Broccoli sprout supplementation during pregnancy prevents brain injury in the newborn rat following placental insufficiency

Behavioural Brain Research 291 (2015) 289–298

Contents lists available at ScienceDirect

Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr

Research report

Broccoli sprout supplementation during pregnancy prevents brain injury in the newborn rat following placental insufficiency A.M. Black a , E.A. Armstrong a , O. Scott a , B.J.H. Juurlink b , J.Y. Yager a,∗ a b

Pediatric Neurosciences, Department of Pediatrics and Centre for Neuroscience, University of Alberta, Edmonton, Alberta, Canada Department of Anatomy and Cell Biology, University of Saskatchewan, Saskatchewan, Canada

h i g h l i g h t s • • • •

Placental insufficiency causes fetal growth restriction (FGR). FGR is associated with developmental disability and increased cerebral palsy (CP). Broccoli sprouts is a natural health product and a potent phase II enzyme inducer. Broccoli sprouts prevents FGR induced behavioral and pathologic alterations.

a r t i c l e

i n f o

Article history: Received 23 October 2014 Received in revised form 13 May 2015 Accepted 16 May 2015 Available online 23 May 2015 Keywords: Intrauterine growth restriction Cerebral palsy Broccoli sprout Developmental delay Preventative therapy

a b s t r a c t Chronic placental insufficiency and subsequent intrauterine growth restriction (IUGR) increase the risk of hypoxic-ischemic encephalopathy in the newborn by 40 fold. The latter, in turn, increases the risk of cerebral palsy and developmental disabilities. This study seeks to determine the effectiveness of broccoli sprouts (BrSp), a rich source of the isothiocyanate sulforaphane, as a neuroprotectant in a rat model of chronic placental insufficiency and IUGR. Placental insufficiency and IUGR was induced by bilateral uterine artery ligation (BUAL) on day E20 of gestation. Dams were fed standard chow or chow supplemented with 200 mg of dried BrSp from E15 – postnatal day 14 (PD14). Controls received Sham surgery and the same dietary regime. Pups underwent neurologic reflex testing and open field testing, following which they were euthanized and their brains frozen for neuropathologic assessment. Compared to Sham, IUGR pups were delayed in attaining early reflexes and performed worse in the open field, both of which were significantly improved by maternal supplementation of BrSp (p < 0.05). Neuropathology revealed diminished white matter, ventricular dilation, astrogliosis and reduction in hippocampal neurons in IUGR animals compared to Sham, whereas broccoli sprout supplementation improved outcome in all histological assessments (p < 0.05). Maternal dietary supplementation with BrSp prevented the detrimental neurocognitive and neuropathologic effects of chronic intrauterine ischemia. These findings suggest a novel approach for prevention of cerebral palsy and/or developmental disabilities associated with placental insufficiency. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Neonatal hypoxic-ischemic encephalopathy is a leading cause of neurodevelopmental disabilities, inclusive of cerebral palsy and

Abbreviations: BrSp, broccoli sprout; BUAL, bilateral uterine artery ligation; IUGR, intra-uterine growth restriction; GFAP, glial fibrillary acidic protein; MBP, myelin basic protein; DTC, dithiocarbamate; ITC, isothiocyanate. ∗ Corresponding author at: Pediatric Neurosciences, Department of Pediatrics, University of Alberta, 3-469 Edmonton Clinic Health Academy, 11405 87 Avenue, Edmonton, AB T6G 1C9, Canada. Tel.: +1 780 964 1161; fax: +1 8883531163. E-mail address: [email protected] (J.Y. Yager). http://dx.doi.org/10.1016/j.bbr.2015.05.033 0166-4328/© 2015 Elsevier B.V. All rights reserved.

mental retardation [1]. Placental insufficiency resulting in growth restriction of the fetus is a significant predisposing ante-partum risk factor, shown to increase the probability of hypoxic-ischemic encephalopathy 40-fold [2]. Evidence indicates that up to 90% of cases of cerebral palsy and developmental disability occur prior to birth [2–4]. This, combined with the toxicity of conventional medications to the immature brain [5,6], highlights the need for neuroprotective strategies that are safe, efficacious and preventive in nature. Broccoli sprouts are a rich source of sulforaphane, a potent isothiocyanate shown to be protective in models of oxidative stress and inflammation [7–9]. Our study tested the effectiveness of this natural product in a rodent model of placental insufficiency.

290

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

We asked whether maternal supplementation with BrSp during gestation and the early newborn period reduces adverse neurodevelopmental effects associated with placental insufficiency, and whether improvement in neuropathologic outcome would coincide with functional improvement.

