DOI 10.1007/s12035-017-0534-2

Substantial decreases in the incidences of spina bifida andanencephaly have occurred in the USA and Canada sincethose countries mandated folic acid fortification of grain prod-ucts in 1998 [10, 11]. However, folic acid fortification is notmandated in most countries, including China. Chinese womenof childbearing age generally have low levels of dietary folateintake and plasma folate concentration [12], and the incidenceof NTDs is threefold to fivefold higher in Northern China thanin the USA [13].Many countries, whether or not theymandatefolic acid fortification of grain products, recommend folic acidsupplementation in women of childbearing age should beginprior to conception. Some of those countries including Chinarecommend that maternal folic acid supplementation shouldbe maintained until the end of the first trimester, but it isrecommended in the USA and Canada that supplementationshould be continued until the end of pregnancy [14]. It isevident that there is no universal recommendation for the du-ration of maternal folic acid supplementation after conception.Therefore, the present study mainly focused on discoveringthe better duration period for maternal folate supplementationthat would be more beneficial for offspring’s neurobehavioraldevelopment.

The present study investigated the effects of maternal ges-tat ion fol ic acid supplementat ion on offspring ’sneurodevelopment by surface righting reflex, negative geotax-is, and Morris water maze (MWM) tests. Surface rightingreflex and negative geotaxis test are two tests that are mostwidely used to assess the early sensory-motor function prior toweaning [15]; MWM test is one of the most frequently labo-ratory tests to access learning and memory ability for ratsorder than postnatal day 30 (PND30) [16, 17]. Furthermore,in order to indicate the effects of maternal folate status onaltering the offspring’s pathological changes in hippocampus,ultrastructure changes were detected by transmission electronmicroscopy.

Results

Characteristics of Dams and Offspring

All dams gained weight, had no detectable morbidity, andsurvived until delivery. There were no between-group differ-ences in prenatal body weight or conception rate, live birthrate, stillbirth rate, absorbed embryo rate, or the sex distribu-tion of offspring within groups (Table 1). Maternal folic acidsupplementation throughout pregnancy raised body weight,but had no effect on brain weight or brain weight-to-bodyweight ratio, in neonatal offspring (Table 1).

Folate Intake Affected Serum Folate and Plasma HcyConcentrations

Serum folate and plasma Hcy concentrations were determinedin dams at baseline (day -7), gestation (day 10) and delivery(day 21–25 (PND0)). Serum folate concentration increased inresponse to folic acid supplementation (repeated measuresanalysis F = 45.960, P = 0.000), but it returned to baselinewhen supplementation stopped in the folate-supplemented di-et short-period (FD-S) group (Fig. 1a). Folic acid deficiencyraised plasma Hcy concentration, but no significant differ-ences were found in plasma Hcy concentration among thosetwo folate-supplemented groups and folate-normal diet (ND)group (repeated measures analysis F = 36.694, P = 0.000)(Fig. 1b).

Maternal Folic Acid Supplementation ImprovedSensory-Motor Development in Offspring Priorto Weaning

All four groups showed a progressive improvement fromPND4 to PND8 in the surface righting reflex and negative

Table 1 Characteristics of dams and offspring

Measure DD group ND group FD-S group FD-L group F/χ2 value P

Maternal (n) 12 12 12 12

Conception (n, %) 8 (66.7) 10 (83.6) 9 (75) 8 (66.7) 1.160 0.763

Prenatal body weight (g) 334.50 ± 27.69 336.90 ± 24.41 340.78 ± 27.87 327.38 ± 17.51 0.476 0.701

Offspring (n) 127 165 168 108

Live birth (n, %) 112 (88.2) 151 (91.5) 145 (86.3) 99 (91.7) 3.206 0.361

Stillbirth (n, %) 13 (10.2) 13 (7.9) 17 (10.1) 8 (7.4) 1.076 0.789

Absorbed embryo (n, %) 2 (1.6) 1 (0.6) 6 (3.6) 1 (0.9) 4.019 0.225

Male offspring (n, %) 52 (46.4) 64 (42.4) 70 (48.3) 43 (43.4) 1.327 0.723

Body weight of neonatal offspring (g) 6.61 ± 0.73 6.48 ± 0.80 6.27 ± 0.77 7.06 ± 0.79 6.380 0.000*

Brain weight of neonatal offspring (g) 0.239 ± 0.024 0.228 ± 0.031 0.229 ± 0.028 0.242 ± 0.013 2.650 0.051

Brain weight-to-body weight ratio (%) 3.66 ± 0.54 3.55 ± 0.51 3.71 ± 0.50 3.46 ± 0.40 1.866 0.138

Data are expressed as proportion or mean ± SD according to the data distribution

*P < 0.05 for differences within four groups

Mol Neurobiol (2018) 55:2676–2684 2677

geotaxis tests, but this improvement differed between gendersand diet groups.

