- Email: [email protected]
Transgenerational effects of neonatal hypoxia-ischemia in progeny
Int. J. Devl Neuroscience 31 (2013) 398–405
Contents lists available at SciVerse ScienceDirect
International Journal of Developmental Neuroscience journal homepage: www.elsevier.com/locate/ijdevneu
Transgenerational effects of neonatal hypoxia-ischemia in progeny Smitha K. Infante, Harriett C. Rea, J.R. Perez-Polo ∗ Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, Galveston, TX 77555, United States
a r t i c l e
i n f o
Article history: Received 19 September 2012 Received in revised form 6 February 2013 Accepted 12 February 2013 Keywords: Hypoxia ischemia Epigenetics Behavior Gender differences Transgenerational effects
a b s t r a c t Neonatal hypoxia-ischemia (HI) affects 60% of low birth weight infants and up to 40% of preterm births. Cell death and brain injury after HI have been shown to cause long-lasting behavioral deficits. By using a battery of behavioral tests on second generation 3-week-old rodents, we found that neonatal HI is associated with behavioral outcomes in the progeny of HI-affected parents. Our results suggest an epigenetic transfer mechanism of some of the neurological symptoms associated with neonatal HI. Elucidating the transfer of brain injury to the next generation after HI calls attention to the risks associated with HI injury and the need for proper treatment to reverse these effects. Assessing the devastating extent of HI’s reach serves as a cautionary tale to the risks associated with neonatal HI, and provides an incentive to create improved therapeutic measures to treat HI. © 2013 ISDN. Published by Elsevier Ltd. All rights reserved.
1. Introduction Perinatal hypoxic-ischemic encephalopathy is an outcome after neonatal hypoxia-ischemia (HI), a major cause of neonatal morbidity and mortality, often resulting in developmental neurological deficits such as cerebral palsy, and delayed cognitive and behavioral deficits(Boichot et al., 2006; Zanelli et al., 2008). In the United States, perinatal HI occurs in 0.2–0.4% of term infants, and up to 60% of preterm (<37 weeks) or very low birth weight (<1500 g) infants (Vannucci et al., 1999; Zanelli et al., 2008). Due to the increasing incidence of preterm and low birth weight infants and the lack of adequate treatment for HI, characterization of brain injury after HI remains an extremely relevant area of interest. We tested the hypothesis that the long-lasting impacts of HI on sensory-motor coordination deficits are transmitted to progeny in an established model of rodent perinatal ischemia. In the Vannucci and coworkers (Rice et al., 1981; Vannucci et al., 1988) rodent model of HI there is increased cell death in brain regions associated with cognitive processes by increasing inflammatory cytokine levels followed by activation of AP-1 and NF-B-mediated transcriptional regulation of free radical generating enzymes and the prostaglandin pathway, mimicking aspects of observations in infants at risk (Bockhorst et al., 2010; Fabian et al., 2004, 2007, 2008; Grafe et al., 2008; Hu et al., 2005; Perez-Polo et al., 2011; Qiu et al., 2001, 2004; Smith et al., 2008; Tong et al., 2003; Gill and Perez-Polo, 2008; Hu et al., 2003; Xiaoming et al., 2006). HI also induces edema (Ferrari et al., 2010a,b) and via a BAX
protein mechanism involving nuclear translocation, a more inflammatory cell death outcome as compared to apoptosis (Dicou and Perez-Polo, 2009; Gill et al., 2008, 2009). HI stimulation of nuclear processes is consistent with epigenetic processes being triggered by HI in animal models (Englander et al., 1999; Martin et al., 2005). Epigenetic modifications can be manifested by (1) changes in cell phenotype; (2) regulation of cell lineage progress; and (3) a stably inherited phenotype. Numerous studies have shown the transgenerational effects of exogenous factors (Diamond et al., 1972; Denenberg and Whimbey, 1963). We focused on the third manifestation of epigenetics, that of stably inherited phenotype and assessed the behavioral characteristics of 2nd generation pups of rats that experienced neonatal HI with special attention to genderspecific transgenerational transference. Thus, we compared 2nd generation offspring of naive and sham-treated pups, to offspring of HI-exposed parents. We used similar behavioral assays to those used by Ferrari et al. (2010a,b), based on the sensory-motor coordination impairments reported for HI vs sham-treated pups (Ferrari et al., 2010a,b). We used a battery of three behavioral assessments of sensory-motor coordination: bar holding; wire mesh ascending; and sticky dot tests (Tchekalarova et al., 2005). This study is the first to show that some behavioral deficits observed after HI are transferred to the next generation of animals. More importantly, we showed that male progeny are more susceptible to inheriting HI-induced neurological deficits than female progeny, consistent with an epigenetic mechanism being responsible. 2. Experimental procedures 2.1. Animal care
∗ Corresponding author at: 301 Univ. Blvd., Galveston, TX 77555, United States. Tel.: +1 409 772 3668; fax: +1 409 772-8028. E-mail address: [email protected] (J.R. Perez-Polo). 0736-5748/$36.00 © 2013 ISDN. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijdevneu.2013.02.003
All animal procedures were performed according to the UTMB Animal Care and Use Committee (IACUC)-approved protocol # 9102020. Pregnant Wistar dams were purchased from Charles River at gestation days between E17-20. Dams gave birth
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
399
Fig. 1. Representation of HI model.
when ready, generally on E21 or E22. On postnatal day 2 (P2), the litter was culled to 10 pups to ensure that all pups received an equal chance to nurse. On surgery day, P7 pups were separated from dams and placed on a heating pad set at 37 ◦ C, to mimic the mother’s body temperature. Each pup was weighed, sexed, numbered, and categorized into the following groups: naive, sham, or HI. Naive animals remained with dams and did not undergo surgery or anesthesia, but were weighed, sexed, and numbered on P7 and before tissue harvest. Post surgery analgesics included daily administration of Buprenorphine (Bupernex) and 2 drops of topical anesthetic on suturing to minimize any localized pain for a maximum of three days post-surgery. 2.2. Surgical ischemia procedure Littermates were separated into two condition groups: HI and sham. As shown in the procedure timeline in Fig. 1, pups were placed in 2.5% isofluorane anesthesia chamber for a minimum of 5 min. For surgical procedure, pups were moved onto a 37 ◦ C heating pad throughout the surgery and anesthetized using 1.5–2.5% isofluorane by nose cone in an air/O2 mix. For HI pups, aseptic technique was used to make an anterior off-midline incision in the neck. The left carotid artery was then isolated from surrounding fat and tissue, dually cauterized, and transected between the cauterizations for a permanent ligation. Topical administration of marcaine was provided directly onto the incision, to address any pain or discomfort. The incision was then closed with 6-0 silk suture and cleaned. Topical bitter-flavored skin sealant was applied to the suture to ensure closure of the incision and to deter the dam from removing sutures. Sham animals were also anesthetized using isofluorane as stated above and placed on a heating pad for surgery. An anterior off-midline incision was made, immediately closed with a 6-0 silk suture, and cleaned. Though an incision was made on sham animals, in order to prevent any minor injuries that could result in ischemia/reperfusion of the left hemisphere, the carotid artery was not isolated. Anesthesia timings for each pup were monitored, recorded, and never exceeded 45 min. After surgery, pups were placed back onto heating pads and monitored for any signs of bleeding until they recovered from anesthesia. If a pup does not nurse or shows no signs of mobility post surgery, the animal was euthanized. Awake pups were then returned to dams, where they were mobile and nursing within 15 min, which was our indication that they recovered from the surgical procedure. 2.3. Hypoxia procedure As shown in the procedure timeline in Fig. 1, after 90 min of post-surgery recovery, all pups were separated from their mother. HI pups were placed in a hypoxia (8% O2 ) chamber for 90 min, and shams were placed on a heating pad in normal air levels of oxygen (20% O2 ) for 90 min. In the hypoxia chamber, animals were monitored for signs of excessive discomfort or immobility. Temperature was also monitored and rigorously maintained at 37 ◦ C to insure reproducibility. After 90 min, all pups were returned to their mothers where they remained until they were sacrificed at specific time points up to 72 h after injury. Litters used for breeding of second generation pups for behavior analyses were returned to the animal resource center facility. 2.4. Animal breeding For our behavior analyses, we required second generation pups. Fig. 2 depicts our timeline for acquiring second-generation animals. Three pregnant dams were ordered from Charles River, and at P7, each first generation litter was assigned a condition (HI, sham, or naive), and handled as stated above under surgical procedure and hypoxia. After surgery, pups were returned to their mothers and left under the care of UTMB’s Animal Resource Center. On P21, first generation pups were
B. First generaon First generaon rats receive rats paired for mang hypoxiaWean from ischemia Injury mother and separate by gender
Postnatal Day 7
Postnatal Day 21
Week 10
Second generaon pups weaned from mother Second generaon liers born
E21
Training and Behavior Assessment
Postnatal Day 21
Week 3
Fig. 2. Experimental plan.
weaned from their mother and separated by gender. On Week 6, pups required further separation as they grew, with a maximum of three females or two males per cage. Wistar rats are mature and ready to mate at 10 weeks of age. On Week 10, rats were paired into harems containing one male and two females. Rats were paired as in Table 1. Harems were maintained for up to two weeks, or once a female appeared to be pregnant. Pregnant first-generation females were immediately separated into their own cages to prepare for birth. Generally, second generation pups were born within a month after harems were set up. As before, on P2, litters were culled to ten pups. Second generation pups were allowed to nurse with the mother until P21, when pups were weaned and randomly separated into groups of five per cage. At this stage, pups were not separated by gender. Starting on P21, each pup was handled daily to reduce anxiety from fear and acclimate them to being held. Animals were trained for three days beginning on Day 25, with the final behavior testing day being Day 28.
