Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Gene-environment interactions in neurodevelopment
(USC Thesis Other)
Gene-environment interactions in neurodevelopment
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
GENE-ENVIRONMENT INTERACTIONS
IN NEURODEVELOPMENT
By
Hanke Heun-Johnson
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirement for the Degree
DOCTOR OF PHILOSOPHY
(NEUROSCIENCE)
August 2016
Copyright 2016 Hanke Heun-Johnson
ii
Acknowledgements
Many people have encouraged me during this educational journey, and I am forever
thankful for all the support that each and every one of you has provided.
First and foremost, I am extremely grateful to have Dr. Pat Levitt as my mentor.
Pat, thank you for teaching me to set the bar as high as possible, for your contagious enthusiasm
for science, for your endless patience and optimism, for supporting my venture into the science
policy field, and for creating a lab with the best colleagues I have ever had:
Allison Knoll, Anna Kamitakahara, Ryan Kast, Williams Rodriguez, Kathie Eagleson,
Piper Williams, Juli Wu, Jasmine Plummer, Lisa Padilla, Zhuihui Xie, Amanda Whipple,
Hojean Yoon, Rebecca Southern, Kevin Jiang, and Kiara Sanford. I am truly going to miss being
part of this team. In addition, everyone at the Saban Research Institute provided a very warm
welcome to our lab when we moved to Children’s Hospital several years ago, and I have been
lucky to become friends with so many of them.
I would like to thank (previous) lab members, collaborators and others for their
contributions to my project: Matthew Judson, Feng Wang, Inmaculada Ballesteros Yáñez,
Barbara Thompson, and Shenfeng Qiu for technical help with the cell injection setup; Mariel
Piechowicz for completing cell reconstructions; Hojean Yoon for assistance with spine analysis;
Kevin Jiang for completing the EPM head-dip analysis; Esteban Fernandez for confocal imaging
support; and Darryl Adams and Julie McGinn for animal care at UPC.
Furthermore, I would like to thank my Ph.D. committee members Drs. Ruth Wood,
Aaron McGee, and Bangyan Stiles, for their expertise, support, enthusiasm, and input on my
dissertation project.
iii
I have received tireless encouragement (and many care packages) from my parents
Johannes and Sjouktje, my sister Kim, my brother-in-law Jeroen, my friends in the Netherlands,
and my parents-in-law Janet and Carter. Thank you for all that you do.
I also want to thank all my friends here in Los Angeles for all the fun times together,
summiting mountains, exploring J-Tree and the Sierras, running, hiking, biking, enjoying beach
days, music festivals, ERYC events, happy hours, long nights around campfires, and road trips
up the 395.
Lastly, a special thanks goes to my husband, Tor. I would not have started and I would
not have been able to complete this Ph.D. journey without you. I am grateful for your
encouragement and support, for your excitement about science during our endless conversations,
and for our adventures exploring our amazing SoCal backyard and the rest of the world.
Funding came from the National Institutes for Mental Health (NRSA grant F31 MH100779-02)
iv
Table of Contents
Acknowledgements ii
List of Figures vi
List of Tables vii
List of Abbreviations viii
ABSTRACT 9
CHAPTER 1: Introduction 12
1.1. The (dis)advantage of the stress response 12
1.2. ELS effects on human behavior 14
1.3. Postnatal ELS paradigms for mice 18
1.4. Postnatal ELS effects on mouse brain structure and behavior 22
1.4.1. ELS effects on social-emotional behaviors in mice 23
1.4.2. ELS effects on neuronal morphology in mice 25
1.5. The role of MET in brain development and behavior 26
1.6. Gene-environment interaction effects on human mental health disorders 30
1.7. Gene-environment interaction effects in mice 32
1.8. Gene-environment interaction effects of ELS and Met
+/-
34
CHAPTER 2: Early-life stress paradigm transiently alters maternal behavior, 35
dam-pup interactions, and offspring vocalizations in mice
2.1. Abstract 35
2.2. Introduction 36
2.3. Methods 38
2.4. Results 47
2.4.1. ELS environment transiently alters dam nest entry behavior 47
2.4.2. ELS environment impacts dam nest entry behavior during off-nest periods 51
2.4.3. Dams experiencing ELS environment exhibit normal behaviors 54
after pup weaning
2.4.4. Mothers that had undergone ELS as pups exhibit normal nest entry behavior 56
2.4.5. ELS pups emit more ultrasonic and audible vocalizations 58
2.4.6. Vocalizations occur after nest entry by dam in ELS environment 62
2.5. Discussion 65
2.6. Supplementary material 72
v
CHAPTER 3: Interaction between early-life stress and reduced expression 76
of MET leads to morphological changes in hippocampo-amygdalar
neurons in mice
3.1. Abstract 76
3.2. Introduction 77
3.3. Methods 80
3.4. Results 91
3.4.1. MET protein levels are reduced in hippocampus of Met
+/-
mice 91
3.4.2. ELS has opposite effects on dendritic complexity in wild-type and Met
+/-
mice 92
3.4.3. Spine characteristics are unaffected by Met
+/-
or ELS 97
3.5. Discussion 99
3.6. Supplemental material 108
CHAPTER 4: Early-life stress alters social-emotional behaviors in mice 109
with reduced expression of MET receptor tyrosine kinase
4.1. Abstract 109
4.2. Introduction 110
4.3. Methods 113
4.4. Results 120
4.4.1. Contextual fear memory is impaired as a result of Met
+/-
genotype 120
4.4.2. Early-life stress reduces social interactions in Met
+/-
mice 122
4.4.3. Met
+/-
genotype reduces anxiety-like behaviors 124
4.5. Discussion 128
CHAPTER 5: Discussion and future directions 136
REFERENCES 143
vi
List of Figures
Chapter 2 Page
2.1 ELS environment induces an increase in the number of nest entries by the dam, but
does not affect total on-nest time
48
2.2 Limited bedding transiently decreases nest quality during the ELS period 50
2.3 Representative schematic of the dam’s location in relation to the nest in the home
cage
52
2.4 Limited bedding leads to a transient increase in short on- and off-nest bouts 53
2.5 Dams in limited bedding conditions from P2-P9 do not exhibit altered anxiety-like
and fear learning behaviors after pup weaning
55
2.6 Dams that have undergone ELS as pups do not show altered nest entry behavior in
adulthood when rearing pups
57
2.7 Representative spectrograms of recording of vocalizations 59
2.8 ELS pups emit more vocalizations when the dam is present on the nest during off-
nest periods
60
2.9 Vocalizations of ELS pups are increased immediately after dams enter the nest 63
S2.1 Top view of cage with dam and pups in ELS environment on P4 72
S2.2 Genotype composition of the litter does not affect the number of vocalizations on
P4
73
S2.3 The duration of long, uninterrupted on-nest bouts on P4 is similar for dams in a
control and ELS environment
74
S2.4 The percentage time spent by dams in specific locations (on-nest, off-nest, or on-
nest during an off-nest period) during USV/AV analysis varied per dam, but was
not significantly different between dams in a control and ELS environment
75
Chapter 3
3.1 MET protein expression is approximately 50% in hippocampus of P9 Met
+/-
mice
compared to control mice, and does not change after ELS
91
3.2 ELS has opposite effects in Met
+/-
mice on dendritic complexity of neurons in CA1
ventral hippocampus that project to the basolateral amygdala compared to wild-
type mice
93
3.3 ELS increases dendritic length of basal arbor in Met
+/-
mice 95
3.4 Representative examples of reconstructed CA1 neurons that project to the BLA,
organized by length of basal dendritic arbor
96
3.5 Spine characteristics are not significantly different as a result of ELS, Met
+/-
or a
combination of both factors
98
3.6 Overview of morphological changes due to the differential effects of ELS on
wild-type and Met
+/-
mice
103
S3.1 Locations of BLA tracer injection sites 108
vii
List of Tables
Chapter 2 Page
2.1 Statistical test details of maternal behavior measurements 49
2.2 Statistical test details of vocalization measurements 61
S2.1 Statistical test details of supplementary data 75
Chapter 4
4.1 Summary of behavior test results in experimental groups
128
Chapter 4
4.1 Contextual fear memory (‘test’) is impaired by Met
+/-
genotype 121
4.2 Early-life stress reduces number of social interactions in the six-minute direct
social interaction test
123
4.3 Met
+/-
genotype reduces anxiety-like behaviors on the elevated plus maze 125
4.4 Met
+/-
genotype decreases anxiety-like behavior as measured by unprotected head-
dips in the elevated-plus maze test
127
Chapter 5
5.1 ELS affects levels inflammatory markers in hippocampus and blood samples of adult mice 141
viii
List of Abbreviations
5-HTTLPR 5-Hydroxy-tryptamine transporter-linked polymorphic region
ACTH Adrenocorticotropic hormone
ADHD Attention deficit hyperactivity disorder
AMPA α-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
ASD Autism spectrum disorder
AV Audible vocalization
BLA Basolateral amygdala
CA1 or 3 Cornu ammonis area 1 or 3
CCR7 Chemokine receptor type 7
CNS Central nervous system
CORT Cortisol (humans) or corticosterone (rodents)
CRF Corticotrophin releasing factor
DSI Direct social interaction
EEG Electroencephalogram
ELS Early-life stress
EPM Elevated-plus maze
EZM Elevated-zero maze
GR Glucocorticoid receptor
GRE Glucocorticoid response element
GWAS Genome-wide association study
HC Hippocampus
HGF Hepatocyte growth factor
HPA Hypothalamic-pituitary-adrenocortical
MR Mineralocorticoid receptor
MS Maternal separation
MTHFR Methylenetetrahydrofolate reductase
NMDA N-Methyl-D-aspartic acid
P Postnatal day
PBS Phosphate buffered saline
PFC Prefrontal cortex
PTSD Post-traumatic stress disorder
PVN Paraventricular nucleus
SD Standard deviation
SEM Standard error of the mean
siRNA Short interference RNA
SNP Single nucleotide polymorphism
USV Ultrasonic vocalization
9
Abstract
Evidence from clinical studies has demonstrated a significant relation between early adverse
experiences, noted as early-life stress (ELS), and later risk for emotional, cognitive, and physical
health problems. However, many mental health disorders that have been associated with ELS
have a significant heritability component that is currently unexplained by results from genome-
wide association studies (GWAS). Interactions between multiple genetic and environmental
factors are thought to be partly responsible for this ‘missing heritability’. The MET gene is a
pertinent candidate that may interact with ELS to alter brain structure and behavior, as altered
expression of MET receptor tyrosine kinase protein in mice has shown to affect synaptogenesis,
neuronal growth and connectivity during early development, and behavior in adult mice.
Moreover, a common variant (rs1858830 ‘C’ allele) in the promoter region has been associated
with increased risk of autism spectrum disorder (ASD), as well as altered structural and
functional outcomes in brain imaging studies, which validates research into the role of MET in
brain development.
To study the question whether Met and ELS interact to have enduring effects on neuronal
morphology and behavior in mice, we used a limited bedding and nesting paradigm that
previously has been shown to induce a stress response in mouse pups, and result in lasting effects
on mouse brain structure and function. However, the mechanism by which this paradigm induces
ELS in mice is not completely clear. Extended temporal behavioral analyses of nest entry
behavior by dams and vocalizations by pups suggest that pups in ELS conditions experienced
aversive and painful stimuli due to altered nest entry behavior by the dam. The nest entry
10
behaviors by the dams due to limited resources were transient and not transmitted to the next
generation of females.
ELS was induced with this limited bedding paradigm in both wild-type pups and Met
+/-
pups,
which express 50% of normal MET receptor tyrosine kinase protein in the central nervous
system, and morphological characteristics were assessed in young adult brains. Analyses of
pyramidal neurons that are part of a network regulating complex behaviors that are affected by
ELS, showed that dendritic arbor complexity of neurons that project from the ventral CA1 region
of the hippocampus to basolateral amygdala was reduced in wild-type mice that had experienced
ELS, as well as in Met
+/-
mice raised in control conditions. However, ELS had the opposite effect
in Met
+/-
mice as in wild-type mice, as arbor complexity in these mice was increased compared to
non-stressed Met
+/-
mice, possibly due to the precocious maturational state of neurons as a result
of a reduction in MET protein. The effect on dendritic morphology appears to be selective, as
Met
+/-
, ELS, or the combination of these genetic and environmental factors did not alter spine
characteristics in these specific neurons.
Behavioral analyses of wild-type and Met
+/-
mice showed that Met genotype reduced anxiety-like
behavior and impaired contextual fear memory, while fear acquisition and extinction were
unaffected. Social behavior was impaired by ELS, specifically in Met
+/-
mice. These results
suggest that the combination of these genetic and environmental factors results in an increased
number of affected behavioral domains, and that Met
+/-
mice may be more sensitive to ELS.
11
In summary, the outcomes of this research project show that gene-environment interactions
between Met expression and ELS affect neuronal structure and behavior in mice. This body of
work creates a foundation for future research on the mechanism of this interaction. Furthermore,
the outcomes provide the first justification to study an interaction between MET rs1858830
genotype and early adverse experiences in the human population, and its effects on severity and
risk of mental health disorders.
12
Chapter 1:
Introduction
1.1 The (dis)advantage of the stress response
The stress response is a reaction in an organism to external or internal stressors to adapt to or
cope with a threatening situation at hand. A timely and brief stress response to a concrete danger
is favorable for survival, while extended and excessive stress as a result of a real or perceived
stressor is considered maladaptive. In addition, particularly during early childhood, tolerable
stress can turn into toxic stress when relationships are not available to provide social support
(Shonkoff et al., 2011).
The best characterized stress response mediators are hormones that are released by various
structures in the hypothalamic-pituitary-adrenocortical (HPA) axis. In the hypothalamus,
parvocellular neurons in the medial paraventricular nucleus (PVN) release corticotropin releasing
factor (CRF) and vasopressin into the portal vessel system in the anterior pituitary, subsequently
stimulating the release of adrenocorticotropic hormone (ACTH) into the bloodstream. ACTH
stimulates the release of corticosteroids (e.g. cortisol in humans, or corticosterone in rodents,
both abbreviated with CORT) from the adrenal cortex (Whitnall, 1993).
Corticosteroids exhibit functions as wide-ranging as mobilizing energy substrates and promoting
gluconeogenesis (Sapolsky et al., 2000), to regulating inflammatory mechanisms (McEwen et al.,
1997). When the stressor is removed, CORT plasma levels normally decrease. However, chronic
exposure to stressors results in reduced ability to maintain a normal diurnal glucocorticoid
13
rhythm, and affects cells throughout the body that express receptors for HPA-axis mediators,
including CORT and CRF (Shanks et al., 1995, Windle et al., 2008, Conway-Campbell et al.,
2007).
Corticosteroids also cross the blood-brain-barrier (Stumpf et al., 1989) to activate
mineralocorticoid receptors (MR) and glucocorticoid receptors (GR) in the brain. Several lines of
evidence have implicated the hippocampus (HC) and prefrontal cortex (PFC) as targets as well as
feedback modulators of the HPA-axis, due to high expression of GR and MR receptors (Pryce,
2008) and the observation that lesions or depletion of GR in these areas elevates CORT plasma
levels (Lupien and Lepage, 2001). This, however, also means that these brain areas are
structurally and functionally susceptible to elevated levels of stress hormones. For example,
corticosteroids alter gene expression, whereby MRs and GRs bind to corticosteroids and together
with co-repressors or activators interface with glucocorticoid response elements (GRE) in DNA.
Activation of these receptors in the hippocampus has been shown to alter expression of specific
genes that are involved in, for example, energy metabolism, signal transduction, protein
synthesis, immune responses, and circadian/ultradian transcriptional regulation (Datson et al.,
2001, Wiley et al., 2016, Galon et al., 2002). Additionally, immediate effects of CORT include
altering excitability of neurons by increasing glutamate release, and trafficking glutamate
receptors (NMDA and AMPA) to the postsynaptic membrane (Hascup et al., 2010, Yuen et al.,
2009, Karst et al., 2010, Karst and Joels, 2005).
Another critical hormone in the stress response, CRF, has been shown to affect excitability, long-
term potentiation and neuronal morphology, independent of its role in the release of ACTH from
14
the pituitary (Wang et al., 2011b, Regev and Baram, 2014). CRF is secreted locally by
interneurons, or by neurons in the PVN that project to other brain areas, and acts like a
neuropeptide on a time-scale that falls between the rapid excitability effects and slow genomic
effects of glucocorticoids (Chen et al., 2012). While CRF is a beneficial mediator of stress-
induced plasticity during short-term stress, prolonged exposure to CRF during chronic stress can
lead to detrimental effects on neuronal morphology (Chen et al., 2013)
Thus, an appropriate stress response is beneficial when stress response mediators are briefly
activated and the stressor can be dealt with appropriately, but prolonged exposure to elevated
levels of stress response mediators can induce damage in the same neurons in many brain areas
that regulate the stress response. Additionally, if toxic stress occurs during a time period when
this brain circuitry is being established, i.e. prenatally and during early childhood, subsequent
stressors could lead to deregulation of normal stress response mediator release and response in
later life.
1.2 ELS effects on human behavior
Increased brain plasticity during pre- and postnatal early life allows humans to gradually refine
an intricate network that is able to regulate behaviors in later life. While this plasticity is
beneficial when sensory input is appropriate, a lack of input (for example when a child is
severely neglected), or the presence of deleterious input (in the case of abuse) can be harmful for
the developing brain. ELS results in deregulation of mediators of the HPA-axis, both baseline
levels as well as response to a stressor (Gunnar and Vazquez, 2001, Carlson and Earls, 1997,
Carpenter et al., 2007, Cicchetti et al., 2010).
15
In addition, prenatal and postnatal ELS have been associated with increased risk of childhood
and adult-onset mental health issues.
Several specific postnatal stressors during the first years of life, or adverse childhood
experiences, have been identified to affect mental health in later life. Research questionnaires,
such as the Childhood Trauma Questionnaire (Spinhoven et al., 2014), commonly distinguish
between emotional, physical and sexual abuse, and emotional and physical neglect. Different
adversities often are co-occurring, and multiple adversities in single individuals are associated
with higher risk, earlier onset, and increased severity of mental health disorders in later life
(Green et al., 2010, Bright et al., 2015). As has been deduced from large studies with multiple
adversities and multiple health outcomes, and perhaps caused by frequent co-occurrence of
multiple adversities, specific health outcomes are not necessarily due to specific adversities
(Green et al., 2010, Kessler et al., 2010, Dube et al., 2004, Schilling et al., 2008). For example,
Kessler et al. (2010) and Green et al. (2010) showed associations between (physical and
emotional) abuse, (physical and emotional) neglect, parental substance abuse, parental mental
illness or parental criminality during early childhood, and a higher risk of mood, anxiety, and
substance abuse disorders in adulthood. Many subsets of these results have been reported by
other groups as well (Norman et al., 2012, Anderson et al., 2002, Chapman et al., 2004, Turner et
al., 2006). Abuse or neglect also has been associated with post-traumatic stress disorder (PTSD)
risk (Widom, 1999, Lang et al., 2008), attention deficit hyperactivity disorder (ADHD) (Ouyang
et al., 2008, Kessler et al., 2010), and higher levels of aggressive behaviors, and inflammatory
markers such as interleukin-6 and C-reactive protein (Fanning et al., 2015). Interestingly, a
recent study has shown that moderate stressors during childhood (for example, moving to a new
16
city, or frequent arguments with parents) in a cohort was associated with lower levels of
depressive symptoms after experiencing subsequent stressors during adolescence. Although
specific coping mechanisms were not addressed in this study, theses results suggest that a
tolerable level of stress may be able to help build resilience (Shapero et al., 2015).
A large study, the Bucharest early intervention project, has been ongoing for more than 15 years.
The goal of the study is to determine whether removal of a child from an adverse environment
can prevent or reverse detrimental health effects. The study is unique in that it is a randomized
controlled trial that aims to place institutionalized children with foster families when available,
and that compares these children with children who remain in institutional care, and community
controls (children who were never institutionalized). The results are stratified by the child’s age
of placement with the foster family. The outcomes show that institutionalized children who are
placed which foster families before the age of two generally show better recovery from severe
neglect compared to children who stayed in institutionalized care. This was true for example for
reactivity of the HPA-axis (McLaughlin et al., 2015a), intelligence quotient (Johnson et al.,
2010), attention bias towards positive stimuli (Troller-Renfree et al., 2015) neural activity
between brain regions as measured by electroencephalogram (EEG)-oscillations (Stamoulis et
al., 2015), social behaviors (Almas et al., 2015), and white matter integrity (Bick et al., 2015).
These and many other measurements will be continued in the future to generate valuable follow
up data on the detrimental effects of institutionalized care, and the health benefits of early
reversal of adverse situations.
17
In addition to postnatal stressors, prenatal stressors have been identified that can affect human
brain development and later life mental health (Class et al., 2014, Li et al., 2010, Pearson et al.,
2013, O'Donnell et al., 2014). The prevailing hypothesis is that the fetus is exposed to abnormal
glucocorticoid levels and inflammatory mediators that cross the placenta when the mother is
experiencing severe stress during the pregnancy (O'Donnell et al., 2009, Bilbo and Schwarz,
2009, Weinstock, 2005), and the current emphasis in research is mostly on the timing (specific
trimester) of the stressor (Kim et al., 2015). It is however difficult to discern whether a prenatal
stressor is purely prenatal or only during one specific trimester, as stressful situations could start
suddenly, but are often not resolved (and perhaps become worse) after childbirth. In addition,
many studies rely on self-reported and retrospective questionnaires, which may be inducing
recall bias specifically in mothers with children with health issues.
Regardless of these methodological difficulties, many groups have shown clear associations of
prenatal stressors with a variety of physiological outcomes in children. For example, being in a
war zone during pregnancy, or experiencing psychosocial stress, has been association with lower
birth weight and preterm birth (Maric et al., 2010, Copper et al., 1996, Wadhwa et al., 1993).
Psychosocial stress, including anxiety experienced during pregnancy, as well as negative life
events, is also associated with reduced performance on working memory tests (Entringer et al.,
2009), other cognitive development measures (Bergman et al., 2007, Gutteling et al., 2006), and
attention problems and ADHD in the offspring (Buitelaar et al., 2003, Van den Bergh and
Marcoen, 2004, Van den Bergh et al., 2005). In addition, ADHD diagnosis in children has been
shown to be associated with bereavement due to the loss of a close family member during
pregnancy (Class et al., 2014, Li et al., 2010). Bereavement can lead to increased risk of autism
18
spectrum disorders (ASD) as well (Beversdorf et al., 2005, Class et al., 2014), although other
studies have not observed an association between prenatal stress and ASD (Li et al., 2009, Rai et
al., 2012). Being exposed to a natural disaster during pregnancy was associated with increased
risk of ASD in the child as well (Kinney et al., 2008). Most aforementioned reports on increased
attention issues and ADHD diagnoses specifically acknowledged stressful events during the
second and third trimester, while preterm and lower birth weight in the majority of studies was
associated with third trimester stressful events. Stressors during the third trimester were also
associated with increased risk of ASD, although earlier trimesters, both first and second, were
additionally contributing to increased risk in the same studies. Many reports did not analyze or
mention a specific timing of the stressor during pregnancy, possibly due to the fact that stressors
are often not confined to a specific trimester.
1.3 Postnatal ELS paradigms for mice
Animal models have been developed to help determine the mechanisms by which pre- or
postnatal stressors affect later life behaviors in humans. With the ability to control genetic
background and use standardized housing conditions, it is relatively straightforward to
manipulate the timing, intensity, and type of stressors for mice. Controlling for confounding
factors (including recall bias) in human population studies on the other hand, is more
complicated.
One important factor to consider in determining the effects of stressors on brain development and
behavior is the relatively ‘early’ birth of a mouse in relationship to the maturational state of its
brain. As comprehensively reviewed by Semple et al. (2013), this effectively results in a
19
corresponding overlap of the last trimester of pregnancy of a human with the first postnatal week
in a mouse. At the start of the last trimester in humans, neuronal proliferation is essentially
complete, while neurite outgrowth and gliogenesis continue to occur throughout this period.
Around human birth, brain growth is starting to slow down, while synaptogenesis and
myelination rates are on the rise to peak at two and five years after birth, respectively, slightly
varying depending on brain region. In mice, these processes occur in a similar order, but -except
for neurogenesis- predominantly after birth, and on a compressed time scale (Semple et al.,
2013). For example, when a prenatal stressor during the last trimester of human pregnancy has
been shown to be associated with later life mental health outcomes, a postnatal stressor should be
applied during the first postnatal week in mice to more closely match the brain development
stage that is most affected.
Several postnatal stress paradigms have been developed to induce ELS in mice, for example the
maternal separation (MS) and limited bedding paradigms. In addition, while it is arguably not
considered a ‘true’ stressor, natural variation in maternal care can also provide a continuous scale
of early postnatal stress, from which we can indirectly deduct the effects of optimal versus
suboptimal maternal care. The rationale behind the latter paradigm is that within each rodent
strain, individual dams exhibit differences in the percentage time licking, grooming, and nursing
of their pups in an active arched-back position, that are correlated to the pups’ stress response
(Liu et al., 1997). The pups from dams that exhibit extremely low or high maternal care can
subsequently be selected for follow-up studies. While this paradigm is occasionally used for
mice (Pedersen et al., 2011), it is more often used in rats. Despite its proven usability in the rat
(van Hasselt et al., 2012), both species exhibit large variability and fluctuations in licking and
20
grooming behaviors, and while differences between strains may be easily detected, large within-
strain differences may require long and/or repeated observations (Champagne et al., 2007).
The most widely used postnatal ELS method is MS of the pups from the dam for several (but
most often three) hours per day, during the first one or two postnatal weeks (de Kloet et al.,
2005, Levine, 2005). Pups are taken out of the home cage, and kept either together in a huddle,
or separated from each other when separated from the dam, with the latter being a more severe
stressor than the former. The temperature is often artificially controlled to prevent hypothermia
during the separation. A longer separation period from the mother is called maternal deprivation,
during which pups are separated from their mother for a single 24-hour period. While the
maternal deprivation protocol provides a more precise control over the timing of the stressor in
relation to brain maturational state compared to repeated maternal separation, other factors such
as dehydration and starvation can possibly become confounders in this specific paradigm.