For the surface righting reflex test, the repeated measuresanalysis showed that males had shorter response time thanfemales (F = 14.644, P = 0.000). The folate-deficient diet(DD) group had the longest response time for both genders,and folic acid supplementation improved the righting responsesignificantly (male F = 23.915, P = 0.000 (Fig. 2a); femaleF = 22.936, P = 0.000 (Fig. 2b)). Compared those two folate-supplemented groups, folate-supplemented diet long-periodgroup (FD-L) group had shorter response time than FD-Sgroup in female (Fig. 2b), but no significant differences werefound in male offspring (Fig. 2a).

For the negative geotaxis test, the repeated measures anal-ysis also showed that the DD group had the longest responsetime, and folic acid supplementation improved performancesignificantly (male F = 15.350, P = 0.000 (Fig. 2c); female

F = 7.994, P = 0.000 (Fig. 2d)). Compared those two folate-supplemented groups, FD-L group had shorter response timethan FD-S group in both gender (Fig. 2c, d).

Taken together these results indicate that maternal supple-mentation, especially if continued throughout pregnancy,could improve the early development of sensory-motor func-tion in offspring.

Folic Acid Improved Learning and Memory Abilityin Adolescent and Adult Offspring

TheMWM test was performed to evaluate learning andmemoryability in adolescent (PND45) and adult (PND90) offspring. Thistest was performed on half male and half female, and there is nogender difference (P > 0.05), so the data was not separated bygender. The adolescent pups in all four diet groups progressivelydecreased the escape latency as the days of trial increased(F = 1154.652, P = 0.000). Repeated measures analysis showedthat theDDgroup took the longest time to find the platform; folicacid supplementation decreased escape latency significantly(F = 172.03, P = 0.000) (Fig. 3a). Comparing those two folate-supplemented groups, FD-L group took shorter time to find theplatform compared to FD-S group (Fig. 3a). In spatial probephase of the MWM test, the DD group spent the shortest timein the targeted quadrant (Fig. 3b) and had the least number ofcrossings (Fig. 3c), and folic acid supplementation improvedthose performances significantly (F = 45.548, P = 0.000 for thefirst day; F = 8.779, P = 0.000 for the second day (Fig. 3b))(H = 46.081, P = 0.000 for the first day; H = 53.759,P = 0.000 for the second day (Fig. 3c)). Compared those twofolate-supplemented groups, FD-L group spent longer time in thetargeted quadrant in the first day (Fig. 3b) and had larger numberof crossings (Fig. 3c) for both days. There was some evidencethat folic acid supplementation throughout pregnancy was moreeffective than supplementation limited to the periconceptionalperiod (Fig. 3a–c).

At PND90, the repeated measures analysis showed that DDgroup took the longest time to find the platform, folic acid sup-plementation decreased escape latency, and FD-L group tookshorter time to find the platform compared to FD-S group(F = 239.945, P = 0.000) (Fig. 3d). Also, the DD group spentthe shortest time in the targeted quadrant (Fig. 3e) and had theleast number of crossings (Fig. 3f), and they were significantlyimproved by folic acid supplementation (F = 137.945, P = 0.000for the first day; F = 28.102, P = 0.000 for the second day(Fig. 3e)) (H = 73.241, P = 0.000 for the first day; H = 42.407,P = 0.000 for the second day (Fig. 3f)). Compared those twofolate-supplemented groups, FD-L group spent longer time in thetargeted quadrant in the second day (Fig. 3e) and had largernumber of crossings in the first day (Fig. 3f). There was someevidence that folic acid supplementation throughout pregnancywas more effective than supplementation limited to thepericonceptional period (Fig. 3d–f).

A

B

Fig. 1 Maternal folic acid supplementation raised serum folate andmaternal folic acid deficiency increased plasma homocysteine (Hcy).Dams were randomly assigned (n = 12 rats/group) to a folate-normaldiet (ND) group, a folate-deficient diet (DD) group, a folate-supplemented diet short-period (FD-S) group, and a folate-supplemented diet long-period (FD-L) group. a Maternal serum folateconcentration. b Maternal plasma Hcy concentration. Data areexpressed as mean ± SD (n = 7 rats/group). Comparing with fourgroups at α = 0.05/(comparison times) (asterisk)