3. Behavior analysis Three different behavior assays were conducted to assess sensory-motor coordination impairment in second generation pups. The setup for the mesh ascending test and bar holding test, based on Tchekalarova et al. (2005), was identical to that of Ferrari et al. (2010a,b). All behavior analyses were carried out between 8 am and 4 pm, using the same room and setup for each animal. All animals were in room at least 30 min before testing, to ensure Table 1 First-generation parent crosses for behavior analyses of second-generation pups. Mother
Father
# Total pups
Naive Naive Naive HI Sham HI
Naive Sham HI Sham HI HI
20 10 30 10 20 10
400
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
Weight Overall
A.
Weight (grams)
160 140
**
120 100
Weight (grams)
# **
120 100
140
100
IM x
H
H
IM
F
H
IF
x F
Sh am
Sh am
IM H
IF
x
F
x
H
M
x F
F
ai ve N
ai ve
ai ve
N x
IF H
N
N
ai ve
Sh am
M
IM x
H
H
IM
F x F
Sh am
H
IF
x
Sh am
x
H
M ai ve N
x
F
M
Sh am
ai ve N F N
ai ve
IM
80
x F
**
120
80
ai ve
IM x
H IF H
Male Weights 160
140
N
H
IM
M x F
Sh am
H
IF
x
Sh am
H
M x
N
ai ve
x F ai ve
N
C.
Female Weights 160
Weight (grams)
F
M
Sh am
ai ve N x F ai ve N
B.
IM
80
Fig. 3. Transgenerational effect of HI on pup weight at 3 weeks. (A) Weights of second-generation pups for all crosses. (B) Female weights. (C) Male weights **p < 0.05 when compared to naive × naive; #p < 0.05 when compared to naive × sham. One way ANOVA, Tukey’s multiple comparisons test.
acclimation to novel smells and sights. In between each behavior assay, the pups were placed in their home cage to rest. 3.1. Mesh ascending
the bar, it was immediately moved back to the center. Before testing day, each pup was trained 5 times per day for three consecutive days. On testing day, we performed five separate trials per animal with a resting time of 5 min between trials.
To assess sensory-motor coordination, we used the mesh ascending test. A 10 mm plastic mesh, 45 cm high and 15 cm wide, was secured at the edge of a table and tilted at an angle of 70◦ to come in contact with a sturdy plastic box at the top. A cardboard box filled with padding was placed below the table to protect from injury should a pup fall from the mesh. As a stimulus to ascend, four littermates and food pellets were placed in the plastic box at the top. Animals were placed and acclimated in the top box for 5 min before training and testing. Each pup was individually were placed at the bottom of the mesh and the time necessary to ascend was measured for up to 120 s. Before testing day, each pup was trained 5 times per day for three consecutive days. On testing day, we performed five separate trials per animal with a resting time of 5 min between trials.
3.3. Sticky dot
3.2. Bar holding
4. Results
The bar holding test was utilized as another assay to measure sensory-motor coordination. The setup consists of a wooden bar 1 cm in diameter and 30 cm long suspended 50 cm high above a padded soft surface. Each pup was placed atop the wooden bar and time spent on the bar, as well as grasping with forelimbs, was measured up to 120 s. If the pup was to balance itself on the corners of
4.1. Progeny of HI animals exhibit major sensory-motor coordination impairment in bar holding assessment
To determine the presence of a somatosensory asymmetry, we used the sticky dot test. For this assay, a 1.5 cm long piece of onesided tape was placed on each forepaw of the pup, and the pup was then placed in a clean cage for observation. Three different measurements were recorded for this assay, including paw preference, contact latency, and total time. The first paw contacted by the pup’s teeth is considered the ‘preferred paw’ and was recorded. Contact latency indicates the time until the pup attempts to remove a sticker. Time elapsed until both stickers were removed was also recorded as “total time.” Five separate trials with a resting time of 5 min between trials were conducted for each pup, but pups received no training prior to testing day for this assay.
Nulliparous first-generation rats were crossed at their peak mating age, 10 weeks after birth. Table 1 illustrates the crosses that were made, and the number of pups per cross that were assessed.
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
A.Bar Holding
401
Assesment at 3 weeks
150
Time (Seconds)
***
**
**
100
50
BH - FEMALE
B.
100
50
*** #
50
IM x
H
H
IM
F
H
IF
x F
Sh am
x IF H
Sh am
IM x
H
M F
ai ve
F ai ve N
ai ve
F
x
x
N
IF H
N
N
ai ve
Sh am
M
IM x
H
H
IM
F x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
F
M
Sh am
ai ve N N
ai ve
IM
0
x F
** ***
100
0
ai ve
IM x
H IF H 150
Time (Seconds)
Time (Seconds)
BH - MALE
C.
150
N
H
IM
F x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
N
ai ve
F
M
Sh am
ai ve N x F ai ve N
IM
0
Fig. 4. Bar holding assessment. (A) Bar holding assessment of sensory-motor coordination in pups whose parents experienced HI. (B) Females. (C) Males. ***p < 0.0001 when compared to naive × naive; **p < 0.05 when compared to naive × naive; #p < 0.05 when compared to naive × sham. One way ANOVA, Tukey’s multiple comparisons test.
A minimum of 10 pups were assessed at 3 weeks of age for each cross. When weight gain at 3 weeks, usually associated with developmental progress, was measured (Fig. 3A), there was no significant difference in weight distribution except in the HI x HI cross progeny, which as a group were significantly lower in weight. Additionally, when we separated female and male groups (Fig. 3B, C), female offspring had significantly lower weights compared to males and all other groups. Interestingly, the pups of HI-treated male parents had a wider weight distribution than those from naive or sham parents, consistent with a bimodal distribution suggestive of there being two groups: a responder group vs a non-responder group. Our first behavioral assessment was the bar holding test. Results indicate that all progeny of HI crosses displayed significant behavioral impairment when compared to naive and sham progeny, as seen in Fig. 4. HI × HI progeny did not exhibit significant impairment in this test as progeny with only one HI parent. However, there was a trend of greater impairment in HI × HI than progeny from crosses using naive or sham animals. When separating the data into female and male offspring groups, it is apparent that the males suffer from behavior deficits significantly more than the females irrespective of the gender of the parents. This was also apparent but not significant in HI × HI progeny, though they still performed better than progeny from only one HI parent. 4.2. Male progeny of HI animals exhibit impairment in mesh ascending test The mesh ascending test is another indicator of sensory-motor impairment. For this, the animals were placed at the bottom of a
sturdy plastic mesh, which was placed at a 70◦ angle against a safe, dark box at the top. During the first day of training, fear of falling stimulates these pups to climb up to the box. Based on previous work on first generation HI pups, the pups require more incentive to climb the mesh and into the box as they become conditioned during training (Ferrari et al., 2010a,b). Therefore, we placed the pup’s littermates in the box so that the pup being tested would have a desire to climb up the mesh and enter the safety of the box and littermates. In Fig. 5A, we show a significant impairment only with progeny from HI × HI. However, when separated by gender, male pups from the naive Female × HI Male cross show significant impairment when compared to naive progeny (Fig. 5B and C). Additionally, the impairment in HI × HI progeny is significant in males, whereas females show no significant impairment.
4.3. HI × HI pups exhibit more significant delay in motor coordination during sticky dot assessment Our last test for behavioral deficits consisted of placing stickers onto the forepaws of pups and recording the time it takes for them to acknowledge the stickers by trying to lick them off, which we called contact latency. We also recorded the total time elapsed before the pups removed both stickers from their paws. No training sessions were possible for this assessment, as the animals quickly began to acclimate to the exercise. However, fear was not an issue since the pups were simply placed back into their empty original cage while they addressed the stickers.
402
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
Wire Mesh Ascending Test
A.
*** #
Time (Seconds)
20 15 10 5
IM x
H IF H
C.