2. Materials and methods 2.1. Animals Pregnant female Long–Evans rats (Charles River Laboratories) were used in this study (n = 17). Animals were housed in the Health Sciences Laboratory Animal facility at the University of Alberta. Following a 5-day acclimatization period, animals were timedpregnant whereby a vaginal lavage containing sperm denoted day one of gestation (E1). Pregnant dams were then assigned to one of the following four groups: (1) Sham-operated (Sham); (2) Shamoperated with broccoli sprout supplement (Sham + B); (3) Intra Uterine Growth Restriction (IUGR); (4) IUGR with broccoli sprout supplement (IUGR + B). Rat pups were born spontaneously and reared with their dams. Litters were culled to 8 where more than eight pups were born. Only litters of at least 5 pups were included in the study. Three to six pups from each litter were used to represent each of the four experimental groups throughout the study (n = 88 pups). All animals were maintained on a 12 h light/dark schedule and received food and water ad libitum throughout the study. The Health Sciences Animal Care and Use Committee at the University of Alberta approved all procedures. 2.2. Surgical procedure Bilateral uterine artery ligation (BUAL), adapted from Wigglesworth’s original model of fetal growth restriction, was performed on E20, of a 23 day gestation, to induce chronic placental insufficiency [10]. This stage of brain development for the fetal rat is comparable to human brain development at approximately 22–26 weeks gestational age [11,12]. Pregnant rats were anesthetized with 4% isoflurane (approximately 2% maintenance) in 21% oxygen and received a vertical, low midline abdominal incision, 2–3 cm long. Both uterine arteries, proximal to the uterine bifurcation, were permanently ligated with 4–0 Vicryl coated suture (Ethicon Inc.,Somerville, NJ, USA). After suturing the muscular layer with Vicryl, 0.05 ml of bupivicaine (Sensorcaine by AstraZeneca Can Inc., Ontario, Canada) was administered in a drop wise fashion for analgesia and the skin layer was closed with 5–0 silk suture (Angiotech Surgical Specialties Corp., Reading, PA, USA). Following surgery, the animals were monitored over a 4–6 h to ensure full recovery. Sham-operated rats underwent identical anesthetic and surgical procedures with the exception of BUAL. 2.3. Intrauterine growth restriction and cephalization index assessment We used the IUGR definition previously described by Olivier et al., namely – birth weight ≤ 2 SD below the mean [13]. Birth weights were collected from four naïve litters (n = 56) to ascertain the mean birth weight for our breeding colony and define the criteria for IUGR. The mean birth weight of the naïve animals was 6.28 ± 0.38 g, therefore animals ≤ 5.52 g, at birth, were considered growth restricted. In this study, IUGR rats were rats born to BUAL operated females, with a birth weight ≤ 5.52 g. Sham rats were rats born to sham operated females, with a birth weight > 5.52 g.

On the day of delivery (PD1), newborn rat pups were counted, weighed, sexed and returned to their dam for 48 h to ensure attachment. The cephalization index (CI = head circumference/birth weight) was calculated for individual litters of Sham (n = 10, 5 female and 5 male) and IUGR (n = 14, 7 female and 7 male) as described by Bassan et al. [14], to ensure the neurological significance of the BUAL model. Animals were chosen randomly from all BUAL and Sham litters in the study. 2.4. Broccoli sprout supplementation Broccoli sprouts were prepared as previously described [15]. The appropriate groups received 200 mg/day dried broccoli sprouts as a supplement to their regular chow, in a separate dish, from E15 (beginning of the third trimester) to PD14. Animals were then observed to be sure that they consumed all of the BrSps. Any animal that did not eat BrSp for 2 days was excluded from the study. 2.5. Analysis of isothiocyanates/dithiocarbamates (ITC/DTC) in fetus In order to determine the bioavailability of the sulforaphane to the fetus, additional pregnant dams in their third trimester (n = 6) were intraperitoneally injected with either 50 or 500 ␮g of sulforaphane in 500 ␮l normal saline, or were fed 200 mg of dried broccoli sprouts for two days. Rats were euthanized 1 h after the sulforaphane injection or consumption of broccoli sprouts. Samples of whole fetuses were collected rapidly from uterine horns, frozen on liquid nitrogen, and stored at −80 ◦ C prior to analysis. Assessment of sulforaphane activity is performed via measurement of its collective metabolites, dithiocarbamates (DTC), by a cyclocondensation reaction that has been previously described [16,17]. The cyclocondensation product (1,3-benzodithiole-2-thione) was detected by a Waters Model 996 photodiode array detector at 365 nm. The DTC level of the tissue was adjusted to tissue weight. 2.6. Evaluation of neurobehavioral development and maturation Testing began on PD3 and continued daily through to PD21 to assess the emergence of reflexes, maturation and other sensorimotor behaviors. Rats were allowed to acclimatize to the testing environment for approximately 1 h. To avoid temperature effect, newborn rats were tested in an incubator maintained at 34.5 ◦ C where possible, or under a warm lamp (31 ◦ C) for tests that could not be performed inside the incubator. The following is a brief description of the reflex tests used, which were adapted from Fox and Lubics et al. [18,19]. 1. Righting reflex: An animal positioned on its back will turn over to rest in a normal prone position. 2. Grasp reflex: Palms are stroked with a blunt instrument and the limb flexes to grasp the instrument. The day each pup could grasp the instrument with both fore or hind limbs was the postnatal day of attainment for that reflex. 3. Hind limb placing response: While suspending the pup, the dorsal side of one hind paw is touched to the edge of a flat surface causing the foot to be raised and be placed on the surface. The day each pup performed the placing task with both hind limbs was the day of attainment. 4. Cliff aversion: When placed at the edge of a flat surface with forepaws and head over the edge, the rat it will turn and crawl away from the edge. We scored this test 0 (pup completely disregards cliff), 1 (hesitation or struggle) or 2 (pup immediately

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

averts cliff) according to each pup’s ability. The postnatal day each pup scored a 2 was the day of attainment. 5. Gait: Pups were placed in the center of a white paper circle,15 cm in diameter. The day they began to move off the circle with both forelimbs within 30 s, was recorded. 6. Acceleration righting: Pups were suspended upside-down by fore and hind limbs and released, falling 12 inches into a container lined with foam to cushion the fall. The successful animal will turn and right itself mid-air in order to land on all fours. A partial attempt to right was given a score of 1 and complete acceleration righting was given a score of 2. The day each pup scored a 2 was considered the day of attainment.

291

Canada) was performed to examine the development of myelinated white fibers. Frozen brain sections were post-fixed in formalin (GFAP) or methanol (MBP), cleared and dehydrated in a graded series of ethanol washes, and then washed with 1% hydrogen peroxide to quench endogenous peroxidases. They were subsequently blocked with normal horse serum mixed with Triton X-100 (0.1%), and allowed to incubate with the primary antibody overnight. After rinsing and 30 min of incubation with the secondary antibody, the sections were rinsed again and incubated with an avidin–biotin complex (ABC, Vector Laboratories Inc., Burlingame, CA, USA). The immunoreactivity was visualized with diaminobenzedine tetrahydrochloride (DAB, Vector Laboratories Inc.).