On the other hand, very short separation periods for approximately 15 minutes per day, called
brief maternal separation or handling, are often used as either a secondary control group besides
standard facility rearing, to control for mouse handling during separation protocols, or to actively
decrease the stress response in the pups. This is based on the observation that brief maternal
separation can lead to increased on-nest time (Millstein and Holmes, 2007) and increased licking
and grooming of pups immediately after the pups are placed back with the dam (Kosten and
Kehoe, 2010), since the pups are displaying more ‘demanding behaviors’ such as suckling
(Pereira and Ferreira, 2006). It has been shown that this increase in maternal care results in a
decrease in stress response in the pup after brief maternal separation. However, longer maternal
21
separation cannot be compensated for by increased maternal care. MS is successful in inducing a
stress response and leads to enduring impairments in the pups (Plotsky et al., 2005, Lippmann et
al., 2007, Sanchez et al., 2001), however, models that reduce contact between animals and
handlers during periods of ELS, such as the limited bedding paradigm described below, have
technical and practical advantages.
The limited bedding paradigm during early postnatal periods has been developed more recently,
and results in qualitatively different maternal care, rather than ‘neglect’ due to MS or maternal
deprivation. The limited bedding paradigm, which was first developed for rats, leads to a change
in maternal behavior, and altered stress response and behavior of offspring in later life (Avishai-
Eliner et al., 2001, McLaughlin et al., 2015b, Ivy et al., 2008, Dubé et al., 2015, Cui et al., 2006).
Mice are housed in cages with a wire mesh elevated above the cage floor, to allow for fecal
pellets to fall through, and urine to be absorbed by a thin layer of bedding on the cage floor. The
dams are given a partial nestlet square to build a suboptimal nest, and the dam and pups are
placed in this condition together, generally from postnatal day (P)2 to P9 (Rice et al., 2008,
Molet et al., 2014, Gilles et al., 1996, Schmidt et al., 2011, Ivy et al., 2008). The limited bedding
paradigm has several advantages over MS; it does not require interaction of research personnel
with the dams and pups, while providing a continuous stressor (Gilles et al., 1996, Molet et al.,
2014), and without interrupting feeding patterns of the pups. Rat dams have been shown to
display ‘abusive’ behavior (for example, stepping on, and rough handling pups) in a limited
bedding condition (Raineki et al., 2012) and by reducing nursing and maternal care (Ivy et al.,
2008). Rat pups vocalize more in limited bedding conditions (Raineki et al., 2012). In mice,
maternal care is thought to be similar in quality (as measured by total on-nest time and quality of
22
maternal care), however, it is thought to be ‘fragmented’ as a result of limited bedding (Baram et
al., 2012, Rice et al., 2008). Several groups have reported that the frequency of nest exits
increases (Rice et al., 2008, Baram et al., 2012, Malter Cohen et al., 2013, Gunn et al., 2013,
Yang et al., 2015, Naninck et al., 2015), whereas the total duration of maternal care generally
does not change as a result of limited bedding (Gunn et al., 2013, Naninck et al., 2015, Rice et
al., 2008). Most research groups have only performed short-term interval-based observations,
and presumably without allowing dams to resume a normal nest-entry pattern before starting
observations.
In this dissertation project, I use the limited bedding paradigm to induce ELS in genetically
modified mice. To validate the effectiveness of the paradigm, as well as to determine in more
detail the mechanism behind stress induction in this paradigm, I perform long (four hour)
observations of dam behavior, and recording ultrasonic and audible vocalizations in pups in
relation to the dam’s behaviors. The male pups that have experienced ELS are subsequently used
to determine whether ELS in our genetically modified mice, as discussed in more detail below,
results in interactive effects on neuronal morphology and behavior in young adult mice.
1.4 Postnatal ELS effects on mouse brain structure and behavior
Postnatal ELS paradigms in rodents have been shown to generate long-term changes in stress
response circuitry and downstream biological and behavioral outcomes (van der Kooij et al.,
2015, Kohl et al., 2015, Wang et al., 2011b, Rice et al., 2008, Plotsky et al., 2005, Lippmann et
al., 2007, Loi et al., 2014, Sanchez et al., 2001). In the following section I summarize research
23
that has been done on the effects of ELS on social-emotional behavior and neuronal morphology
in mice.
1.4.1 ELS effects on social-emotional behaviors in mice
As outlined above, ELS is associated with higher risk of anxiety- and mood disorders and PTSD
in humans. ELS in mice has shown to affect social-emotional behaviors related to these
disorders, for example anxiety-like behaviors, social interaction, and fear learning and memory.
While anxiety-like behavior as measured on the elevated plus maze (EPM), is often unchanged
due to ELS, social interactions and recall of fear memory are generally impaired in adult mice
that have experienced ELS as pups.
The effects of ELS on behavior in adult mice have been studied by many groups using the
elevated plus maze (EPM) or elevated-zero maze (EZM), whereby more time spent in open
sections of the maze is indicating a decrease in anxiety-like behavior. Overall, ELS has shown to
be somewhat ineffective in changing anxiety-like behavior as many groups have measured no
change in duration spent in the open arm of an EPM (van der Kooij et al., 2015, Zoicas and
Neumann, 2016, Venerosi et al., 2003, Wang et al., 2012, Niwa et al., 2011, van Heerden et al.,
2010, Liu et al., 2016) and the open section of the EZM (Harrison et al., 2014). However,
increased anxiety-like behavior (Levine et al., 2012, Mehta and Schmauss, 2011, Bouet et al.,
2011, Veenema et al., 2007, Shin et al., 2016), or decreased anxiety-like behavior (Savignac et
al., 2011, Fabricius et al., 2008) after ELS in adult mice have also been reported.
24
Common behavior tests to assess learning, memory and extinction of a fearful event, which are
reliant on activity in brain structures such as the amygdala, are contextual and cued fear
conditioning. In these tests, a specific context or a cue (most often a tone), respectively, is
associated with foot shocks during the training phase. The memory of the association is tested
the following day by measuring the amount of immobility (freezing) while the animal is exposed
to the context or cue without the foot shock. Extinction consists of repeatedly presenting the
context or tone alone during the following days or weeks, and measuring freezing duration
during these sessions as well. ELS due to limited bedding and MS impair both contextual and
cued fear memory in adult mice. Specifically, Wang et al. (2011a), Niwa et al. (2011) and
Kanatsou et al. (2016) reported that ELS reduced freezing when the animals were placed back
into the associated context 24 hours after training. Freezing during the cue was also reduced
(Wang et al., 2011a) or unchanged (Niwa et al., 2011, Zoicas and Neumann, 2016). However,
ELS did not affect the formation of the fear memory during training, nor the extinction of the
fear memory in two recent studies using a contextual (Kanatsou et al., 2016) and cued (Zoicas
and Neumann, 2016) fear conditioning paradigm in mice.
Social behavior is generally impaired after ELS due to limited bedding, as social interaction time
is decreased in a direct social interaction (DSI) test or three-chamber sociability test in adulthood
after ELS (Santarelli et al., 2014, van der Kooij et al., 2015, Raineki et al., 2012, Raineki et al.,
2015). Similarly, MS also has shown to impair social behavior in mice (Tsuda and Ogawa, 2012,
Bouet et al., 2011, Niwa et al., 2011, Venerosi et al., 2003, Tsuda et al., 2014), although several
groups have observed no decrease in social interaction measurements in mice (Harrison et al.,
2014, Franklin et al., 2011, Zoicas and Neumann, 2016, Tsuda et al., 2011).
25
Thus, in general, ELS in mice impairs social behavior, while the effects on anxiety-like behavior
as tested on the EPM is inconclusive, with most groups reporting no effects of ELS, and others
either a decrease or an increase of anxiety-like behavior. Furthermore, although very few studies
have been published on the effects of ELS on fear learning and memory, fear memory is reported
to be impaired after ELS in mixed cued/contextual fear conditioning paradigms.
1.4.2 ELS effects on neuronal morphology in mice
Altered neuronal network functioning, due to changes in neuronal connectivity and/or activity, is
generally thought to be responsible for altered behavior outcomes. The mechanisms behind the
lasting effects of ELS are due to both epigenetic changes, such as DNA methylation, that alter
protein expression (Suderman et al., 2014), as well a permanent adjustment of the course of
structural brain development, for example by morphological changes in pyramidal neurons of the
neocortex and hippocampus (Yang et al., 2015, Liao et al., 2014, Ivy et al., 2010, Brunson et al.,
2005). Specifically, ELS due to limited bedding during the first postnatal weeks of rats and mice
has been shown to induce a reduction in dendritic complexity and length of pyramidal neurons in
cortical areas (Yang et al., 2015) as well as the dorsal hippocampus (Ivy et al., 2010, Brunson et
al., 2005, Liao et al., 2014). MS of mouse pups during the first postnatal weeks had similar
effects as limited bedding (Romano-Lopez et al., 2015, Chocyk et al., 2013, Bock et al., 2005,
Eiland and McEwen, 2012), although application of the stressor in older pups (P14-16) induced
an opposite effect – an increase of dendritic length of cortical and hippocampal neurons
compared to earlier stressors (Bock et al., 2005, Xie et al., 2013). ELS also alters spine
morphology and density. Most studies report a decrease in spine density in hippocampal (Liao et
al., 2014, Wang et al., 2011b, Wang et al., 2013) and cortical neurons (Yang et al., 2015, Bock et
26
al., 2005), although some have observed an increase as a result of applying the stressor from P14
to P16 (Bock et al., 2005, Xie et al., 2013) instead of the first postnatal week. The results
generated in slightly older pups suggest that ELS in older pups leads to opposite effects on
dendritic morphology and spine density, possibly due to a more mature state of neurons.
1.5 The role of MET in brain development and behavior
Altered maturation of neurons has also been shown by modifying expression of several genes
involved in neuronal development, such as Met (Qiu et al., 2014). This gene encodes for the
MET receptor tyrosine kinase, and a reduction in MET leads to precocious maturation of
structural and functional neuronal characteristics in mouse brain (Qiu et al., 2014). In this
context, it would be very interesting to research whether different levels of MET expression
leads to altered brain structure and function in response to ELS, especially considering the role of
MET in synaptogenesis and neuronal growth as described in more detail below. Furthermore, a
genetic variant in the promoter of MET in humans has been associated with changes in brain
structure and function (Rudie et al., 2012), and increases risk of ASD (Campbell et al., 2006).
Thus, these experiments could possibly provide more insight in whether genetic variants in MET
in the human population are involved in resilience or sensitivity to ELS regarding mental health
outcomes.
MET receptor tyrosine kinase is present in brains of the developing mouse and non-human
primate in hippocampus, neocortex and other forebrain structures (Judson et al., 2011a, Judson et
al., 2009), and specifically enriched in developing synapses (Eagleson et al., 2013). Expression
of MET in mouse brain increases dramatically around birth, peaks in the second postnatal week,
27
and gradually decreases to a lower level in adulthood (Judson et al., 2009). Our lab identified a
single nucleotide polymorphism (SNP) in the promoter region of the human MET gene
(rs1858830) that significantly increases risk of ASD due to the presence of one or two ‘C’ alleles
instead of a ‘G’ allele (Campbell et al., 2006). Furthermore, brain imaging studies have shown
that this SNP is associated with altered structure and function in typically developing people.
For example, cortical thickness in parts of the temporal and frontopolar cortex, pre- and
postcentral gyri, and anterior cingulate is decreased in people with one and two ‘C’ alleles in a
dose-dependent manner (Hedrick et al., 2012). Additionally, white matter integrity was
significantly impaired in people with ‘CC’ genotype compared to the ‘GG’ group, and people
with a ‘CC’ genotype on average responded with atypical neural network activation when
looking at emotional faces. These effects were stronger in people with ASD (Rudie et al., 2012).
In post-mortem cerebral cortex of people with a diagnosis of ASD (regardless of genotype), as
well as of typically developing people with a ‘CC’ genotype compared to a ‘GG’ genotype, MET
protein is reduced (Campbell et al., 2007). These data suggest that a combined genetic risk of
reduced MET expression with other non-heritable risk factors will alter the molecular and
structural integrity of social-emotional circuits.
To help elucidate the mechanisms behind the effects of reduced expression of MET in humans,
our lab has developed several genetically altered mice that either express no MET, or have
reduced expression of MET in (parts of) the central nervous system. Complete ablation from the
dorsal pallium, a developmental structure that gives rise to adult brain areas that control many
social and emotional behaviors, results in hypoactive behavior, while heterozygous mice appear
unaffected. In contrast, heterozygous mice expressing approximately 50% of MET protein in the
28
complete central nervous system seem more affected than their homozygous counterparts: for
example, these mice show impaired fear learning and memory in a cued fear conditioning test,
and a mild reduction in anxiety-like behavior on the EPM. Homozygous mice showed a slight
reduction in sociability in the three-chamber test, while the heterozygous mice were no different
than wild-type mice in social behavior tests (Thompson and Levitt, 2015). Interestingly, Okaty et
al. (2015) conditionally deleted MET from serotonergic neurons in the dorsal raphe, which
resulted in reduced sociability in a three-chamber test. These results shows that complete
deletion of MET from a subset of neurons has different effects on behavior than partial deletion
in the complete central nervous system, suggesting that these different but overlapping subsets of
neurons are affected by relative levels of MET in connected and associated circuits.
Our lab has also determined structural changes in mice with complete deletion of Met from the
dorsal pallium. Judson et al. (2010) observed reduced apical dendritic complexity in layers II/III
and V of the anterior cingulate cortex, and an increase or no change in basal dendritic complexity
in basal arbors of layer II/III and V, respectively. The spines on the basal arbors of these neurons
are increased in volume, whereas the spine density and spine length remain stable (Judson et al.,
2010). Reduced Met expression during early development in vivo using short interference RNA
(siRNA) in the CA1 region of the dorsal hippocampus, reduces dendritic complexity and length
of apical arbors, decreases spine density and increases spine head volume on these arbors (Qiu et
al., 2014). In the same study, overexpression of MET resulted in opposite effects, that is, an
increase in apical arbor length and complexity, and an increase in spine density and decrease in
spine head volume. Previous studies by other labs have shown that reduction of MET
phosphorylation by removing its ligand, hepatocyte growth factor, from in vitro cortical slice
29
cultures from mice, decreases dendritic branching and length of pyramidal neurons (Gutierrez et
al., 2004). Likewise, knocking down MET expression with siRNA in dissociated primary
cultures from hippocampus results in reduction of dendritic length of pyramidal neurons (Lim
and Walikonis, 2008). Thus, a reduction of MET protein generally leads to reduction of dendritic
branching in several types of neurons. Subsequent studies into the effects homozygous and
heterozygous MET knockout mice show that these mice show altered interlaminar excitatory
drive in the neocortex (Qiu et al., 2011).
As mentioned above, Qiu et al. (2014) also demonstrated that MET is important for functional
maturation of neurons, a hypothesis that was first raised by Judson et al. (2011b). Reduction or
ablation of MET from CA1 neurons during development showed precocious expression levels at
the synapse of AMPA and NMDA receptor subunits and more mature electrophysiological
properties during the second postnatal week, that are normally seen in wild-type mice during the
fourth postnatal week (Qiu et al., 2014). These results suggest that reduced expression of MET
results in precocious neuronal maturation, which could determine how sensitive neurons are to
environmental impacts such as ELS, based on MET expression levels. Such a gene-environment
interaction may affect resilience to ELS regarding mental health outcomes.
30
1.6 Gene-environment interaction effects on human mental health disorders
Genome-wide association studies (GWAS) have been extremely helpful in determining genetic
variants or SNPs that are associated with increased risk of mental health disorders. For many
disorders, such as ASD and schizophrenia, SNPs have been identified that increase risk, although
the effect size of each of these is small. For others disorders, for example major depressive
disorder, it has been much harder to identify SNPs (Sharma et al., 2015) although the largest
studies may still be underpowered (Cross-Disorder Group of the Psychiatric Genomics
Consortium, 2013). Twin and family studies however, have found a significant heritability
component to developing depression (Sullivan et al., 2000) and other disorders. When SNPs
have been identified that contribute to increased risk of disorders, the contribution of these
variants in addition to rare mutations, does not explain heritability fully. The gap between these
types of studies, as well as the findings that environmental factors play a large role in the
etiology of many mental health disorders, are one of the driving forces behind the hypothesis that
gene-environment, as well as polygenic, interactions will explain part of this so-called ‘missing
heritability’ (Duncan et al., 2014, Sharma et al., 2015).
In a recent paper, Kim and Lee (2016) comprehensively reviewed different genetic variants that
have been identified and interact with ELS to either increase risk of mental health disorders, or
affect the onset or severity of a disorder. An example of a gene polymorphism that interacts with
ELS to affect the severity of a disorder is the gene encoding for methylenetetrahydrofolate
reductase, MTHFR; one or two ‘T’ alleles at rs1801133 leads to a quicker recurrence of
depression symptoms during follow-up of a cohort of patients with recurrent major depressive
disorder who experienced ELS in childhood (Lok et al., 2013). Another example is a
31
polymorphism in the FKBP5 gene, which in combination with ELS conveys a higher risk of
more severe symptoms when diagnosed with PTSD (Binder et al., 2008). Examples of genetic
variants that interact with ELS to increase risk of a disorder in the general population are in the
BDNF, NR3C1, and CRHR1 genes to increase risk of major depressive disorder (Gatt et al.,
2009, Bet et al., 2009, Bradley et al., 2008), in the MAOA gene to increase anti-social behavior
(Caspi et al., 2002), and in the COMT gene to increase risk of anxiety disorders (Baumann et al.,
2013). In addition, after a well-known study by Caspi et al. (2003), many reports have been
published on the effects of genetic variants in the 5-hydroxy-tryptamine transporter-linked
polymorphic region (5-HTTLPR) on depression. However, Duncan et al. (2011, 2014) have
scrutinized these results, and conclude that many reports on this and other gene-environment
interactions (including several that are mentioned above) are showing a high false discovery rate
of novel interactions due to a candidate gene-environment approach, as well as a very heavy
publication bias of follow-up studies. Their general recommendation is to initiate gene-
environment studies using genes that have already been shown to increase risk of a disorder in
multiple GWAS, and subsequently identify environmental factors that can increase or alter the
association between these disorders. This however, may prevent identification of a SNP that
interacts with a common environmental factor that has a completely opposite effect on different
genotypes, since the SNP association with a certain outcome may have be diluted in the larger
population that has not been stratified by the environmental factor. Nonetheless, it is favorable to
base candidate gene-environment studies on SNPs that have been shown to increase risk of
disorders or traits in the general population, rather than to find a SNP in the gene of a favorite
protein that has shown to only be involved in a potentially relevant biological process based on
the arguments provided by Duncan et al. (2014).
32
1.7 Gene-environment interaction effects in mice
Similar to human studies, gene-environment interaction research in mice is either based on
existing risk genes from GWAS, or genes/proteins involved in a biological process related to the
stress response or brain development. I will highlight several studies that have measured
interaction effects between several genes and ELS on neuronal morphology or relevant
behavioral outcomes.
While ELS and genetic manipulations alone have been known to alter neuronal morphology, not
many research groups have combined both independent variables to assess changes in neuronal
morphology. Wang et al. (2011b) first showed that knocking out CRF in the mouse forebrain
reversed a reduction in spine density on apical dendrites in dorsal CA3 area of the hippocampus
that was observed as a result of ELS and CRF knock out alone, confirming earlier reports that
CRF is involved in stress-induced remodeling of neurons (Brunson et al., 2001). The same group
showed that nectin-3 overexpression similarly restored a reduced spine density in CA3
hippocampus resulting from ELS (Wang et al., 2013). In the latter study ELS downregulated
nectin-3 via activation of the CRF receptor, suggesting that either preventing the activation of the
CRF receptor by knocking out CRF, or supplementing reduced nectin-3 expression can prevent
or reverse ELS-induced changes in spine density. Only the first study also included behavioral
results, which suggested that the CRF knockout mice were not as sensitive to ELS regarding
impairments in short-term and long-term spatial memory as tested with a Y-maze and Morris
water maze, respectively.
33
Regarding social-emotional behavior outcomes, gene-environment interactions involving ELS
have been studied more widely, with genetically altered mice generally showing either a lack of
response or increased sensitivity to ELS. An example of the latter situation, Harrison et al.
(2014) showed that either MS or knockout of CCR7 (chemokine receptor type 7) did not alter
normal preference for social novelty, while the combination of both factors lead to a blunting of
that preference. In another study, Sachs et al. (2013) observed an increase of anxiety-like
behavior as measured in the open field, due to either ELS or a knock-in mutation in Tph2
(tryptophan hydroxylase 2). ELS, however, was not able to induce the same behavioral change in
Tph2 knock-in mice. The same group also found a main effect on contextual fear conditioning as
a result of MS, while genetically altered Tph2 knock-in mice responded to ELS in a similar way
as wild-type mice (Sachs et al., 2013). A lack of response to ELS in genetically altered mice has
also been observed by Ognibene et al. (2007) who showed that wild-type pups that had just
experienced ELS show reduced social motivation, while this result is not observed in Reln
knockout mice that lack expression of reelin.
More complex interaction effects between different mouse genotypes and ELS also have been
found, for example on contextual fear memory after overexpression of the mineralocorticoid
receptor (Kanatsou et al., 2016), and on social investigation after knocking out the estrogen
receptor-beta (!ERKO)(Tsuda et al., 2014).
34
1.8 Gene-environment interaction effects of ELS and Met
+/-
The goal of this dissertation project is to determine the effects of a gene-environment interaction
between Met and ELS on brain structure and function. Specifically, Met
+/-
mice (with 50%
expression of MET protein in the central nervous system) and wild-type mice will be subjected to
ELS using a limited bedding and nesting ELS protocol, and I will measure neuronal morphology
characteristics of specific projection neurons from the ventral CA1 of the hippocampus to the
basolateral amygdala (Chapter 3), as well changes in several social-emotional behaviors
(Chapter 4). In addition, whereas the effectiveness of this ELS paradigm has been proven, the
stress-inducing mechanism is only partly known; therefore I will first perform analyses of dam
and pup behavior to gain more insight into the pups’ experience during this ELS paradigm
(Chapter 2).
The results obtained in this dissertation project will provide insight into whether reduced
expression of MET in mice, encoded by a gene that is associated with a higher risk of ASD and
altered brain structure and function in humans, results in an altered response to ELS regarding
neuronal morphology and behavioral measures. A differential response in Met
+/-
mice provides a
basis for studying the precise mechanism through which an altered response occurs.
Furthermore, it would provide a foundation to research whether the presence of a ‘C’ or ‘G’
allele at rs1858830 in the MET promoter in the human population alters susceptibility to ELS,
and whether this common variant interacts with ELS to increase risk or severity of mental health
disorders.
35
Chapter 2:
Early-life stress paradigm transiently alters maternal behavior,
dam-pup interactions, and offspring vocalizations in mice
2.1. Abstract
Animal models can help elucidate the mechanisms through which early-life stress (ELS)
has pathophysiological effects on the developing brain. One model that has been developed for
rodents consists of limiting the amount of bedding and nesting material during the first postnatal
weeks of pup life. This ELS environment has been shown to induce ‘abusive’ behaviors by rat
dams towards pups, while mouse dams have been hypothesized to display ‘fragmented care’.
Here we show, using long observation periods of C57Bl/6J mice, that dams with limited
resources from postnatal day (P)2 to P9 preserved regular long on-nest periods, and instead
increased the number of discrete dam-pup interactions during regular off-nest periods.
Immediately after dams entered the nest during off-nest periods in this ELS environment, pups
responded to these qualitatively different interactions with an increased number of ultrasonic
vocalizations (USV) and audible vocalizations (AV), communication signals that have been
associated with aversive and painful stimuli. After returning to control conditions, nest entry
behaviors normalized, and dams did not show altered anxiety-like or contextual fear learning
behaviors after pup weaning. Furthermore, female mice that had experienced ELS as pups did
not show atypical nest entry behaviors in control conditions in adulthood, suggesting that these
specific maternal behaviors are not learned during the ELS period. The results suggest that
36
atypical responses of both mother and pups during exposure to this ELS environment likely
contribute to long-term negative outcomes, and that these responses more closely resemble the
effects of limited bedding on rat dams and pups than was previously suggested. Discerning how
different early-life stressors mediate changes in maternal-pup interactions can help elucidate the
mechanisms of ELS on brain development and behavior.
2.2. Introduction
Accumulating evidence from both prospective and retrospective clinical studies has
demonstrated a significant relation between early adverse experiences, noted as early-life stress
(ELS) or toxic stress, and later risk for cognitive, emotional and physical health problems
(Shonkoff and Levitt, 2010, Shonkoff, 2012, McEwen and McEwen, 2015, Moffitt, 2013). While
long on clinical description, major challenges remain in determining the mechanisms through
which ELS has both immediate and long-term pathophysiological effects. These include
difficulties in determining the extent to which an individual must be exposed to ELS to elicit
long-lasting responses, the influence of genetic factors in modulating the response to ELS,
sensitive periods during which exposure has its most powerful impact on developing biological
systems, and the factors that influence individual differences in response to ELS.
Animal models provide an opportunity to address biological mechanisms underlying the impact
of ELS (Lyons et al., 2010, Moriceau et al., 2010, Joëls and Baram, 2009, Molet et al., 2014).
Several models have been developed in rodents to induce ELS in a controlled fashion. Methods
include separating pups from their dam for various periods of time during postnatal periods
37
(maternal separation) (de Kloet et al., 2005, Levine, 2005), and the introduction of limited
nesting materials and bedding during early postnatal periods (Rice et al., 2008, Molet et al.,
2014, Gilles et al., 1996, Schmidt et al., 2011, Ivy et al., 2008). Reports document that both
maternal separation and limited bedding generate long-term changes in stress response circuitry
and downstream biological and behavioral outcomes (van der Kooij et al., 2015, Kohl et al.,
2015, Wang et al., 2011, Rice et al., 2008, Plotsky et al., 2005, Lippmann et al., 2007, Loi et al.,
2014, Sanchez et al., 2001). Because the reproducibility of obtained rodent behavior can be
impacted significantly by human factors (Sorge et al., 2014), models that reduce contact during
periods of ELS may have technical, as well as practical advantages.
Limited bedding during early postnatal periods has been a more recent addition to research
focusing on ELS. This paradigm, first developed for rats, leads to altered maternal behavior that
ultimately correlates with disruption of offspring stress hormone response, brain excitability, and
adult behavior (Avishai-Eliner et al., 2001, McLaughlin et al., 2015, Ivy et al., 2008, Dubé et al.,
2015, Cui et al., 2006). Rat dams respond to limited bedding by displaying ‘abusive’ behavior
(stepping on, and rough handling pups) (Raineki et al., 2012) and by reduced nursing and
maternal care (Ivy et al., 2008). Rat pups respond with increased vocalizations (Raineki et al.,
2012). In mice, maternal care is thought to be qualitatively unaltered (in total on-nest time and
quality of maternal care), albeit fragmented, as a result of limited bedding (Baram et al., 2012,
Rice et al., 2008).