WM - FEMALE
10 5
15 10 5
IM x
H
H
IM
F
H
IF
x F
Sh am
H
IF
x
Sh am
H
M x
ai ve
x
N
ai ve
F
F ai ve N
N
N x
IF H
F
M
Sh am
ai ve
IM x
H
H
IM
F x F
Sh am
IM H
IF
x
Sh am
H
M x
ai ve N
x F ai ve
F
M
Sh am
ai ve N x F
IM
0
N
ai ve
*** **
Time (Seconds)
15
0
N
#
WM - MALE 20
20
Time (Seconds)
H
IM
F x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
N
ai ve
F
M
Sh am
ai ve N x F ai ve N
B.
IM
0
Fig. 5. Mesh ascending assessment. (A) Mesh ascending assessment of sensory-motor coordination in pups whose parents experienced HI. (B) Females. (C) Males. ***p < 0.0001 when compared to naive × naive; One way ANOVA, Tukey’s multiple comparisons test. **p < 0.05 when compared to naive × naive; #p < 0.05 when compared to naive × sham. Bartlett’s test for equal variances shows significance, but otherwise there’s no significance naive F × naive M vs naive F × HI M.
As seen in Fig. 6A, no significant differences were observed when comparing progeny with at least one HI parent, and progeny with only naive or sham parents, even when the data was separated by gender (Fig. 6B and C). Of note is the fact that contact latency was significantly shorter in males from sham Female × HI Male crosses when compared to males from naive Female × HI Male crosses. Fig. 7A indicates that no significant differences were observed between naive/sham pups and pups from only one HI parent with regards to total time elapsed before both stickers were removed. Interestingly, as with the wire mesh assay, the HI × HI appeared to perform slower than naive and sham progeny. In addition, as with our previous behavior assays, males performed worse than females when compared to sham and naive crosses, as seen in Fig. 7B and C. An additional point to note is that the distribution of values is much wider in animals with HI parents than pups from naive and sham crosses.
5. Discussion Overall, HI may lead to epigenetic changes that manifest in the next generation as behavioral deficits. In all animals, fear is always a strong motivating factor in survival situations, and the resulting response can often confound any sensory-motor coordination impairments. For this reason, all animals were trained for five times a day for three days before the actual behavioral test. Assessing the progress and behavior of animals during the training period can also be a good indicator of memory impairment.
Our results show that in general male progeny are more sensitive to the epigenetic transfer of the behavioral deficits measured here that are associated with HI when compared to females. This may be attributed to the fragility of the Y chromosome and its susceptibility to modifications such as methylation (Hill and Fitch, 2012). Sex-specific neuroprotection has been found in numerous injury paradigms, including neonatal HI, in both rodent models and humans (Giordano et al., 2011; Hill and Fitch, 2012). This is consistent with our observation that there are significant deficits in second-generation males. A primary factor for cell death after HI is oxidative stress. It was recently reported that paraoxonase 2, a potent antioxidant, is expressed at 2× higher levels in females than males. This may also explain the gender differences in the manifestation of behavioral deficits. In addition, it is important to note that imprinting occurs at different stages based on gender, so perhaps at the stage of our injury model, males are more susceptible to brain injury than females (Wagner et al., 1995). In our data based on the training days for the Bar Holding Assessment (not shown), unimpaired animals learn by the test day. When compared to naive/shams, far fewer of the injured animals’ progeny learn by the test day. Another important result to note is that HI × HI progeny fared better than pups with only one HI parent. Further analysis is required before these results can be interpreted. Though they fared better at the Bar Holding Assessment, HI × HI progeny showed significant deficits in the mesh ascending assessment. Though this is primarily a motor coordination assay, this also indicates social behavior. For example, their uninterest and unwillingness to mingle after entering the box and mingling with
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
403
Sticky Dot Contact Latency
A. 10
Time (Seconds)
8
**
6
**
4 2
IM x
H IF H
C.
SD CL - Female
SD CL - Male
8
10 8
Time (Seconds)
6 4 2
4 2
IM x
H
H
IM
F
H
IF
x F
Sh am
IM
x
Sh am
H
M x F
IF
F ai ve
H
x F N
ai ve
ai ve
N
x H
N
N
ai ve
Sh am
M
IM H
IM H IF
x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F ai ve
F
M
Sh am
ai ve N x F N
ai ve
F
0
IM
0
N
**
6
x
Time (Seconds)
H
IM
F x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
N
ai ve
F
M
Sh am
ai ve N x F ai ve N
B.
IM
0
Fig. 6. Sticky dot assessment. (A) Sticky dot assessment of sensory-motor coordination in pups whose parents experienced HI. (B) Females. (C) Males. **p < 0.05 when compared to naive F × HI M. One way ANOVA, Tukey’s multiple comparisons test.
their siblings may indicate antisocial behavior resulting from the oxidative stress damage suffered by both parents. The limitations of our sticky dot test include the fact that animals cannot be trained before being assessed. Therefore, fear may still be a variable in these tests. However, fear would have led the animals to pay less attention to the stickers and more attention to finding an exit from their cage, but this was certainly not the case. Interestingly, however, we saw significant differences between sham F x HI M and naive F × HI M, and between HI F × HI M and naive F × HI M, which indicate that the HI male may also have an effect on progeny behavior. However, this was only seen in the sticky dot assessment and, as stated before, this test has its limitations. It would certainly be interesting to assess the epigenetic modifications in the male that appear to be transferred to another generation. This indicates that there may be at least a small amount of gonadal epigenetic modifications in the HI parent. As shams of our rodent HI model are exposed to anesthesia, we were curious about whether anesthesia affects sensory-motor coordination in the second generation as well. Therefore, we assessed naive pups and compared them to shams. As expected, there were minimal differences between progeny from naives and/or shams. Some crosses only had one litter assessed, so the results may be litter-specific. However, the pattern of behavioral deficits observed in pups from HI parents cannot be coincidental. Our behavioral results are consistent with HI affecting the germ line but lack direct evidence implicating HI as a transgenerational epigenetic process. For example, the observed behavioral deficits of the next generation could be due to epigenetic modifications arising from cross-talk among the nervous, endocrine, and reproductive
systems. There is evidence that throughout development, there is cross-talk among the nervous, endocrine, digestive, immune, and reproductive systems (Marchetti et al., 1990) and that the nervous system and endocrine system are seamlessly integrated and interact, especially in response to environmental stress (Dronca et al., 2012). Although according to our data, the impairment appeared to be due to HI effects on the primordial germ line, another important consideration is the impact of HI on female uterine environment, as well as the psychological effects on the female, which may affect parenting methods on both male and female offspring or selectively on one gender. It is well known that maternal neglect can lead to anxiety and stress issues in the progeny. This may be a factor to be considered in future studies, where one could assess the effects of postnatal maternal behavior after HI, by cross-fostering HI second generation pups to a healthy sham after birth. It would be interesting to study the effects of global neonatal hypoxia-ischemia on the second and third generation of animals. In the case of global HI, the second generation’s germ cells would be directly exposed to HI, therefore analysis of a third generation is necessary to prove epigenetic inheritance as opposed to direct epigenetic effects of global HI. We were able to observe the effects of HI in the first generation and relate it to indirect transfer of behavioral phenotype to the next generation. Assessments of DNA methylation patterns, histone modifications protein acetylation or genomic alterations introduced by DNA repair mechanisms reversins epigenetic DNA modifications after HI in first and second-generation animals of this model could confirm the presence and mechanisms of action of epigenetic influence
404
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
Sticky Dot Total Time
A. Time (Seconds)
40 30 20 10
IM x
IF H
SD TT - Male 40
Time (Seconds)
40 30 20 10
30 20 10
IM x
H
H
IM
F
H
IF
x F
Sh am
IM
x
Sh am
H
M x F
IF
F ai ve
H
x F N
ai ve
ai ve
N
x H
N
N
ai ve
Sh am
M
IM H
IM H IF
x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
N
ai ve
F
M
Sh am
ai ve N x F ai ve N
F
0
IM
0
x
Time (Seconds)
H
IM
F
H x F
Sh am
x IF H
C.
SD TT - Female
B.