2.7. Signs of maturation 2.12. GFAP cell counting 1. Auditory startle/eye opening: A loud sharp noise will cause a rat to startle when auditory processing emerges. The first day a clapping sound produced an observable startle response, defined as a whole body twitch, was considered the day of attainment [20]. The first postnatal day both eyes were opened was recorded for eye opening maturation. 2. Posture: Beginning on PD12, rat pups were observed for stance and limb placement. The first day the pups could maintain a stance where all four limbs were held comfortably beneath their body during both rest and locomotion was considered attainment day of a mature posture [18,21]. 2.8. Open field behavior On PD21, animals were video-monitored for 5 min in an open field to observe various motor and exploratory behaviors. Pups were placed in the center of a 45 cm × 45 cm × 30 cm Plexiglas box, the floor of which was divided into 4 cm × 4 cm areas. Parameters measured include: ambulation (number of squares crossed), rearing (both forepaws leave the ground), grooming behavior, head lifting (lifting the head and neck in an exploratory manner) and defecation [19,22]. 2.9. Histopathology On PD21, rat pups were euthanized with 5% isoflurane in 30% oxygen and balanced nitrogen. They were then decapitated, their brains were rapidly removed, flash frozen in iso-pentane, and stored at −80 C◦ . Coronal brain sections (10 ␮m) were cut on a cryostat (Leica, cryocut 1800) at −17 C◦ at the level of the anterior commissure, mamillary bodies, and through the midbrain. Sections were mounted on slides and stored at −20 ◦ C. They were subsequently were used for immuno-histochemistry or stained with hematoxylin and eosin (H&E) for cell counting and cortical measurements. Animals used for histopathology were chosen randomly from all litters used for reflex testing. 2.10. Hippocampal cell counting Sample sizes of at least 8 brains were sectioned per experimental group for all measures in the study. H&E stained tissue sections were used to assess the number of healthy-looking hippocampal pyramidal neurons in the CA1 and CA3 sectors. The method used to count these areas is described elsewhere [23]. 2.11. Immunohistochemistry Brain tissue sections were stained with glial fibrillary acidic protein (GFAP) (Z0334; DakoCytomation, Glostrup, Denmark) to investigate the astrocytic response to injury. Staining with myelin basic protein (MBP) (SMI 94; Cedarlane Laboratories Ltd., Ont.,

Images of GFAP stained coronal sections were taken at 400× magnification in the areas of the corpus callosum and cingulum with a Spot Flex Camera (Diagnostic Instruments, Sterling Heights MI) attached to a Leica ATC 2000 microscope (Leica, Buffalo, NY) and Spot 4.5 software (Diagnostic Instruments, Sterling Heights MI). For the corpus callosum, images were taken on either side of the midline, alternating between left and right hemispheres. For the cingulum, the tip of the cingulate peak was positioned to the lowest point of view in the microscope field so the image would reflect a portion of the cingular white matter projections just superior to the peak, alternating hemispheres between sections. Reactive astrocytes denoted by their enlarged cell bodies and highlighted processes were marked on a grid overlay of the collected images using Adobe Photoshop CS2 version 9.0.2. 2.13. White matter thickness, ventricle area and densitometry Whole brain images of MBP-stained sections were taken with a Spot Flex Camera (Diagnostic Instruments, Sterling Heights MI) attached to a Leica GZ6E stereoscope (Leica Microsystems, Richmond Hill, ON, Canada) and Spot 4.5 software (Diagnostic Instruments, Sterling Heights, MI). All measurements were obtained using Image J version 1.41 calibrated with a Kodak No. 3 Calibrated Step Tablet and scanned with an Epson Expression 1680 Professional scanner. Both anterior and posterior sections were measured for corpus callosum thickness and cingulate peak thickness. Measurements for the corpus callosum were taken at the midline of each section and cingulum thickness was measured from the most visibly superior fibers in the peak to the most inferior of the corpus callosum in both hemispheres of each section and averaged for each animal. Ventricular area assessment and densitometry of myelin staining were also performed on the MBP-stained sections. For densitometry, a 0.035 mm2 area of the corpus callosum immediately beside the midline was measured. The same area of the dorsal cortex beside the midline that did not show MBP reactivity was also measured to determine background staining levels. This background level was then subtracted from the optical density (OD) of the corpus callosum, in the same section. 2.14. Statistical analyses Where possible, an experimenter blinded to the treatment groups scored the behavioral testing; however, due to the observable appearance of growth restricted animals, this was not always feasible. The video-monitored tests of righting, acceleration righting and open field were scored by a blind experimenter. Between-group comparisons were analyzed with SPSS Statistics using the student t-test, one-way Analysis of Variance (ANOVA), two-way ANOVA and the Levene test for equality of error variances

292

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

or homogeneity of variances. Tukey HSD or Games-Howell post hoc tests were used where necessary (SPSS Statistics 17.0; SPSS Inc., Chicago, IL, USA). Since most of the data gathered from this study violates the equal variance assumption for two-way ANOVA, we used a oneway ANOVA and the Games-Howell post hoc tests in addition to two-way ANOVA statistics. Data are expressed as mean ± standard deviation (SD). A p value of ≤0.05 was considered to be statistically significant. 3. Results The total number of animals utilized in testing was 88 in all groups combined. This represented animals from 15 litters. The maximum number of animals in each litter prior to culling was 17, with a mean number of 10 pups. Each litter was culled to 8, and only 6 pups were unable to be used due to size (too small or too big). The breakdown of males was 43, and females were 45. No differences in outcomes were found between sexes. In addition, Table 1 includes all animals utilized, as well as a determination of why pups were excluded, or those that died following birth. 3.1. ITC/DTC analysis In fetuses of dams that received 50 or 500 ␮g intraperitoneal injections of sulforaphane, mean DTC concentration was 3.72 ± 0.73 and 27.0 ± 8.31 pmol/mg, respectively. Mean DTC concentration in fetuses of dams fed 200 mg broccoli sprouts was 31.9 ± 4.64 pmol/mg. These findings indicate that 200 mg of broccoli sprouts fed to pregnant dams, provide the equivalent of approximately 500 ␮g of sulforaphane. 3.2. Intrauterine growth restriction and cephalization index assessment Bilateral uterine artery ligation (BUAL) resulted consistently resulted in intrauterine growth restriction (IUGR). IUGR was determined to be those rat pups with birth weights less than 2 standard deviations below the mean birth weight of our sham (naïve) animals. The sham newborns used in this comparison, mean birth weight was 6.71 ± 0.12 g; IUGR was therefore determined to be a birth weight of ≤5.52 g. The mean birth weight of our IUGR rat pups was 4.41 ± 0.20 g, significantly less than sham (p < 0.0001). Broccoli sprout supplementation had no effect of birth weight of the rat pups in any of the four groups tested. With respect to cephalization, the IUGR groups had significantly smaller head circumference (3.42 ± 0.07 cm) compared to the Sham animals (4.17 ± 0.06 cm) (p < 0.0001). Values of CI were significantly