As a forerunner of future gene by environment (G x E) experiments, the present study was
designed to understand more fully how limited resources in the home cage of mice alter normal
38
behavior of the dam, and how vocal responses of mouse pups change as a result of these
behavioral alterations. Measures of nest entry behaviors were obtained in the home cage, and
fear learning and anxiety-like behavior were assessed in the dam to examine potential long-
lasting impact on dam behavior that could induce a residual stress response in pups after ending
the ELS period. The data suggest that the pups are affected by, and responsive to, atypical
physical contact with the dam, and that ELS is induced through atypical contact in this limited
resources model in mice.
2.3. Methods
2.3.1. Animals
All animal procedures were approved by the Institutional Animal Care and Use Committee at the
University of Southern California. C57Bl/6J mice were housed in a temperature- and humidity-
controlled vivarium (20-22 °C, 40-60% humidity) that was maintained on a 12-hour light/dark
cycle (with lights on at 6 am), in standard ventilated JAG mouse cages (Allentown Inc., NJ) with
ad libitum access to regular rodent chow and filtered water in drinking bottles. Cages were
cleaned weekly, with ample Alpha-Dri bedding (Shepherd Specialty Papers, MI) and one
standard pulped cotton fiber nestlet square (Ancare Corp., NY). All experiments were performed
using second litters of each dam in the study, which facilitates larger litter size and reduces the
variations in maternal care that we have observed with inexperienced, first-time mothers.
Breeding was initiated with mice of approximately eight weeks of age. During breeding, three to
four females were housed with one male. The males were removed from the cage with females
after 10 days, and on approximately embryonic day 16, the females were placed in a clean cage
39
with one nestlet square and housed singly until the pups were born. Pups were weaned on
postnatal day (P)21. In the present study, data from one dam and litter were removed from all
analyses due to seizures exhibited by the dam.
Data reported here were generated for ongoing G x E studies, focusing on the Met receptor
tyrosine kinase gene. MET is implicated in synapse maturation in social-emotional forebrain
circuitry (Judson et al., 2010, Qiu et al., 2011, Qiu et al., 2014) and is a risk gene for autism
spectrum disorder (Campbell et al., 2006, Peng et al., Konopka, G., 2014). The experimental
dams were generated by crossing Met
fx/fx
females with Met
fx/fx
male mice, resulting in
homozygous Met
fx/fx
mice, in which exon 16 of the Met gene is flanked by loxP sites. These dams
do not express Cre, and by all measures (Judson et al., 2009, Judson et al., 2010, Qiu et al., 2011,
Qiu et al., 2014) are no different than wild-type animals. Experimental Met
fx/fx
dams were mated
with Nestin
Cre
males (generated by crossing Nestin
Cre
males with C57Bl/6J females) to produce
litters with both control (Met
fx/+
) and heterozygous Nestin
Cre
/Met
fx/+
(Met
+/-
) pups. Met
+/-
pups
express approximately 50% of MET protein levels of control brains (Thompson and Levitt,
2015). The proportion of Met
+/-
pups in each culled litter ranged from one to five, out of five
pups, with an average of 2.65 Met
+/-
pups. To investigate potential intergenerational effects, six
Met
fx/fx
female mice that had experienced ELS as pups were paired with C57BL/6J wild-type
males to produce Met
fx/+
pups. These second generation pups were reared under control
conditions. All mice were genotyped after weaning according to a previously published protocol
(Judson et al., 2009) with minor modifications: the final elongation step in the Nestin
Cre
reaction
was seven minutes, and the PCR product was 320 base pairs. For the Met
fx
reaction, the duration
of the denaturation step during the amplification cycles was set at one minute.
40
2.3.2. Early-life stress paradigm
The ELS procedure was implemented based on Rice et al (2008). The center floor of the cage
was covered with approximately 50 grams of Alpha-Dri bedding to absorb urine, and a stainless
steel raised wire floor with 10 x 10 mm square openings and 1 mm wire diameter (Cat#
RWF75JMV, Allentown Inc., NJ) was inserted above the bedding. Dams were not able to
retrieve the bedding to incorporate in their nests. Two-thirds (1.8 g) of a standard pulped cotton
fiber nestlet square was provided in each ELS cage. Supplementary Figure 2.1 shows an ELS
cage with dam and pups on P4, two days after starting the ELS period. Control cages had
approximately 160g identical bedding material and a full nestlet square, and both cage set-ups
had identical access to food and water. Dams were alternately assigned to control or ELS
conditions based on the time of birth of pups. On the morning of the second day (P2) after birth
(P0), the experimental dams were weighed and subsequently placed into ELS or control cages.
The pups were removed by hand one-by-one from the nest, weighed, and placed onto the wire
floor of the ELS cage or on the bedding of the control cage after determining their sex using
genital pigmentation intensity (Wolterink-Donselaar et al., 2009). Three male and two female
pups were placed in each cage; the remaining pups in the litter were euthanized. The dam and
litter were left undisturbed for 7.5 days until the afternoon of P9, when the dam and pups were
removed from the ELS cage, weighed, and placed into a control cage with ample Alpha-Dri
bedding and one nestlet square. Cage changes were carried out on P16, and pups were weaned on
P21. Dams remained in their respective home cages until behavioral testing. Litters from control
dams were similarly culled to three males and two females on P2, and received a regular clean
cage with one nestlet square on P2, P9, and P16.
41
2.3.3. Video recording and analysis of dam behaviors
Experimental dams and respective litters, in their home cage, were moved to a separate, quiet
room within the vivarium suite at 11 am, and left undisturbed for one hour before starting video
recording from noon until 4 pm. Water (in a drinking bottle) and food were available ad libitum
during the recording period. The mice were recorded through the side cage wall on P4, P8 and
P12. Several measures were obtained from the video recordings, including the frequency of the
dam leaving and entering the nest (defined as the moment both hind paws are leaving or entering
the nest, respectively), the total time spent on-nest and off-nest, the duration of each bout, and
the duration of the longest uninterrupted on-nest period. An on-nest period was defined as a
period that includes at least one on-nest bout of 900 seconds or longer - but included brief off-
nest bouts shorter than 450 seconds. An off-nest period was defined as at least 450 seconds of
combined off- and on-nest bouts, without any on-nest bouts longer than 900 seconds. In addition,
the short on-nest bouts during an off-nest period were analyzed separately due to the initial
observation that many pup vocalizations occurred during this situation; this was termed ‘on-nest
during off-nest period’.
2.3.4. Nest quality score
The nest quality was visually scored through the sidewall of the cage on P2, P4, P9, P12 and
P21, at least 48 hours after providing dams with new nesting material. We determined the nest
quality score for 37 ELS and 37 control litters for an average of 3.00 and 2.89 of the
aforementioned time points, respectively. The scoring methodology is based on a protocol by
Hess et al. (2008), and ranges from zero to five. A score of zero means that the dam had not
manipulated the nestlet square, while a score of one indicates that the nestlet square had been
42
manipulated or interacted with, but a specific nest is not evident. A score of two is given when
the nesting material is present on top of the bedding (or the grid in ELS cages), but no substantial
walls are visible (“flattened saucer shape”). The nest quality score is three when the walls are
just below “half of a hollow sphere”, and four when they “reach or are higher than half of a
hollow sphere”. A score of five is given when the nest resembles a “complete dome” with a
small exit opening. In addition, 0.25 points can be added for each quarter of the nest that has a
higher score than the base score of the nest.
2.3.5. Ultrasonic and audible vocalization recording and analysis
Eighteen litters (nine control and nine ELS litters) were recorded with ultrasonic audio recording
equipment and software (UltraSoundGate 116Hb, condenser microphone CM16/CMPA,
AviSASLab Pro software, Avisoft bioacoustics (Germany)) to determine the counts and timing
of USV and AV by the pups in relation to the dam’s location. The microphone was suspended
slightly above a temporary cage lid in which a hole was present right above the litter
(approximately five inches), at the start of the one-hour acclimation period. The video- and audio
recordings were synchronized by quickly tapping a pen against the cage in close proximity to the
microphone immediately before and after the analysis period. Audio was recorded for the full
four-hour duration of the video recording period. One complete on-nest period and one complete
off-nest period was analyzed for vocalizations (vocalization analysis time averages 138.2 ± 42.9
minutes (mean ± standard deviation, with a range of 50.9-226.6 minutes). The USV and AV
were scored by visually analyzing the audio spectrogram and generating a time stamp of the
vocalizations (using Ocenaudio and Praat software) that corresponds with the time stamp of
video analysis of the dam’s nest entries and location. For this study, an AV (squeal) was defined
43
as a broadband sound formed by regularly spaced harmonics covering the larger part of the audio
frequency spectrum of 0-120 kHz, including the human audible range of 0-20 kHz. USV were
defined as any type of narrowband call at a frequency in the ultrasonic range (>20 kHz).
Vocalizations that occurred when pups were temporarily outside of the nest were excluded. The
average number of vocalizations was determined for the total time that a dam spent in each
location (on-nest, off-nest, or on-nest during an off-nest period). In addition, we analyzed the
number of vocalizations in relation to the dam’s entry in or exit from the nest; these numbers
were presented as the average number of vocalizations per nest entry or exit. One outlier was
removed from the control group (in the minus 20 to minus 10 second time bin) in the temporal
analysis of USV in relation to the dam’s nest entry, using Grubbs outlier detection (alpha =
0.05). The average USV values per nest entry for this time bin for all control dams are: 0, 0, 0,
1.22, 0, 12.33*, 0.09, 0, 0, with the outlier indicated with an asterisk. To successfully perform a
repeated measures two-way ANOVA, all time points from this specific litter were removed from
the USV vs. entry time point analysis only. Pup genotype (proportion of Met
+/-
compared to wild-
type pups in a litter) did not affect the average number of USV or AV per minute in either
control or ELS conditions (Supplementary Figure 2.2).
2.3.6. Behavioral tests of dams
Following pup weaning, each dam was housed singly for three days while being acclimated daily
to handling in preparation for elevated-plus maze testing on the fourth day after weaning. After
the elevated-plus maze test, the dams were housed in sets of three by experimental group (control
or ELS environment) until the contextual fear conditioning test, which was performed seven to
15 days after pup weaning. All behavioral tests of the dams were performed during the light
44
cycle in the afternoon, in designated mouse behavioral testing rooms that were within the same
vivarium corridor as the rooms in which all mice were housed. Dams received a clean cage with
ample bedding and one nestlet square after completing the elevated-plus maze test. The person
running behavioral tests was blind to the environment experienced by each dam.
2.3.6.1. Elevated-plus maze
The elevated-plus maze apparatus (San Diego Instruments Inc., CA) was located on the floor in
the center of a 10 x 8 ft. room with white walls and even LED lighting throughout the room. The
arms of the elevated-plus maze measured 30 cm long, 6.5 cm wide, with 3 mm elevated edges
along the open arms. The floor of the maze and the walls of the closed arms were opaque white
and opaque black, respectively. The height of the maze was 40 cm above the floor, and the walls
of the closed arms extended 14 cm from the maze surface. The light intensity at the end of the
open arm, the center of the maze, and the end of the closed arm was 6-7 lux, 3 lux, and 0-1 lux,
respectively.
On the three days prior to the elevated-plus maze test, mice were acclimated to human handling
and to the opaque glass beaker used for transporting mice from the cage to the maze, for one
hour a day. Acclimation consisted of transporting the home cage to a room adjacent to the
behavior room, placing each mouse into the beaker and letting them exit the beaker, as well as
letting mice freely explore the beaker with the cage top closed. On testing day, mice were placed
in a quiet room (light intensity of 8 lux) adjacent to the behavior testing room for three hours
prior to testing. Between each mouse being tested, the maze was cleaned with 70% isopropyl
alcohol, and water, and allowed to dry. Mice were transported to the elevated-plus maze room in
45
a beaker, and the beaker was placed at a 45° downward angle in the center of the maze, facing
away from the experimenter and the door. Mice were allowed to exit the beaker spontaneously,
and the experimenter subsequently exited the room and closed the door. Beaker exit times
averaged 85.7 ± 83.9 seconds (mean ± standard deviation), and the latency to exit the beaker was
not correlated with measures of anxiety-like behavior. The five-minute test session began once
the tail base of the mouse entered the center of the maze. Time spent in the open and closed arms
and the center as well as the number of entries into the arms was recorded with overhead video
cameras (SuperExwave SS-E473, Sony) and analyzed using automated TopScan tracking
software (CleverSys Inc., VA). The time spent in one area on the maze was defined as the
moment the center- and tail base markers of the mouse crossed the border of the area until both
markers crossed the border of an adjacent area.
2.3.6.2. Contextual fear conditioning
Contextual fear conditioning tests were performed over two days. One hour before testing, the
dams were transported to a room (light intensity of 200 lux) adjacent to the fear conditioning
room (400 lux). During the test, freezing behavior was recorded using near-infrared light and
cameras (#NIR-100 infrared light source, Med Associates Inc., VT; #SCA640-71fm camera,
Basler, Germany) in double-enclosed fear conditioning chambers 24 x 30 x 22 cm in size, with a
metal grid floor, and plexiglass and aluminum walls (#ENV065FPU-M, Med Associates Inc.,
VT). On day one, mice were acclimated to the fear conditioning chamber for three minutes, after
which five x two-second foot shocks of 0.5 mA were administered to the mice via the metal grid
floor (#ENV-414S stimulator/scrambler, Med Associates Inc., VT). Freezing duration was
recorded during the three minute acclimation period (‘baseline’), and during four 220-second
46
inter-trial intervals and one 220-second post-shock period (average of five periods reported as
‘training’). Between tests, the chambers were cleaned with 70% isopropyl alcohol, followed by
water. After training, dams were returned to their respective home cages for 24 hours. On day
two, the dams were placed into the same fear conditioning chambers for eight minutes while
freezing behavior (‘test’) was recorded and analyzed. Freezing behavior analysis on both days
was performed in real-time using Video Freeze software (Med Associates, Inc., VT), with a
linear method of observation, a motion threshold of 18, and a minimum freezing duration of 30
frames (one second).
2.3.7. Statistical analyses
All data were analyzed using GraphPad Prism 6 (Graphpad Software, Inc.). The average number
of dam nest entries per hour, the total duration of the dam’s presence on the nest, and the nest
quality differences were analyzed using a two-way ANOVA (environment x postnatal age of
pup). A two-way repeated measures ANOVA was used to analyze fear conditioning data
(environment x baseline/training/test), USV and AV counts (environment x location of dam),
time spent by dam in location category during vocalization analysis (environment x location of
dam), the timing of USV and AV in relation to timing of the dam’s entry (environment x time
point relative to dam’s nest entry/exit), and the frequency of on- and off-nest bouts of different
durations (environment x bout duration, for on- and off-nest bouts separately). A Pearson
correlation analysis was performed to analyze the correlation between the number of Met
+/-
pups
in a litter and the number of USV and AV on P4. The duration of complete on-nest bouts, as well
as EPM measures (percentage time spent in arm and number of arm entries) was analyzed using
an unpaired Student’s t-test (ELS environment vs. control). The number of nest entries on P4 of
47
next-generation females was compared to control and ELS dams using a one-way ANOVA.
Bonferroni post-hoc analyses were performed on ANOVA tests with significant main or
interaction effects (alpha = 0.05). Data are presented as mean ± standard error of the mean
(SEM).
2.4. Results
2.4.1. ELS environment transiently alters dam nest entry behavior
Analysis of dam behavior during the four-hour recording sessions confirmed previous reports
(Rice et al., 2008) that the ELS environment results in an increase in the number of nest entries
by the dam, while the total duration of on-nest time is similar between the control and ELS
groups (Figure 2.1, and Table 2.1 for statistical test details of all maternal behavior
measurements in the Results section). Post-hoc tests reveal that on P4 and P8, the number of nest
entries are increased as a result of the ELS environment (p < 0.001 and p < 0.01, respectively),
while on P12, three days after returning dams and litters to cages with normal bedding, the nest
entry frequency of ELS dams returned to control levels (p > 0.999). The time spent on the nest
on P12 decreased compared to P4 and P8 (p < 0.001 and p < 0.01, respectively) for dams in ELS
as well as in control conditions. The nest quality score was affected by both the ELS
environment and age of the litter (Figure 2.2). Post-hoc tests indicate that the nest quality score is
lower during, and at the end of the ELS period (P4 and P9, respectively) as a result of the ELS
environment (p < 0.001), but returned to control levels on P12 (p > 0.999). These results show
that the ELS environment transiently alters dam nest entry behavior and nest quality.
48
Figure 2.1: ELS environment induces an increase in the number of nest entries by the dam,
but does not affect total on-nest time. (A) The average number of nest entries per hour between
noon and 4pm (light phase) is increased on P4 and P8 during the ELS period (P2-P9) due to the
ELS environment, while the average number of nest entries by ELS dams on P12 are comparable
to dams in control conditions (B) ELS environment does not affect the total percentage of on-
nest time from noon to 4pm (light phase) on P4, P8 and P12. The total on-nest time on P12 does,
however, decrease in both control and ELS conditions compared to earlier time points.
Individual data points represent independent dams, and data are presented as mean ± SEM. Per
group on: P4, n = 23; P8, n =10-12; P12, n = 7. ** p < 0.01, *** p < 0.001.
49
Table 2.1: Statistical test details of maternal behavior measurements
Figure Measurement F or t statistic p value
1A Number of nest entries
Environment: F1,76 = 23.65
Litter age: F2,76 = 10.77
Interaction: F2,76 = 6.09
p < 0.001
p < 0.001
p < 0.01
1B Duration of on-nest time
Environment: F1,76 = 0.28
Litter age: F2,76 = 15.04
Interaction: F2,76 = 0.72
n.s.
p < 0.001
n.s.
2 Nest quality score
Environment: F1,203 = 59.38
Litter age: F4,203 = 106.90
Interaction: F4,203 = 25.64
p < 0.001
p < 0.001
p < 0.001
4A
On-nest bout duration P4
Environment: F1,44 = 38.25
Bout duration: F9,396 = 14.41
Interaction: F9,396 = 10.60
p < 0.001
p < 0.001
p < 0.001
Off-nest bout duration P4
Environment: F1,44 = 39.42
Bout duration: F9,396 = 21.39
Interaction: F9,396 = 18.09
p < 0.001
p < 0.001
p < 0.001
4B
On-nest bout duration P12
Environment: F1,12 = 0.09
Bout duration: F9,108 = 5.28
Interaction: F9,108 = 2.41
n.s.
p < 0.001
p < 0.05
Off-nest bout duration P12
Environment: F1,12 = 0.03
Bout duration: F9,108 = 3.05
Interaction: F9,108 = 3.05
n.s.
p <0.001
n.s.
5A EPM open arm time t18 = 0.40 n.s.
5B
EPM open arm entries t18 = 0.84 n.s.
EPM closed arm entries t18 = 1.37 n.s.
5C Contextual fear conditioning
Environment: F1,19 = 3.18
Trial: F2,38 = 262.8
Interaction: F2,38 = 0.47
n.s.
p < 0.001
n.s.
6A Number of nest entries – next generation P4 Environment: F2,49 = 23.88 p < 0.001
6B
On-nest bout duration – next generation P4
Environment: F1,27 = 0.02
Bout duration: F9,234 = 5.16
Interaction: F9,234 = 0.51
n.s.
p < 0.001
n.s.
Off-nest bout duration – next generation P4
Environment: F1,27 = 0.02
Bout duration: F9,234 = 4.69
Interaction: F9,234 = 0.89
n.s.
p < 0.001
n.s.
50
Figure 2.2: Limited bedding transiently decreases nest quality during the ELS period. The
ELS conditions only affected nest quality during the ELS period when limited nesting material
was available; after the ELS dams return to control conditions, the nest quality was similar to
that of control dams. The nest quality was assessed at least 48 hours after providing the dams
with new nesting materials. Data are presented as mean ± SEM. Per group on: P2, n = 23-24;
P4, n = 28-29; P9, n = 23-24; P12, n = 11; P21, n = 20. *** p < 0.001.
51
2.4.2. ELS environment impacts dam nest entry behavior during off-nest periods
The extended video recording time allowed us to examine in more detail the temporal changes in
the patterns of nest entry behavior by each dam. As has been shown previously in rats
(Kobayashi et al., 1997), we observed that in control conditions, dams alternated longer on-nest
periods with shorter off-nest periods (see representative pattern in Figure 2.3A). During long on-
nest periods, dams slept or cared for the pups by nursing, grooming, and licking them. During
shorter off-nest periods, dams spent most of the time eating and drinking, while leaving the
litters relatively undisturbed. Self-grooming and nest building occurred both during on- and off-
nest periods. Visualization of the nest-entry patterns showed that dams in an ELS environment
exited and re-entered the nest more frequently than control dams, but only during off-nest
periods (see representative pattern in Figure 2.3B). Further analyses showed that, despite the
increased overall number of nest entries by dams in an ELS environment, the average duration of
long, uninterrupted on-nest periods was statistically similar between control and ELS dams
(Supplementary Figure 2.3, p = 0.09, Supplementary Table 2.1). Additionally, on-nest bouts
longer than 50 seconds occurred at a similar frequency during the four-hour recording time in
control and ELS dams; only short on-nest bouts that are ≤ 50 seconds were observed more
frequently in the ELS group compared to the control group (Figure 2.4A, p < 0.001). Similarly,
only the frequency of off-nest bouts ≤ 25 seconds was significantly increased as a result of the
ELS environment (Figure 2.4A, p < 0.001). On P12, three days after the ELS period, the
frequency of shorter on- and off-nest bouts was largely the same between dams in ELS and
control conditions, except for a small but significant difference in the frequency of one on-nest
duration bin (Figure 2.4B, p < 0.01 for on-nest bout duration bin of 200-400 sec). These analyses
52
of maternal behavior during the post-ELS period show that the altered nest entry behavior of the
dam is a transient effect due to the ELS environment.
Figure 2.3: Representative schematic of the dam’s location in relation to the nest in the
home cage. Black color indicates that the dam is off-nest, while on-nest time is represented in
white. Complete on-nest periods (dotted lines) are alternated with off-nest periods (solid lines)
by dams in both the control and ELS environment. (A) Typical pattern of on/off-nest behavior by
a dam in a control and (B) ELS environment in a four-hour recording period during the afternoon
of the light phase. An ELS environment induces an increase in the number of nest entries by the
dam, but only during off-nest periods.
53
Figure 2.4: Limited bedding leads to a transient increase in short on- and off-nest bouts.
(A) The number of short on- and off-nest bouts (≤ 50 seconds and ≤ 25 seconds in duration,
respectively) by the dam is increased as a result of the ELS environment on P4. (B) After
returning the dams to control conditions on P9, the frequency of on- and off-nest bouts of
different durations on P12 are no different between the dams that experienced an ELS
environment and control dams, except for a small but significant difference in the frequency of
one on-nest duration bin (201-400 seconds). Data are presented as mean ± SEM.
Per group on: P4, n = 23; P12, n = 7. ** p < 0.01, *** p < 0.001.
54
2.4.3. Dams experiencing ELS environment exhibit normal behaviors after pup
weaning
The focus in animal ELS models has been predominantly on determining short- and long-term
effects on the offspring. To our knowledge, mouse dam behavior after being in an ELS
environment with limited nesting materials has not been reported in the literature. To investigate
whether the ELS environment impacts dam behavior following pup weaning on P21, we
evaluated anxiety-like and contextual fear learning behaviors of the dams. We did not observe a
difference in the percentage of time spent in the open arm (p = 0.695) and open arm entries (p =
0.413) between control dams and dams in the ELS environment (Figure 2.5A). Additionally, the
number of closed arm entries was unchanged due to the ELS environment (Figure 2.5B, p =
0.187), suggesting that there was no difference in overall activity levels between control and ELS
dams during this test.
We next evaluated fear learning and memory in the same dams using a contextual fear
conditioning paradigm. We did not find a significant main effect of ELS environment on
baseline contextual fear, fear acquisition, or fear memory (Figure 2.5C). The behavioral analyses
indicate that dams do not exhibit changes in anxiety-like and contextual fear learning behavior as
a result of the ELS environment while raising their pups.
55
Figure 2.5: Dams in limited bedding conditions from P2-P9 do not exhibit altered anxiety-
like and fear learning behaviors after pup weaning. ELS environment does not result in
changes in (A) the time the dams spent in the open arms of the elevated-plus maze or (B) in the
number of arm entries. (C) Contextual fear learning (‘training’) or memory (‘test’) was not
affected as a result of limited bedding, as measured here by the amount of freezing in a specific
context that is associated with foot shocks. Individual data points represent independent dams,
and data are presented as mean ± SEM. Control, n = 11, ELS environment, n = 9-10.
56
2.4.4. Mothers that had experienced ELS as pups exhibit normal nest entry behavior
To investigate whether ELS has intergenerational effects on nest entry behavior, we analyzed
maternal behavior of female offspring that had pups and were subsequently housed in control
conditions. The number of nest entries on P4 of these dams was comparable to control
(p > 0.999) but different from ELS dams (p < 0.001) of the previous generation (Figure 2.6A). In
addition, no differences were measured in the number of on- and off-nest bouts of different
duration between control dams and dams that had experienced ELS as pups (Figure 2.6B). These
results indicate that nest entry and exit behaviors are not affected by early pup experiences
during P2-P9 as a result of the ELS environment (i.e. they are not learned from the dam during
this time period).
57
Figure 2.6: Dams that have experienced ELS as pups do not show altered nest entry
behavior in adulthood when rearing pups. (A) The average number of nest entries per hour
during the afternoon on P4 in females that have experienced ELS as pups is similar to the nest
entry behavior of controls. Both control dams and next generation dams enter the nest less
frequently than dams in ELS conditions. Note that control and ELS environment data are the
same as in Figure 2.1. Individual data points represent individual dams. (B) A more detailed
analysis of the frequency of on- and off-nest bouts of different duration shows that next
generation dams do not differ in their nest entry behavior from control dams. Control data are the
same as in Figure 2.4A. Data are presented as mean ± SEM. Control & ELS environment, n =
23; Next generation, n = 6. *** p < 0.001.
58
2.4.5. ELS pups emit more ultrasonic and audible vocalizations
To gain insight into the response of the pups to altered dam nest entry behavior due to the ELS
environment, we analyzed ultrasonic vocalizations (USV) and audible vocalizations (AV) – see
Figure 2.7 for example spectrograms – for one complete off-nest period and one complete on-
nest period per video recording session on P4. We observed both USV and AV during on- and
off-nest periods in both control and ELS litters. Throughout the on-nest and off-nest periods, we
did not observe differences in the number of USV or AV between the control and ELS litters.