Sh am
H
M x
ai ve
N
N
x F ai ve
F
M
Sh am
ai ve N x F ai ve N
IM
0
Fig. 7. Sticky dot assessment, total time, showed a larger spread of values with progeny of HI parents than from naive/sham progeny. (A) Total time assessments do not indicate any significant differences between crosses, even when separating data into female (B) and male (C) groups. One way ANOVA, Tukey’s multiple comparisons test.
on the inheritance of behavioral deficits observed in the second generation. These results also address the importance of male parent contributions to offspring beyond the classical gene base pair sequence-directed changes. The increased spread in outcomes observed in offspring behavior may also reflect developmental processes relative to chromatin configuration. Acknowledgements This work was supported by grant PO1 HDO039833 from NICHD and grant PT074693P9 from the DoD. References Boichot, C., Walker, P.M., Durand, C., Grimaldi, M., Chapuis, S., Gouyon, J.B., Brunotte, F., 2006. Term neonate prognoses after perinatal asphyxia: contributions of MR imaging, MR spectroscopy, relaxation times, and apparent diffusion coefficients. Radiology 239 (3), 839–848. Bockhorst, K.H., Narayana, P.A., Dulin, J., Liu, R., Rea, H.C., Wosik, H.K., Perez-Polo, J.R., 2010. Normobaric hyperoximia increases hypoxia-induced cerebral injury: DTI study in rats. Journal of Neuroscience Research 88 (April (5)), 1146–1156, PMCID:PMC 19885827. Denenberg, V.H., Whimbey, A.E., 1963. Behavior of adult rats is modified by the experiences their mothers had as infants. Science 142, 1192–1193. Diamond, M.C., Rosenzweig, M.R., Bennett, E.L., et al., 1972. Effects of environmental enrichment and impoverishment on rat cerebral cortex. Journal of Neurobiology 3 (1), 47–64. Dicou, E., Perez-Polo, J.R., 2009. Bax-an emerging role in ectopic cell death. International Journal of Developmental Neuroscience 27 (June (4)), 299–304, PMCID: PMC 19460623. Dronca, Roxana, S., Leontovich, Alexey, A., Nevala, Wendy, K., Markovic, Svetomir, N., 2012. Personalized therapy for metastatic melanoma: could timing be everything? Future Oncology 8 (11), 1401–1406.
Englander, E.W., Greeley Jr., G.H., Wanger Jr., G., Perez-Polo, J.R., Lee, H.-M., 1999. Hypoxia-induced mitochondrial and nuclear DNA damage in the rat brain. JNR 58, 262–269. Fabian, R.H., Perez-Polo, J.R., Kent, T.A., 2008. Perivascular nitric oxide and superoxide in neonatal cerebral hypoxia-ischemia. American Journal of Physiology: Heart and Circulatory Physiology 295 (October (4)), H1809–H1814, PMCID:PMC2593505. Fabian, R.H., Kent, T.A., Perez-Polo, J.R., 2004. Extracellular superoxide concentration increases following cerebral hypoxia but does not affect cerebral blood flow. International Journal of Developmental Neuroscience 22 (4), 225–230. Fabian, R.H., Perez-Polo, J.R., Kent, T.A., 2007. A decoy oligonucleotide inhibiting NFB binding to the IgGB consensus site reduces cerebral injury and apoptosis in neonatal hypoxic-ischemic encephalopathy. Journal of Neuroscience Research 85 (7), 1420–1426. Ferrari, D.C., Nesic-Taylor, O.B., Perez-Polo, J.R., 2010a. Oxygen resuscitation does not ameliorate neonatal hypoxia/ischemia-induced cerebral edema. Journal of Neuroscience Research 88, 2056–2065, PMCID:PMC 20143414. Ferrari, D.C., Nesic, O., Perez-Polo, J.R., 2010b. Perspectives on neonatal hypoxia/ischemia-induced edema formation. Neurochemical Research 35, 1957–1965, http://dx.doi.org/10.1007/s11064-010-0308-y, PMID:21136160. Gill, M.B., Bockhorst, K.H., Narayana, P.A., Perez-Polo, J.R., 2008. Bax shuttling after neonatal hypoxia-ischemia: hyperoxia effects. Journal of Neuroscience Research 86 (December (16)), 3584–3604, PMCID:PMC2585158. Gill, M.B., Perez-Polo, J.R., 2009. Bax shuttling after rotenone treatment of neuronal primary cultures: effects on cell death phenotypes. Journal of Neuroscience Research 87 (July (9)), 2047–2065, PMCID: PMC 19224578. Gill, M.B., Perez-Polo, J.R., 2008. Hypoxia ischemia-mediated cell death in neonatal rat brain. Neurochemical Research 33 (December (12)), 2379–2389, PMID 18408999. Giordano, G., Cole, T.B., Furlong, C.E., Costa, L.G., 2011. Paraoxonase 2 (PON2) in the mouse central nervous system: a neuroprotective role? Toxicology and Applied Pharmacology 256, 369. Grafe, M.R., Woodworth, K.N., Noppens, K., Perez-Polo, J.R., 2008. Long-term histological outcome after post-hypoxic treatment with 100% or 40% oxygen in a model of perinatal hypoxic-ischemic brain injury. IJDN 26 (February (1)), 119–124, PMCID:PMC2291019. Hill, C.A., Fitch, R.H., 2012. Sex differences in mechanisms and outcome of neonatal hypoxia-ischemia in rodent models: implications for sex-specific
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405 neuroprotection in clinical neonatal practice. Neurology Research International 2012. Hu, X., Nesic-Taylor, O., Qiu, J., Rea, H.C., Fabian, R., Rassin David, K., Perez-Polo, J.R., 2005. Activation of nuclear factor-B signaling pathway by interleukin1 after hypoxia/ischemia in neonatal rat hippocampus and cortex. Journal of Neurochemistry 93, 26–37. Hu, X., Qiu, J., Grafe, M.R., Rea, H.C., Rassin, D.K., Perez-Polo, J.R., 2003. Bcl-2 family members make different contributions to cell death in hypoxia and/or hyperoxia in rat cerebral cortex. IJDN 21 (7), 371–378. Xiaoming, H., Rea, H.C., Wiktorowicz, J.E., Perez-Polo, J.R., 2006. Proteomic analysis of hypoxia/ischemia-induced alteration of cortical development and dopamine neurotransmission in neonatal rat. Journal of Proteome Research 5 (9), 2396–2404. Marchetti, B., et al., 1990. Cross-talk communication in the neuroendocrinereproductive-immune axis. Age-dependent alterations in the common communication networks. Annals of the New York Academy of Sciences 594, 309. Martin, S.S., Regino Perez-Polo, J., Noppens, K.M., Grafe, M., 2005. Biphasic changes in the levels of poly(ADP-ribose) polymerase-1 and caspase 3in the immature brain following hypoxia-ischemia. IJDN 23 (8), 673–686. Perez-Polo, J.R., Reilly, C.B., Rea, H.C., 2011. Oxygen resuscitation after hypoxia ischemia stimulates prostaglandin pathway in rat cortex. IJDN 29 (6), 639–644, PM: 21514373. Qiu, J.-X., Grafe, M.R., Schmura, S.M., Glasgow, J., Kent, T.A., Rassin, D.K., Perez-Polo, J.R., 2001. Differential NF-B regulation of bcl-x gene expression in hippocampus and basal forebrain in response to hypoxia. Journal of Neuroscience Research 64, 223–234.
405
Qiu, J., Hu, X., Nesic, O., Grafe, M.R., Rassin, D.K., Perez-Polo, J.R., 2004. Effects of NFB oligonucleotide “Decoys” on gene expression in p7 rat hippocampus after hypoxia/ischemia. Journal of Neuroscience Research 77 (1), 108–118. Rice 3rd, J.E., Vannucci, R.C., Brierley, J.B., 1981. The influence of immaturity on hypoxic-ischemic brain damage in the rat. Annals of Neurology 9 (February), 131. Smith, K.E., Keeney, S., Zhang, L., Perez-Polo, R., Rassing, D.K., 2008. The association of early blood oxygenation with child development in preterm infants with acute respiratory disorders. IJDN 26 (February (1)), 125–131, PMCID:PMC2288779. Tchekalarova, J., Kubova, H., Mares, P., 2005. Postnatal caffeine exposure: effects on motor skills and locomotor activity during ontogenesis. Behavioural Brain Research 160 (1), 99–106. Tong, L.Q.T., Toliver-Kinsky, D., Rassin, K., Werrbach-Perez, J.R., Perez-Polo, 2003. Hyperoxia increases AP-1 DNA binding in rat brain. Neurochemical Research 28, 111–115. Vannucci, R.C., Lyons, D.T., Vasta, F., 1988. Regional cerebral blood flow during hypoxia-ischemia in immature rats. Stroke 19 (February), 245. Vannucci, R.C., Connor, J.R., Mauger, D.T., Palmer, C., Smith, M.B., Towfighi, J., Vannucci, S.J., 1999. Title Rat model of perinatal hypoxic-ischemic brain damage. Journal of Neuroscience Research 55 (2), 158–163. Wagner, A.K., Fabio, A., Puccio, A.M., Hirschberg, R., Zafonte, R.D., Marion, D.W., 1995. Gender associations with cerebrospinal fluid glutamate and lactate/pyruvate levels after severe traumatic brain injury. Critical Care Medicine 33 (2), 407–413. Zanelli, S.A., Stanley, D.P., Kaufman, D., 2008. Hypoxic–Ischemic Encephalopathy. http://emedicine.medscape.com/article/973501-overview
Contents lists available at SciVerse ScienceDirect
International Journal of Developmental Neuroscience journal homepage: www.elsevier.com/locate/ijdevneu
Transgenerational effects of neonatal hypoxia-ischemia in progeny Smitha K. Infante, Harriett C. Rea, J.R. Perez-Polo ∗ Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, Galveston, TX 77555, United States
a r t i c l e
i n f o
Article history: Received 19 September 2012 Received in revised form 6 February 2013 Accepted 12 February 2013 Keywords: Hypoxia ischemia Epigenetics Behavior Gender differences Transgenerational effects
a b s t r a c t Neonatal hypoxia-ischemia (HI) affects 60% of low birth weight infants and up to 40% of preterm births. Cell death and brain injury after HI have been shown to cause long-lasting behavioral deficits. By using a battery of behavioral tests on second generation 3-week-old rodents, we found that neonatal HI is associated with behavioral outcomes in the progeny of HI-affected parents. Our results suggest an epigenetic transfer mechanism of some of the neurological symptoms associated with neonatal HI. Elucidating the transfer of brain injury to the next generation after HI calls attention to the risks associated with HI injury and the need for proper treatment to reverse these effects. Assessing the devastating extent of HI’s reach serves as a cautionary tale to the risks associated with neonatal HI, and provides an incentive to create improved therapeutic measures to treat HI. © 2013 ISDN. Published by Elsevier Ltd. All rights reserved.