higher in the Sham groups (0.62 ± 0.01 cm/g) than in the IUGR groups (0.79 ± 0.02 cm/g) (p < 0.0001). 3.3. Exclusions and mortality The BUAL surgery was tolerated well by all dams. In order to estimate mortality, a one-way ANOVA with Tukey post hoc was done to compare litter sizes between all four experimental groups, as well as a naïve group of litters born in our colony during the same time period. We found no difference in litter size between Sham, Sham + B and the naïve group (p > 0.05; 13.0 ± 3.4; 13.0 ± 2.9; 14.4 ± 2.2) and no difference between IUGR and IUGR + B litter size (p > 0.05; 6.7 ± 3.6; 6.6 ± 2.6). IUGR and IUGR + B litters were significantly smaller than the sham groups (p < 0.01). Of the 10 IUGR/IUGR + B litters used in this study (54 pups), 2 animals were excluded for being above the determined weight for IUGR. 3.4. Neurobehavioral/reflex and maturation assessment Two-way ANOVAs showed a main effect of treatment in every test we used to evaluate the pups. A main effect of diet was observed in forelimb grasping, hind limb placing and cliff aversion and an interaction was observed in the forelimb grasping and posture tests (Table 1). Similar to the weight data, with the exception of gait, all tests were significant for the Levene statistic (overall, p ≤ 0.043; data not shown). We therefore compared means using one-way ANOVA and the Games-Howell post hoc test where an effect of diet or an interaction was observed. Forelimb grasp reflex: There was no difference between Sham and Sham + B groups (p = 0.914), a significant difference in performance between IUGR and all other groups (p < 0.0001), and no difference between IUGR + B and either Sham or Sham + B (p ≥ 0.174) (Fig. 1A). Hind limb placing response: IUGR animals were delayed compared to Sham (p < 0.0001), and IUGR + B also showed poorer performance than Sham groups (p ≤ 0.029). Although animal behavior was not returned to Sham level in this test, IUGR + B animals were still significantly improved over IUGR (p = 0.009) (Fig. 1B). Cliff aversion: IUGR animals were delayed compared to both Sham groups (p ≤ 0.017), and IUGR + B animals were improved over IUGR (p < 0.0001); IUGR + B were no different from Sham + B (p = 0.174) and significantly different from the Sham group (p < 0.0001) (Fig. 1C). Posture: Attainment of mature posture was delayed for IUGR animals compared to Sham and Sham + B (p < 0.0001). No difference was recorded between Sham and Sham + B groups (p = 0.972). Improved attainment of normal posture for the IUGR + B animals

Table 1 Results of early reflex behavior and maturation tests. Test

n (male, female) Righting PD7a Forelimb graspb Hind limb graspb Hind limb placingb Cliff aversionb Gaitb Postureb Eye openingb Auditoryb Acceleration-rightingb a b

In seconds. Postnatal day of appearance.

Results (mean ± SD)

Main effects (p value)

SHAM

SHAM + B

IUGR

IUGR + B

Treatment

Diet

Interaction

16 (7M, 9F) 1.3 ± 0.4 2.9 ± 0.3 4.3 ± 0.9 5.5 ± 1.2 6.6 ± 0.7 7.8 ± 1.1 14.3 ± 0.6 16.2 ± 0.5 11.9 ± 0.7 15.3 ± 1.0

20 (10M, 10F) 1.3 ± 0.3 3.0 ± 0.3 3.8 ± 0.8 5.1 ± 0.9 5.0 ± 1.0 7.1 ± 1.0 14.4 ± 0.6 16.4 ± 0.6 12.4 ± 0.8 15.6 ± 0.9

27 (14M, 13F) 5.2 ± 4.8 4.6 ± 1.0 6.8 ± 2.2 8.1 ± 1.4 7.7 ± 1.5 8.9 ± 1.3 16.5 ± 0.9 16.6 ± 0.9 13.7 ± 1.2 18.0 ± 2.1

25 (12M, 13F) 4.6 ± 4.8 3.3 ± 0.9 6.1 ± 2.1 6.8 ± 1.5 5.5 ± 0.9 8.8 ± 1.1 15.7 ± 0.9 16.7 ± 0.6 14.0 ± 0.7 17.2 ± 1.0

0.000 0.000 0.000 0.000 0.002 0.000 0.000 0.028 0.000 0.000

0.715 0.000 0.124 0.002 0.000 0.121 0.067 0.360 0.094 0.413

0.708 0.000 0.848 0.118 0.291 0.229 0.024 0.903 0.547 0.085

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

A

†

10

* 4

2

PD of Appearance

PD of Appearance

6

0 SHAM+B

C

IUGR

†* 8 6 4 2

SHAM

IUGR+B

†

8

20

*

* 6 4 2 0

PD of Appearance

PD of Appearance

†

0 SHAM

10

B

293

SHAM+B

D

IUGR

†

IUGR+B

†*

15

10

5

0 SHAM

SHAM+B

IUGR

IUGR+B

SHAM

SHAM+B

IUGR

IUGR+B

Fig. 1. Reflex test performed from our neurologic/maturation test battery: (A) forelimb grasp; (B) hind limb placing; (C) cliff aversion; and (D) posture. Results are expressed as mean ± SD. † Significantly different from Sham groups; *Significantly different from IUGR.