However, the average number of USV and AV was significantly higher in ELS litters compared
to control litters when the dam was on-nest during off-nest periods (Figure 2.8, and Table 2.2 for
statistical test details of all vocalization measures in Results section). Specifically, post-hoc
analyses revealed 1) an increase of USV and AV during this period in ELS litters compared to
control litters (p < 0.001), and 2) an increase of USV and AV compared to regular on-nest and
off-nest time in the ELS group (p < 0.001 and p < 0.01, respectively). In contrast, the number of
USV and AV did not change as an effect of the dam's location in the control group (p > 0.999).
Due to slight variability in the duration of on- and off-nest periods between mice, the number of
vocalization was normalized to the total time spent in each category, providing a measure of
average vocalizations per unit time. Note that no significant overall differences were observed
between control and dams in ELS conditions in the duration of each of these categories
(Supplementary Figure 2.4, Supplementary Table 2.1), and all individual dams spent time in
each category.
59
Figure 2.7: Representative spectrograms of recording of vocalizations. Ultrasonic
vocalizations (narrowband vocalizations in the ultrasonic range of 20-120 kHz) and audible
vocalizations (‘squeals’ with harmonics covering the larger part of the 0-120 kHz frequency
spectrum, including the human audible range of 0-20 kHz) were observed regardless of when the
dam was present or absent from the nest, in both control and ELS conditions.
60
Figure 2.8: ELS pups emit more vocalizations when the dam is present on the nest during
off-nest periods. Vocalizations are normalized to time spent in each location in relation to the
nest (see also Supplementary Figure 2.4). The number of (A) USV and (B) AV is increased when
the dam is on-nest during off-nest periods. Individual data points represent independent litters,
and data are presented as mean ± SEM. n = 9 litters per group. *** p < 0.001.
61
Table 2.2: Statistical test details of vocalization measurements
Figure Measurement F statistic p value
8A Ultrasonic vocalizations
Environment: F1,16 = 9.54
Dam’s location: F2,32 = 8.82
Interaction: F2,32 = 9.59
p < 0.01
p < 0.001
p < 0.001
8B Audible vocalizations
Environment: F1,16 = 4.59
Dam’s location: F2,32 = 4.00
Interaction: F2,32 = 5.32
p < 0.05
p < 0.05
p < 0.05
9A Ultrasonic vocalizations & nest entry
Environment: F1,15 = 5.10
Timing: F12,180 = 6.17
Interaction: F12,180 = 4.74
p < 0.05
p < 0.001
p < 0.001
9B Audible vocalizations & nest entry
Environment: F1,16 = 13.26
Timing: F12,192 = 1.97
Interaction: F12,192 = 1.84
p < 0.01
p < 0.05
p < 0.05
9C Ultrasonic vocalizations & nest exit
Environment: F1,16 = 0.25
Timing: F11,176 = 1.41
Interaction: F11,176 = 1.32
n.s.
n.s.
n.s.
9D Audible vocalizations & nest exit
Environment: F1,16 = 0.84
Timing: F11,176 = 1.34
Interaction: F11,176 = 0.93
n.s.
n.s.
n.s.
62
2.4.6. Vocalizations occur after nest entry by dam in ELS environment
To determine whether the dam’s entry in, or exit from the nest induced pup vocalizations, the
average number of vocalizations per entry or exit was analyzed in relation to the timing of these
nest entries and exits by the dams. In relation to dam entry, the analysis revealed significant main
and interaction effects of environment and the temporal relation to the dam’s entry on the
expression of USV (Figure 2.9A). Post-hoc tests show an increase in the average USV per entry
during the first 10 seconds after the dam enters the nest in the ELS environment (p < 0.05). For
AV, we also observed significant main and interaction effects of environment and the temporal
relation to the dam’s entry (Figure 2.9B). Specifically, the average number of AV per nest entry
is increased between 10-20 seconds and 50-60 seconds after the dam enters the nest in the ELS
environment (p < 0.01 and p < 0.05, respectively). In contrast, in relation to the exit of dams
from the nest, there was no effect of environment on the average USV or AV per exit (Figure
2.9C). The increase in pup vocalizations following nest entry by dams in an ELS environment
suggests that the manner in which dams enter the nest and initially interact with the pups is
qualitatively different from dams in control conditions.
63
64
(figure on previous page)
Figure 2.9: Vocalizations of ELS pups are increased immediately after dams enter the nest.
The average number per nest entry during the analysis period of (A) USV and (B) AV are
increased at different time periods during the first minute after dams enter the nest as a result of
the ELS environment. During the remaining time of the analysis (> 60 seconds after each nest
entry), no difference in average vocalizations per entry between the two groups was observed.
No significant effect of the ELS environment was observed on the number of (C) USV and (D)
AV before or after dams exit the nest. Data are presented as mean ± SEM. & indicates the
control group in which one data point in this time bin was determined to be an outlier, which has
been removed from this time bin (see methods section for more details). n = 9 litters per group.
** p < 0.01, *** p < 0.001.
65
2.5. Discussion
The present study provides a detailed analysis of mouse dam and pup behavior as a result of
experiencing a one-week period with limited access to nesting and bedding material. We show
that this ELS environment increases the average number of nest entries of the dam during the
off-nest period, while on-nest periods are unaffected. Furthermore, there were no lasting effects
on dam non-maternal behaviors that include anxiety-like behavior and fear learning, nor
intergenerational effects on nest entry behavior. Finally, in this ELS environment, pups increase
their vocalization rate immediately after dams enter the nest during off-nest time, suggesting that
atypical physical interactions between the dam and pup contribute to the induction of stress and
subsequent long-lasting neurobiological and physiological effects in this ELS paradigm.
2.5.1. Disruption of typical dam-pup interaction patterns due to ELS environment
The analyses show that an ELS environment with limited bedding and nesting materials prompts
the dam to return repeatedly to the nest during off-nest periods when pups are normally left
relatively undisturbed, thereby increasing the number of discrete physical interactions. This ELS
environment, however, does not induce apparent changes in interactions during typical
uninterrupted on-nest periods. Our results replicate and elaborate on findings by several groups
who reported that the frequency of nest exits increased (Rice et al., 2008, Baram et al., 2012,
Malter Cohen et al., 2013, Gunn et al., 2013, Yang et al., 2015, Naninck et al., 2015), whereas
the total duration of maternal care is not altered during the ELS period (Gunn et al., 2013,
Naninck et al., 2015, Rice et al., 2008). In contrast with these studies, as well as with our results,
two groups report a decreased (Malter Cohen et al., 2013) and increased (Yang et al., 2015)
duration of total on-nest time. We found that four-hour continuous recording sessions, as well as
66
minimal interference with the home cage – including acclimation to the recording room, which
allows dams to resume their natural pattern of on/off nest behavior prior to recording – lead to
highly reproducible nest entry behaviors. However, by limiting the interference with the home
cage, the nest structures, particularly in control cages, obscured viewing discrete dam-pup
interactions, other than exiting and entering the nest. This prevented us from analyzing
differences in maternal care (e.g. licking and grooming) in more detail.
Under control conditions, dams alternate long on-nest periods with shorter off-nest periods
(Kobayashi et al., 1997), and transitions between several maternal and non-maternal behaviors
occur in stereotyped strain-specific patterns (Carola et al., 2011). An alternating on/off-nest
pattern is necessary to balance the needs for increased energy intake to sustain lactation, nursing
pups, and sleeping. This results, during early stages of lactation, in the elimination of increased
activity during the dark phase compared to the light phase that is normally observed in mice
(Gamo et al., 2013). An increased maternal ventral temperature due to pup presence, as well as
the satiety state of the pups, has been implicated as the trigger for rat dams to leave the nest in
this stereotypical pattern (Croskerry et al., 1978, Leon et al., 1978, Woodside et al., 1980, Stern
and Azzara, 2002, Stern and Keer, 2002). An alternative hypothesis is that the patterns of nest
entry and exit patterns by dams are modulated by a motivational switch between maternal
behaviors (e.g. nursing pups, licking and grooming, nest building) and consumption of
food/water (and other non-maternal behaviors such as activity). Consistent with this, mice have
been shown to bar press in order to take care of pups (Hauser and Gandelman, 1985) and to
receive food and water, with these behaviors generally being mutually exclusive. It is possible
that the grid in this ELS paradigm (or a plastic cage floor often used in rat studies), on which the
67
dam resides is sufficiently aversive for the dam to interrupt the consumption of food and water,
or the search for nesting materials, and return to the nest for temporary alleviation.
Regardless of causality, it is clear that the ELS paradigm used in the current study creates an
atypical behavioral pattern in the dam that increases discrete interactions between the dam and
pups. This is in contrast to the commonly used maternal separation model, which relies on the
absence of maternal care to evoke a stress response in the pups. Maternal separation can lead to
increased on-nest time (Millstein and Holmes, 2007) and increased licking and grooming of pups
immediately after the pups are placed back with the dam (Kosten and Kehoe, 2010), due to an
increase in “demanding behaviors” such as suckling, by the pups (Pereira and Ferreira, 2006). It
has been shown that this increase in maternal care results in a decrease in pup stress response
after repeated daily short-term isolation (15 min), whereas increased maternal care cannot
compensate for the effects of daily periods of long-term isolation on the stress response
(commonly 3h or longer). While the maternal separation paradigm is successful in inducing a
stress response and leads to enduring behavioral impairments in the pup (Plotsky et al., 2005,
Lippmann et al., 2007, Sanchez et al., 2001), the ELS paradigm with limited nesting materials
has several advantages; it does not require daily human interaction with the dams and pups,
while providing the potential for a continuous stressor to the pups (Gilles et al., 1996, Molet et
al., 2014), and it does not interrupt regular food intake patterns of the pups due to uninterrupted
on-nest periods. Additionally, it is important to note that maternal separation and the ELS
paradigm with limited nesting materials may be simulating different types of stressors, namely
neglect and abuse, respectively.
68
2.5.2. Impact of ELS environment on pup vocalizations in relation to dam behavior
Abusive behavior has been referred to as the de facto stressor in limited bedding studies using
rats (Roth and Sullivan, 2005, Raineki et al., 2010, Raineki et al., 2012). Since reports of mouse
studies did not reveal any changes in total on-nest time and licking/grooming behaviors (Baram
et al., 2012) previously observed in rats (Braw et al., 2009, Dalle Molle et al., 2012, Raineki et
al., 2012, Raineki et al., 2010, Roth and Sullivan, 2005), it was thought that limited nesting
materials induces stress in mice through different means, namely fragmented maternal care. Our
results, however, indicate that ELS mouse pups are negatively affected by increased nest entries
as they vocalize immediately after the dam enters the nest, while the longer on-nest periods are
unaffected. Although the pups could be conveying positive affect in response to the return of the
dam, we conclude that this is not likely, since, per nest entry, control pups do not vocalize more
immediately after the dam returns. Increased vocalizations have been reported in the limited
bedding model in rats (Roth and Sullivan, 2005, Raineki et al., 2010, Raineki et al., 2012),
although it was not clear from these studies when these vocalizations occur in relation to the
dam’s nest entry. The present data demonstrate this temporal relation in mice.
Mouse USV accompany salient behaviors such as affiliative, aggressive, and courtship behaviors
(Sirotin et al., 2014, Portfors and Perkel, 2014), although the meaning of specific vocalization
patterns is not clear. Several groups have investigated the saliency of USV fragments (Holfoth et
al., 2014), the ability of mice to discriminate spectrotemporal patterns of USV (Neilans et al.,
2014), and cortical plasticity depending on relevant USV features for specific environmental
conditions (e.g. pup vocalizations for lactating dams) (Shepard et al., 2015). Rodent pups
vocalize when they fall out of the nest, which induces pup retrieval by the dam (Noirot, 1972,
69
Ehret and Haack, 1982, Branchi et al., 2006, Okabe et al., 2013, Ehret, 2005). In addition, male
odors, tactile stimuli and low temperature can induce USV in mouse pups (Branchi et al., 1998).
Pup vocalizations may also act as inhibitory signals to alert the dam, as female rats that are
deafened or have pups that are being prevented from making calls are more likely to injure their
offspring (Stern, 1997, White et al., 1992). Dams interrupt licking behavior when pups vocalize,
and frequently switch to other behaviors such as nest building (Noirot, 1966). Increased numbers
of AV and USV have also been reported when mouse pups were exposed to a variety of painful
stimuli (Haack et al., Williot, J. F., 1983, Williams et al., 2008, Delwig et al., 2012, Tsuzuki et
al., 2012, Kurejova et al., 2010, Han et al., 2005). Many of these studies used one or a limited
number of ultrasonic frequency recording channels, which does not allow for discrimination
between narrowband USV at the selected frequency range, and broadband AV that include (and
extends beyond) the same specific frequency range. However, several studies specifically
mention the occurrence of both AV and USV as a response to painful stimuli (Haack et al.,
Williot, J. F., 1983, Williams et al., 2008, Delwig et al., 2012, Han et al., 2005), and moreover,
Delwig et al. (2012) attributes an increased number of pup USV as response to aversive stimuli
(e.g. bright light) while AV additionally occur when the stimulus becomes painful (e.g. tail
pinch).
The current studies cannot formally exclude that a fraction of USV and AV are generated by the
dam, which may be affected by the physical properties of the metal grid. We believe this is
unlikely, however, because we would have observed an increase of USV and/or AV specifically
when the dam was outside of the nest, on the grid in limited bedding conditions. It also is
possible that limited nesting and bedding materials (i.e. a reduced nest quality) alters
70
vocalization rates by changing pup body temperature (Branchi et al., 1998), but there is no
difference in the number of USV or AV when the dam is off-nest and the pups are most exposed
to the environment.
2.5.3. Transient effects of ELS environment on dam behavior
After returning the dam and litter to control conditions on P9, nest entry behavior exhibited by
the dams that experienced the ELS environment was comparable to control dams. This is
consistent with other findings of a lack of impact due to (prenatal) limited nesting materials (Ivy
et al., 2008, Bolton et al., 2013). Furthermore, the ELS environment does not appear to impact
long-term non-maternal behavior of the dams, as our data show that anxiety-like and fear
learning behaviors in dams were normal after weaning their pups. These results, and the
observation that nest-entry frequency by the dam is normalized at P12, indicate that stressors
experienced by the pups are only present during, and not after ending the limited bedding
conditions due to residual effects of this paradigm on the dam’s behavior. This knowledge will
help future studies with delineating the stress exposure when analyzing the effects of stressors on
specific stages of brain development. Ivy et al. (2008) have shown a similar lack of effect on
anxiety-like behavior in rat dams as measured on the elevated-plus maze, although they did
observe a decrease in center time in the open-field test, as well as altered physiological stress
markers. Other studies using prolonged maternal separation in rats show no change in the dams
in anxiety-like behavior (Eklund et al., 2009, Aguggia et al., 2013), exploration, risk assessment,
risk taking or shelter seeking (Daoura et al., 2010). However, additional behavioral tests would
need to be done to assess maternal and other general behaviors to rule out lasting effects of
limited nesting materials on the dam.
71
Furthermore, we found that female offspring that underwent ELS as pups did not exhibit altered
nest entry behavior in adulthood. This suggests that the stereotypical nest entry pattern that we
observed in this study is not learned from the dam during the first postnatal week, but rather is an
expression of innate behavior that is transiently altered as a result of different environmental
conditions (e.g. the amount of bedding and nesting material). It would be interesting, however, in
light of our results that maternal behavior quickly returns to normal after returning to control
conditions on P9, to test whether an extension of the ELS time window until P21 (weaning)
would negatively impact intergenerational nest entry behavior.
In summary, an environment of limited bedding and nesting materials is an effective early-life
stressor that consistently evokes vocal responses in mouse pups that have been associated with
aversive and painful stimuli. It appears that the effects of this ELS environment on mouse dams
and pups more closely resemble the effects of limited bedding on rat dams and pups than was
previously suggested. Our results confirm and extend previous findings that the mouse ELS
paradigm used in the present study alternates normal maternal care (during long on-nest periods)
and ‘abusive’ behaviors by the dam (through an increased number of physical and/or painful
interactions between the dam and pups) with minimal human interference. ELS can be evoked by
many different mechanisms in both rodents and humans, and detailed analyses of animal models
can help effectively model the impact of stressors on brain development in humans.
72
2.6. Supplementary Material
Supplementary Figure 2.1. Top view of cage with dam and pups in ELS environment on
P4. On P2, the dam and pups are transferred to the ELS environment: a cage with a metal wire
grid that is suspended above the cage floor, with sparse bedding on the bottom of the cage to
absorb urine. The dam receives approximately two-thirds (1.8 g) of a standard nestlet square to
build a small nest. On P9, at the end of the ELS period, the dam and pups are transferred to a
standard cage with regular bedding and a full nestlet square.
73
Supplementary Figure 2.2. Genotype composition of the litter does not affect the number of
vocalizations on P4. For ELS and control litters separately, as well as combined (shown here),
the average number of USV and AV during the observation period did not correlate with the
percentage of Met
+/-
pups in the litter (ELS and control combined: USV: r = 0.30, R
2
= 0.09, p =
0.22 and AV: r = 0.19, R
2
= 0.04, p = 0.45. For ELS and control separately: ELS USV: r = 0.52,
R
2
= 0.27, p = 0.15 and ELS AV: r = 0.26, R
2
= 0.07, p = 0.50; Control USV: r = 0.40, R
2
= 0.16,
p = 0.28 and Control AV: r = 0.24, R
2
= 0.06, p = 0.53). Litters were culled to three males and
two females on P2, blind to genotype. Each individual data point represents one independent
litter with mixed genotypes.
74
Supplementary Figure 2.3. The duration of long, uninterrupted on-nest bouts on P4 is
similar for dams in a control and ELS environment. For each dam only the longest complete
on-nest bout during the four-hour recording window is included in this analysis. Dams that only
had long on-nest bouts that already commenced at the start of the recording session, or continued
until after the recording session, were excluded. Individual data points represent independent
dams, and data are presented as mean ± SEM.
75
Supplementary Table 2.1: Statistical test details of supplementary data
Figure Measurement F or t statistic p value
S3 Duration on-nest bouts P4 t40 = 1.71 n.s.
S4 Time & location on/off-nest
Environment: F1,16 = 0.35
Dam’s location: F2,32 = 148.80
Interaction: F2,32 = 2.11
n.s.
p < 0.001
n.s.
Supplementary Figure 2.4. The percentage time spent by dams in specific locations (on-
nest, off-nest, or on-nest during an off-nest period) during USV/AV analysis varied per
dam, but was not significantly different between dams in a control and ELS environment.
Data in Figure 2.8 have been normalized to the absolute time spent by each dam in each category
to account for these small variations. Data are presented as mean ± SEM.
76
Chapter 3:
Interaction between early-life stress
and reduced expression of MET leads to morphological changes
in hippocampo-amygdalar neurons in mice
3.1. Abstract
Research incorporating interactions between genetic and environmental factors, such as
early-life stress (ELS) can provide more insight in the mechanisms involved in the developing
brain and neurodevelopmental disorders.
Previously, our lab has identified a functional single nucleotide polymorphism
(SNP; rs1858830 ‘C’ allele) in the promoter region of the gene encoding the MET receptor
tyrosine kinase, that has been associated with increased risk of autism spectrum disorder (ASD).
Partial or complete deletion of Met in mouse brain results in altered morphology in hippocampus
and cortical areas. In addition, CA1 hippocampal neurons with reduced MET expression show
precocious maturational characteristics during the first two postnatal weeks. While ELS has
previously shown to affect neuronal morphology in mice, it is unknown how ELS would affect
morphological characteristics of neurons that are functionally and structurally more mature, due
to a reduction of MET protein in the brain.
Here, we introduced ELS from postnatal day (P)2 to P9 using a limited bedding and
nesting ELS paradigm in both wild-type mice, and Met
+/-
mice that have 50% of MET protein
expression in the central nervous system. In young adult mice, we analyzed neuronal
77
morphology of pyramidal neurons in the CA1 region of the ventral hippocampus that project to
the basolateral amygdala (BLA). We observed decreased basal and apical dendritic complexity
of these specific projection neurons in wild-type mice as a result of ELS. Non-stressed Met
+/-
mice also showed a reduction in dendritic complexity. Interestingly, ELS in Met
+/-
mice had the
opposite effect on neuronal arbor complexity, as it increased apical and basal arbor complexity in
Met
+/-
mice. Spine morphology on the basal arbor on the other hand was unaffected by ELS,
Met
+/-
, or the combination of both factors. We hypothesize that the responsiveness to ELS in
neurons with reduced expression of Met is altered due to a precocious maturational state. These
results provide a novel insight into the interaction of well-delineated genetic and environmental
factors on neuronal morphology in connections that are relevant to complex brain functions.
3.2. Introduction
Our growing understanding of neurodevelopmental disorder pathophysiology is primarily based
on separate contributions of genetic and environmental factors. For example, autism spectrum
disorder (ASD) research has identified many genetic variations that can increase risk of
developing this neurodevelopmental disorder (De Rubeis and Buxbaum, 2015), while several
environmental impacts have also been associated with higher risk (Windham et al., 2006,
Roberts et al., 2007, Volk et al., 2014). For ASD and other developmental disorders, prenatal
stressful events have been identified as potential risk factor (Garner et al., 2012, Shonkoff et al.,
2012, Beversdorf et al., 2005, Kinney et al., 2008, Ward, 1990). These data, considered in the
context of a more variable stress response in children with ASD (Corbett et al., 2009, Spratt et
al., 2012) suggest that experiments incorporating genetic and environmental factors, such as
78
early-life stress (ELS), can provide more insight in the mechanisms involved in the developing
brain.
Our laboratory identified a functional single nucleotide polymorphism (SNP; rs1858830 ‘C’
allele) in the promoter region of the gene encoding the MET receptor tyrosine kinase (Campbell
et al., 2006), a receptor enriched at developing synapses (Eagleson et al., 2013) in hippocampal,
neocortical and other forebrain structures during normal circuit formation in mice and non-
human primates (Judson et al., 2011a, Judson et al., 2009). The SNP is associated with increased
risk of ASD and reduces MET transcription by approximately 50% in in vitro assays (Campbell
et al., 2006, Jackson et al., 2009). Post-mortem studies have revealed that MET protein is
significantly reduced in temporal cortex of individuals with ASD (Campbell et al., 2007,
Voineagu et al., 2011), as well as in individuals with a ‘CC’ genotype but without a diagnosis of
ASD (Campbell et al., 2007). Furthermore, human neuroimaging studies show atypical neural
network activation to emotional faces, and white matter integrity in typically developing
adolescents with a ‘CC’ genotype, but with a stronger effect size in those with ASD (Rudie et
al., 2012). Moreover, a rare heterozygous deletion of the MET gene in a family resulted in severe
autism in the proband and milder social, language, and attention impairments in a sibling
(Lambert et al., 2014). These data suggest that a combined genetic risk of reduced MET
expression with other non-heritable risk factors will alter the molecular and structural integrity of
social-emotional circuits.
Complete deletion of Met in the dorsal pallium in mice leads to complex structural changes in
pyramidal neurons of the neocortex (Judson et al., 2010). Mice with complete or partial deletion
79
of Met in the central nervous system exhibit altered interlaminar excitatory drive in neocortex
(Qiu et al., 2011), and results of a recent study suggest that, during the first postnatal weeks, CA1
neurons with reduced MET expression are in a functionally and morphologically more mature
state compared to wild-type neurons (Qiu et al., 2014). Similarly, ELS has been shown to induce
morphological changes in pyramidal neurons of the neocortex and hippocampus in mice and rats
(Yang et al., 2015, Liao et al., 2014, Ivy et al., 2010, Brunson et al., 2005).
In this study we introduced ELS that targets circuitry in which MET functions during
development, namely the cornus ammonis area 1 (CA1) of the hippocampus, in both wild-type
and Met
+/-
mice with 50% of MET protein expression in the central nervous system. Specifically,
we analyzed pyramidal neurons in the CA1 region of the ventral hippocampus, with projections
to the basolateral amygdala (BLA), regions that are also affected by ELS (Danielewicz and Hess,
2014, Dubé et al., 2015, Malter Cohen et al., 2013, Moriceau et al., 2009, Réus et al., 2011).
Here, ELS decreased basal and apical dendritic complexity of these specific ventral CA1 to BLA
projection neurons in wild-type mice. Non-stressed Met
+/-
mice showed a similar reduction in
dendritic complexity. Surprisingly, ELS had opposite effects (i.e. an increase in complexity) on
neuronal morphology in Met
+/-
mice. We hypothesize that the precocious maturational state due
to a reduction of MET protein alters the responsiveness of these neurons to ELS. While these
projection neurons represent a small fraction of an intricate network, these results provide a
novel insight into the interaction of well-delineated genetic and environmental factors on
neuronal morphology in connections that are relevant to complex brain functions.
80
3.3. Methods
3.3.1. Animals
All animal procedures were approved by the Institutional Animal Care and Use Committee at the
University of Southern California. C57BL/6J mice were housed in a temperature- and humidity-
controlled vivarium (20-22 °C, 40-60% humidity) that was maintained on a 12-hour light/dark
cycle (with lights on at 6 am), in standard ventilated JAG mouse cages (Allentown Inc., NJ) with
ad libitum access to regular rodent chow and filtered water in drinking bottles. Cages were
cleaned weekly, with ample Alpha-Dri bedding (Shepherd Specialty Papers, MI) and one
standard pulped cotton fiber nestlet square (Ancare Corp., NY). All experiments were performed
using second litters of each dam in the study, which facilitates larger litter size and reduces the
variations in maternal care that we have observed with inexperienced, first-time mothers.
Breeding was initiated with mice of approximately eight weeks of age. During breeding, three to
four females were housed with one male. The males were removed from the cage after 10 days,
and the females were placed in a clean cage with one nestlet square and housed singly from
approximately embryonic day 16 until the pups were born. Pups were weaned on postnatal day
(P)21. Only males were used for this study.
To generate control and Met
+/-
experimental mice, Met
fx/fx
dams were mated with Nestin
Cre
males
(generated by crossing Nestin
Cre
males with C57Bl/6J females) to produce litters with both
control (Met
fx/+
) and heterozygous Nestin
Cre
/Met
fx/+
(Met
+/-
) pups. In Met
+/-
pups, exon 16 of one
Met allele is flanked by loxP sites, and subsequent expression of Cre will create mice with
heterozygous expression of Met. The dams of the experimental mice do not express Cre, and by
all measures are no different than wild-type animals. All mice were genotyped after weaning
81
according to a previously published protocol (Judson et al., 2009) with minor modifications: the
final elongation step in the Nestin
Cre
reaction was seven minutes, and the PCR product was 320
base pairs. For the Met
fx
reaction, the duration of the denaturation step during the amplification
cycles was set at one minute.