1. Introduction Perinatal hypoxic-ischemic encephalopathy is an outcome after neonatal hypoxia-ischemia (HI), a major cause of neonatal morbidity and mortality, often resulting in developmental neurological deficits such as cerebral palsy, and delayed cognitive and behavioral deficits(Boichot et al., 2006; Zanelli et al., 2008). In the United States, perinatal HI occurs in 0.2–0.4% of term infants, and up to 60% of preterm (<37 weeks) or very low birth weight (<1500 g) infants (Vannucci et al., 1999; Zanelli et al., 2008). Due to the increasing incidence of preterm and low birth weight infants and the lack of adequate treatment for HI, characterization of brain injury after HI remains an extremely relevant area of interest. We tested the hypothesis that the long-lasting impacts of HI on sensory-motor coordination deficits are transmitted to progeny in an established model of rodent perinatal ischemia. In the Vannucci and coworkers (Rice et al., 1981; Vannucci et al., 1988) rodent model of HI there is increased cell death in brain regions associated with cognitive processes by increasing inflammatory cytokine levels followed by activation of AP-1 and NF-B-mediated transcriptional regulation of free radical generating enzymes and the prostaglandin pathway, mimicking aspects of observations in infants at risk (Bockhorst et al., 2010; Fabian et al., 2004, 2007, 2008; Grafe et al., 2008; Hu et al., 2005; Perez-Polo et al., 2011; Qiu et al., 2001, 2004; Smith et al., 2008; Tong et al., 2003; Gill and Perez-Polo, 2008; Hu et al., 2003; Xiaoming et al., 2006). HI also induces edema (Ferrari et al., 2010a,b) and via a BAX
protein mechanism involving nuclear translocation, a more inflammatory cell death outcome as compared to apoptosis (Dicou and Perez-Polo, 2009; Gill et al., 2008, 2009). HI stimulation of nuclear processes is consistent with epigenetic processes being triggered by HI in animal models (Englander et al., 1999; Martin et al., 2005). Epigenetic modifications can be manifested by (1) changes in cell phenotype; (2) regulation of cell lineage progress; and (3) a stably inherited phenotype. Numerous studies have shown the transgenerational effects of exogenous factors (Diamond et al., 1972; Denenberg and Whimbey, 1963). We focused on the third manifestation of epigenetics, that of stably inherited phenotype and assessed the behavioral characteristics of 2nd generation pups of rats that experienced neonatal HI with special attention to genderspecific transgenerational transference. Thus, we compared 2nd generation offspring of naive and sham-treated pups, to offspring of HI-exposed parents. We used similar behavioral assays to those used by Ferrari et al. (2010a,b), based on the sensory-motor coordination impairments reported for HI vs sham-treated pups (Ferrari et al., 2010a,b). We used a battery of three behavioral assessments of sensory-motor coordination: bar holding; wire mesh ascending; and sticky dot tests (Tchekalarova et al., 2005). This study is the first to show that some behavioral deficits observed after HI are transferred to the next generation of animals. More importantly, we showed that male progeny are more susceptible to inheriting HI-induced neurological deficits than female progeny, consistent with an epigenetic mechanism being responsible. 2. Experimental procedures 2.1. Animal care
∗ Corresponding author at: 301 Univ. Blvd., Galveston, TX 77555, United States. Tel.: +1 409 772 3668; fax: +1 409 772-8028. E-mail address: [email protected] (J.R. Perez-Polo). 0736-5748/$36.00 © 2013 ISDN. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijdevneu.2013.02.003
All animal procedures were performed according to the UTMB Animal Care and Use Committee (IACUC)-approved protocol # 9102020. Pregnant Wistar dams were purchased from Charles River at gestation days between E17-20. Dams gave birth
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
399
Fig. 1. Representation of HI model.
when ready, generally on E21 or E22. On postnatal day 2 (P2), the litter was culled to 10 pups to ensure that all pups received an equal chance to nurse. On surgery day, P7 pups were separated from dams and placed on a heating pad set at 37 ◦ C, to mimic the mother’s body temperature. Each pup was weighed, sexed, numbered, and categorized into the following groups: naive, sham, or HI. Naive animals remained with dams and did not undergo surgery or anesthesia, but were weighed, sexed, and numbered on P7 and before tissue harvest. Post surgery analgesics included daily administration of Buprenorphine (Bupernex) and 2 drops of topical anesthetic on suturing to minimize any localized pain for a maximum of three days post-surgery. 2.2. Surgical ischemia procedure Littermates were separated into two condition groups: HI and sham. As shown in the procedure timeline in Fig. 1, pups were placed in 2.5% isofluorane anesthesia chamber for a minimum of 5 min. For surgical procedure, pups were moved onto a 37 ◦ C heating pad throughout the surgery and anesthetized using 1.5–2.5% isofluorane by nose cone in an air/O2 mix. For HI pups, aseptic technique was used to make an anterior off-midline incision in the neck. The left carotid artery was then isolated from surrounding fat and tissue, dually cauterized, and transected between the cauterizations for a permanent ligation. Topical administration of marcaine was provided directly onto the incision, to address any pain or discomfort. The incision was then closed with 6-0 silk suture and cleaned. Topical bitter-flavored skin sealant was applied to the suture to ensure closure of the incision and to deter the dam from removing sutures. Sham animals were also anesthetized using isofluorane as stated above and placed on a heating pad for surgery. An anterior off-midline incision was made, immediately closed with a 6-0 silk suture, and cleaned. Though an incision was made on sham animals, in order to prevent any minor injuries that could result in ischemia/reperfusion of the left hemisphere, the carotid artery was not isolated. Anesthesia timings for each pup were monitored, recorded, and never exceeded 45 min. After surgery, pups were placed back onto heating pads and monitored for any signs of bleeding until they recovered from anesthesia. If a pup does not nurse or shows no signs of mobility post surgery, the animal was euthanized. Awake pups were then returned to dams, where they were mobile and nursing within 15 min, which was our indication that they recovered from the surgical procedure. 2.3. Hypoxia procedure As shown in the procedure timeline in Fig. 1, after 90 min of post-surgery recovery, all pups were separated from their mother. HI pups were placed in a hypoxia (8% O2 ) chamber for 90 min, and shams were placed on a heating pad in normal air levels of oxygen (20% O2 ) for 90 min. In the hypoxia chamber, animals were monitored for signs of excessive discomfort or immobility. Temperature was also monitored and rigorously maintained at 37 ◦ C to insure reproducibility. After 90 min, all pups were returned to their mothers where they remained until they were sacrificed at specific time points up to 72 h after injury. Litters used for breeding of second generation pups for behavior analyses were returned to the animal resource center facility. 2.4. Animal breeding For our behavior analyses, we required second generation pups. Fig. 2 depicts our timeline for acquiring second-generation animals. Three pregnant dams were ordered from Charles River, and at P7, each first generation litter was assigned a condition (HI, sham, or naive), and handled as stated above under surgical procedure and hypoxia. After surgery, pups were returned to their mothers and left under the care of UTMB’s Animal Resource Center. On P21, first generation pups were
B. First generaon First generaon rats receive rats paired for mang hypoxiaWean from ischemia Injury mother and separate by gender
Postnatal Day 7
Postnatal Day 21
Week 10
Second generaon pups weaned from mother Second generaon liers born
E21
Training and Behavior Assessment
Postnatal Day 21
Week 3
Fig. 2. Experimental plan.
weaned from their mother and separated by gender. On Week 6, pups required further separation as they grew, with a maximum of three females or two males per cage. Wistar rats are mature and ready to mate at 10 weeks of age. On Week 10, rats were paired into harems containing one male and two females. Rats were paired as in Table 1. Harems were maintained for up to two weeks, or once a female appeared to be pregnant. Pregnant first-generation females were immediately separated into their own cages to prepare for birth. Generally, second generation pups were born within a month after harems were set up. As before, on P2, litters were culled to ten pups. Second generation pups were allowed to nurse with the mother until P21, when pups were weaned and randomly separated into groups of five per cage. At this stage, pups were not separated by gender. Starting on P21, each pup was handled daily to reduce anxiety from fear and acclimate them to being held. Animals were trained for three days beginning on Day 25, with the final behavior testing day being Day 28.