over IUGR was noted (p = 0.037); however, IUGR + B were still delayed compared to the Sham groups (p < 0.0001) (Fig. 1D). 3.5. Open field behavior Two-way ANOVA of open field data showed a main effect of treatment in ambulation, head lifts and defecation, a main effect of diet in head lifts only, and an interaction in ambulation, head lifts and grooming (Table 2). A comparison of means by one-way ANOVA and Tukey HSD or Games-Howell post hoc tests were done where appropriate, and showed IUGR animals were significantly more active than Sham groups in ambulation (p ≤ 0.002) with no difference in activity between the Sham and Sham + B groups (p = 0.105). Broccolitreated IUGR animals performed significantly better than IUGR animals with respect to ambulation (p = 0.001) and were no different from Sham (p ≥ 0.115) (Fig. 2A). Similarly for head-lifting behavior, IUGR animals were overactive in comparison to Sham groups (p < 0.0001). Broccoli-treated IUGR animals were less active than IUGR (p = 0.027) and were no different than Sham (p = 0.056) (Fig. 2B). Two-way ANOVA shows a significant interaction for grooming. One-way comparison showed no differences between any of the groups for grooming (p ≥ 0.194). Likewise, defecation showed a main effect of treatment, but no differences with one-way comparisons (p ≥ 0.064). This discrepancy is probably due to the very low occurrence of grooming or defecation within any of the groups.

p = 0.973). IUGR + B had significantly more cells in CA1 than IUGR (85.2 ± 8.2, n = 13; 70.3 ± 5.4, n = 13; p < 0.0001); however IUGR + B counts did not reach the level of Sham animals (p < 0.0001) (Fig. 3A and Fig. 4). CA3 cell counts resulted in a main effect of treatment (p < 0.0001), no effect of diet (p = 0.169), but a significant interaction (p < 0.03). Again, one-way ANOVA comparisons showed no difference between Sham groups (152.8 ± 19.5, 148.7 ± 13.3; p = 0.946). IUGR + B pups had significantly more cells than IUGR (134.7 ± 11.8, 117.2 ± 17; p = 0.029) and were not statistically different from Sham (p = 0.057) (Fig. 3B). 3.7. GFAP cell counting For the corpus callosum, two-way ANOVA resulted in a main effect of treatment, a main effect of diet and a significant interaction (p < 0.0001, p < 0.0001 and p = 0.001, respectively, Fig. 5A). Mean cell counts were 11.8 ± 3.0, 10.1 ± 2.1, 18.9 ± 1.8 and 11.9 ± 2.4 for Sham, Sham + B, IUGR and IUGR + B respectively. For the cingulum area, two-way ANOVA gave a main effect of treatment and of diet (p < 0.0001, p = 0.002), but no interaction (p = 0.088) (Fig. 5B). Mean counts in the same order as above were 29.4 ± 6.0, 26.8 ± 5.6, 42.1 ± 4.9 and 33.8 ± 4.5. Although the interaction was not significant by two-way, one-way ANOVA shows the IUGR + B group is significantly different from IUGR (p = 0.005). IUGR animals had more reactive astrocytes in both the corpus callosum and cingulum white matter areas than Sham controls. Broccoli sprout-treated IUGR animals had significantly fewer reactive astrocytes in these areas than IUGR alone (Fig. 6).

3.6. Hippocampal cell counting 3.8. White matter thickness, densitometry and ventricle area Two-way ANOVA for CA1 cell counts resulted in a main effect of treatment (p < 0.0001), a main effect of diet (p < 0.0001) and a significant interaction (p < 0.002). A comparison of means by one-way ANOVA and Tukey’s post hoc test showed no difference between Sham and Sham + B groups (99.4 ± 5.6, n = 8; 100.8 ± 5.9, n = 8;

Corpus callosum thickness measurements in the anterior rat brain revealed a main effect of treatment (p < 0.0001), a main effect of diet (p = 0.009) and no interaction (p = 0.318) by two-way ANOVA. Comparison of means by one-way ANOVA and Tukey’s

294

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

Table 2 Results from the open field behavior test. Test

n (male, female) Ambulationa Head liftsb Rearingb Groomingb Defecationb

Main effects (p value)

SHAM

SHAM + B

IUGR

IUGR + B

Treatment

Diet

Interaction

16 (8M, 8F) 67.6 ± 23.8 13.9 ± 4.4 35.3 ± 9.6 1.7 ± 1.4 2.3 ± 1.0

20 (10M, 10F) 87.1 ± 21.6 13.5 ± 3.8 38.4 ± 8.8 0.9 ± 0.8 1.7 ± 1.1

26 (13M, 13F) 115 ± 24.1 24.0 ± 6.8 36.8 ± 9.6 1.0 ± 0.8 1.3 ± 1.4

24 (12M, 12F) 86.0 ± 29.5 18.5 ± 7.2 33.0 ± 12.2 1.3 ± 0.8 1.2 ± 1.6

0.000 0.000 0.393 0.519 0.011

0.390 0.022 0.850 0.182 0.268

0.000 0.045 0.127 0.010 0.447

Squares Crossed (5 min)

Number of squares. Number of times behavior was observed.