3.3.2. Early life stress paradigm
The ELS procedure was implemented based on Rice et al. (2008). The center floor of the cage
was covered with approximately 50 grams of Alpha-Dri bedding to absorb urine, and a stainless
steel raised wire floor with 10 x 10 mm square openings and 1 mm wire diameter (Cat#
RWF75JMV, Allentown Inc., NJ) was inserted above the bedding. Dams were not able to
retrieve the bedding to incorporate in their nests. Two-thirds (1.8 g) of a standard pulped cotton
fiber nestlet square was provided in each ELS cage. Control cages had approximately 160 g
identical bedding material and a full nestlet square, and both cage set-ups had identical access to
food and water. Dams were alternately assigned to control or ELS conditions based on the time
of birth of pups. On the morning of the second postnatal day (P2) after birth (P0), the
experimental dams were weighed and subsequently placed into ELS or control cages. The pups
were removed by hand one-by-one from the nest, weighed, and placed onto the wire floor of the
ELS cage or on the bedding of the control cage after determining their sex using genital
pigmentation intensity (Wolterink-Donselaar et al., 2009). Three male and two female pups were
placed in each cage; the remaining pups in the litter were euthanized. The dam and litter were
left undisturbed for 7.5 days until the afternoon of P9, when the dam and pups were removed
from the ELS cage, weighed, and placed into a control cage with ample Alpha-Dri bedding and
one nestlet square. Cage changes were carried out on P16, and pups were weaned on P21. Litters
82
from control dams were similarly culled to three males and two females on P2, and received a
regular clean cage with one nestlet square on P2, P9, P16, and P21. Following weaning, the mice
were housed with same-sex littermates until intra-amygdalar surgery on P55 with weekly cage
changes. After surgery, the mice were placed in clean cages with their original cage mates until
P60, at which age they were used for morphological evaluation.
3.3.3. Immunoblot analysis of MET protein expression in hippocampus
P9 mice were decapitated, and whole hippocampus was subsequently dissected on ice, snap
frozen in liquid nitrogen, and stored at -80˚C until processing. Tissues were homogenized using a
1 ml glass dounce and pestle (‘tight’, Wheaton) in 10:1 (volume buffer : weight tissue) ice cold
homogenization buffer (10 mM Tris-HCl, pH7.4, 1% SDS, 1% protease inhibitor cocktail
(Sigma #8340), 1% phosphatase inhibitor 2 (Sigma #5726)). The homogenate was centrifuged
for 15 minutes at 1,000 x g at 4˚C, and the supernatant transferred to an Eppendorf tube. After
taking out a small aliquot for protein quantification (Bio-Rad DC Protein Assay), 5x final sample
buffer was added to the supernatant, and heated at 100˚C for 5 minutes. The samples were then
centrifuged for 5 minutes at 13,000 rpm in a standard tabletop microcentrifuge, and the
supernatant was transferred to a new tube. The samples were stored at -20˚C. The following day,
samples were diluted with 1x final sample buffer to load 40 ug total protein in 10 ul of each
sample per lane on a fresh 7.5% acrylamide/bis gel. After separating proteins on the gel, the
samples were transferred onto nitrocellulose membrane for 17 hours at 30V in a 4˚C cold room.
Membranes were subsequently blocked with 5% blotto (Cell Signaling #9999S) in TBST buffer
(50mM Tris HCl, 0.15M NaCl, 0.05% Tween 20) for 2 x 30 minutes, and incubated with anti-
MET antibody (Santa Cruz Biotechnology #8057, 1:3000) in 5% blotto/TBST at 4˚C overnight.
83
Immunoblots were washed 8 x 5 minutes in 5% blotto/TBST, incubated for one hour with
secondary antibody (Jackson ImmunoResearch #715-035-150, 1:5000) in 5% blotto/TBST,
washed 8 x 5 minutes in TBST, and incubated with Femto chemiluminescent substrate
(ThermoFisher #34095). The signal was captured and analyzed using a CCD camera (UVP
BioImaging System) and VisionWorksLS Image Acquisition and Analysis software
(VisionWorks). The density of the correctly sized protein band was determined and adjusted for
background signal using background subtraction. The immunostaining process was repeated for
anti-α-Tubulin protein (primary antibody #CP06 EMD Millipore, 1:200,000), and densitometry
values of MET were normalized to α-Tubulin measurements. To assure a linear detection range
of relative changes in protein expression, antibody concentrations were tested at different
concentrations and with different protein quantities prior to analysis of experimental tissues.
3.3.4. Intra-amygdalar injection of fluorescent tracer
At age P55, 29 WT (15 control and 14 ELS) and 31 Met
+/-
(15 control and 16 ELS) mice were
weighed, and anesthetized with an intraperitoneal injection of a mixture of 100 mg/kg Ketamine
and 10 mg/kg Xylazine in phosphate buffered saline (PBS). After ensuring the mice were deeply
anesthetized by the absence of a corneal reflex and lack of responsiveness to toe pinch, the heads
of the mice were shaved and cleaned with iodine followed by ethanol, after applying opthalamic
ointment to prevent drying of the cornea. The mice are susequently immobilized in a stereotaxic
apparatus (Stoelting Co. #51615) using blunt ear bars, on a 39˚C Deltaphase Isothermal heating
pad. To prevent hypoxia during surgery, the tongue is pulled to the side before securing the
mouth into the mouth piece of the stereotaxic apparatus, and supplemental oxygen is supplied
through a tube in close proximity to the mouth. A small incision is made in the skin covering the
84
skull with sterile scissors and the skin is retracted to expose the skull. After establishing the
appropriate stereotaxic position (-1.34 mm caudal, 3.3 mm lateral, from Bregma), small holes are
drilled in the skull over the injection sites in both hemispheres. A borosilicate micropipette with
filament (1.5mm OD x 0.86mm ID, Warner Instruments #G150F-3) was pulled with a Sutter P-
1000, and scored and cut back one quarter with a final opening of ~30 uM as measured with a
Microforge Narishige MF-830. The pulled micropipette was held in place by a probe holder
(PicoNozzle kit 5430-ALL, World Precision Instruments), loaded with red fluorescent beads
(Lumafluor, Inc.), and was slowly lowered into the basolateral amygdala, 4.4 mm ventral from
the dura mater. Approximately 50-100 nl of tracer was expelled using a Narishige IM-200
Microinjector with 10 psi pressure created by compressed nitrogen. The micropipette remained
in the final injection position in the brain for 10 minutes, after which it was slowly raised to
prevent dye leakage during retraction. The injection was then repeated in the opposite
hemisphere. The order of injecting left and right hemispheres was alternated between mice. After
the injections, the skin was sewn close with silk sutures, cleaned, and covered with triple
antibiotic ointment. An analgesic (Ketoprofen) was injected subcutaneously at a dose of 5
mg/kg. The mouse was subsequently removed from the stereotaxic apparatus and placed on its
side on a 39˚C Deltaphase Isothermal heating pad to recover from anesthesia. After recovery
from anesthesis, mice were housed individually overnight in a cage with clean bedding in the
standard housing vivarium. On the following day, the mice were reunited with their original
litter- and cage mates that all underwent surgery the previous day. The sutures remained in place
until perfusion procedures at P60, at which time the fluorescent dye had reached the target areas
(ventral CA1 of hippocampus). All mice appeared alert and responsive after recovery from
anesthesia and surgery, and none of them showed adverse reactions.
85
3.3.5. Perfusion and preparation of slices for cell injections
At age P60, the mice were weighed, and anesthetized with an intraperitoneal injection of a
mixture of 100 mg/kg Ketamine and 10 mg/kg Xylazine in PBS. A deep level of anesthesia was
ensured by monitoring loss of corneal reflex and lack of responsiveness to toe pinch. The mice
were immobilized, and an incision was made into the abdomen and diaphragm, which exposes
the heart to allow for transcardial perfusion with fixative. The fixative solution was prepared
freshly on each perfusion day as described in Dumitriu et. al (2011), and kept on ice. After
inserting a needle into the left ventricle, a cut was made into the right atrium with small scissors,
and ~5 ml of 1% paraformaldehyde (PFA) was first pumped through the body at a rate of 5
ml/min using a perfusion pump (Minipuls, Gilson). The perfusion solution was then switched to
60 ml 4% PFA + 0.125% gluteraldehyde. After perfusion, the brain was removed from the skull,
and divided into three segments in the coronal plane using a brain block with 1 mm divisions, at
4 mm and 7 mm from the frontal pole. The anterior and posterior brain segments containing the
prefrontal cortex and hippocampus, respectively, were post-fixed in 4% PFA/0.125%
gluteraldehyde for four hours on ice. After post-fixation, the segments were rinsed with PBS, and
sectioned on a vibratome (#VT1200S, Leica,) in ice cold PBS at a thickness of 250 μm, with a
speed of 0.1 mm/sec and amplitude of 0.70mm. The frontal segment was cut coronally, and the
posterior segment containing the hippocampus was cut horizontally. Sections were kept in 0.1%
NaN
3
in PBS at 4˚C until cell injections.
3.3.6. Evaluation of amgydalar injection site
After perfusion, the middle segment of the brain containing the amygdala was post-fixed in 4%
PFA/0.125% gluteraldehyde overnight at 4˚C, then placed into sucrose solutions of increasing
86
concentration of 10%, 20% and 30% sucrose in PBS, for 24 hours at 4˚C for each step. The
brains were then frozen in dry ice-cooled isopentane, and stored at -80˚C until sectioning on a
cryostat. Sections of 20 μm were mounted on glass microscope slides with Prolong Gold with
DAPI, and imaged with a Zeiss Axioplan 2 microscope to evaluate the location of the fluorescent
tracer injection in the basolateral amygdala. An image of the section with the largest surface of
fluorescent tracer was captured using a 20x objective and AxioCam MRm camera, and the
images were stitched together using Adobe Photoshop. A translucent image derived from a
digital version of the Paxinos Mouse Brain Atlas (Franklin and Paxinos, 2007) corresponding to
the correct coronal section was added as a layer on top of the collated image of the brain. The
shape and edges of the amygdala (determined by a change in cell density with DAPI staining)
was used as a primary determinant of the correct anterior-posterior level. Only brains with
fluorescent dye present in the BLA and not in adjacent structures (e.g. amgydalar or external
capsules, lateral amygdala, central amygdala) were used for subsequent cell fill injections and
analysis. In 16 wild-type brains (seven control and nine ELS) and 14 Met
+/-
brains (six control
and eight ELS) from 16 independent litters the fluorescent tracer was located correctly in the
BLA. All groups (Control, ELS, Met
+/-
, ELS × Met
+/-
) were represented equally among the
anterior-posterior axis of the basolateral amygdala (Supplemental Figure 3.1).
3.3.7. Cell injections
To visualize tracer-positive cells in layer 2/3 of the prelimbic cortex and the ventral CA1 of the
hippocampus, pyramidal cells were injected with Lucifer Yellow dye. The slices were transferred
to the stage of a Leica DM LFSA microscope and immobilized using a slice harp (#SHD-27H/2,
Warner Instruments) in PBS. Lucifer Yellow dye (#L-453 ThermoFisher) was prepared by
87
dissolving the Lucifer Yellow CH Lithium salt at 8% in 0.1M tris-HCl (pH 7.5), and filtering the
solution through a Millix GV filter (0.22 μm, Millipore). The dye was injected using a
borosilicate micropipette with filament (1.0mm OD x 0.78mm ID), and an Ag wire micropipette
holder (#MEW-F10B, Warner Instruments) connected to a Kation Scientific M1200
microiontophoresis unit (modified range of 0-50 nA). The micropipette was diagonally advanced
into the tissue slices using a micromanipulator (Sutter #MP-225). Tracer-positive cells in the
correct brain area were identified using a Cy3 fluorescent filter through a 40x Leica 0.80 NA
water immersion objective, and injected with 0.1-5 nA of continuous negative current for
approximately five minutes, until the neuron was completely filled as evaluated intermittently
using a Lucifer Yellow fluorescent filter. Adjacent vibratome slices were used for whole cell and
spine analyses. For the former, cells were selected that were approximately five cell layers deep,
while for the latter, more superficial cells were injected. Due to suboptimal perfusion and/or
post-fixation, cells in the prelimbic cortex did not fill completely, particularly because of dye
leaks; therefore there were an insufficient number of cells from prelimbic cortex to analyze
statistically.
3.3.8. Whole cell imaging, reconstruction and analysis
Brain slices with dye-injected cells were mounted between a glass microscope slide and a #1.5
glass coverslip (#48393194, VWR) in Prolong Gold Antifade Mountant (ThermoFisher
#P36930), with 2 spacers of 120 μm thickness each (EMS Diasum #70327-13S) to prevent
compression of the sections. The slices were cured at least 24 hours before imaging. Neurons are
visualized using an Zeiss Axioscope equipped with automatic X-Y stage and Z meter driven by a
Neurolucida system (MicroBrightField), and an epifluorescent light source with FITC and CY3
88
filters under a 20X objective. Morphology of Lucifer Yellow filled neurons is captured with a
high sensitivity monochrome camera (Retiga 2000R, Qimaging). The reconstructed neurons are
analyzed using NeuroExplorer (NeuroLucida) to obtain total dendritic length of apical and basal
arbors, as well as dendritic complexity by performing a Scholl analysis with three-dimensional
spheres spaced 10 μm apart starting at the cell body.
3.3.9. Spine imaging, reconstruction and analysis
To obtain consistent images of spines on CA1 neurons, the initial signal from Lucifer Yellow
dye that was injected into the neurons was amplified using an antibody against Lucifer Yellow.
The brain slices were incubated with polyclonal Rb anti-Lucifer Yellow antibody (#A5750, Life
Technologies, at 1/3000 in antibody solution (5% sucrose, 2% BSA, 1% triton in PBS))
for 5 days at 4˚C. After 3 x 5 min washes with PBS, the slices were transferred to Biotin-SP-
conjugated AffiniPure Dk anti-rabbit secondary antibody (#711-065-152, Jackson
ImmunoResearch, at 1/300 in antibody solution) for two hours at room temperature. The slices
were washed again for 3 x 5 min with PBS, and incubated with Alexa Fluor 488-conjugated
streptavidin (#S-11223, Life Technologies, at 1/1000 in PBS) for two hours at room temperature.
After a final wash step of 3 x 5 min with PBS, the slices were dabbed dry and mounted between
a glass microscope slide and #1.5 glass coverslip (#48393194, VWR) in Prolong Diamond
(ThermoFisher #P36970) with 2 spacers of x 120 μm thickness each (EMS Diasum #70327-13S)
to prevent compression of the sections. The slices were cured at room temperature for 24 hours
before imaging. Images of spines were obtained with a Zeiss confocal microscope (LSM 710)
with 63x 1.40 NA oil-immersion objective. An optical zoom of 1.5x was used to obtain images
with a voxel size of 101 nm (x) x 101 nm (y) x 160 nm (z-step size) and a total image size of
89
103.4 x 103.4 μm (1024 x 1024 pixels). The pixel dwell time was 3.15 μs with 4x averaging. The
laser power of the 488 nm laser was kept as low as possible to prevent bleaching of the signal
(under 6%). The pinhole was set at 1 AU, and the gain at 680 V, and the images were captured in
16-bit Tiff format.
Images were processed with attenuation correction and 3D deconvolution using AutoDeblur
(Media Cybernetics, Inc., MD) and converted back to 16-bit Tiff with Fiji to prepare for spine
detection using NeuronStudio software. Spine measurements were averaged per image from two
individual neurons, that contained spine segments located >50 um from the cell body and of
second or higher order (with an average dendritic length of 115 μm per image). Dendrite
segments that were close to the top or bottom z-planes of the image stack were excluded. Spines
were automatically detected using NeuronStudio, and subsequently adjusted manually for correct
spine placement and detection. Manually added spines were excluded from volumetric spine
analysis (with RayBurst) due to an inherently difference in contrast detection in NeuronStudio
(using an average of the whole dendrite segment for automatic detection versus local contrast
around the spine for manual detection) which results in an overestimation of spine volume. The
manual spines were however included in spine length and spine density analysis.
3.3.10. Statistical analyses
Data were analyzed using GraphPad Prism 6 and SPSS/PASW 22 (IBM). MET protein
expression, apical and basal dendritic length, spine volume, spine density, and spine length were
analyzed using a two-way ANOVA (genotype x environment). A three-way repeated measures
ANOVA was used to analyze dendritic complexity and length for Sholl analysis (genotype x
90
environment x distance from soma). Tukey’s multiple comparisons tests were performed on
ANOVA tests (alpha = 0.05) when appropriate. Data are presented as mean ± standard error of
the mean (SEM).
91
3.4. Results
3.4.1. MET protein levels are reduced in hippocampus of Met
+/-
mice
Analysis of immunoblot densitometry results showed a main effect of genotype on the levels of
MET protein in the hippocampus of P9 (Figure 3.1, F
1,34
=194.1, p < 0.001) and P21 (F
1,27
=
93.14, p < 0.001), while ELS did not have an effect (P9: F
1,34
=1.08, p = 0.307; P21: F
1,27
=0.09,
p = 0.764). Post-hoc analyses indicated a significant decrease in MET expression in Met
+/-
and
Met
+/-
mice that had experienced ELS (p < 0.001).
Figure 3.1. MET protein expression is approximately 50% in hippocampus of P9 Met
+/-
mice compared to control mice, and does not change after ELS. (A) Representative
immunoblots with anti-MET and anti-α-tubulin antibody labeling of hippocampal tissue from P9
control (wild-type) mice compared to Met
+/-
mice, and control compared to ELS mice. (B) All
raw MET densitometry values from immunoblots were corrected for background signal, and
normalized to respective tubulin densitometry values on the same blot, and subsequently
normalized to the average of normalized values from control mice. Data are presented as mean ±
SEM. Individual data points are measurements from individual pups. *** p < 0.001
A
92
3.4.2. ELS has opposite effects on dendritic complexity in wild-type and Met
+/-
mice
Factorial repeated measures ANOVA of Sholl analysis showed that for the basal arbor, an
interaction effect exists between genotype and environment on the number of dendrite
intersections, and this interaction effect is not uniform across the basal dendritic tree, analyzed
by examining the interaction between genotype, environment and distance from the soma (Figure
3.2A, F
2.5,49.8
= 5.406, p < 0.01; Mauchly's test indicated that the assumption of sphericity had
been violated, thus a Greenhouse-Geisser correction was applied to df). Subsequent analyses
revealed that ELS decreased basal arbor complexity at 70-110 μm from the cell body (p < 0.05)
compared to mice in control conditions. In addition, in Met
+/-
mice that experienced ELS an
increased basal arbor complexity was observed at 50-110 μm from the cell body compared to
both ELS and Met
+/-
groups (p < 0.05). Therefore, the basal arbor complexity in the Met
+/-
x ELS
mice was not significantly different from control mice at any distance from the soma, suggesting
that ELS has opposite effects on arbor complexity in wild-type and Met
+/-
mice.
Sholl analysis revealed a similar interaction effect regarding apical arbor complexity, with an
interaction between genotype, environment and distance from the soma on the number of
dendritic intersections (Figure 3.2B, F
4.7,89.3
= 2.431, p < 0.05, using a Greenhouse-Geisser
correction). Similar to the basal arbor, ELS reduced apical arbor complexity at 140 μm and 160-
190 μm from the cell body (p < 0.05), and Met
+/-
also resulted in reduced complexity of the
apical arbor at 180-190 μm from the cell body (p < 0.05) compared to wild-type mice.
Additionally, Met
+/-
mice that had experienced ELS were not significant different in apical arbor
complexity compared to non-stressed wild-type mice.
93
Figure 3.2: ELS has opposite effects in Met
+/-
mice on dendritic complexity of neurons in
CA1 ventral hippocampus that project to the basolateral amygdala compared to wild-type
mice. (A) Basal and (B) apical arbor complexity is reduced as a result of ELS and Met
+/-
as
shown here by a decrease in the number of intersections in Sholl analysis, whereas ELS in Met
+/-
mice leads to an increase of complexity compared to non-stressed Met
+/-
mice. See Results
section for specific significant differences between groups at each distance from the soma. Data
are presented as mean ± SEM. Number of animals per group: Control n = 5; ELS n = 7;
Met
+/-
n = 5-6; Met
+/-
× ELS n = 6. Each n is an average of 2.3 neurons per animal (range 1-4).
* p < 0.05.
94
We also observed an interaction effect of genotype and environment on dendritic length of basal
arbors (F
1,20
=14.27, p < 0.01). Post-hoc tests revealed that ELS increased basal arbor length in
Met
+/-
mice compared to non-stressed Met
+/-
mice and wild-type mice that experienced ELS
(Figure 3.3A, p < 0.05). Despite changes in complexity in sections of the apical arbors, the apical
dendritic length (Figure 3.3B), as well as the total dendritic length (Figure 3.3C) did not
significantly change as a result of ELS, Met
+/-
or the combination of the two factors. Examples of
representative neuron tracings with basal lengths close to the 25
th
, 50
th
, and 75
th
percentile are
presented in Figure 3.4.
95
Figure 3.3: ELS increases dendritic length of basal arbor in Met
+/-
mice. Dendritic length of
the basal arbor (A) is increased as a result of ELS in Met
+/-
mice compared to non-stressed Met
+/-
mice or wild-type mice that had experienced ELS, while (B) dendritic length of the apical arbor
and (C) total dendritic length of the complete neuronal arbor is unaffected. Data are presented as
mean ± SEM. Number of animals per group: Control n = 5; ELS n = 7; Met
+/-
n = 5-6;
Met
+/-
× ELS n = 6. Each individual data point is an average of 2.3 neurons per animal
(range 1-4). * p < 0.05.
Control ELS Met
+/-
Met
+/-
x ELS
0
1000
2000
3000
4000
Basal dendrite length (µm)
A
*
Control ELS Met
+/-
Met
+/-
x ELS
0
500
1000
1500
2000
2500
Apical dendrite length (µm)
B
Control ELS Met
+/-
Met
+/-
x ELS
0
1000
2000
3000
4000
5000
Dendrite length (µm)
C
96
Figure 3.4: Representative examples of reconstructed CA1 neurons that project to the
BLA, organized by length of basal dendritic arbor. Three examples per experimental group
with basal arbor lengths close to the 25
th
, 50
th
and 75
th
percentile in their respective experimental
groups are provided to show the distribution of basal lengths.
97
3.4.3. Spine characteristics are unaffected by Met
+/-
or ELS
We next analyzed whether spine characteristics of dendritic spines on the basal arbor of CA1
pyramidal neurons were affected by ELS, Met
+/-
, or a combination of Met
+/-
and ELS. We found
no main effect of ELS on spine volume (Figure 3.5A, F
1,22
= 2.01, p = 0.170), spine density
(Figure 3.5B, F
1,22
= 0.80, p = 0.381) nor spine length (Figure 3.5C, F
1,22
= 0.01, p = 0.939).
Similarly, Met
+/-
had no main effect on these measurements (volume: F
1,22
= 0.42, p = 0.526,
density: F
1,22
= 0.003, p = 0.959, length: F
1,22
= 1.38, p = 0.253), nor did we find an interaction
effect (volume: F
1,22
= 0.0007, p = 0.980, density: F
1,22
= 0.13, p = 0.717, length: F
1,22
= 0.60, p =
0.445).
98
Figure 3.5: Spine characteristics are not significantly different as a result of ELS, Met
+/-
or
a combination of both factors. No significant effects of ELS, Met
+/-
or a combination of both
factors were measured on (A) spine head volume, (B) spine length, or (C) spine density.
99
3.5. Discussion
The current study shows that either Met
+/-
genotype or ELS alone results in decreased dendritic
complexity and length of pyramidal neurons in the ventral CA1 of the hippocampus that project
to the BLA. Interestingly, ELS resulted in opposite effects on dendritic complexity (i.e. an
increase) in Met
+/-
mice compared to wild-type mice. The effect appears to be selective, as Met
+/-
,
ELS, or the combination of these genetic and environmental factors did not alter spine
characteristics. Furthermore, ELS did not alter neuronal morphology directly through changing
MET protein expression, as protein levels in hippocampus on P9 of mice that had experienced
were similar to non-stressed mice.
Several research groups have previously shown that reducing expression of the MET receptor or
inhibiting its phosphorylation can affect dendritic morphology in rodents. For example,
knockdown of Met with short interference RNA (siRNA) in dissociated cultures from rat
hippocampus reduced dendritic length and number of primary dendrites of pyramidal neurons,
with a lower expression of MET correlating to a larger effect (Lim and Walikonis, 2008).
Another study shows that two days of inhibiting MET phosphorylation in vitro with an antibody
against its ligand, hepatocyte growth factor (HGF), decreased dendritic branching and length of
basal and apical pyramidal neurons in cerebral cortex slices from mice (Gutierrez et al., 2004).
Furthermore, our lab has examined the effects of complete ablation of Met in the dorsal pallium.
Apical dendritic complexity in both layer II/III and layer V of the anterior cingulate cortex was
reduced, whereas in the basal arbor an increase or no change in basal dendritic complexity was
observed in layer II/III and V, respectively. A recent in vivo study using siRNA to reduce the
expression of MET in developing mouse hippocampus also showed a reduction in dendritic
100
complexity of pyramidal neurons (Qiu et al., 2014). In this study, siRNA was introduced in the
dorsal hippocampus on embryonic day 14.5, and dendritic morphology of transfected CA1
neurons was assessed on P22-25. The complexity and length of apical arbors –but not the basal
arbors– were reduced compared to the control group (Qiu et al., 2014). In multiple studies,
therefore, a reduction in MET protein (phosphorylation) has growth-inhibiting effects on the
length and complexity of dendrites, although some studies show complex combination of both
positive and negative changes in the basal and apical arbors of certain pyramidal neurons,
respectively (Judson et al., 2010).
The aforementioned reports by Judson et al. (2010) and Qiu et al. (2014) additionally show
changes in spine morphology. Specifically, MET ablation in the dorsal pallium leads to an
increase in spine head volume of spines on basal dendrites in layers II/III and V in the anterior
cingulate cortex. No change in spine density or spine length was observed (Judson et al., 2010).
Partial reduction of MET expression in the dorsal CA1 region of the hippocampus also led to an
increase of spine head volume of spines on the apical arbor in four-week old mice, as well as a
decrease in spine density (Qiu et al., 2014). We were not able to replicate these findings in our
current study, as we did not observe changes in spine head volume, spine density, or spine
length. A possible explanation for this unexpected result is that the neurons in this study did not
represent the general population of neurons that previously have been shown to be affected by
Met genotype (or ELS as discussed below). Analyzing additional tracer-negative neurons to
compare to our current results would not suffice however, as it cannot be excluded that these
neurons project to a different part of the BLA that did not contain the tracer injection.