2678 Mol Neurobiol (2018) 55:2676–2684

Fig. 3 Folic acid improved learning and memory ability in adolescentand adult offspring. Diets and pups were the same as in Fig. 2. a Escapelatency during spatial acquisition phase of Morris water maze (MWM)test on PND45. b Time in the targeted quadrant during spatial probe phaseof MWM test on PND45. c Number of crossings during spatial probephase on PND45. d Escape latency during spatial acquisition phase ofMWM test on PND90. e Time in the targeted quadrant during spatialprobe phase of MWM test on PND90. f Number of crossings during

spatial probe phase on PND90. Comparisons between different groupswere performed by repeated measures ANOVA for escape latency (dataare expressed as mean ± SD), one-way ANOVA for time in the targetedquadrant (data are expressed as mean ± SD), and Mann-Whitney test fornumber of crossings (data are expressed as median (P25, P75)) (n = 30/group). Comparing with four groups at α = 0.05/(comparisons times)(asterisk). §P < 0.05 compared with DD group. #P < 0.05 comparedwith ND group. &P < 0.05 compared with FD-S group

Fig. 2 Maternal folic acid supplementation improved sensory-motor de-velopment in infant offspring. Dams were fed as described in Fig. 1.Fifteen female and 15 male pups were selected randomly from each dietgroup (3–4 offspring from each dam) at PND4 and used for 5 consecutivedays of surface righting reflex and negative geotaxis testing. a Surface

righting reflex test for male offspring. b Surface righting reflex test forfemale offspring. c Negative geotaxis test for male offspring. d Negativegeotaxis test for female offspring. Data are expressed as mean ± SD(n = 15 offspring for each gender/group). Comparing with four groupsat α = 0.05/(comparison times) (asterisk)

Mol Neurobiol (2018) 55:2676–2684 2679

Taken together, these results indicate that maternal supple-mentation, especially if continued throughout pregnancy,could be more effective on improving the learning and mem-ory ability in adolescent and adult offspring.

Folic Acid ProtectedHippocampal Ultrastructure in AdultOffspring

The DD group showed obvious pathological changes in hip-pocampal neurons’ cell bodies, such as condensation and mar-gination of nuclear chromatin and blurring of the membranes

of the nuclear envelope. Organelles were few and swollen inthis group: mitochondria were swollen and had disorderedcristae, rough endoplasmic reticulum was expanded severely,and Golgi body was not evident. However, these pathologicalchanges were prevented in the FD-L group (Fig. 4a).

Ultrastructural changes were also observed in myelinatedaxons, synapses, and capillaries. Myelinated axons with com-pact layer of myelin lamellae were observed in the FD-Lgroup, while many myelin sheaths surrounding axons wereloosened with slighted disrupted layers of myelin lamellaewidespread and vacuoles formed indicating decompacted

Fig. 4 Folic acid preventedpathological changes in theultrastructure of the hippocampusin adult offspring. Dams were fedas described in Fig. 1, andhippocampal samples of 4-month-old male offspring ratsfrom DD group and FD-L group(n = 2/ group) were collected fortransmission electron microscopeexamination. a Ultrastructure ofneurons in hippocampus. Scalebars are 5.0 μm for the topimages and 1.0 μm for the bottomimages. b Ultrastructural changesof myelin and axons inhippocampus (15 visual fields/rat). Arrow shows impairedmyelinated axon withdecompacted myelin. c G-ratiodata are expressed as mean ± SD.d Distribution of g-ratios. Dataare expressed as percentage. eElectron microscopic images ofsynapses. Scale bar = 1.0 μm. fQuantity of synapses (6000×).Data is expressed as median withminimum to maximum (15 visualfields/rat). g Width of synapticclefts. Data are expressed asmean ± SD (15 synapses/rat). hUltrastructural changes incapillaries in hippocampus.Arrow shows abnormal capillarywith obvious edema. Scalebar = 1.0 μm. *P < 0.05compared with DD group

2680 Mol Neurobiol (2018) 55:2676–2684

myelin in the DD group (Fig. 4b). The g-ratio (axon diameterdivided by total fiber diameter including the myelin sheath)was calculated to indicate alteration in myelin thickness. A g-ratio higher than 0.77 may slow conduction velocity [18].Compared to the FD-L group, the DD group had a higher g-ratio (t = 4.44, P = 0.043) (Fig. 4c), and tended to have agreater percentage of g-ratios above 0.77 (χ2=3.849,P = 0.05) (Fig. 4d). Synapses were fewer in the DD groupcompared with the FD-L group (Z = −3.419, P = 0.000)(Fig. 4e, f). However, no significant difference was found insynaptic cleft width (t = 1.922,P = 0.183) (Fig. 4g). Unlike theFD-L group, the DD group showed edema around capillaries,indicating a pathological change in the capillary wall structurefor the DD group (Fig. 4h).