150

A

†

35

* 100

50

B

†

30

Head Lifts (5 min)

a b

Results (mean ± SD)

*

25 20 15 10 5 0

0 SHAM

SHAM+B

IUGR

SHAM

IUGR+B

SHAM+B

IUGR

IUGR+B

Fig. 2. Open field measures of activity done on PD 21: (A) ambulation; and (B) head-lifting. † Significantly different from Sham; results are expressed as mean ± SD *Significantly different from IUGR.

post hoc test showed IUGR animals had a significant reduction in corpus callosum compared to Sham (p = 0.001). The IUGR + B group showed significant improvement in thickness compared to IUGR (p = 0.033) and was not different from Sham in this measure (p = 0.494) (Fig. 7A). The posterior corpus callosum thickness measurements gave a main effect of treatment (p < 0.0001), but no effect of diet (p = 0.279) and no interaction (p = 0.14). One-way comparisons show no difference between IUGR + B and control groups (p ≥ 0.096), but also no difference compared to IUGR (p = 0.205) (Fig. 7B). Thickness measurements of the cingulate peak in the anterior rat brain resulted in a main effect of treatment (p < 0.0001), a main effect of diet (p < 0.0001) and a significant interaction (p = 0.001). Mean comparisons by one-way ANOVA and Tukey’s post hoc revealed the cingulate peak thickness of the IUGR group to be reduced significantly compared to Sham controls (p < 0.0001). The IUGR + B measure was significantly improved over IUGR (p < 0.0001) and not different from Sham controls (p = 0.983) (Fig. 8A). Posterior cingulate peak thickness gave a main effect of treatment (p < 0.0001), a main effect of diet (p < 0.0001), but no interaction (p = 0.115). One-way mean comparisons showed a significant reduction in posterior cingulate peak thickness of the IUGR

A

200 †

150

*

100

Cells / 0.09 mm 2

Cells / 0.09 mm 2

200

group compared to controls (p < 0.002). IUGR + B was significantly improved over IUGR (p < 0.0001) and not different from Sham controls (p = 0.999) (Fig. 8B). Densitometry measurements gave a main effect of treatment (p = 0.002), no effect of diet (p = 0.087) and a significant interaction (p = 0.025). A one-way ANOVA with Tukey post hoc showed a significantly weaker MBP signal in IUGR sections compared with Sham (p = 0.001). IUGR + B had a significantly stronger MBP signal compared with IUGR (p = 0.016) and was not different compared to Sham controls (0.721) (Fig. 9). Area measurements of ventricles gave a main effect of treatment (p < 0.0001), no effect of diet (p < 0.071), but a significant interaction (p < 0.004) and significant Levene test (p = 0.003). One-way ANOVA with the Games-Howell post hoc test revealed significantly larger ventricle areas in IUGR sections compared to Sham (p < 0.0001). The BrSp-supplemented IUGR group had significantly smaller ventricle areas than the IUGR group (p = 0.025), although not to the level of Sham (Fig. 10). We also examined cortical thickness and hippocampus lengths to rule out any differences in the size of brain structures that may be due to growth restriction in general. We did not find any differences that would confound the validity of our results. Fig. 11 shows

B †

150

*

100

50 0

50 0

SHAM SHAM+B

IUGR

IUGR+B

SHAM SHAM+B

IUGR

IUGR+B

Fig. 3. Cell counts in CA1 (A) and CA3 (B) of the hippocampus. Results are expressed as mean ± SD, † Significantly different from Sham; *Significantly different from IUGR.

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

295

Fig. 4. CA1 photomicrograph at PD21: (A) Sham; (B) Sham + B; (C) IUGR; and (D) IUGR + B. Pyknotic cells are still visible in IUGR sections, indicated by the black arrows.

A

50

†

20

*

15 10 5

Cells / 0.09 mm 2

Cells / 0.09 mm 2

25

B

†

*

40 30 20 10 0

0 SHAM

SHAM+B

IUGR

IUGR+B

SHAM

SHAM+B

IUGR

IUGR+B

Fig. 5. Cell counts for reactive astrocytes in the (A) corpus callosum and (B) cingular projections at PD21. Results are expressed as mean ± SD, † Significantly different from Sham; *Significantly different from IUGR.

photomicrographs of MBP-stained sections illustrating differences in parameters measured and reported here. 4. Discussion Improved survivability of premature and growth restricted infants has increased awareness of the risk associated with placental insufficiency to the immature brain. In this regard, our study has established that maternal consumption of broccoli sprouts during pregnancy and the first two postnatal weeks, protects against the neurodevelopmental difficulties, both structural and functional, associated with an placental insufficiency and IUGR as a phenotype. From a neuropathological perspective, IUGR animals that received BrSp treatment had thicker white matter bundles, denser myelination and less ventricular dilation than untreated IUGR animals. In most cases the IUGR + B group was no different from Sham controls. Moreover, astrocytic reactivity was significantly reduced

in BrSp supplemented IUGR animals, indicating a lesser extent of inflammation. With regard to gray matter alterations, IUGR animals supplemented with BrSp demonstrated improved cell counts in both CA1 and CA3 of the hippocampus in comparison to IUGR animals not treated with BrSp. Granted the cell counting was not done using stereologic technique. However, only cells with a complete nucleus were counted, thereby likely eliminating any bias or duplicate cell counts. Moreover, the veracity of our results would suggest that utilization this technique would not have altered the outcome significantly. Our histopathological findings were further supported by improved functional outcome. BrSp significantly enhanced IUGR performance on forelimb grasping, hind limb placing, cliff avoidance and posture. IUGR + B pups also showed marked improvement on open field testing, behaving no differently from Sham pups. Identified limitations within our study may point to the relatively short recovery period that was tested. In this regard,

296

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

Fig. 6. Photomicrographs showing reactive astrocytes (black arrows) in the corpus callosum. (A) Sham; (B) Sham + B; (C) IUGR; and (D) IUGR + B.