101
Similar to the effects of a reduction of MET protein, research groups have shown that ELS can
decrease dendritic complexity and length of pyramidal neurons in rodents. ELS due to limited
bedding during the first postnatal weeks induces a reduction in dendritic complexity and/or
length of pyramidal neurons in the cortex (Yang et al., 2015) as well as in dorsal hippocampus
(Ivy et al., 2010, Brunson et al., 2005, Liao et al., 2014) in later life. The present data show
similar effects of ELS on complexity of ventral hippocampo-amygdala neurons. Maternal
separation of rat pups during the first postnatal weeks had similar effects as limited bedding
(Romano-Lopez et al., 2015, Chocyk et al., 2013, Bock et al., 2005, Eiland and McEwen, 2012),
although application of the stressor in older pups induced an opposite effect on dendritic length
(an increase) of cortical and hippocampal neurons compared to earlier stressors (Bock et al.,
2005, Xie et al., 2013). ELS also can induce changes in spine morphology and number. Most
studies report a decrease in spine density in hippocampal (Liao et al., 2014, Wang et al., 2011,
Wang et al., 2013) and cortical neurons (Yang et al., 2015, Bock et al., 2005), although some
have observed mixed results or an increase as a result of applying the stressor during later time
periods (Bock et al., 2005, Xie et al., 2013). These two studies provide early support for the
hypothesis that ELS in older pups leads to opposite effects on dendritic morphology and spine
density. If neurons in hippocampus of Met
+/-
mice develop precociously, the effects of ELS
during the first week in these mice would mimic more closely the effects of ELS in the second
week in wild-type mice; thus, in Met
+/-
mice in our study, ELS would indeed lead to an increase
in dendritic complexity as we have observed.
In support of this explanation, Qiu et al. (2014) demonstrated that MET is important for
maturation of neurons, a hypothesis that was first proposed by Judson et al. (2011b).
102
Qiu et al. demonstrated that CA1 neurons that had reduced or no expression of MET showed
more mature electrophysiological properties, as well as precocious expression levels of AMPA
and NMDA receptor subunits at the synaptic surface (Qiu et al., 2014). Mature characteristics
were visible at the end of the first and second postnatal week, but normalized to control levels on
P26-28. These results suggest that reduced expression of MET results in a different neuronal
maturational state than in wild-type neurons expressing the MET receptor at regular levels.
Therefore, how would precociously mature neurons respond to ELS? The research by Bock et al.
(2005) and Xie et al. (2013) show that ELS by separating pups from the dam on P14 through P16
increases dendritic length, complexity, and spine density. The results in the current study show
indeed that ELS in Met
+/-
mice results in an increase of dendritic complexity compared to the
potent effect of ELS on decreasing wild-type neuron complexity, suggesting that the maturational
state renders Met
+/-
mice differentially responsive to ELS (Figure 3.6).
103
Figure 3.6: Overview of morphological changes due to the differential effects of ELS on
wild-type and Met
+/-
mice. Non-stressed Met
+/-
mice display decreased complexity of ventral
CA1 neurons in the hippocampus that project to the BLA. ELS decreases dendritic complexity of
these neurons in wild-type mice, but increases complexity in Met
+/-
mice, possibly due the altered
maturational state of neurons as a result of reduced MET protein expression.
104
To test whether our results are indeed due to altered maturational properties of the brain, we
could determine whether a later ELS time period in wild-type mice (e.g. starting in the second or
third postnatal week) can evoke similar morphological effects as in Met
+/-
mice during the first
week. Likewise, overexpression of MET in the nervous system could make neurons less mature,
rendering ELS that starts in the second or third postnatal weeks perhaps as effective in
remodeling dendritic morphology as ELS in the first postnatal week in wild-type mice. Qiu et al.
(2014) found that local overexpression of MET in the hippocampus had opposite effects on
several electrophysiological properties compared to a reduction of MET, suggesting that the
neurons with higher expression of MET are in a less mature state.
Most, if not all, ELS studies completing morphological analyses in the hippocampus have
focused on the dorsal hippocampus, most likely due to practical issues of the non-coronal
orientation of neurons in the ventral hippocampus that require creating horizontal sections.
However, the ventral hippocampus has denser connections with regions that are involved in
emotional behaviors (Strange et al., 2014) that are known to be affected by ELS (Bock et al.,
2014). The dorsal hippocampus is connected with areas that are mainly regulating spatial
behaviors (Strange et al., 2014). Additionally, different molecular markers have been found in
the dorsal and ventral hippocampus, which suggest that the dorsal and ventral regions indeed
should be considered anatomically distinct structures (Dong et al., 2009). However, although
molecular markers are demarcated precisely, the transition from ventral to dorsal hippocampus
regarding differential connectivity is more continuous than was previously assumed (Strange et
al., 2014).
105
In addition to differences between the dorsal and ventral hippocampus, two types of neurons
exist within the CA1 and subiculum of the ventral hippocampus. These two neuronal types have
distinct firing patterns, morphology, and projection areas. Neurons that project to the amygdala
are found only in ventral CA1 regions of the hippocampus (Kim and Spruston, 2012), are ‘late-
bursting’ neurons (previously called ‘regular-spiking’), and lack a distinct tuft in the apical arbor,
while the basal arbor is more complex (Graves et al., 2012). The second type is called a ‘late-
bursting’ neuron that has a well-developed apical tuft. Two streams of information are processed
through these distinct neurons, with the neurons projecting to the amygdala being part of an
indirect pathway carrying ‘emotional information’, receiving information from the entorhinal
cortex, to the dentate gyrus, the proximal CA3, to the basal dendrites of neurons in the distal
CA1 (close to the subiculum) (Kim and Spruston, 2012, Graves et al., 2012). A reduction in the
basal dendritic complexity in our study due to ELS or Met
+/-
suggests that these morphological
changes could modulate emotional states, such as anxiety-like behaviors. In addition, many
ventral CA1 neurons that are in close proximity to the subiculum (and thus more likely to be part
of the indirect pathway) have collateral axons that project simultaneously to mPFC and
amygdala (Ishikawa and Nakamura, 2006), and anxiety-like behaviors are regulated by
projections from the ventral hippocampus to the medial prefrontal cortex (Adhikari et al., 2010).
It is reasonable to expect that these neurons that regulate anxiety-like behavior bifurcate and also
project to the amygdala. A recent study from Ciocchi et al. (2015) seems to contradict this
hypothesis, concluding that only non-bifurcating CA1 neurons that project to mPFC are
responsible for regulating anxiety-like behavior, rather than neurons that bifurcate and also
project to the amygdala. However, this conclusion may be based on incomplete data, as the target
area studied in the ventral hippocampus is located too rostral to detect a sufficient number of
106
neurons in CA1 that project to the amygdala. The analysis appears to have low statistical power
for this specific projection group. We suggest that the projection neurons analyzed in the present
study may be able to modulate emotional behaviors such as anxiety-like behavior, and that both
ELS and Met
+/-
could affect these behaviors by altering morphological features of these and
other neurons.
The current findings of altered morphology in projection neurons from the ventral CA1 to BLA
reflect what may be a larger impact on neurons in multiple circuits impacted by ELS or a
reduction of MET protein. The specific neurons that are the focus of this study are a small part of
complex circuits that adapt to altered input due to intrinsic and extrinsic factors. For example,
brainstem autonomic neurons, which receive input from key forebrain regions, express MET and
may be involved in the response to ELS early in development. The continuing focus on areas
such as the PFC, BLA and HC in ELS research (including the current study) are likely a result of
elaborating on early positive morphological findings in the field, and the intriguing association of
these brain areas to altered behaviors. Future techniques will hopefully allow for high-throughput
analysis of changes in neuronal morphology in all other brain areas as a result of environmental
and genetic factors.
In summary, the present research study shows that the level of MET expression in the brain
affects morphological changes in response to adverse environments, such as ELS. Because a
large fraction of the human population carries a single nucleotide polymorphism (rs1858830 ‘C’
allele) that has been shown to reduce MET expression in vitro (Campbell et al., 2006), and
reduced expression of MET has been associated with ASD (Campbell et al., 2007), clinical
107
studies that examine the effect of ELS on the etiology of neurodevelopmental disorders should
take MET genotype (‘GG’, ‘GC’ or ‘CC’) into account as potential moderator of phenotype
effect size.
108
3.6. Supplemental Material
Supplemental Figure 3.1. Locations of BLA tracer injection sites. (A) Image of a
representative successful tracer injection on a coronal section indicated with a blue arrow in the
basolateral amygdala, with schematic overlay of brain area (Franklin and Paxinos, 2007). (B) All
groups (Control, ELS, Met
+/-
, Met
+/-
x ELS) were represented equally among the anterior-
posterior axis (measured as distance from Bregma) of the basolateral amygdala (F
3,26
= 0.19, p =
0.900). (C) The location along the anterior-posterior axis of the center of the injection side for
each brain that was included in the final morphometric analysis.
109
Chapter 4:
Early-life stress alters social-emotional behaviors in mice with
reduced expression of MET receptor tyrosine kinase
4.1. Abstract
Early adverse experiences during childhood (e.g. abuse or neglect) increase risk of
anxiety, mood, and post-traumatic stress disorders (PTSD) in adulthood, and specific gene-
environment interactions may further increase risk. A common promoter variant in humans that
reduces expression of MET in vitro (rs1858830 ‘C’ allele) has been associated with increased risk
of autism spectrum disorder (ASD) in several independent cohorts, as well as with altered
functional and structural brain connectivity in the general population. Inhibition or reduced
expression of MET in mice leads to precocious maturation of neurons, increased intralaminal
cortical connectivity, reduced neuronal arbor complexity, and reduced synaptogenesis.
Previously, behavioral tests of Met
+/-
mice, with 50% of MET protein expression in the central
nervous system, have shown reduced fear acquisition and memory, and a mild reduction in
anxiety-like behavior. It is unknown whether early-life stress (ELS) during the peak of MET
protein expression will affect these and other behavioral measurements in Met
+/-
mice.
Here, we induced ELS in wild-type and Met
+/-
mice using a limited bedding and nesting
ELS paradigm from postnatal day (P)2 to P9. In adult mice, we assessed anxiety-like behaviors
on the elevated-plus maze (EPM), social interaction behaviors with a direct social interaction test
(DSI) and fear acquisition, memory and extinction using a contextual fear conditioning
110
paradigm. Our results reveal that Met
+/-
genotype resulted in a reduction of anxiety-like
behaviors as well as an impairment of contextual fear memory, and ELS reduced the number of
social interactions, specifically in Met
+/-
mice.
These results show that ELS and altered Met brain expression affect different behavioral
outcomes in mice, thereby increasing the number of affected behavioral domains in Met
+/-
mice
that experience ELS. Moreover, Met
+/-
mice that have experienced ELS showed a reduction in
the number of social interactions compared to non-stressed Met
+/-
mice, suggesting that these
mice may be more responsive to the effects of ELS. The results warrant further research of the
mechanisms behind this gene-environment interaction, and provide preliminary support for
examining a potential interaction effect between early adverse experiences and the rs1858830
‘C’ allele on (mental) health outcomes in the human population.
4.2. Introduction
ELS during childhood, for example abuse or neglect, is associated with increased risk of PTSD,
and anxiety and mood disorders in adulthood (Green et al., 2010, Lang et al., 2008, Widom,
1999). Additionally, family and twin studies show that the heritability of these disorders is
approximately 0.3-0.4 (Sullivan et al., 2000, Hettema et al., 2001, Stein et al., 2002).
Subsequent gene-wide association studies (GWAS) and rare variant and mutation analyses have
resulted in discovery of several genetic risk factors associated with, for example, PTSD (Almli et
al., 2014), but a large fraction of heritability of many mental health disorders is still unexplained
by single factors, reflecting opportunities for discovery of additional gene-environment,
polygenic, or more complex interaction effects (Duncan et al., 2014, Sharma et al., 2015).
111
Animal studies can help elucidating interaction effects of early-life stress (ELS) with genetic
variations in a controlled fashion, since both the genetic background and environment can be
held constant by using inbred animals and standard housing conditions. The outcomes can
provide insight into the mechanisms behind altered brain development, as well as provide
support for research of specific interactions in clinical studies. Several genes that are associated
with brain disorders and diseases have been shown before to interact with ELS to affect
behavior, for example reelin, the mineralocortiocoid receptor and estrogen receptor-beta
(Ognibene et al., 2007, Kanatsou et al., 2016, Tsuda et al., 2014).
Here, we focus on the effects of ELS and reduced expression of Met, a gene encoding for the
MET receptor tyrosine kinase. A single nucleotide polymorphism (SNP) in the promoter region
of the MET gene in humans is associated with increased risk of ASD in multiple independent
cohorts (the rs1858830 ‘C’ allele) (Campbell et al., 2006). Interestingly, in people without a
diagnosis of autism spectrum disorder (ASD), the presence of the ‘C’ allele is associated with
altered functional brain connectivity in response to social stimuli (faces with emotional
expressions) (Rudie et al., 2012). This study also showed an interaction of ASD diagnosis and
‘C’ allele dosage, suggesting other (environmental) factors may affect functional and diagnostic
outcome. The ‘C’ risk variant alters expression of MET in a reporter plasmid in several cell lines
in vitro by approximately 50% (Campbell et al., 2006). In previous behavioral studies in our lab,
partial deletion of MET in the central nervous system (CNS) results in reduced anxiety-like
behavior, and impaired cued fear learning and memory, while complete deletion of MET in the
CNS results in mild sociability impairments (Thompson and Levitt, 2015). Moreover, deletion of
MET from a subset of neurons results in different behavioral outcomes. Deletion of MET from
112
the dorsal pallium (Thompson and Levitt, 2015) or serotonergic neurons (Okaty et al., 2015)
results in decreased activity levels and decreased sociability, respectively. These combined
results suggest that the interaction between cells with differential levels of MET expression can
affect circuitry that modulates social-emotional behavior in unique ways. This is also illustrated
by the observation that, in humans, MET expression is highly differential between the temporal
and frontal cortices in postmortem samples from the general population, whereas in people
diagnosed with ASD, expression levels of MET are not different between these two brain areas
(Voineagu et al., 2011).
ELS in mice has been shown to affect social-emotional behaviors related to anxiety- and mood
disorders and PTSD in humans. For example, many research groups have studied anxiety-like
behaviors, social interaction, and fear learning and memory in mice. While anxiety-like
behaviors, for example as measured on the elevated plus maze (EPM), are often unaffected after
ELS (Veenema et al., 2007, Zoicas and Neumann, 2016), social interactions and fear memory are
generally impaired in adult mice that had experienced ELS (van der Kooij et al., 2015, Wang et
al., 2011).
These results prompted the question whether reduced expression of Met in the CNS interacts
with ELS to produce altered social-emotional behavior in mice. We observed that Met
+/-
genotype resulted in a reduction of anxiety-like behaviors, as well as an impairment of
contextual fear memory, but not fear acquisition or extinction. ELS reduced the number of social
interactions, specifically in Met
+/-
mice. General activity levels were not affected by genotype or
ELS. These results suggest that Met
+/-
mice may be more responsive to the effects of ELS.
113
Furthermore, genetic and environmental factors in this gene-environment interaction study had
differential impacts on different social-emotional behaviors, and thus the combination of both
factors resulted in more affected behavioral domains than each factor alone.
4.3. Methods
4.3.1. Animals
All animal procedures were approved by the Institutional Animal Care and Use Committee at the
University of Southern California. To generate control and Met
+/-
experimental mice
(backcrossed onto a C57Bl/6 background), Met
fx/fx
dams were mated with Nestin
Cre
males
(generated by crossing Nestin
Cre
males with C57Bl/6J females) to produce litters with both
control (Met
fx/+
) and heterozygous Nestin
Cre
/Met
fx/+
(Met
+/-
) pups. In Met
+/-
pups, exon 16 of one
Met allele is flanked by loxP sites, and subsequent expression of Cre will create mice with
heterozygous expression of Met. The dams of the experimental mice do not express Cre, and by
all measures are no different than wild-type animals. All mice were genotyped after weaning
according to a previously published protocol (Judson et al., 2009) with minor modifications: the
final elongation step in the Nestin
Cre
reaction was seven minutes, and the PCR product was 320
base pairs. For the Met
fx
reaction, the duration of the denaturation step during the amplification
cycles was set at one minute.
The mice were housed in a temperature- and humidity-controlled vivarium (20-22 °C, 40-60%
humidity) that was maintained on a 12-hour light/dark cycle (with lights on at 6 am), in standard
ventilated JAG mouse cages (Allentown Inc., NJ) with ad libitum access to regular rodent chow
114
and filtered water in drinking bottles. All experiments were performed using second litters of
each dam in the study, which facilitates larger litter size and reduces the variations in maternal
care that we have observed with inexperienced, first-time mothers. Breeding was initiated with
mice of approximately eight weeks of age. During breeding, three to four females were housed
with one male. The males were removed from the cage after 10 days, and the females were
placed in a clean cage with one nestlet square and housed singly from approximately embryonic
day 16 until the pups were born. Only males were used for behavioral tests in this study.
4.3.2. Early-life stress paradigm
The ELS procedure was implemented based on a method by Rice et al. (2008) The center floor
of the cage was covered with approximately 50 grams of Alpha-Dri bedding (Shepherd Specialty
Papers, MI) to absorb urine, and a stainless steel raised wire floor with 10 x 10 mm square
openings and 1 mm wire diameter (Cat# RWF75JMV, Allentown Inc., NJ) was inserted above
the bedding. Dams were not able to retrieve the bedding to incorporate in their nests. Two-thirds
(1.8 g) of a standard pulped cotton fiber nestlet square (Ancare Corp., NY) was provided in each
ELS cage. Control cages had approximately 160g Alpha-Dri bedding and a full nestlet square,
and both cage set-ups had identical access to food and water. Dams were alternately assigned to
control or ELS conditions based on the time of birth of pups. On the morning of the second day
(P2) after birth (P0), the experimental dams were weighed and subsequently placed into ELS or
control cages. The pups were removed by hand one-by-one from the nest, weighed, and placed
onto the wire floor of the ELS cage or on the bedding of the control cage after determining their
sex using genital pigmentation intensity (Wolterink-Donselaar et al., 2009). Three male and two
female pups were placed in each cage; the remaining pups in the litter were euthanized. The dam
115
and litter were left undisturbed for 7.5 days until the afternoon of P9, when the dam and pups
were removed from the ELS cage, weighed, and placed into a control cage with ample Alpha-Dri
bedding and one nestlet square. Cage changes were carried out on P16, and pups were weaned on
P21. Litters from control dams were similarly culled to three males and two females on P2, and
received a regular clean cage with one nestlet square on P2, P9, P16 and P21. On P21, after
weaning, tagging and tailing, mice were housed with their original littermates until behavioral
testing. Cage cleaning after weaning was performed on a regular weekly schedule, while
ensuring that mice did not receive a clean cage during the two-day period before any behavioral
test.
4.3.3. Behavioral tests
Anxiety-like behavior was measured first using the elevated-plus maze (EPM; P58-67), followed
by the direct social interaction test (DSI; P68-74), and contextual fear conditioning (starting at
P74-80), with at least one week between tests for individual mice. A general activity test was
performed in a different, subsequent cohort (P52-55). All behavioral tests were performed during
the light cycle in the afternoon, in designated mouse behavioral testing rooms that were within
the same vivarium corridor as the rooms in which all mice were housed. While acclimating for
behavioral tests, the mice had ad libitum access to food and water. When cage mates were tested
sequentially due to the availability of a single apparatus (EPM and DSI), animals that had been
tested were placed in a clean, temporary holding cage with bedding, food and water, until all
animals were tested and able to return to their home cage. The person running behavioral tests
was blind to genotype and environment experienced by each mouse.
116
4.3.3.1. Anxiety-like behavior: elevated-plus maze
The EPM apparatus (San Diego Instruments Inc., CA) was located on the floor in the center of a
10 x 8 ft. room with white walls and even LED lighting throughout the room. The arms of the
EPM measured 30 cm long, 6.5 cm wide, with 3 mm elevated edges along the open arms. The
floor of the maze and the walls of the closed arms were opaque white and opaque black,
respectively. The height of the maze was 40 cm above the floor, and the walls of the closed arms
extended 14 cm from the maze surface. The light intensity at the end of the open arm, the center
of the maze, and the end of the closed arm was 6-7 lux, 3 lux, and 0-1 lux, respectively.
Five days prior to the EPM test, mice were acclimated to human handling and to the opaque
glass beaker used for transporting mice from the cage to the maze, for a total of one hour a day
until testing day. Acclimation consisted of transporting the home cage to a room adjacent to the
behavior room, placing each mouse into the clean beaker and letting them exit, as well as letting
mice freely explore the beaker with the cage top closed for approximately ten minutes. On
testing day, mice were placed in a quiet room (light intensity of 8 lux) adjacent to the behavior
testing room for three to four hours prior to testing. Between each mouse being tested, the maze
was cleaned with 70% isopropyl alcohol, and water, and allowed to dry. Mice were transported
to the EPM room, and the beaker was placed at a 45° downward angle in the center of the maze,
facing away from the experimenter and the door. Mice were allowed to exit the beaker
spontaneously, and the experimenter subsequently exited the room and closed the door. Beaker
exit times averaged 59.2 ± 72.9 seconds (mean ± SD), and the latency to exit the beaker was not
correlated with measures of anxiety-like behavior. The five-minute test session began once the
tail base of the mouse entered the center of the maze. Time spent in the open and closed arms
117
and the center as well as the number of entries into the arms was recorded with overhead video
cameras (SuperExwave SS-E473, Sony) and analyzed using automated TopScan tracking
software (CleverSys Inc., VA). The time spent in one area on the maze was defined as the
moment the center- and tail base markers of the mouse crossed the border of the area until both
markers crossed the border of an adjacent area. Furthermore, as an additional measure of
anxiety-like behavior, the number of head dips were analyzed (Fernandez Espejo, 1997) by
counting the number of times the mice extended their head and dipped down over the edge of the
open arm, while either with their tail base in the center or closed arm (protected head dip) or with
their tail base in the open arm (unprotected head dip).
4.3.3.2. Social behavior: direct social interaction test
Experimental mice were placed in an adjacent room (200 lux) to the behavioral testing room (70
lux) two to three hours before the start of the DSI test. Male juvenile 129S1/SvImJ mice (P26-
P28) were being used as stimulus mice in the DSI test, and placed into an adjacent room one to
two hours before the test to acclimate. Stimulus mice were used two or three times while ranging
in age between P26 and P28, up to one time each day. The weight of the stimulus mice averaged
14.4 ± 1.35g (mean ± SD; range: 11.5-18.7g), and experimental mice weighed 27.7 ± 2.2g (mean
± SD; range: 19.6-29.3g), on testing day.
For the DSI test, each experimental mouse was transported from the home cage to the test room
in an empty, clean standard cage (JAG, Allentown, PA) and placed into the center of the DSI
chamber (31 x 20 x 20 cm in size, with clear plexiglas walls and floor, placed on a white
surface). Mice were allowed to acclimate to the DSI chamber for ten minutes, after which a
118
juvenile was introduced for a six-minute interaction period. The acclimation and interaction
period were recorded with one overhead and one side video camera (SuperExwave SS-E473,
Sony). None of the experimental mice or juvenile stimulus mice attempted to escape from the
DSI chamber, or showed aggression towards the other mouse. Between successive tests, the DSI
chamber was cleaned with 70% isopropyl, and water, and allowed to dry. Side and overhead
video recordings were imported into Final Cut Pro 10, and synchronized and stacked to allow for
viewing both perspectives simultaneously, and subsequently imported into Anvil 5 software for
manual annotation of sniffing bout duration and frequency. Social interaction was defined as the
time period during which the experimental adult mouse was in close contact (~ 0-5 mm) and
sniffing any body part of the juvenile mouse. The juveniles did not sniff or approach the adults
experimental mice during the DSI tests.
4.3.3.3. Fear learning and memory: contextual fear conditioning test
Contextual fear conditioning tests were completed during a seven-week period. One hour before
testing, mice were transported to a room (light intensity of 200 lux) adjacent to the fear
conditioning room (400 lux). During the test, freezing behavior was recorded using near-infrared
light and cameras (#NIR-100 light source, Med Associates Inc., VT; #SCA640-71fm camera,
Basler, Germany) in double-enclosed fear conditioning chambers 24 x 30 x 22 cm in size, with a
metal grid floor, and plexiglas and aluminum walls (#ENV065FPU-M, Med Associates Inc.,
VT). On day one, mice were acclimated to the fear conditioning chamber for three minutes, after
which five x two-second foot shocks of 0.5 mA were administered to the mice via the metal grid
floor (#ENV-414S stimulator/scrambler, Med Associates Inc., VT). Freezing duration was
recorded during the three minute acclimation period (‘baseline’), and during four 220-second
119
inter-trial intervals and one final 220-second period (average of five periods reported as
‘training’). Between tests, the chambers were cleaned with 70% isopropyl alcohol, followed by
water. After training, mice were returned to their respective home cages for 24 hours. On day
two (‘test’), mice were placed into the same fear conditioning chamber without receiving shocks,
for eight minutes while freezing behavior was recorded and analyzed. ‘Extinction’ existed of
weekly repeats of the ‘test’ protocol during the following six weeks. Freezing behavior analysis
was performed in real-time using Video Freeze software (Med Associates, Inc., VT), with a
linear method of observation, a motion threshold of 18, and a minimum freezing duration of 30
frames (one second).
4.3.3.4. General activity: activity chamber
Activity levels were determined in a separate cohort of mice aged P52 to P55, that had not
undergone any other (behavioral) tests. Mice were placed in dark activity chambers (28 x 28 x 20
cm in size, clear plexiglas walls inside a slightly larger enclosure (Med Associates Inc., VT),
while infrared beam breaks in the X, Y and Z direction were analyzed using SOF-811 Activity
Monitor automated software (MedAssociates Inc., VT), determining total distance traveled and
ambulatory time during the complete testing period (30 minutes). Between tests, chambers were
cleaned with 70% isopropyl alcohol, followed by water, and allowed to dry.