Discussion

In the present study, as compared to the maternal folate-normal group, the folate-deficient group showed increases ofplasma Hcy concentration in dams and inhibition of neurobe-havioral development in offspring. The effects of maternalfolate deficiency on offspring were expressed as delayedsensory-motor reflex development in infancy, impaired spatiallearning and memory ability in adolescence and adulthood,and pathological changes in the ultrastructure of hippocampusin adulthood.

Previous studies found that maternal micronutrients status[19], especially folate and related B vitamins, may play im-portant roles in neurobehavioral development of offspring [3,4, 9]. In particular, maternal folate deficiency may impairlearning and memory ability in offspring [6]. There are mul-tiple mechanisms that may explain the effects of maternalfolate deficiency observed in the present study. Folate pro-motes methylation of Hcy to methionine through one carbonmetabolism [20]. By this mechanism, folate deficiency maybe the cause of HHcy [21], which may increase oxidativestress and apoptosis leading to brain damage [7, 22, 23].Increased maternal plasma Hcy may increase Hcy concentra-tion in the fetus to levels that persistently alter neurobehavioraldevelopment [4]. HHcy, along with oxidative stress, may im-pair central nervous system (CNS) functions by damagingblood-brain barrier [24, 25] and inhibiting axon myelination[18], which may induce neurobehavioral defects [26, 27].Besides, maternal folate deficiency may inhibit synaptic for-mation to reduce spatial learning and memory ability [28] inthis study.

Compared to the folate-normal group, folic acid supple-mentation increased serum folate concentration in dams andthen improved sensory-motor function in infant offspring aswell as learning and memory ability in adolescent and adultoffspring. The folate-normal diet (AIN-93G) contains 2.1 mgfolic acid/kg, and it is considered to meet the general nutrition

requirements of rats [29]. The folate-supplemented dietcontained 3.5 mg folic acid/kg diet and therefore added1.4 mg folic acid/kg diet compared to the folate-normal diet.This supplementation in rats is equivalent to consumption of a400-μg folic acid tablet daily in people consuming a healthydiet.

Although folate plays important roles in preventing NTDs,recommendations about folic acid intake for women of child-bearing age differ among countries, and whether folic acidsupplementation should be continued throughout pregnancyis controversial [14]. In the present study, as compared to FD-S group, FD-L group increased serum folate concentration to ahigher stable level in dams, and improved offspring’s neuro-behavioral development to some extent. The neurobehavioralperformances in offspring improved by prolong maternal folicacid supplementation throughout pregnancy were expressedas partly improved sensory-motor reflex development in in-fancy, and partly improved spatial learning and memory abil-ity in adolescence and adulthood.

The present study provides evidence that folic acid supple-mentation throughout pregnancy is more beneficial for theneurobehavioral development of offspring than supplementa-tion limited to the periconceptional period. There may be sev-eral mechanisms accounting for the apparent superiority ofcontinuing supplementation throughout pregnancy. The firstpotential mechanism is that folic acid may have persistentbeneficial effect on brain development long after neural tubeclosure since the CNS begins to form neuroepithelium aroundgestational day 11 in rats, and then the caudal portion of neuraltube gives rise to medulla, pons, spinal cord [30], and brainregions such as hippocampus, striatum, auditory, and visualcortices through rapid morphogenesis and synaptogenesis inthe late fetal and early postnatal period [9]. The second poten-tial mechanism is that folic acid supplementation during thepericonceptional period may induce persistent change in phe-notype and thereby need a sustained high levels of folatewhich might lead to brain injury if the need is not met bycontinued folic acid supplementation [31, 32]. For example,in the present study, the maternal serum folate concentration inthe FD-S group dropped once the supplementation stopped,which may have created a relative deficiency of folate supplyfor fetuses that had become accustomed to very high folatelevels. Although further investigation still needed, DNAmethylation may partly explain the improved neurobehavioraldevelopment benefited by maternal folic acid supplementa-tion. DNA methylation patterns, which may be altered bymaternal folic acid intake through one-carbon metabolism,are established during embryonic development [33] and con-sidered to be one of the epigenetic mechanisms that couldunderline fetal programming and brain development [9].Transient inhibition of DNAmethylation during infancy couldinduce long-lasting synaptic plasticity deficits, and impairlearning and memory ability in adulthood of mice [34].

Mol Neurobiol (2018) 55:2676–2684 2681

Barua’s study indicated that maternal folic acid status couldmodulate the expression of several genes involved in neuronalpathways through DNA methylation to impact behavior out-comes in mice offspring [35].