A

0.6

* † 0.3

cc thickness (mm)

cc thickness (mm)

0.6

B †

0.3

0.0

0.0 SHAM SHAM+B

IUGR

SHAM SHAM+B

IUGR+B

IUGR

IUGR+B

Fig. 7. White matter measurements of the corpus callosum thickness of (A) anterior and (B) posterior sections of the brain at PD21. Results are expressed as mean ± SD, † Significantly different from Sham; *Significantly different from IUGR.

the findings persist to day 21 postnatal, which is weaning within our rodent population, and likely comparable to early infancy. In this regard, aspects of later onset ‘learning deficits’ may not be picked up, and whether the preventive nature of the BrSps is permanent requires attention. Our laboratory is currently undertaking long term studies to determine exactly these questions. Additional research around the mechanistic

A

*

1.0 Cingulate Height (mm)

Cingulate Height (mm)

1.0

aspects of the positive findings also requires further investigation, and were outside the boundaries of the current investigation. Important aspects of the contribution to prevention from the maternal vs fetal side, measurement of the inflammatory response, and production of oxidant and anti-oxidants under circumstances of placental insufficiency and BrSp therapy are certainly warranted.

† 0.5

B * †

0.5

0.0

0.0 SHAM SHAM+B

IUGR

IUGR+B

SHAM SHAM+B

IUGR

IUGR+B

Fig. 8. Height of the peak of the cingulum projections in (A) anterior and (B) posterior sections at PD21. Results are expressed as mean ± SD, † Significantly different from Sham; *Significantly different from IUGR.

0.4

297

1.5

0.3

†

*

0.2

Area (mm2)

OD (Absorbance Units)

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

†

0.1

1.0

* 0.5

0.0 SHAM

SHAM+B

IUGR

IUGR+B

0.0 Fig. 9. Mean OD for myelin basic protein-stained sections. Results are expressed as mean ± SD, † Significantly different from Sham; *Significantly different from IUGR.

The therapeutic benefit of broccoli sprouts is attributable to its high content of glucoraphanin, the precursor to the isothiocyanate, sulforaphane. The latter increases the expression of the transcription factor, Nrf2, which induces anti-oxidant enzymes such as glutathione transferases and thioredoxin. These enzymes are collectively important in endogenous scavenging of reactive oxygen species [24]. With respect to the cerebrovascular system, sulforaphane has been reported to protect the blood brain barrier and improve cognitive outcome after traumatic brain injury, modulate neuroinflammation, reduce infarct volume following MCA occlusion and improve cell counts and neurologic deficit scores in a model of intracerebral hemorrhage [25–29]. In order to explain these findings, a number of groups sought to investigate the unique pattern of gene induction and inhibition which takes place in neurons and astrocytes in response to Nrf2 induction. Activation of Nrf2 was found to regulate the expression

SHAM

SHAM+B

IUGR

IUGR+B

Fig. 10. Ventricular areas of (A) Sham; (B) Sham + B; (C) IUGR; and (D) IUGR + B. Results are expressed as mean ± SD, † Significantly different from Sham; *Significantly different from IUGR.

of dozens of genes involved in cell-cycle events, protein modification, calcium homeostasis, intracellular phosphorylation cascades, synapse formation and conduction, neurotransmitter release, and response to injury. Detoxifying enzymes, as well as genes associated with energy production were both upregulated, suggesting enhanced coupling of metabolism and energy utilization between astrocytes and neurons. Furthermore, genes associated with extracellular interaction were induced, attesting to modification and perhaps greater efficiency of glia-neuronal interactions when Nrf2 is induced [30–32]. The above findings may provide the underlying mechanism whereby sulforaphane serves as a neuroprotective agent. However, further exploration of this approach as a long-term solution is warranted.

Fig. 11. Photomicrographs of myelin basic protein-stained images indicating thickness of myelinated bundles in: (A) Sham; (B) Sham + B; (C) IUGR; and (D) IUGR + B. The black arrows indicate an example of normal-thickness myelinated bundles.

298

A.M. Black et al. / Behavioural Brain Research 291 (2015) 289–298

Given that the majority of insults resulting in cerebral palsy and developmental disability occur prior to the onset of labor and delivery, alternative approaches to therapy must be considered. Certainly a majority of causes of cerebral palsy have hypoxia or vascular compromise of blood flow from the placenta, as a final common pathway. While not all result in fetal growth restriction, the latter risk factor increases the likelihood of cerebral palsy. Safe and efficacious approaches to treating pregnant mothers must be explored, in order to prevent injury to the fetus. In summary, we showed that broccoli sprout supplementation in the last trimester of gestation and first two weeks of life, lessened the behavioral and pathologic effects of placental insufficiency. These findings suggest a novel approach to preventing developmental disabilities associated with perinatal brain injury that can be provided safely to mother and fetus. Acknowledgements This research was supported by grants from the Canadian Institutes of Health Research, Heart and Stroke Foundation of Alberta, NWT and Inuvit, the Toronto Hospital for Sick Kids Foundation and NeuroDevNet (JYY). A. Black was supported by grants from the Alberta Heritage Foundation for Medical Research, the Hotchkiss Brain Institute, and the Maternal Fetal Network of CIHR. The authors thank Dr. Paul Talalay for measurements of DTC in fetal rats. References [1] Ferriero DM. Neonatal brain injury. N Engl J Med 2004;351:1985–95. [2] Badawi N, Kurinczuk JJ, Keogh JM, Alessandri LM, O’Sullivan F, Burton PR, Pemberton PJ, Stanley FJ. Antepartum risk factors for newborn encephalopathy: the Western Australian case-control study. Br Med J 1998;317:1549–53. [3] Jacobsson B, Hagberg G. Antenatal risk factors for cerebral palsy. Best Pract Res Clin Obstet Gynaecol 2004;18:425–36. [4] Low JA. Determining the contribution of asphyxia to brain damage in the neonate. J Obstet Gynaecol Res 2004;30:276–86. [5] Ikonomidou C, Bosch F, Miksa M, Bittigau P, Vockler J, Dikranian K, Tenkova TI, Stefovska V, Turski L, Olney JW. Blockade of NMDA receptors and apoptotic neurodegeneration in the developing brain. Science 1999;283:70–4. [6] Olney JW, Wozniak DF, Jevtovic-Todorovic V, Farber NB, Bittigau P, Ikonomidou C. Drug-induced apoptotic neurodegeneration in the developing brain. Brain Pathol 2002;12:488–98. [7] Bahadoran Z, Mirmiran P, Hosseinpanah F, Hedayati M, Hosseinpour-Niazi S, Azizi F. Broccoli sprouts reduce oxidative stress in type 2 diabetes: a randomized double-blind clinical trial. Eur J Clin Nutr 2011;65:972–7. [8] Akhlaghi M, Bandy B. Dietary broccoli sprouts protect against myocardial oxidative damage and cell death during ischemia-reperfusion. Plant Foods Hum Nutr 2010;65:193–9. [9] Munday R, Mhawech-Fauceglia P, Munday CM, Paonessa JD, Tang L, Munday JS, Lister C, Wilson P, Fahey JW, Davis W, Zhang Y. Inhibition of urinary bladder carcinogenesis by broccoli sprouts. Cancer Res 2008;68:1593–600. [10] Wigglesworth JS. Fetal growth retardation. Animal model: uterine vessel ligation in the pregnant rat. Am J Pathol 1974;77:347–50. [11] Rice D, Barone Jr S. Critical periods of vulnerability for the developing nervous system: evidence from humans and animal models. Environ Health Perspect 2000;108(Suppl. 3):511–33.