3. Behavior analysis Three different behavior assays were conducted to assess sensory-motor coordination impairment in second generation pups. The setup for the mesh ascending test and bar holding test, based on Tchekalarova et al. (2005), was identical to that of Ferrari et al. (2010a,b). All behavior analyses were carried out between 8 am and 4 pm, using the same room and setup for each animal. All animals were in room at least 30 min before testing, to ensure Table 1 First-generation parent crosses for behavior analyses of second-generation pups. Mother
Father
# Total pups
Naive Naive Naive HI Sham HI
Naive Sham HI Sham HI HI
20 10 30 10 20 10
400
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
Weight Overall
A.
Weight (grams)
160 140
**
120 100
Weight (grams)
# **
120 100
140
100
IM x
H
H
IM
F
H
IF
x F
Sh am
Sh am
IM H
IF
x
F
x
H
M
x F
F
ai ve N
ai ve
ai ve
N x
IF H
N
N
ai ve
Sh am
M
IM x
H
H
IM
F x F
Sh am
H
IF
x
Sh am
x
H
M ai ve N
x
F
M
Sh am
ai ve N F N
ai ve
IM
80
x F
**
120
80
ai ve
IM x
H IF H
Male Weights 160
140
N
H
IM
M x F
Sh am
H
IF
x
Sh am
H
M x
N
ai ve
x F ai ve
N
C.
Female Weights 160
Weight (grams)
F
M
Sh am
ai ve N x F ai ve N
B.
IM
80
Fig. 3. Transgenerational effect of HI on pup weight at 3 weeks. (A) Weights of second-generation pups for all crosses. (B) Female weights. (C) Male weights **p < 0.05 when compared to naive × naive; #p < 0.05 when compared to naive × sham. One way ANOVA, Tukey’s multiple comparisons test.
acclimation to novel smells and sights. In between each behavior assay, the pups were placed in their home cage to rest. 3.1. Mesh ascending
the bar, it was immediately moved back to the center. Before testing day, each pup was trained 5 times per day for three consecutive days. On testing day, we performed five separate trials per animal with a resting time of 5 min between trials.
To assess sensory-motor coordination, we used the mesh ascending test. A 10 mm plastic mesh, 45 cm high and 15 cm wide, was secured at the edge of a table and tilted at an angle of 70◦ to come in contact with a sturdy plastic box at the top. A cardboard box filled with padding was placed below the table to protect from injury should a pup fall from the mesh. As a stimulus to ascend, four littermates and food pellets were placed in the plastic box at the top. Animals were placed and acclimated in the top box for 5 min before training and testing. Each pup was individually were placed at the bottom of the mesh and the time necessary to ascend was measured for up to 120 s. Before testing day, each pup was trained 5 times per day for three consecutive days. On testing day, we performed five separate trials per animal with a resting time of 5 min between trials.
3.3. Sticky dot
3.2. Bar holding
4. Results
The bar holding test was utilized as another assay to measure sensory-motor coordination. The setup consists of a wooden bar 1 cm in diameter and 30 cm long suspended 50 cm high above a padded soft surface. Each pup was placed atop the wooden bar and time spent on the bar, as well as grasping with forelimbs, was measured up to 120 s. If the pup was to balance itself on the corners of
4.1. Progeny of HI animals exhibit major sensory-motor coordination impairment in bar holding assessment
To determine the presence of a somatosensory asymmetry, we used the sticky dot test. For this assay, a 1.5 cm long piece of onesided tape was placed on each forepaw of the pup, and the pup was then placed in a clean cage for observation. Three different measurements were recorded for this assay, including paw preference, contact latency, and total time. The first paw contacted by the pup’s teeth is considered the ‘preferred paw’ and was recorded. Contact latency indicates the time until the pup attempts to remove a sticker. Time elapsed until both stickers were removed was also recorded as “total time.” Five separate trials with a resting time of 5 min between trials were conducted for each pup, but pups received no training prior to testing day for this assay.
Nulliparous first-generation rats were crossed at their peak mating age, 10 weeks after birth. Table 1 illustrates the crosses that were made, and the number of pups per cross that were assessed.
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
A.Bar Holding
401
Assesment at 3 weeks
150
Time (Seconds)
***
**
**
100
50
BH - FEMALE
B.
100
50
*** #
50
IM x
H
H
IM
F
H
IF
x F
Sh am
x IF H
Sh am
IM x
H
M F
ai ve
F ai ve N
ai ve
F
x
x
N
IF H
N
N
ai ve
Sh am
M
IM x
H
H
IM
F x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
F
M
Sh am
ai ve N N
ai ve
IM
0
x F
** ***
100
0
ai ve
IM x
H IF H 150
Time (Seconds)
Time (Seconds)
BH - MALE
C.
150
N
H
IM
F x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
N
ai ve
F
M
Sh am
ai ve N x F ai ve N
IM
0
Fig. 4. Bar holding assessment. (A) Bar holding assessment of sensory-motor coordination in pups whose parents experienced HI. (B) Females. (C) Males. ***p < 0.0001 when compared to naive × naive; **p < 0.05 when compared to naive × naive; #p < 0.05 when compared to naive × sham. One way ANOVA, Tukey’s multiple comparisons test.
A minimum of 10 pups were assessed at 3 weeks of age for each cross. When weight gain at 3 weeks, usually associated with developmental progress, was measured (Fig. 3A), there was no significant difference in weight distribution except in the HI x HI cross progeny, which as a group were significantly lower in weight. Additionally, when we separated female and male groups (Fig. 3B, C), female offspring had significantly lower weights compared to males and all other groups. Interestingly, the pups of HI-treated male parents had a wider weight distribution than those from naive or sham parents, consistent with a bimodal distribution suggestive of there being two groups: a responder group vs a non-responder group. Our first behavioral assessment was the bar holding test. Results indicate that all progeny of HI crosses displayed significant behavioral impairment when compared to naive and sham progeny, as seen in Fig. 4. HI × HI progeny did not exhibit significant impairment in this test as progeny with only one HI parent. However, there was a trend of greater impairment in HI × HI than progeny from crosses using naive or sham animals. When separating the data into female and male offspring groups, it is apparent that the males suffer from behavior deficits significantly more than the females irrespective of the gender of the parents. This was also apparent but not significant in HI × HI progeny, though they still performed better than progeny from only one HI parent. 4.2. Male progeny of HI animals exhibit impairment in mesh ascending test The mesh ascending test is another indicator of sensory-motor impairment. For this, the animals were placed at the bottom of a
sturdy plastic mesh, which was placed at a 70◦ angle against a safe, dark box at the top. During the first day of training, fear of falling stimulates these pups to climb up to the box. Based on previous work on first generation HI pups, the pups require more incentive to climb the mesh and into the box as they become conditioned during training (Ferrari et al., 2010a,b). Therefore, we placed the pup’s littermates in the box so that the pup being tested would have a desire to climb up the mesh and enter the safety of the box and littermates. In Fig. 5A, we show a significant impairment only with progeny from HI × HI. However, when separated by gender, male pups from the naive Female × HI Male cross show significant impairment when compared to naive progeny (Fig. 5B and C). Additionally, the impairment in HI × HI progeny is significant in males, whereas females show no significant impairment.
4.3. HI × HI pups exhibit more significant delay in motor coordination during sticky dot assessment Our last test for behavioral deficits consisted of placing stickers onto the forepaws of pups and recording the time it takes for them to acknowledge the stickers by trying to lick them off, which we called contact latency. We also recorded the total time elapsed before the pups removed both stickers from their paws. No training sessions were possible for this assessment, as the animals quickly began to acclimate to the exercise. However, fear was not an issue since the pups were simply placed back into their empty original cage while they addressed the stickers.
402
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
Wire Mesh Ascending Test
A.
*** #
Time (Seconds)
20 15 10 5
IM x
H IF H
C.