In conclusion, the present study shows that maternal folatedeficiency impairs neurobehavioral development in offspring,and those impairments can be prevented bymaternal folic acidsupplementation. Moreover, compared with supplementationlimited to the periconceptional period, prolonged maternalfolic acid supplementation throughout pregnancy maintainsserum folate concentration at a higher stable level in dams,and would be more beneficial for offspring’s neurobehavioraldevelopment to some extent, expressed as improved sensory-motor reflex development in infancy and improved spatiallearning and memory ability in adolescence and adulthood.This supports the recommendation that women of childbear-ing age supplement their diet with folic acid from pre-conception until the end of pregnancy, not only for preventingNTDs but also for providing further beneficial effects on theneurobehavioral development of offspring.

Materials and Methods

Rats and Diets

The Tianjin Medical University Animal Ethics Committeeapproved the experimental protocols of this study(TMUaEC2015001). Three-month-old female Sprague-Dawley rats (Charles River Laboratories, Beijing, China) wereassigned randomly into four groups (12 rats/ group): (1)folate-normal diet (ND) group fed the folate-normal dietthroughout pregnancy; (2) folate-deficient diet (DD) groupfed the folate-deficient diet throughout pregnancy; (3) folate-supplemented diet short-period (FD-S) group fed the folate-supplemented diet from mating for 10 consecutive days andthen fed the folate-normal diet for remainder days of pregnan-cy; and (4) folate-supplemented diet long-period (FD-L)group fed the folate-supplemented diet throughout pregnancy

(Fig. 5). The dams were mated with male rats with a female-to-male ratio 4:1. After delivery of their pups, all dams werefed the folate-normal diet. Some of the offspring were fed thefolate-normal diet until 4 months old. All rats were housed in aspecific pathogen-free facility at 24 ± 2 °C with 12-h light/dark cycle and allowed ad libitum access to food and water.

Diets were purchased from TestDiet (St. Louis, MO, USA).The folate-normal diet (AIN-93 diet) contained 2.1 mg folicacid/kg of food. The folate-deficient and folate-supplementeddiets were based on the AIN-93 diet but contained, respective-ly, 0.1 and 3.5 mg folic acid/kg. The folate-supplemented dietadded 1.4 mg folic acid/kg diet compared to the folate-normaldiet, which in rats is equivalent to consumption of a 400-μgfolic acid supplement daily on consuming a healthy diet inhuman [14].

Angular venous blood was collected from dams in coagu-lant tubes or potassium-EDTA tubes, respectively, centrifugedat 3000×g for 10 min to obtain serum and plasma, and thenstored at −80 °C. Some of the offspring were sacrificed atPND0; the brain tissue was rapidly collected with liquid ni-trogen flash-freezing and stored at −80 °C for future assay.The other offspring were used for neurobehavioral or ultra-structural examinations. For the latter, 4-month-old offspringwere euthanized with sodium pentobarbital (100 mg/kg, i.p.),left brain tissue was flash-frozen with liquid nitrogen andstored at −80 °C, and right brain tissue was fixed with 2.5%glutaraldehyde for transmission electron microscopy analysis.

Folate and Hcy Assays

Folate concentration was determined by competitive protein-binding assay with automated chemiluminescence system(Immulite 2000 Xpi; Siemens, Berlin, Germany), accordingto the manufacturer’s instructions, in serum samples dilutedfivefold with 0.9% saline. The automated chemiluminescencesystem detected all types of folate with detection limits of 1–24 ng/ml [36]. Hcy was measured in undiluted plasma sam-ples by enzymatic cycling assay using the Auto-ChemistryAnalyzer (DIRUI; Changchun, China), according to

Fig. 5 Flowchart for theintervention with maternal diet

2682 Mol Neurobiol (2018) 55:2676–2684

manufacturer’s instructions [37], with detection limits of 3–50 μmol/l. The homocysteine reagent used was supplied byMedicalSystem (Ningbo, China).

Neurobehavioral Testing

Thirty pups (equal number of female and male) were random-ly selected from each group (3–4 offspring from each dam) atPND4 for five consecutive days of surface righting reflex andnegative geotaxis testing. The surface righting reflex was ex-amined daily from PND4 to PND8 to study sensory-motorfunction in offspring prior to weaning [38]. The pups wereplaced supine on a flat surface, and the time for them to turnover and restore normal prone position was recorded.