4.3.4. Statistical analyses
All data were analyzed using GraphPad Prism 6 (Graphpad Software). Distance traveled in
activity chamber, DSI social interaction duration, DSI number of social interactions, EPM time
in arm, EPM arm entry, and fear conditioning freezing were analyzed using a two-way ANOVA
120
(gene x environment). Tukey’s post-hoc analyses were performed after ANOVA tests when
appropriate (alpha = 0.05). A Pearson correlation analysis was performed to analyze the
correlation between open arm time on the EPM and the number of unprotected head dips, and
regression analysis to determine differences between the correlations. Data are presented as
mean ± standard error of the mean (SEM).
4.4. Results
4.4.1. Contextual fear memory is impaired as a result of Met
+/-
genotype
To find out whether Met genotype or ELS affects fear acquisition or memory, we performed a
contextual fear conditioning test. We observed a main effect of genotype on fear memory (‘test’)
whereby the amount of freezing was reduced in Met
+/-
mice (Figure 4.1A, F
1,62
= 6.53, p < 0.05),
but no main effect of environment (F
1,62
= 0.23, p = 0.64), or an interaction effect between the
two factors (F
1,62
= 0.53, p = 0.47). Within ELS-group comparisons did not reveal statistical
differences between genotypes; Met
+/-
mice only showed a trend towards a significant decrease
in freezing compared with the control group (p = 0.075). No main effects for genotype (F
1,62
=
2.58, p = 0.114), environment (F
1,62
= 0.16, p = 0.689) or interaction effects (F
1,62
= 1.031, p =
0.314) were observed on fear acquisition (‘training’), or the extinction of fear memory (Figure
4.1B, e.g. for week 1, genotype: F
1,62
= 2.78, p = 0.101; environment: F
1,62
= 0.06, p = 0.808;
interaction: F
1,62
= 2.41, p = 0.126).
121
Figure 4.1: Contextual fear memory (‘test’) is impaired by Met
+/-
genotype.
(A) A main effect of genotype was observed that reduced fear memory (the amount of freezing
24 hours after fear acquisition) in a contextual fear conditioning paradigm. (B) Neither genotype
nor ELS had an effect on freezing during six weekly fear extinction trials. Data are presented as
mean ± SEM. Number of animals per group: Control n = 21; ELS n = 11; Met
+/-
n = 15; Met
+/-
×
ELS n = 19.
122
4.4.2. Early-life stress reduces social interactions in Met
+/-
mice
A DSI test was performed to measure changes in social interaction as a result of ELS or Met
genotype. We observed that ELS had a main effect on the number of social interactions initiated
by the experimental adult during the six minute DSI test (Figure 4.2A, F
1,62
= 11.59, p < 0.01),
whereas Met genotype or the interaction between the two factors did not affect this behavior
(gene: F
1,62
= 0.15, p = 0.700, interaction: F
1,62
= 0.86, p = 0.357). Within-genotype comparisons
for each genotype revealed a significant effect of ELS in Met
+/-
mice (p < 0.05), but not in wild-
type mice. In contrast to the number of social interactions, total sniffing duration was not altered
as a result of genotype (Figure 4.2B, F
1,62
= 0.00, p = 0.970), ELS (F
1,62
= 0.97, p = 0.327), or an
interaction between the two factors (F
1,62
= 0.04, p = 0.847).
123
Figure 4.2: Early-life stress reduces number of social interactions in the six-minute direct
social interaction test. (A) A main effect of ELS was observed on the number of social
interactions, and post-hoc tests between revealed a decrease as a result of ELS in Met
+/-
mice
compared to non-stressed Met
+/-
mice. (B) Neither genotype nor ELS had an effect on the
duration of social interactions. Data are presented as mean ± SEM. Number of animals per
group: Control n = 21; ELS n = 11; Met
+/-
n = 15; Met
+/-
× ELS n = 19.
* p < 0.05 for within-genotype effect of ELS.
124
4.4.3. Met
+/-
genotype reduces anxiety-like behaviors
Mice were tested for changes in anxiety-like behavior using three different measures on the
EPM: time spent in the open arm, open arm entries, and the number of head dips. Genotype had
a main effect on the duration of time spent in the open arm (Figure 4.3A, F
1,62
= 5.46, p < 0.05),
but no main effect of environment (F
1,62
= 1.44, p = 0.234) or interaction effect was observed
(F
1,62
= 0.60, p = 0.443). Within-ELS group comparisons did not show a difference between
genotypes. A similar result was present for the number of open arm entries, due to a main effect
of genotype only (Figure 4.3B, F
1,62
= 7.17, p < 0.01). No effect of genotype or ELS was
observed on the number of closed arm entries (Figure 4.3B, genotype: F
1,62
= 2.78, p = 0.101;
ELS: F
1,62
= 2.090, p = 0.153; interaction: F
1,62
= 0.05, p = 0.824), suggesting ELS or genotype
did not affect general activity levels. This was further confirmed by a lack of change in total
distance traveled in the activity chamber in a different, independent cohort (Figure 4.3C
genotype: F
1,55
= 0.26, p = 0.615; ELS: F
1,55
= 0.42, p = 0.522; interaction: F
1,55
= 0.17, p =
0.678).
125
Figure 4.3: Met
+/-
genotype reduces
anxiety-like behaviors on the elevated
plus maze. A main effect of genotype was
observed on (A) time spent in the open arms
and (B) open arm entries in the EPM test.
(C) General activity levels were not changed
due to genotype or ELS. Data are presented
as mean ± SEM. Number of animals per
group: Control n = 21; ELS n = 11;
Met
+/-
n = 15; Met
+/-
× ELS n = 19.
126
In addition to time spent in open arm of the EPM, we analyzed the number of protected and
unprotected head dips as a measure of anxiety-like behavior. We observed a main effect of
genotype on the number of unprotected head dips (i.e. with tail base in the open arm; Figure
4.4B, F
1,62
= 6.97, p < 0.05). Within-ELS group comparisons did not show a difference between
genotypes. The number of protected head dips (i.e. with tail base in the center area or closed
arm) was not altered due to genotype (F
1,62
= 0.01, p = 0.919), or environment (F
1,62
= 1.42, p =
0.238). Because an increase in open arm time logically allows for an increased number of
unprotected head dips, we compared the relationship between open arm time and the number of
unprotected head dips in wild-type and Met
+/-
mice (non-stressed and stressed values combined).
For both groups, as expected, a significant linear positive correlation existed between the two
factors (Figure 4.4C, wild-type: r = 0.98, Met
+/-
: r = 0.90, both p < 0.001). However, the slopes
between the two groups were different (wild-type = 0.42; Met
+/-
= 0.62, p < 0.001), suggesting
that in the Met
+/-
mice, increased open arm time predicted a larger increase of unprotected head
dips compared to wild-type mice. The results of all behavior tests are summarized in Table 4.1.
127
Figure 4.4: Met
+/-
genotype decreases anxiety-like behavior as measured by unprotected
head-dips in the elevated-plus maze test. (A) Whereas the number of protected head-dips (tail
base in center or closed arm of the maze) did not change as a result of ELS or genotype, (B) a
main effect of genotype was observed on the number of unprotected head dips (body in open arm
of the maze compared to wild-type mice. Data are presented as mean ± SEM. (C) The increase in
number of unprotected head-dips by Met
+/-
mice is only partially explained by time spent in the
open arm, as a higher number of unprotected head dips is predicted based on open-arm time in
these mice compared to wild-type mice. Non-ELS and ELS values were combined for (C).
Number of animals per group: Control n = 21; ELS n = 11; Met
+/-
n = 15; Met
+/-
× ELS n = 19.
128
Table 4.1: Summary of behavior test results in experimental groups. The outcomes and
direction of the difference in specific measurements (⬆ = increased, ⬇ = decreased,
⟷ = unchanged) are indicated for anxiety-like behavior, social behavior and fear conditioning
measurements.
Met
+/-
ELS
Met
+/-
× ELS
Anxiety-like behavior
Open arm duration EPM
⬆
a
⟷
⬆
a
Open arm entries EPM
⬆
a
⟷
⬆
a
Unprotected head dips EPM
⬆
a
⟷
⬆
a
Protected head dips EPM ⟷ ⟷ ⟷
Social behavior
Number of social interactions ⟷
⬇
b
⬇
bc
Duration of social interactions ⟷ ⟷ ⟷
Fear conditioning
Contextual fear acquisition ⟷ ⟷ ⟷
Contextual fear memory
⬇
ad
⟷
⬇
a
Contextual fear extinction ⟷ ⟷ ⟷
a – main effect of genotype, and b – main effect of ELS.
c – decreased, and d – trend in reduction, compared to appropriate comparison group
4.5. Discussion
The current study shows that reduced MET expression in the CNS significantly affects anxiety-
like behaviors and contextual fear memory, and ELS reduces the number of social interactions,
specifically in Met
+/-
mice. This suggests that brain circuits involved in social behaviors in Met
+/-
mice show a more pronounced response to early environmental perturbations compared to wild-
type mice, and that Met
+/-
mice that have experienced ELS are affected in more behavioral
domains than wild-type mice or non-stressed Met
+/-
mice. Altered expression or inhibition of
MET has previously been implicated in modulating the maturational state of neurons (Qiu et al.,
2014), intralaminal cortical connectivity (Qiu et al., 2011), neuronal morphology (Judson et al.,
129
2010, Qiu et al., 2014), and synaptogenesis (Eagleson et al., 2016). Moreover, a common
promoter variant in the MET gene has been associated with increased risk of ASD in several
independent cohorts (Campbell et al., 2006), as well as with altered structural and functional
brain connectivity in the human population (Rudie et al., 2012). This suggests that expression
levels of MET may be able to modulate the impact of ELS on later life behavior through altering
brain circuitry important in regulating social-emotional behaviors.
4.5.1. Effect of reduced MET expression on behavior
Our findings of behavioral changes due to Met
+/-
alone (without ELS), compared to non-stressed
wild-type mice are in line with results from a recent study that assessed the effects of Met
+/-
on
behaviors (Thompson and Levitt, 2015). In our study, a significant main effect of genotype on
fear memory was observed, with a trend towards a decrease in freezing time in non-stressed
Met
+/-
mice compared to non-stressed wild-type mice. Thompson and Levitt (2015) reported a
similar but more pronounced effect on both fear memory and fear learning, albeit in a cued fear
conditioning task. The latter task is known to test behaviors that are dependent on overlapping
but not identical brain circuitry (Smith and Bulkin, 2014), and although the results are in a
similar direction (impaired fear memory), the lack of effect on fear acquisition may be due to
these differences in cued versus contextual test characteristics. In the same study, Met
+/-
mice did
not show a change in sociability and social novelty preference in the three-chamber test, i.e.
these mice showed a normal increased preference for mice vs. an object, and stranger mice
versus familiar mice (Thompson and Levitt, 2015). Although we assessed social behavior using a
different approach, the DSI test, our results are consistent with this report, since we did not
observe a difference in the number of social interactions in non-stressed Met
+/-
mice compared to
130
wild-type mice. Interestingly, a recent study by (Okaty et al., 2015), in which MET was deleted
selectively from serotonergic neurons in the dorsal raphe, resulted in reduced sociability in a
three-chamber test. These mice did not display a normal preference for the stranger mouse
compared to an object, and the social interaction time in this test was reduced compared to wild-
type mice. This shows that complete deletion of MET from a subset of neurons has different
effects on behavior than partial deletion in the complete central nervous system, suggesting that
these different but overlapping subsets of neurons are affected by relative levels of MET in
associated circuits. This idea is strengthened by different results in social behaviors in studies
when Met is completely (compared to partially) deleted from the central nervous system in mice
(Thompson and Levitt, 2015). Lastly, regarding anxiety-like behavior, we observed a main
effect of genotype on measures in the EPM as a result of Met genotype. Despite this significant
main effect, within-ELS group comparisons revealed that open arm duration and number of
head-dips in the EPM by non-stressed Met
+/-
mice was not significantly different from wild-type
mice. Thompson and Levitt (2015) measured an increase in center time (but not open-arm time)
in the EPM by Met
+/-
mice, suggesting that these mice when raised in control conditions exhibit
mildly reduced anxiety-like behavior.
4.5.2. Effect of ELS on behavior
The effects of ELS on behavior in adult mice have been studied widely, with many groups
reporting on effects on anxiety-like behavior as tested using an EPM or elevated-zero maze
(EZM). ELS has shown to be somewhat ineffective in changing anxiety-like behavior as several
group have measured no change in duration spent in the open arm of an EPM (van der Kooij et
al., 2015, Zoicas and Neumann, 2016, Venerosi et al., 2003, Wang et al., 2012, Niwa et al., 2011,
131
van Heerden et al., 2010, Liu et al., 2016) and the open section of EZM experiments (Harrison et
al., 2014). However, other groups reported increased (Levine et al., 2012, Mehta and Schmauss,
2011, Bouet et al., 2011, Veenema et al., 2007, Shin et al., 2016), or decreased anxiety-like
behavior (Savignac et al., 2011, Fabricius et al., 2008). Our results are consistent with a lack of
effect of ELS alone on anxiety-like behavior as measured on the EPM, considering ELS in wild-
type mice did not affect time spent in the open arm, nor did it change the number of protected
and unprotected head dips.
Multiple studies examining the effect of ELS induced by limited bedding in mice and rats
generally have shown a decrease in social interaction time in either a DSI test or three-chamber
sociability test in adulthood (Santarelli et al., 2014, van der Kooij et al., 2015, Raineki et al.,
2012, Raineki et al., 2015). Similarly, maternal separation (MS) has been shown to induce
impairment of social behavior in mice (Tsuda and Ogawa, 2012, Bouet et al., 2011, Niwa et al.,
2011, Venerosi et al., 2003, Tsuda et al., 2014), although several groups have reported no change
in the duration of social interaction (Harrison et al., 2014, Franklin et al., 2011, Zoicas and
Neumann, 2016, Tsuda et al., 2011). Consistent with these results, we observed a main effect of
ELS on the number of social interactions.
A few groups have described effects of maternal separation and limited bedding on (contextual
and cued) fear learning, memory and extinction in adult mice. Specifically, Wang et al. (2011),
Niwa et al. (2011) and Kanatsou et al. (2016) reported that ELS reduced freezing in a specific
context (24 hours after combining a tone and foot shock in the same context). Freezing after
presenting the actual cue (the tone) 24 hours after fear learning, was also reduced (Wang et al.,
132
2011) or unchanged (Niwa et al., 2011, Zoicas and Neumann, 2016). However, ELS did not
affect acquisition, nor extinction of a fear memory in two recent studies using a contextual
(Kanatsou et al., 2016) and cued (Zoicas and Neumann, 2016) fear conditioning paradigm in
mice. The lack of impairment of fear memory due to ELS in our current study is unexpected,
although our methodology was slightly different from the studies mentioned here, as we induced
a contextual association with the foot shock in absence of a tone, and the behavioral response to
these two protocols have been shown to rely on overlapping but not identical brain circuits
(Smith and Bulkin, 2014).
Thus, in general, ELS in mice impairs social behavior, while the effects on anxiety-like behavior
as tested on the EPM is inconclusive, with most studies showing no effects of ELS. Furthermore,
although not widely reported, fear memory is generally impaired after ELS in mixed
cued/contextual fear conditioning paradigms.
4.5.3. Genotype × ELS interaction effects
Our results show that Met
+/-
mice are more impacted by ELS than wild-type mice regarding
certain social behavior measurements. Specifically, Met
+/-
mice that had experienced ELS
showed a reduction in social interactions compared to non-stressed Met
+/-
mice. In addition,
Met
+/-
genotype impaired contextual fear memory, and reduced anxiety-like behavior. Whether
this translates into a maladaptive (i.e. a ‘two-hit’ impact of genotype and environment) or
actually beneficial adaptation to a detrimental environment is not clear from this study,
especially due to the somewhat contradicting outcome of a reduced number of social interactions
in combination with reduced anxiety-like behavior. However, even though the behavioral
133
constructs of anxiety-like behavior and social behavior partially overlap, these two behaviors are
not directly correlated to each other. In addition, these single tests of anxiety-like and social
behavior provide a starting point for future behavioral studies including open field test, light-dark
test, marble burying, and three-camber sociability and social novelty tests. Furthermore,
assessment of other behavioral constructs, such as learned helplessness/depression-like behavior
with a Porsolt forced swim test would create a more complete behavioral profile of the effects of
Met
+/-
genotype and ELS. The results presented here, however do show that social behavior in
mice with altered MET expression is affected in a different way than wild-type mice by ELS; an
observation that could have clinical impact considering variation in MET expression levels in the
human population, and especially regarding the association of SNPs in the MET promoter with
ASD (Campbell et al., 2006).
Several other groups have shown changes in behavioral measures in genetically altered mice
after ELS, while ELS by itself did not have the same effects (Harrison et al., 2014). A different
interaction type between genotype and environment has been observed in other studies, in which
ELS had an effect by itself, but did not affect genetically altered mice to the same extent as wild-
type mice (Ognibene et al., 2007, Sachs et al., 2013). More complex interactions between
genotype and ELS also have been reported, for example where ELS and genotype had opposite
effects on contextual fear memory (Kanatsou et al., 2016) and social investigation (Tsuda et al.,
2014), whereas the combination of both factors resulted in the elimination of this difference in
these studies. In gene-environment interaction studies, complex interactions between genotype
and ELS seem to be the norm when examining social-emotional behaviors, rather than
‘straightforward’ additive effects or non-responsiveness of genetically altered mice, which
134
makes interpretations of outcomes more difficult.
Regardless of complexity of interpretation, these results from animal studies can provide support
for examining interactions between genotype and ELS in the human population. One caveat is
that most research studies however, including the current study, use (partial) knock-out of genes
that have been implicated in human disorders, rather than genetic variants that may affect
transcription in more nuanced ways in different parts of the brain as a result of specific impacts.
For example, as stated earlier, relative and differential levels of MET expression in different
brain areas in humans appear to be just as important as absolute levels throughout the brain
(Voineagu et al., 2011). Therefore, a general reduction of MET levels in the central nervous
system of Met
+/-
mice as has been done here, is only partially comparable with genetic variants in
the human MET gene. Nonetheless, outcomes in the current study provide support to further
research mechanisms behind altered sensitivity or resilience to environmental impacts as a result
of reduced MET expression.
Additionally, the validity of behavioral tests that measure social-emotional behaviors in mice
continues to be subject of debate, i.e. are changes in test outcomes of mouse behaviors
meaningful (predictive validity); do behaviors in mice and humans change based on similar
underlying causes (construct validity); and are the anxiety-like, social, and fear learning
behaviors similar as we observe them in humans (face validity)? These issues have been
discussed at length in the literature (Blanchard et al., 2013, Crawley, 2007, Barron and Robinson,
2008, Dzirasa and Covington, 2012, Hånell and Marklund, 2014) and tests and interpretations
have been refined in the past to take into account confounders, e.g. levels of general activity.
135
Needless to say, whereas current behavioral tests are generally accepted despite suboptimal
translation to human behavior, animal behavioral tests should continue to be subjected to
scrutiny and further testing, and the research fields utilizing these tests should remain open to
adjusting behavioral tests based on evidence that challenges validity.
In conclusion, our results provide support to further study the impact of environmental stressors,
such as ELS, on behavioral measures in mice with altered expression of MET receptor tyrosine
kinase. We found that anxiety-like behavior is reduced, and contextual fear memory is impaired
in Met
+/-
mice, and that Met
+/-
mice that have experienced ELS show impaired social behavior
compared to non-stressed Met
+/-
mice. This implies that the expression level of MET in the brain
can modulate the responsiveness of brain circuitry to ELS; a relevant outcome in light of the
existence of a common variant in the promoter region of the MET gene in the general population
that has been shown to affect MET protein expression. It would be very interesting to determine
potential interaction effects on risk and severity of (mental) health outcomes between protein-
expression altering SNPs in the MET gene and ELS in humans. Similarly, in future GWAS
studies examining ASD and other neurodevelopmental disorders, one should stratify the cohort
by ELS history; perhaps carriers of one or two ‘C’ risk alleles at rs1858830 respond differently
to ELS in terms of functional outcomes.
136
Chapter 5:
Discussion and future directions
Common genetic variations and toxic stress during early childhood are associated with increased
risk of many mental health disorders in later life. In addition, interactions between genetic and
environmental factors have also been shown to increase risk or severity of disorders, and will
explain part of the gap that currently exists between twin studies and the outcomes of more
recent GWAS. Identifying these interactions between ELS and common genetic variants will
help elucidate the mechanisms behind ELS and possibly allow for targeted prevention and
treatment of health disorders in later life.
Animal models provide an opportunity to determine effects of gene-environment interactions on
brain structure and function. One particular ELS paradigm that limits the amount of bedding and
nesting materials during early life is being used by a growing number of research groups, but the
stress-inducing mechanism is not well understood. It has been suggested that maternal care is
‘fragmented’ due to increased number of nest entries. To test this assumption, I analyzed
temporal patterns of the dam’s nest entry behaviors in this ELS paradigm and the pups’ response
to this altered behavior. I showed that immediately after the dam enters the nest, pups emit more
audible and ultrasonic vocalizations, signals associated with aversive and painful stimuli.
Surprisingly, extended periods on the nest appeared unaltered, suggesting that the method of
stress-induction is not due to ‘fragmented care’, but rather through painful stimuli (stepping on
and rough handling pups) that alter sensory stimulation of the pups. Additionally, the nest entry
behaviors by the dams due to limited resources are transient and not transmitted to the next
137
generation of females. Thus, this ELS paradigm provides an ‘abusive’ stressor that is present
only in the period during which limited resources are available. The paradigm thus may be used
reliably to determine effects of ELS on brain development in mice.
The subsequent goal of my project was to determine specific gene-environment effects of
reduction in Met expression and ELS on brain structure and function in mice, using a limited
bedding ELS paradigm. A reduction of MET protein in the central nervous system, as well as
ELS, leads to a reduction in dendritic complexity of specific projection neurons from the ventral
CA1 of the hippocampus to the basolateral amygdala in young adult mice. When Met
+/-
mice
were subjected to ELS, dendritic complexity was increased compared to non-stressed Met
+/-
mice. The findings, while initially surprising, indicate that Met
+/-
neurons exhibit the opposite
effect of ELS alone -an increase in dendritic arbor complexity- in response to ELS compared to
wild-type neurons, possibly as a result of an altered maturational state. In behavioral tests, Met
genotype had a main effect on anxiety-like behavior and contextual fear memory, while fear
acquisition and extinction were unaffected. Social behavior was impaired by ELS, specifically in
Met
+/-
mice. These results suggest that the combination of these genetic and environmental
factors results in an increased number of affected behavioral domains, and that Met
+/-
mice may
be more sensitive to ELS. In summary, the level of MET protein expression in the brain alters
the development of circuitry that regulates certain social-emotional behaviors in response to
different environments during early brain development.
Complex experimental designs and multiple comparisons in gene-environmental interaction
studies create an inherent challenge to deduct biological meaning from observations. When ELS
138
has an impact in genetically altered mice only (i.e. increased sensitivity compared to wild-type
mice), this result may be interpreted as either a ‘two-hit’ situation with ‘worse’ behavioral
outcomes. Alternatively, the genetically altered mice may actually be more successful in
adapting to the environment created by ELS. With regard to behavioral test outcomes, this
question can only be fully answered if one knows the valence of, for example, anxiety-like
behavior on the elevated plus maze. Increased anxiety is generally considered to be a
maladaptive outcome; however, anxiety behaviors are on a spectrum with a normative mean, and
some level of anxiety is desirable to increase focus, successfully detect threats, and protect
against indiscriminate behavior. When raised in an adverse early life environment, increased
anxiety may be beneficial for survival or health status. Therefore, animal studies in the context of
an altered environment have acceptable limitations in that one can determine a change in
behavior; the functional significance of altered behavior due to a gene-environment interaction
needs to be measured in the human population before one can determine whether the adaptation
is beneficial or detrimental for overall health or survival, especially in an early adverse
environment.
While the behavioral test results suggest that the combination of these genetic and environmental
factors results in an increased number of affected behavioral domains, and that social behavior in
Met
+/-
mice is more affected by ELS than in wild-type mice, the morphological changes in a
specific population of projection neurons show a complex interaction effect. The dual approach
to study structure and function in the same gene-environment experiment in this dissertation
project is, however, not to determine whether a change in neuronal morphology of one type of
projection neuron is responsible for behavior changes. Optogenetic approaches have shown that
139
specific behaviors are altered as a result of artificially activating or inhibiting certain projection
neurons (Felix-Ortiz et al., 2013, Ciocchi et al., 2015). Complex behaviors, however, are a result
of the integration of many different connections throughout the brain as a whole. Determining
structural changes in one specific type of projection neuron solely provides data addressing
whether neurons with reduced levels of MET are in general differentially sensitive to ELS.
Technical advances in high-throughput morphological analyses, in combination with neuronal
activity markers and/or real-time neuronal activity measures, may be able to address changes on
a whole-brain scale. This will provide opportunities to more comprehensibly integrate
knowledge on the relation between brain structure and function.
As the aims of my research project are mostly descriptive in nature, the next step is to elucidate
mechanisms underlying the observed effects. The stress response is known to affect the organism
as a whole, and therefore multiple mechanisms are mostly likely involved. Together, these
mechanisms create a situation in which ELS can differentially affect neurons and other cells that
have altered expression levels of MET protein. Our lab is currently pursuing several directions
simultaneously. A desire in the clinical research community to develop objective biomarkers for
ELS is driving our efforts to study the effects of ELS on inflammatory mediators and markers.
F
2
-isoprostanes and F
4
-neuroprostanes are general and neuron-specific markers of oxidative
stress, respectively. Preliminary results in this research project from tissues obtained after
behavioral studies show that ELS increases levels of F
2
-Isoprostanes and F
4
-neuroprostanes in
hippocampus in adult wild-type mice (main effect of ELS on F
2
-Isoprostanes: F
1,20
= 48.19, p <
0.001; main effect of ELS on F
4
-neuroprostanes: F
1,20
= 5.66, p < 0.05; trend of interaction effect
on F
4
-neuroprostanes: F
1,20
= 3.92, p = 0.062). Whereas Met
+/-
mice responded similarly to ELS
140
as wild-type mice with increased levels of F
2
-Isoprostanes (Figure 5.1A), F
4
-neuroprostanes were
not increased as a result of ELS in these mice (Figure 5.1B). Frontal cortex tissue surprisingly
did not show any changes in F
2
-Isoprostanes or F
4
-neuroprostanes due to genotype or ELS
(Figure 5.1C and D). A second cohort of young adult mice (P60) that were not subjected to any
behavioral tests showed a similar main effect of ELS on of F
2
-isoprostanes in blood as in
hippocampus (Figure 5.1E, main effect of ELS: F
1,14
= 6.88, p < 0.05). The hippocampus and
frontal cortex tissues that correspond to these blood samples are currently being analyzed, and
we anticipate that the blood values of F
2
-isoprostanes will correlate to measurements of
hippocampal tissue F
2
-isoprostanes based on our previous results. These preliminary data
indicate that ELS increases long-term general inflammation in hippocampal tissue and in blood
of these mice, but neurons specifically may not be affected in Met
+/-
mice as they are in wild-type
mice.