WM - FEMALE
10 5
15 10 5
IM x
H
H
IM
F
H
IF
x F
Sh am
H
IF
x
Sh am
H
M x
ai ve
x
N
ai ve
F
F ai ve N
N
N x
IF H
F
M
Sh am
ai ve
IM x
H
H
IM
F x F
Sh am
IM H
IF
x
Sh am
H
M x
ai ve N
x F ai ve
F
M
Sh am
ai ve N x F
IM
0
N
ai ve
*** **
Time (Seconds)
15
0
N
#
WM - MALE 20
20
Time (Seconds)
H
IM
F x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
N
ai ve
F
M
Sh am
ai ve N x F ai ve N
B.
IM
0
Fig. 5. Mesh ascending assessment. (A) Mesh ascending assessment of sensory-motor coordination in pups whose parents experienced HI. (B) Females. (C) Males. ***p < 0.0001 when compared to naive × naive; One way ANOVA, Tukey’s multiple comparisons test. **p < 0.05 when compared to naive × naive; #p < 0.05 when compared to naive × sham. Bartlett’s test for equal variances shows significance, but otherwise there’s no significance naive F × naive M vs naive F × HI M.
As seen in Fig. 6A, no significant differences were observed when comparing progeny with at least one HI parent, and progeny with only naive or sham parents, even when the data was separated by gender (Fig. 6B and C). Of note is the fact that contact latency was significantly shorter in males from sham Female × HI Male crosses when compared to males from naive Female × HI Male crosses. Fig. 7A indicates that no significant differences were observed between naive/sham pups and pups from only one HI parent with regards to total time elapsed before both stickers were removed. Interestingly, as with the wire mesh assay, the HI × HI appeared to perform slower than naive and sham progeny. In addition, as with our previous behavior assays, males performed worse than females when compared to sham and naive crosses, as seen in Fig. 7B and C. An additional point to note is that the distribution of values is much wider in animals with HI parents than pups from naive and sham crosses.
5. Discussion Overall, HI may lead to epigenetic changes that manifest in the next generation as behavioral deficits. In all animals, fear is always a strong motivating factor in survival situations, and the resulting response can often confound any sensory-motor coordination impairments. For this reason, all animals were trained for five times a day for three days before the actual behavioral test. Assessing the progress and behavior of animals during the training period can also be a good indicator of memory impairment.
Our results show that in general male progeny are more sensitive to the epigenetic transfer of the behavioral deficits measured here that are associated with HI when compared to females. This may be attributed to the fragility of the Y chromosome and its susceptibility to modifications such as methylation (Hill and Fitch, 2012). Sex-specific neuroprotection has been found in numerous injury paradigms, including neonatal HI, in both rodent models and humans (Giordano et al., 2011; Hill and Fitch, 2012). This is consistent with our observation that there are significant deficits in second-generation males. A primary factor for cell death after HI is oxidative stress. It was recently reported that paraoxonase 2, a potent antioxidant, is expressed at 2× higher levels in females than males. This may also explain the gender differences in the manifestation of behavioral deficits. In addition, it is important to note that imprinting occurs at different stages based on gender, so perhaps at the stage of our injury model, males are more susceptible to brain injury than females (Wagner et al., 1995). In our data based on the training days for the Bar Holding Assessment (not shown), unimpaired animals learn by the test day. When compared to naive/shams, far fewer of the injured animals’ progeny learn by the test day. Another important result to note is that HI × HI progeny fared better than pups with only one HI parent. Further analysis is required before these results can be interpreted. Though they fared better at the Bar Holding Assessment, HI × HI progeny showed significant deficits in the mesh ascending assessment. Though this is primarily a motor coordination assay, this also indicates social behavior. For example, their uninterest and unwillingness to mingle after entering the box and mingling with
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
403
Sticky Dot Contact Latency
A. 10
Time (Seconds)
8
**
6
**
4 2
IM x
H IF H
C.
SD CL - Female
SD CL - Male
8
10 8
Time (Seconds)
6 4 2
4 2
IM x
H
H
IM
F
H
IF
x F
Sh am
IM
x
Sh am
H
M x F
IF
F ai ve
H
x F N
ai ve
ai ve
N
x H
N
N
ai ve
Sh am
M
IM H
IM H IF
x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F ai ve
F
M
Sh am
ai ve N x F N
ai ve
F
0
IM
0
N
**
6
x
Time (Seconds)
H
IM
F x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
N
ai ve
F
M
Sh am
ai ve N x F ai ve N
B.
IM
0
Fig. 6. Sticky dot assessment. (A) Sticky dot assessment of sensory-motor coordination in pups whose parents experienced HI. (B) Females. (C) Males. **p < 0.05 when compared to naive F × HI M. One way ANOVA, Tukey’s multiple comparisons test.
their siblings may indicate antisocial behavior resulting from the oxidative stress damage suffered by both parents. The limitations of our sticky dot test include the fact that animals cannot be trained before being assessed. Therefore, fear may still be a variable in these tests. However, fear would have led the animals to pay less attention to the stickers and more attention to finding an exit from their cage, but this was certainly not the case. Interestingly, however, we saw significant differences between sham F x HI M and naive F × HI M, and between HI F × HI M and naive F × HI M, which indicate that the HI male may also have an effect on progeny behavior. However, this was only seen in the sticky dot assessment and, as stated before, this test has its limitations. It would certainly be interesting to assess the epigenetic modifications in the male that appear to be transferred to another generation. This indicates that there may be at least a small amount of gonadal epigenetic modifications in the HI parent. As shams of our rodent HI model are exposed to anesthesia, we were curious about whether anesthesia affects sensory-motor coordination in the second generation as well. Therefore, we assessed naive pups and compared them to shams. As expected, there were minimal differences between progeny from naives and/or shams. Some crosses only had one litter assessed, so the results may be litter-specific. However, the pattern of behavioral deficits observed in pups from HI parents cannot be coincidental. Our behavioral results are consistent with HI affecting the germ line but lack direct evidence implicating HI as a transgenerational epigenetic process. For example, the observed behavioral deficits of the next generation could be due to epigenetic modifications arising from cross-talk among the nervous, endocrine, and reproductive
systems. There is evidence that throughout development, there is cross-talk among the nervous, endocrine, digestive, immune, and reproductive systems (Marchetti et al., 1990) and that the nervous system and endocrine system are seamlessly integrated and interact, especially in response to environmental stress (Dronca et al., 2012). Although according to our data, the impairment appeared to be due to HI effects on the primordial germ line, another important consideration is the impact of HI on female uterine environment, as well as the psychological effects on the female, which may affect parenting methods on both male and female offspring or selectively on one gender. It is well known that maternal neglect can lead to anxiety and stress issues in the progeny. This may be a factor to be considered in future studies, where one could assess the effects of postnatal maternal behavior after HI, by cross-fostering HI second generation pups to a healthy sham after birth. It would be interesting to study the effects of global neonatal hypoxia-ischemia on the second and third generation of animals. In the case of global HI, the second generation’s germ cells would be directly exposed to HI, therefore analysis of a third generation is necessary to prove epigenetic inheritance as opposed to direct epigenetic effects of global HI. We were able to observe the effects of HI in the first generation and relate it to indirect transfer of behavioral phenotype to the next generation. Assessments of DNA methylation patterns, histone modifications protein acetylation or genomic alterations introduced by DNA repair mechanisms reversins epigenetic DNA modifications after HI in first and second-generation animals of this model could confirm the presence and mechanisms of action of epigenetic influence
404
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405
Sticky Dot Total Time
A. Time (Seconds)
40 30 20 10
IM x
IF H
SD TT - Male 40
Time (Seconds)
40 30 20 10
30 20 10
IM x
H
H
IM
F
H
IF
x F
Sh am
IM
x
Sh am
H
M x F
IF
F ai ve
H
x F N
ai ve
ai ve
N
x H
N
N
ai ve
Sh am
M
IM H
IM H IF
x F
Sh am
H
IF
x
Sh am
H
M x
ai ve N
x F
N
ai ve
F
M
Sh am
ai ve N x F ai ve N
F
0
IM
0
x
Time (Seconds)
H
IM
F
H x F
Sh am
x IF H
C.
SD TT - Female
B.