The negative geotaxis test was also performed daily fromPND4 to PND8 to assess sensory-motor function as the orientingresponse andmovement expressed in opposition to clues of grav-itational vector [15]. The pups were placed facing downwardwith their hindlimbs in the middle of an apparatus, which wasa rough flat timber surface inclined at 30° angle. The time takento turn 180° and begin to climb upward was recorded.

TheMWM test was conducted on adolescent (PND45) andadult (PND90) offspring in a tank measuring 1.5 m in diam-eter and 50 cm in height. The tank contained 30 cm of waterand was divided into four equal quadrants. The test had aspatial acquisition phase and a subsequent spatial probe phase.In the spatial acquisition phase, an escape platform (10 cm indiameter) was located in the north-east quadrant (targetedquadrant) of the tank and submerged 1.5 cm below the watersurface. The rats were trained for 2 days as the tank was filledwith water and the platform was visible. From the third to fifthday, black ink was poured into the water to make the platforminvisible. Each rat was given 90 s to reach the platform, timewas recorded as escape latency, and the rat remained 15 s atthe platform for memory strengthening. If the rat could notfind the platform within 90 s, then it was gently placed onplatform for 15 s, and escape latency was recorded as 90 s.In this phase of the MWM, the rats were released at the centerof the pool wall in each quadrants and escape latency wasanalyzed by the average of four quadrants as a standard per-formance measurement. In the spatial probe phase of theMWM, the platform was removed from the tank and the ratswere released from the southwest quadrant only. The timespent in the targeted quadrant and the number of crossingsof the place where platform previously had been located wererecorded [17].

Transmission Electron Microscopy

Samples of hippocampal tissue were collected from 4-month-old male offspring of the DD and FD-L groups (n = 2/group)for ultrastructural examination. The samples were cut into 2-mm pieces and fixed with 2.5% glutaraldehyde at 4 °C

overnight and osmium tetraoxide for 2 h, dehydrated in gradedacetone series, and embedded in araldite and dodecenylsuccinic anhydride mixture for 48 h at 65 °C. Ultrathin sec-tions (50–70 nm) were cut with an ultramicrotome, stainedwith uranyl acetate and lead citrate, and imaged with a trans-mission electron microscope (HT7700; Hitachi, Tokyo,Japan).

Statistical Analysis

The data were expressed as proportion, mean ± SD or P50(P25, P75) according to the data distribution. Comparisonsbetween different groups were performed by one-wayANOVA or two-tailed Student’s t test (for normally distribut-ed data), Mann-Whitney test (for skewed distribution), andPearson chi-square or Fisher’s exact test (for frequencies anal-ysis). Repeated measures data were compared using the re-peated measures ANOVA. The statistical software packageSPSS 19.0 was used to evaluate differences within groups,which were considered statistically significant at P < 0.05.

Acknowledgments This research was supported by a grant from theNational Natural Science Foundation of China (Nos. 81472967 and81602849).

Compliance with Ethical Standards Tianjin Medical UniversityAnimal Ethics Committee approved the experimental protocols of thisstudy (TMUaEC2015001).

Conflict of Interest The authors declare that they have no conflicts ofinterest.

References

1. Safi J, Joyeux L, Chalouhi GE (2012) Periconceptional folate defi-ciency and implications in neural tube defects. J Pregnancy 2012:295083. doi:10.1155/2012/295083

2. Wallingford JB, Niswander LA, Shaw GM, Finnell RH (2013) Thecontinuing challenge of understanding, preventing, and treatingneural tube defects. Science 339:1222002. doi:10.1126/science.1222002

3. Gueant JL, Namour F, Gueant-Rodriguez RM, Daval JL (2013)Folate and fetal programming: a play in epigenomics? TrendsEndocrinol Metab 24:279–289. doi:10.1016/j.tem.2013.01.010

4. Blaise SA, Nedelec E, Schroeder H, Alberto JM, Bossenmeyer-Pourie C, Gueant JL, Daval JL (2007) Gestational vitamin B defi-ciency leads to homocysteine-associated brain apoptosis and altersneurobehavioral development in rats. Am J Pathol 170:667–679.doi:10.2353/ajpath.2007.060339

5. Craciunescu CN, Brown EC, Mar MH, Albright CD, Nadeau MR,Zeisel SH (2004) Folic acid deficiency during late gestation de-creases progenitor cell proliferation and increases apoptosis in fetalmouse brain. J Nutr 134:162–166

6. Jadavji NM, Deng L, Malysheva O, Caudill MA, Rozen R (2015)MTHFR deficiency or reduced intake of folate or choline in preg-nant mice results in impaired short-term memory and increasedapoptosis in the hippocampus of wild-type offspring.Neuroscience 300:1–9. doi:10.1016/j.neuroscience.2015.04.067