[12] Yager JY. Animal models of hypoxic-ischemic brain damage in the newborn. Semin Pediatr Neurol 2004;11:31–46. [13] Olivier P, Baud O, Evrard P, Gressens P, Verney C. Prenatal ischemia and white matter damage in rats. J Neuropathol Exp Neurol 2005;64:998–1006. [14] Bassan H, Bassan M, Pinhasov A, Kariv N, Giladi E, Gozes I, Harel S. The pregnant spontaneously hypertensive rat as a model of asymmetric intrauterine growth retardation and neurodevelopmental delay. Hypertens Pregnancy 2005;24:201–11. [15] Noyan-Ashraf MH, Sadeghinejad Z, Juurlink BH. Dietary approach to decrease aging-related CNS inflammation. Nutr Neurosci 2005;8:101–10. [16] Shapiro TA, Fahey JW, Wade KL, Stephenson KK, Talalay P. Chemoprotective glucosinolates and isothiocyanates of broccoli sprouts: metabolism and excretion in humans. Cancer Epidemiol Biomarkers Prev 2001;10: 501–8. [17] Ye L, Dinkova-Kostova AT, Wade KL, Zhang Y, Shapiro TA, Talalay P. Quantitative determination of dithiocarbamates in human plasma, serum, erythrocytes and urine: pharmacokinetics of broccoli sprout isothiocyanates in humans. Clin Chim Acta 2002;316:43–53. [18] Fox WM. Reflex-ontogeny and behavioural development of the mouse. Anim Behav 1965;13:234–41. [19] Lubics A, Reglodi D, Tamas A, Kiss P, Szalai M, Szalontay L, Lengvari I. Neurological reflexes and early motor behavior in rats subjected to neonatal hypoxic-ischemic injury. Behav Brain Res 2005;157:157–65. [20] Beard JL, Felt B, Schallert T, Burhans M, Connor JR, Georgieff MK. Moderate iron deficiency in infancy: biology and behavior in young rats. Behav Brain Res 2006;170:224–32. [21] Bekoff A, Trainer W. The development of interlimb co-ordination during swimming in postnatal rats. J Exp Biol 1979;83:1–11. [22] Kiss P, Hauser D, Tamas A, Lubics A, Racz B, Horvath ZS, Farkas J, Zimmermann F, Stepien A, Lengvari I, Reglodi D. Changes in open-field activity and noveltyseeking behavior in periadolescent rats neonatally treated with monosodium glutamate. Neurotoxicol Res 2007;12:85–93. [23] Colbourne F, Corbett D. Delayed postischemic hypothermia: a six month survival study using behavioral and histological assessments of neuroprotection. J Neurosci 1995;15:7250–60. [24] Fahey JW, Talalay P. Antioxidant functions of sulforaphane: a potent inducer of Phase II detoxication enzymes. Food Chem Toxicol 1999;37:973–9. [25] Zhao J, Moore AN, Clifton GL, Dash PK. Sulforaphane enhances aquaporin-4 expression and decreases cerebral edema following traumatic brain injury. J Neurosci Res 2005;82:499–506. [26] Zhao J, Kobori N, Aronowski J, Dash PK. Sulforaphane reduces infarct volume following focal cerebral ischemia in rodents. Neurosci Lett 2006;393: 108–12. [27] Zhao J, Moore AN, Redell JB, Dash PK. Enhancing expression of Nrf2driven genes protects the blood brain barrier after brain injury. J Neurosci 2007;27:10240–8. [28] Innamorato NG, Rojo AI, Garcia-Yague AJ, Yamamoto M, de Ceballos ML, Cuadrado A. The transcription factor Nrf2 is a therapeutic target against brain inflammation. J Immunol 2008;181:680–9. [29] Dash PK, Zhao J, Orsi SA, Zhang M, Moore AN. Sulforaphane improves cognitive function administered following traumatic brain injury. Neurosci Lett 2009;460:103–7. [30] Lee JM, Calkins MJ, Chan K, Kan YW, Johnson JA. Identification of the NFE2-related factor-2-dependent genes conferring protection against oxidative stress in primary cortical astrocytes using oligonucleotide microarray analysis. J Biol Chem 2003;278:12029–38. [31] Kraft AD, Johnson DA, Johnson JA. Nuclear factor E2-related factor 2-dependent antioxidant response element activation by tert-butylhydroquinone and sulforaphane occurring preferentially in astrocytes conditions neurons against oxidative insult. J Neurosci 2004;24:1101–12. [32] Kwak MK, Cho JM, Huang B, Shin S, Kensler TW. Role of increased expression of the proteasome in the protective effects of sulforaphane against hydrogen peroxide-mediated cytotoxicity in murine neuroblastoma cells. Free Radic Biol Med 2007;43:809–17.