Sh am
H
M x
ai ve
N
N
x F ai ve
F
M
Sh am
ai ve N x F ai ve N
IM
0
Fig. 7. Sticky dot assessment, total time, showed a larger spread of values with progeny of HI parents than from naive/sham progeny. (A) Total time assessments do not indicate any significant differences between crosses, even when separating data into female (B) and male (C) groups. One way ANOVA, Tukey’s multiple comparisons test.
on the inheritance of behavioral deficits observed in the second generation. These results also address the importance of male parent contributions to offspring beyond the classical gene base pair sequence-directed changes. The increased spread in outcomes observed in offspring behavior may also reflect developmental processes relative to chromatin configuration. Acknowledgements This work was supported by grant PO1 HDO039833 from NICHD and grant PT074693P9 from the DoD. References Boichot, C., Walker, P.M., Durand, C., Grimaldi, M., Chapuis, S., Gouyon, J.B., Brunotte, F., 2006. Term neonate prognoses after perinatal asphyxia: contributions of MR imaging, MR spectroscopy, relaxation times, and apparent diffusion coefficients. Radiology 239 (3), 839–848. Bockhorst, K.H., Narayana, P.A., Dulin, J., Liu, R., Rea, H.C., Wosik, H.K., Perez-Polo, J.R., 2010. Normobaric hyperoximia increases hypoxia-induced cerebral injury: DTI study in rats. Journal of Neuroscience Research 88 (April (5)), 1146–1156, PMCID:PMC 19885827. Denenberg, V.H., Whimbey, A.E., 1963. Behavior of adult rats is modified by the experiences their mothers had as infants. Science 142, 1192–1193. Diamond, M.C., Rosenzweig, M.R., Bennett, E.L., et al., 1972. Effects of environmental enrichment and impoverishment on rat cerebral cortex. Journal of Neurobiology 3 (1), 47–64. Dicou, E., Perez-Polo, J.R., 2009. Bax-an emerging role in ectopic cell death. International Journal of Developmental Neuroscience 27 (June (4)), 299–304, PMCID: PMC 19460623. Dronca, Roxana, S., Leontovich, Alexey, A., Nevala, Wendy, K., Markovic, Svetomir, N., 2012. Personalized therapy for metastatic melanoma: could timing be everything? Future Oncology 8 (11), 1401–1406.
Englander, E.W., Greeley Jr., G.H., Wanger Jr., G., Perez-Polo, J.R., Lee, H.-M., 1999. Hypoxia-induced mitochondrial and nuclear DNA damage in the rat brain. JNR 58, 262–269. Fabian, R.H., Perez-Polo, J.R., Kent, T.A., 2008. Perivascular nitric oxide and superoxide in neonatal cerebral hypoxia-ischemia. American Journal of Physiology: Heart and Circulatory Physiology 295 (October (4)), H1809–H1814, PMCID:PMC2593505. Fabian, R.H., Kent, T.A., Perez-Polo, J.R., 2004. Extracellular superoxide concentration increases following cerebral hypoxia but does not affect cerebral blood flow. International Journal of Developmental Neuroscience 22 (4), 225–230. Fabian, R.H., Perez-Polo, J.R., Kent, T.A., 2007. A decoy oligonucleotide inhibiting NFB binding to the IgGB consensus site reduces cerebral injury and apoptosis in neonatal hypoxic-ischemic encephalopathy. Journal of Neuroscience Research 85 (7), 1420–1426. Ferrari, D.C., Nesic-Taylor, O.B., Perez-Polo, J.R., 2010a. Oxygen resuscitation does not ameliorate neonatal hypoxia/ischemia-induced cerebral edema. Journal of Neuroscience Research 88, 2056–2065, PMCID:PMC 20143414. Ferrari, D.C., Nesic, O., Perez-Polo, J.R., 2010b. Perspectives on neonatal hypoxia/ischemia-induced edema formation. Neurochemical Research 35, 1957–1965, http://dx.doi.org/10.1007/s11064-010-0308-y, PMID:21136160. Gill, M.B., Bockhorst, K.H., Narayana, P.A., Perez-Polo, J.R., 2008. Bax shuttling after neonatal hypoxia-ischemia: hyperoxia effects. Journal of Neuroscience Research 86 (December (16)), 3584–3604, PMCID:PMC2585158. Gill, M.B., Perez-Polo, J.R., 2009. Bax shuttling after rotenone treatment of neuronal primary cultures: effects on cell death phenotypes. Journal of Neuroscience Research 87 (July (9)), 2047–2065, PMCID: PMC 19224578. Gill, M.B., Perez-Polo, J.R., 2008. Hypoxia ischemia-mediated cell death in neonatal rat brain. Neurochemical Research 33 (December (12)), 2379–2389, PMID 18408999. Giordano, G., Cole, T.B., Furlong, C.E., Costa, L.G., 2011. Paraoxonase 2 (PON2) in the mouse central nervous system: a neuroprotective role? Toxicology and Applied Pharmacology 256, 369. Grafe, M.R., Woodworth, K.N., Noppens, K., Perez-Polo, J.R., 2008. Long-term histological outcome after post-hypoxic treatment with 100% or 40% oxygen in a model of perinatal hypoxic-ischemic brain injury. IJDN 26 (February (1)), 119–124, PMCID:PMC2291019. Hill, C.A., Fitch, R.H., 2012. Sex differences in mechanisms and outcome of neonatal hypoxia-ischemia in rodent models: implications for sex-specific
S.K. Infante et al. / Int. J. Devl Neuroscience 31 (2013) 398–405 neuroprotection in clinical neonatal practice. Neurology Research International 2012. Hu, X., Nesic-Taylor, O., Qiu, J., Rea, H.C., Fabian, R., Rassin David, K., Perez-Polo, J.R., 2005. Activation of nuclear factor-B signaling pathway by interleukin1 after hypoxia/ischemia in neonatal rat hippocampus and cortex. Journal of Neurochemistry 93, 26–37. Hu, X., Qiu, J., Grafe, M.R., Rea, H.C., Rassin, D.K., Perez-Polo, J.R., 2003. Bcl-2 family members make different contributions to cell death in hypoxia and/or hyperoxia in rat cerebral cortex. IJDN 21 (7), 371–378. Xiaoming, H., Rea, H.C., Wiktorowicz, J.E., Perez-Polo, J.R., 2006. Proteomic analysis of hypoxia/ischemia-induced alteration of cortical development and dopamine neurotransmission in neonatal rat. Journal of Proteome Research 5 (9), 2396–2404. Marchetti, B., et al., 1990. Cross-talk communication in the neuroendocrinereproductive-immune axis. Age-dependent alterations in the common communication networks. Annals of the New York Academy of Sciences 594, 309. Martin, S.S., Regino Perez-Polo, J., Noppens, K.M., Grafe, M., 2005. Biphasic changes in the levels of poly(ADP-ribose) polymerase-1 and caspase 3in the immature brain following hypoxia-ischemia. IJDN 23 (8), 673–686. Perez-Polo, J.R., Reilly, C.B., Rea, H.C., 2011. Oxygen resuscitation after hypoxia ischemia stimulates prostaglandin pathway in rat cortex. IJDN 29 (6), 639–644, PM: 21514373. Qiu, J.-X., Grafe, M.R., Schmura, S.M., Glasgow, J., Kent, T.A., Rassin, D.K., Perez-Polo, J.R., 2001. Differential NF-B regulation of bcl-x gene expression in hippocampus and basal forebrain in response to hypoxia. Journal of Neuroscience Research 64, 223–234.
405
Qiu, J., Hu, X., Nesic, O., Grafe, M.R., Rassin, D.K., Perez-Polo, J.R., 2004. Effects of NFB oligonucleotide “Decoys” on gene expression in p7 rat hippocampus after hypoxia/ischemia. Journal of Neuroscience Research 77 (1), 108–118. Rice 3rd, J.E., Vannucci, R.C., Brierley, J.B., 1981. The influence of immaturity on hypoxic-ischemic brain damage in the rat. Annals of Neurology 9 (February), 131. Smith, K.E., Keeney, S., Zhang, L., Perez-Polo, R., Rassing, D.K., 2008. The association of early blood oxygenation with child development in preterm infants with acute respiratory disorders. IJDN 26 (February (1)), 125–131, PMCID:PMC2288779. Tchekalarova, J., Kubova, H., Mares, P., 2005. Postnatal caffeine exposure: effects on motor skills and locomotor activity during ontogenesis. Behavioural Brain Research 160 (1), 99–106. Tong, L.Q.T., Toliver-Kinsky, D., Rassin, K., Werrbach-Perez, J.R., Perez-Polo, 2003. Hyperoxia increases AP-1 DNA binding in rat brain. Neurochemical Research 28, 111–115. Vannucci, R.C., Lyons, D.T., Vasta, F., 1988. Regional cerebral blood flow during hypoxia-ischemia in immature rats. Stroke 19 (February), 245. Vannucci, R.C., Connor, J.R., Mauger, D.T., Palmer, C., Smith, M.B., Towfighi, J., Vannucci, S.J., 1999. Title Rat model of perinatal hypoxic-ischemic brain damage. Journal of Neuroscience Research 55 (2), 158–163. Wagner, A.K., Fabio, A., Puccio, A.M., Hirschberg, R., Zafonte, R.D., Marion, D.W., 1995. Gender associations with cerebrospinal fluid glutamate and lactate/pyruvate levels after severe traumatic brain injury. Critical Care Medicine 33 (2), 407–413. Zanelli, S.A., Stanley, D.P., Kaufman, D., 2008. Hypoxic–Ischemic Encephalopathy. http://emedicine.medscape.com/article/973501-overview