Mol Neurobiol (2018) 55:2676–2684 2683

7. Jadavji NM, Deng L, Leclerc D, Malysheva O, Bedell BJ, CaudillMA, Rozen R (2012) Severe methylenetetrahydrofolate reductasedeficiency in mice results in behavioral anomalies with morpholog-ical and biochemical changes in hippocampus. Mol Genet Metab106:149–159. doi:10.1016/j.ymgme.2012.03.020

8. Czeizel AE, Dudas I (1992) Prevention of the first occurrence ofneural-tube defects by periconceptional vitamin supplementation. NEngl J Med 327:1832–1835. doi:10.1056/nejm199212243272602

9. McGarel C, Pentieva K, Strain JJ, McNulty H (2015) Emergingroles for folate and related B-vitamins in brain health across thel i f e cyc l e . P r o c Nu t r Soc 74 : 46–55 . do i : 10 . 1017 /S0029665114001554

10. Centers for Disease Control and Prevention (2004) Spina bifida andanencephaly before and after folic acid mandate—United States,1995-1996 and 1999-2000. MMWR Morb Mortal Wkly Rep 53:362–365

11. De Wals P, Tairou F, Van Allen MI, Soo-Hong U, Lowry RB,Sibbald B, Evans JA, Van den Hof MC et al (2007) Reduction inneural-tube defects after folic acid fortification in Canada. N Engl JMed 357:135–142. doi:10.1056/NEJMoa067103

12. Zhao Y, Hao L, Zhang L, Tian Y, Cao Y, Xia H, Deng Y, Wang Tet al (2009) Plasma folate status and dietary folate intake amongChinese women of childbearing age. Matern Child Nutr 5:104–116.doi:10.1111/j.1740-8709.2008.00172.x

13. Chitayat D, Matsui D, Amitai Y, Kennedy D, Vohra S, Rieder M,Koren G (2016) Folic acid supplementation for pregnant womenand those planning pregnancy: 2015 update. J Clin Pharmacol 56:170–175. doi:10.1002/jcph.616

14. Gomes S, Lopes C, Pinto E (2016) Folate and folic acid in thepericonceptional period: recommendations from official health or-ganizations in thirty-six countries worldwide and WHO. PublicHealth Nutr 19:176–189. doi:10.1017/S1368980015000555

15. Fitt ing S, Booze RM, Mactutus CF (2007) Neonatalintrahippocampal gp120 injection: an examination early in devel-opment. Neurotoxicology 28:101–107. doi:10.1016/j.neuro.2006.07.014

16. Vorhees CV, Williams MT (2014) Value of water mazes forassessing spatial and egocentric learning and memory in rodentbasic research and regulatory studies. Neurotoxicol Teratol 45:75–90. doi:10.1016/j.ntt.2014.07.003

17. Liu Z, Jiang M, Kang T, Miao D, Gu G, Song Q, Yao L, Hu Q et al(2013) Lactoferrin-modified PEG-co-PCL nanoparticles for en-hanced brain delivery of NAP peptide following intranasal admin-istration. Biomaterials 34:3870–3881. doi:10.1016/j.biomaterials.2013.02.003

18. Tiwari SK, Agarwal S, Chauhan LK, Mishra VN, Chaturvedi RK(2015) Bisphenol-A impairs myelination potential during develop-ment in the hippocampus of the rat brain. Mol Neurobiol 51:1395–1416. doi:10.1007/s12035-014-8817-3

19. Benton D, ILSI Europe a.i.s.b.l. (2008) Micronutrient status, cog-nition and behavioral problems in childhood. Eur J Nutr 47(Suppl3):38–50. doi:10.1007/s00394-008-3004-9

20. Bhatia P, Singh N (2015) Homocysteine excess: delineating thepossible mechanism of neurotoxicity and depression. FundamClin Pharmacol 29:522–528. doi:10.1111/fcp.12145

21. Li D, Pickell L, Liu Y, Wu Q, Cohn JS, Rozen R (2005) Maternalmethylenetetrahydrofolate reductase deficiency and low dietary fo-late lead to adverse reproductive outcomes and congenital heartdefects in mice. Am J Clin Nutr 82:188–195

22. Obeid R, Herrmann W (2006) Mechanisms of homocysteine neu-rotoxicity in neurodegenerative diseases with special reference todementia. FEBS Lett 580:2994–3005. doi:10.1016/j.febslet.2006.04.088

23. Dhobale M, Joshi S (2012) Altered maternal micronutrients (folicacid, vitamin B(12)) and omega 3 fatty acids through oxidative

stress may reduce neurotrophic factors in preterm pregnancy. JMatern Fetal Neonatal Med 25:317–323. doi:10.3109/14767058.2011.579209

24. Kalani A, Kamat PK, Givvimani S, Brown K, Metreveli N, TyagiSC, Tyagi N (2014) Nutri-epigenetics ameliorates blood-brain bar-rier damage and neurodegeneration in hyperhomocysteinemia: roleof folic acid. J Mol Neurosci 52:202–215. doi:10.1007/s12031-013-0122-5

25. Abbott NJ (2013) Blood-brain barrier structure and function and thechallenges for CNS drug delivery. J Inherit Metab Dis 36:437–449.doi:10.1007/s10545-013-9608-0

26. Fan LW, Chen RF, Mitchell HJ, Lin RC, Simpson KL, Rhodes PG,Cai Z (2008) Alpha-phenyl-n-tert-butyl-nitrone attenuateslipopolysaccharide-induced brain injury and improves neurologicalreflexes and early sensorimotor behavioral performance in juvenilerats. J Neurosci Res 86:3536–3547. doi:10.1002/jnr.21812

27. Franco-Pons N, Torrente M, Colomina MT, Vilella E (2007)Behavioral deficits in the cuprizone-induced murine model of de-myelination/remyelination. Toxicol Lett 169:205–213. doi:10.1016/j.toxlet.2007.01.010

28. Yang HY, Liu Y, Xie JC, Liu NN, Tian X (2015) Effects of repet-itive transcranial magnetic stimulation on synaptic plasticity andapoptosis in vascular dementia rats. Behav Brain Res 281:149–155. doi:10.1016/j.bbr.2014.12.037

29. Reeves PG, Nielsen FH, Fahey GC Jr (1993) AIN-93 purified dietsfor laboratory rodents: final report of the American Institute ofNutrition ad hoc writing committee on the reformulation of theAIN-76A rodent diet. J Nutr 123:1939–1951

30. Dwyer JB, McQuown SC, Leslie FM (2009) The dynamic effectsof nicotine on the developing brain. Pharmacol Ther 122:125–139.doi:10.1016/j.pharmthera.2009.02.003

31. Burdge GC, Lillycrop KA (2014) Environment-physiology, dietquality and energy balance: the influence of early life nutrition onfuture energy balance. Physiol Behav 134:119–122. doi:10.1016/j.physbeh.2013.12.007

32. Godfrey KM, Lillycrop KA, Burdge GC, Gluckman PD, HansonMA (2007) Epigenetic mechanisms and the mismatch concept ofthe developmental origins of health and disease. Pediatr Res 61:5R–10R. doi:10.1203/pdr.0b013e318045bedb

33. Irwin RE, Pentieva K, Cassidy T, Lees-Murdock DJ, McLaughlinM, Prasad G, McNulty H, Walsh CP (2016) The interplay betweenDNA methylation, folate and neurocognitive development.Epigenomics 8:863–879. doi:10.2217/epi-2016-0003

34. Subbanna S, Nagre NN, Shivakumar M, Basavarajappa BS (2016)A single day of 5-azacytidine exposure during development inducesneurodegeneration in neonatal mice and neurobehavioral deficits inadult mice. Physiol Behav 167:16–27. doi:10.1016/j.physbeh.2016.08.036

35. Barua S, Kuizon S, Brown WT, Junaid MA (2016) DNA methyla-tion profiling at single-base resolution reveals gestational folic acidsupplementation influences the epigenome of mouse offspring cer-ebellum. Front Neurosci 10:168. doi:10.3389/fnins.2016.00168

36. Li W, Liu H, Yu M, Zhang X, Zhang M, Wilson JX, Huang G(2015) Folic acid administration inhibits amyloid beta-peptide ac-cumulation in APP/PS1 transgenic mice. J Nutr Biochem 26:883–891. doi:10.1016/j.jnutbio.2015.03.009

37. Song X, Fan X, Li X, Kennedy D, Pang L, QuanM, Chen X, Gao Jet al (2014) Serum levels of BDNF, folate and homocysteine: inrelation to hippocampal volume and psychopathology in drug na-ive, first episode schizophrenia. Schizophr Res 159:51–55. doi:10.1016/j.schres.2014.07.033

38. Mesquita AR, Pego JM, Summavielle T, Maciel P, Almeida OF,Sousa N (2007) Neurodevelopment milestone abnormalities in ratsexposed to stress in early life. Neuroscience 147:1022–1033. doi:10.1016/j.neuroscience.2007.04.007

2684 Mol Neurobiol (2018) 55:2676–2684