141
Figure 5.1: ELS affects levels of inflammatory markers in hippocampus and blood of adult
mice. F
2
-isoprostanes and F
4
-neuroprostanes were measured in (A, B) hippocampus, (C, D), and
(E) blood of wild-type and Met
+/-
mice that experienced control or ELS conditions. Main effects
of ELS were observed on F
2
-isoprostanes and F
4
-neuroprostanes in hippocampus and F
2
-
isoprostanes in blood. Within-genotype comparisons revealed increased measurements in
hippocampal tissue of mice that had experienced ELS. Individual data points represent individual
mice for blood measurements, and tissues from 2-4 pups combined (> 65 mg of tissue) for
hippocampus and frontal cortex. Data are presented as mean ± SEM. *** p < 0.001, ** p < 0.01,
* p <0.05 for within-genotype effect of ELS.
142
Another project that can provide insight into the mechanisms behind the observed results aims to
determine altered protein expression in brain tissue. Using a 4-plex ‘isobaric tags for relative and
absolute quantitation’ (iTRAQ) protocol, we are currently measuring relative protein expression
changes in adult cortical tissue from wild-type and Met
+/-
mice that have experienced ELS. The
results will give unbiased insight into lasting protein changes due to gene-environment
interactions. A logical follow-up from data indicating altered expression of specific proteins as a
result of ELS, is to measure epigenetic changes (e.g. DNA methylation, histone modification)
that could result in changes in protein expression in later life, even after the original stressor
factor has been removed.
In summary, the outcomes of my dissertation project show that gene-environment interactions
between Met expression and ELS affect neuronal structure and behavior in mice. This body of
work creates a foundation for future research on the mechanism of this interaction. Furthermore,
the outcomes provide a rationale to study an interaction between MET rs1858830 genotype and
early adverse experiences in the human population, and its effects on severity and risk of mental
health disorders.
143
References
Adhikari, A., Topiwala, M.A. and Gordon, J.A. (2010). Synchronized activity between the
ventral hippocampus and the medial prefrontal cortex during anxiety. Neuron. 65, 257-
269.
Aguggia, J.P., Suarez, M.M. and Rivarola, M.A. (2013). Early maternal separation:
neurobehavioral consequences in mother rats. Behavioural brain research. 248, 25-31.
Almas, A.N., Degnan, K.A., Walker, O.L., Radulescu, A., Nelson, C.A., Zeanah, C.H., et al.
(2015). The Effects of Early Institutionalization and Foster Care Intervention on
Children's Social Behaviors at Age 8. Social development. 24, 225-239.
Almli, L.M., Fani, N., Smith, A.K. and Ressler, K.J. (2014). Genetic approaches to
understanding post-traumatic stress disorder. The international journal of
neuropsychopharmacology. 17, 355-370.
Anderson, P.L., Tiro, J.A., Price, A.W., Bender, M.A. and Kaslow, N.J. (2002). Additive impact
of childhood emotional, physical, and sexual abuse on suicide attempts among low-
income African American women. Suicide & life-threatening behavior. 32, 131-138.
Avishai-Eliner, S., Gilles, E.E., Eghbal-Ahmadi, M., Bar-El, Y. and Baram, T.Z. (2001). Altered
regulation of gene and protein expression of hypothalamic-pituitary-adrenal axis
components in an immature rat model of chronic stress. Journal of neuroendocrinology.
13, 799-807.
Baram, T.Z., Davis, E.P., Obenaus, A., Sandman, C.A., Small, S.L., Solodkin, A., et al. (2012).
Fragmentation and unpredictability of early-life experience in mental disorders. The
American journal of psychiatry. 169, 907-915.
Barron, A.B. and Robinson, G.E. (2008). The utility of behavioral models and modules in
molecular analyses of social behavior. Genes, brain, and behavior. 7, 257-265.
Baumann, C., Klauke, B., Weber, H., Domschke, K., Zwanzger, P., Pauli, P., et al. (2013). The
interaction of early life experiences with COMT val158met affects anxiety sensitivity.
Genes, brain, and behavior. 12, 821-829.
Bergman, K., Sarkar, P., O'Connor, T.G., Modi, N. and Glover, V. (2007). Maternal stress during
pregnancy predicts cognitive ability and fearfulness in infancy. Journal of the American
Academy of Child and Adolescent Psychiatry. 46, 1454-1463.
Bet, P.M., Penninx, B.W., Bochdanovits, Z., Uitterlinden, A.G., Beekman, A.T., van Schoor,
N.M., et al. (2009). Glucocorticoid receptor gene polymorphisms and childhood adversity
are associated with depression: New evidence for a gene-environment interaction.
American journal of medical genetics. 150B, 660-669.
144
Beversdorf, D.Q., Manning, S.E., Hillier, A., Anderson, S.L., Nordgren, R.E., Walters, S.E., et
al. (2005). Timing of prenatal stressors and autism. Journal of autism and developmental
disorders. 35, 471-478.
Bick, J., Zhu, T., Stamoulis, C., Fox, N.A., Zeanah, C. and Nelson, C.A. (2015). Effect of early
institutionalization and foster care on long-term white matter development: a randomized
clinical trial. JAMA pediatrics. 169, 211-219.
Bilbo, S.D. and Schwarz, J.M. (2009). Early-life programming of later-life brain and behavior: a
critical role for the immune system. Frontiers in behavioral neuroscience. 3, 14.
Binder, E.B., Bradley, R.G., Liu, W., Epstein, M.P., Deveau, T.C., Mercer, K.B., et al. (2008).
Association of FKBP5 polymorphisms and childhood abuse with risk of posttraumatic
stress disorder symptoms in adults. Journal of the American Medical Association. 299,
1291-1305.
Blanchard, D.C., Summers, C.H. and Blanchard, R.J. (2013). The role of behavior in
translational models for psychopathology: functionality and dysfunctional behaviors.
Neuroscience and biobehavioral reviews. 37, 1567-1577.
Bock, J., Gruss, M., Becker, S. and Braun, K. (2005). Experience-induced changes of dendritic
spine densities in the prefrontal and sensory cortex: correlation with developmental time
windows. Cerebral cortex. 15, 802-808.
Bock, J., Rether, K., Groger, N., Xie, L. and Braun, K. (2014). Perinatal programming of
emotional brain circuits: an integrative view from systems to molecules. Frontiers in
neuroscience. 8, 11.
Bolton, J.L., Huff, N.C., Smith, S.H., Mason, S.N., Foster, W.M., Auten, R.L., et al. (2013).
Maternal stress and effects of prenatal air pollution on offspring mental health outcomes
in mice. Environmental health perspectives. 121, 1075-1082.
Bouet, V., Lecrux, B., Tran, G. and Freret, T. (2011). Effect of pre- versus post-weaning
environmental disturbances on social behaviour in mice. Neuroscience letters. 488, 221-
224.
Bradley, R.G., Binder, E.B., Epstein, M.P., Tang, Y., Nair, H.P., Liu, W., et al. (2008). Influence
of child abuse on adult depression: moderation by the corticotropin-releasing hormone
receptor gene. Archives of general psychiatry. 65, 190-200.
Branchi, I., Santucci, D., Vitale, A. and Alleva, E. (1998). Ultrasonic vocalizations by infant
laboratory mice: a preliminary spectrographic characterization under different conditions.
Developmental psychobiology. 33, 249-256.
Branchi, I., Santucci, D. and Alleva, E. (2006). Analysis of ultrasonic vocalizations emitted by
infant rodents. Current Protocols in Toxicology. 13.12.11-13.12.14.
145
Braw, Y., Malkesman, O., Merenlender, A., Dagan, M., Bercovich, A., Lavi-Avnon, Y., et al.
(2009). Divergent maternal behavioral patterns in two genetic animal models of
depression. Physiology & behavior. 96, 209-217.
Bright, M.A., Knapp, C., Hinojosa, M.S., Alford, S. and Bonner, B. (2015). The Comorbidity of
Physical, Mental, and Developmental Conditions Associated with Childhood Adversity:
A Population Based Study. Maternal and child health journal.
Brunson, K.L., Eghbal-Ahmadi, M., Bender, R., Chen, Y. and Baram, T.Z. (2001). Long-term,
progressive hippocampal cell loss and dysfunction induced by early-life administration of
corticotropin-releasing hormone reproduce the effects of early-life stress. Proceedings of
the National Academy of Sciences. 98, 8856-8861.
Brunson, K.L., Kramar, E., Lin, B., Chen, Y., Colgin, L.L., Yanagihara, T.K., et al. (2005).
Mechanisms of late-onset cognitive decline after early-life stress. The Journal of
neuroscience. 25, 9328-9338.
Buitelaar, J.K., Huizink, A.C., Mulder, E.J., de Medina, P.G. and Visser, G.H. (2003). Prenatal
stress and cognitive development and temperament in infants. Neurobiology of aging. 24
Suppl 1, S53-60; discussion S67-58.
Campbell, D.B., Sutcliffe, J.S., Ebert, P.J., Militerni, R., Bravaccio, C., Trillo, S., et al. (2006). A
genetic variant that disrupts MET transcription is associated with autism. Proceedings of
the National Academy of Sciences. 103, 16834-16839.
Campbell, D.B., D'Oronzio, R., Garbett, K., Ebert, P.J., Mirnics, K., Levitt, P., et al. (2007).
Disruption of cerebral cortex MET signaling in autism spectrum disorder. Annals of
neurology. 62, 243-250.
Carlson, M. and Earls, F. (1997). Psychological and neuroendocrinological sequelae of early
social deprivation in institutionalized children in Romania. Annals of the New York
Academy of Sciences. 807, 419-428.
Carola, V., Mirabeau, O. and Gross, C.T. (2011). Hidden Markov model analysis of maternal
behavior patterns in inbred and reciprocal hybrid mice. PloS one. 6, e14753.
Carpenter, L.L., Carvalho, J.P., Tyrka, A.R., Wier, L.M., Mello, A.F., Mello, M.F., et al. (2007).
Decreased adrenocorticotropic hormone and cortisol responses to stress in healthy adults
reporting significant childhood maltreatment. Biological psychiatry. 62, 1080-1087.
Caspi, A., McClay, J., Moffitt, T.E., Mill, J., Martin, J., Craig, I.W., et al. (2002). Role of
genotype in the cycle of violence in maltreated children. Science. 297, 851-854.
Caspi, A., Sugden, K., Moffitt, T.E., Taylor, A., Craig, I.W., Harrington, H., et al. (2003).
Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene.
Science. 301, 386-389.
146
Champagne, F.A., Curley, J.P., Keverne, E.B. and Bateson, P.P. (2007). Natural variations in
postpartum maternal care in inbred and outbred mice. Physiology & behavior. 91, 325-
334.
Chapman, D.P., Whitfield, C.L., Felitti, V.J., Dube, S.R., Edwards, V.J. and Anda, R.F. (2004).
Adverse childhood experiences and the risk of depressive disorders in adulthood. Journal
of affective disorders. 82, 217-225.
Chen, Y., Andres, A.L., Frotscher, M. and Baram, T.Z. (2012). Tuning synaptic transmission in
the hippocampus by stress: the CRH system. Frontiers in cellular neuroscience. 6, 13.
Chen, Y., Kramar, E.A., Chen, L.Y., Babayan, A.H., Andres, A.L., Gall, C.M., et al. (2013).
Impairment of synaptic plasticity by the stress mediator CRH involves selective
destruction of thin dendritic spines via RhoA signaling. Molecular psychiatry. 18, 485-
496.
Chocyk, A., Bobula, B., Dudys, D., Przyborowska, A., Majcher-Maslanka, I., Hess, G., et al.
(2013). Early-life stress affects the structural and functional plasticity of the medial
prefrontal cortex in adolescent rats. The European journal of neuroscience. 38, 2089-
2107.
Cicchetti, D., Rogosch, F.A., Gunnar, M.R. and Toth, S.L. (2010). The differential impacts of
early physical and sexual abuse and internalizing problems on daytime cortisol rhythm in
school-aged children. Child development. 81, 252-269.
Ciocchi, S., Passecker, J., Malagon-Vina, H., Mikus, N. and Klausberger, T. (2015). Selective
information routing by ventral hippocampal CA1 projection neurons. Science. 348, 560-
563.
Class, Q.A., Abel, K.M., Khashan, A.S., Rickert, M.E., Dalman, C., Larsson, H., et al. (2014).
Offspring psychopathology following preconception, prenatal and postnatal maternal
bereavement stress. Psychological medicine. 44, 71-84.
Conway-Campbell, B.L., McKenna, M.A., Wiles, C.C., Atkinson, H.C., de Kloet, E.R. and
Lightman, S.L. (2007). Proteasome-dependent down-regulation of activated nuclear
hippocampal glucocorticoid receptors determines dynamic responses to corticosterone.
Endocrinology. 148, 5470-5477.
Copper, R.L., Goldenberg, R.L., Das, A., Elder, N., Swain, M., Norman, G., et al. (1996). The
preterm prediction study: maternal stress is associated with spontaneous preterm birth at
less than thirty-five weeks' gestation. American journal of obstetrics and gynecology.
175, 1286-1292.
Corbett, B.A., Schupp, C.W., Levine, S. and Mendoza, S. (2009). Comparing cortisol, stress, and
sensory sensitivity in children with autism. Autism research. 2, 39-49.
Crawley, J.N. (2007). Mouse behavioral assays relevant to the symptoms of autism. Brain
pathology. 17, 448-459.
147
Croskerry, P.G., Smith, G.K. and Leon, M. (1978). Thermoregulation and the maternal
behaviour of the rat. Nature. 273, 299-300.
Cross-Disorder Group of the Psychiatric Genomics Consortium. (2013). Genetic relationship
between five psychiatric disorders estimated from genome-wide SNPs. Nature genetics.
45, 984-994.
Cui, M., Yang, Y., Yang, J., Zhang, J., Han, H., Ma, W., et al. (2006). Enriched environment
experience overcomes the memory deficits and depressive-like behavior induced by early
life stress. Neuroscience letters. 404, 208-212.
Dalle Molle, R., Portella, A.K., Goldani, M.Z., Kapczinski, F.P., Leistner-Segal, S., Salum, G.A.,
et al. (2012). Associations between parenting behavior and anxiety in a rodent model and
a clinical sample: relationship to peripheral BDNF levels. Transl. psychiatry. 2, e195.
Danielewicz, J. and Hess, G. (2014). Early life stress alters synaptic modification range in the rat
lateral amygdala. Behavioural brain research. 265, 32-37.
Daoura, L., Hjalmarsson, M., Oreland, S., Nylander, I. and Roman, E. (2010). Postpartum
behavioral profiles in Wistar rats following maternal separation - altered exploration and
risk-assessment behavior in MS15 dams. Frontiers in behavioral neuroscience. 4, 37.
Datson, N.A., van der Perk, J., de Kloet, E.R. and Vreugdenhil, E. (2001). Identification of
corticosteroid-responsive genes in rat hippocampus using serial analysis of gene
expression. The European journal of neuroscience. 14, 675-689.
de Kloet, E.R., Sibug, R.M., Helmerhorst, F.M. and Schmidt, M.V. (2005). Stress, genes and the
mechanism of programming the brain for later life. Neuroscience and biobehavioral
reviews. 29, 271-281.
De Rubeis, S. and Buxbaum, J.D. (2015). Genetics and genomics of autism spectrum disorder:
embracing complexity. Human molecular genetics. 24, R24-31.
Delwig, A., Logan, A.M., Copenhagen, D.R. and Ahn, A.H. (2012). Light evokes melanopsin-
dependent vocalization and neural activation associated with aversive experience in
neonatal mice. PloS one. 7, e43787.
Dong, H.W., Swanson, L.W., Chen, L., Fanselow, M.S. and Toga, A.W. (2009). Genomic-
anatomic evidence for distinct functional domains in hippocampal field CA1.
Proceedings of the National Academy of Sciences. 106, 11794-11799.
Dubé, C.M., Molet, J., Singh-Taylor, A., Ivy, A., Maras, P.M. and Baram, T.Z. (2015). Hyper-
excitability and epilepsy generated by chronic early-life stress. Neurobiology of stress. 2,
10-19.
Dube, S.R., Williamson, D.F., Thompson, T., Felitti, V.J. and Anda, R.F. (2004). Assessing the
reliability of retrospective reports of adverse childhood experiences among adult HMO
members attending a primary care clinic. Child abuse & neglect. 28, 729-737.
148
Dumitriu, D., Rodriguez, A. and Morrison, J.H. (2011). High-throughput, detailed, cell-specific
neuroanatomy of dendritic spines using microinjection and confocal microscopy. Nature
protocols. 6, 1391-1411.
Duncan, L.E. and Keller, M.C. (2011). A critical review of the first 10 years of candidate gene-
by-environment interaction research in psychiatry. The American journal of psychiatry.
168, 1041-1049.
Duncan, L.E., Pollastri, A.R. and Smoller, J.W. (2014). Mind the gap: why many geneticists and
psychological scientists have discrepant views about gene-environment interaction (GxE)
research. The American psychologist. 69, 249-268.
Dzirasa, K. and Covington, H.E., 3rd. (2012). Increasing the validity of experimental models for
depression. Annals of the New York Academy of Sciences. 1265, 36-45.
Eagleson, K.L., Milner, T.A., Xie, Z. and Levitt, P. (2013). Synaptic and extrasynaptic location
of the receptor tyrosine kinase met during postnatal development in the mouse neocortex
and hippocampus. The Journal of comparative neurology. 521, 3241-3259.
Eagleson, K.L., Lane, C.J., McFadyen-Ketchum, L., Solak, S., Wu, H.H. and Levitt, P. (2016).
Distinct intracellular signaling mediates C-MET regulation of dendritic growth and
synaptogenesis. Developmental neurobiology.
Ehret, G. and Haack, B. (1982). Ultrasound recognition in house mice: key-stimulus
configuration and recognition mechanism. Journal of Comparative Physiology. 148, 245-
251.
Ehret, G. (2005). Infant rodent ultrasounds - a gate to the understanding of sound
communication. Behavior genetics. 35, 19-29.
Eiland, L. and McEwen, B.S. (2012). Early life stress followed by subsequent adult chronic
stress potentiates anxiety and blunts hippocampal structural remodeling. Hippocampus.
22, 82-91.
Eklund, M.B., Johansson, L.M., Uvnas-Moberg, K. and Arborelius, L. (2009). Differential
effects of repeated long and brief maternal separation on behaviour and neuroendocrine
parameters in Wistar dams. Behavioural brain research. 203, 69-75.
Entringer, S., Buss, C., Kumsta, R., Hellhammer, D.H., Wadhwa, P.D. and Wust, S. (2009).
Prenatal psychosocial stress exposure is associated with subsequent working memory
performance in young women. Behavioral neuroscience. 123, 886-893.
Fabricius, K., Wortwein, G. and Pakkenberg, B. (2008). The impact of maternal separation on
adult mouse behaviour and on the total neuron number in the mouse hippocampus. Brain
structure & function. 212, 403-416.
Fanning, J.R., Lee, R., Gozal, D., Coussons-Read, M. and Coccaro, E.F. (2015). Childhood
trauma and parental style: Relationship with markers of inflammation, oxidative stress,
149
and aggression in healthy and personality disordered subjects. Biological psychology.
112, 56-65.
Felix-Ortiz, A.C., Beyeler, A., Seo, C., Leppla, C.A., Wildes, C.P. and Tye, K.M. (2013). BLA
to vHPC inputs modulate anxiety-related behaviors. Neuron. 79, 658-664.
Fernandez Espejo, E. (1997). Structure of the mouse behaviour on the elevated plus-maze test of
anxiety. Behavioural brain research. 86, 105-112.
Francis, D.D. and Meaney, M.J. (1999). Maternal care and the development of stress responses.
Current opinion in neurobiology. 9, 128-134.
Franklin, T.B., Linder, N., Russig, H., Thony, B. and Mansuy, I.M. (2011). Influence of early
stress on social abilities and serotonergic functions across generations in mice. PloS one.
6, e21842.
Galon, J., Franchimont, D., Hiroi, N., Frey, G., Boettner, A., Ehrhart-Bornstein, M., et al. (2002).
Gene profiling reveals unknown enhancing and suppressive actions of glucocorticoids on
immune cells. FASEB journal : official publication of the Federation of American
Societies for Experimental Biology. 16, 61-71.
Gamo, Y., Troup, C., Mitchell, S.E., Hambly, C., Vaanholt, L.M. and Speakman, J.R. (2013).
Body temperatures and physical activity of female mice during lactation. The Journal of
experimental biology. 216, 3751-3761.
Garner, A.S. and Shonkoff, J.P. (2012). Early childhood adversity, toxic stress, and the role of
the pediatrician: translating developmental science into lifelong health. Pediatrics. 129,
e224-231.
Gatt, J.M., Nemeroff, C.B., Dobson-Stone, C., Paul, R.H., Bryant, R.A., Schofield, P.R., et al.
(2009). Interactions between BDNF Val66Met polymorphism and early life stress predict
brain and arousal pathways to syndromal depression and anxiety. Mol Psychiatry. 14,
681-695.
Gilles, E.E., Schultz, L. and Baram, T.Z. (1996). Abnormal corticosterone regulation in an
immature rat model of continuous chronic stress. Pediatric neurology. 15, 114-119.
Graves, A.R., Moore, S.J., Bloss, E.B., Mensh, B.D., Kath, W.L. and Spruston, N. (2012).
Hippocampal pyramidal neurons comprise two distinct cell types that are
countermodulated by metabotropic receptors. Neuron. 76, 776-789.
Green, J.G., McLaughlin, K.A., Berglund, P.A., Gruber, M.J., Sampson, N.A., Zaslavsky, A.M.,
et al. (2010). Childhood adversities and adult psychiatric disorders in the national
comorbidity survey replication I: associations with first onset of DSM-IV disorders.
Archives of general psychiatry. 67, 113-123.
Gunn, B.G., Cunningham, L., Cooper, M.A., Corteen, N.L., Seifi, M., Swinny, J.D., et al. (2013).
Dysfunctional astrocytic and synaptic regulation of hypothalamic glutamatergic
150
transmission in a mouse model of early-life adversity: relevance to neurosteroids and
programming of the stress response. The Journal of neuroscience. 33, 19534-19554.
Gunnar, M.R. and Vazquez, D.M. (2001). Low cortisol and a flattening of expected daytime
rhythm: potential indices of risk in human development. Development and
psychopathology. 13, 515-538.
Gutierrez, H., Dolcet, X., Tolcos, M. and Davies, A. (2004). HGF regulates the development of
cortical pyramidal dendrites. Development. 131, 3717-3726.
Gutteling, B.M., de Weerth, C., Zandbelt, N., Mulder, E.J., Visser, G.H. and Buitelaar, J.K.
(2006). Does maternal prenatal stress adversely affect the child's learning and memory at
age six? Journal of abnormal child psychology. 34, 789-798.
Haack, B., Markl, H. and Ehret, G. (1983). "Sound communication between parents and
offspring," in The Auditory Psychobiology of the Mouse, ed. Williot, J. F. (Charles C
Thomas Pub Ltd), 57-97.
Han, J.S., Bird, G.C., Li, W., Jones, J. and Neugebauer, V. (2005). Computerized analysis of
audible and ultrasonic vocalizations of rats as a standardized measure of pain-related
behavior. Journal of neuroscience methods. 141, 261-269.
Hånell, A. and Marklund, N. (2014). Structured evaluation of rodent behavioral tests used in
drug discovery research. Frontiers in behavioral neuroscience. 8, 252.
Harrison, E.L., Jaehne, E.J., Jawahar, M.C., Corrigan, F. and Baune, B.T. (2014). Maternal
separation modifies behavioural and neuroendocrine responses to stress in CCR7
deficient mice. Behavioural brain research. 263, 169-175.
Hascup, E.R., Hascup, K.N., Stephens, M., Pomerleau, F., Huettl, P., Gratton, A., et al. (2010).
Rapid microelectrode measurements and the origin and regulation of extracellular
glutamate in rat prefrontal cortex. Journal of neurochemistry. 115, 1608-1620.
Hauser, H. and Gandelman, R. (1985). Lever pressing for pups: evidence for hormonal influence
upon maternal behavior of mice. Hormones and behavior. 19, 454-468.
Hedrick, A., Lee, Y., Wallace, G.L., Greenstein, D., Clasen, L., Giedd, J.N., et al. (2012).
Autism risk gene MET variation and cortical thickness in typically developing children
and adolescents. Autism research. 5, 434-439.
Hess, S.E., Rohr, S., Dufour, B.D., Gaskill, B.N., Pajor, E.A. and Garner, J.P. (2008). Home
improvement: C57BL/6J mice given more naturalistic nesting materials build better
nests. Journal of the American Association for Laboratory Animal Science : JAALAS. 47,
25-31.
Hettema, J.M., Neale, M.C. and Kendler, K.S. (2001). A review and meta-analysis of the genetic
epidemiology of anxiety disorders. The American journal of psychiatry. 158, 1568-1578.
151
Holfoth, D.P., Neilans, E.G. and Dent, M.L. (2014). Discrimination of partial from whole
ultrasonic vocalizations using a go/no-go task in mice. The Journal of the Acoustical
Society of America. 136, 3401.
Ishikawa, A. and Nakamura, S. (2006). Ventral hippocampal neurons project axons
simultaneously to the medial prefrontal cortex and amygdala in the rat. Journal of
neurophysiology. 96, 2134-2138.
Ivy, A.S., Brunson, K.L., Sandman, C. and Baram, T.Z. (2008). Dysfunctional nurturing
behavior in rat dams with limited access to nesting material: a clinically relevant model
for early-life stress. Neuroscience. 154, 1132-1142.
Ivy, A.S., Rex, C.S., Chen, Y., Dube, C., Maras, P.M., Grigoriadis, D.E., et al. (2010).
Hippocampal dysfunction and cognitive impairments provoked by chronic early-life
stress involve excessive activation of CRH receptors. The Journal of neuroscience. 30,
13005-13015.
Jackson, P.B., Boccuto, L., Skinner, C., Collins, J.S., Neri, G., Gurrieri, F., et al. (2009). Further
evidence that the rs1858830 C variant in the promoter region of the MET gene is
associated with autistic disorder. Autism research. 2, 232-236.
Joëls, M. and Baram, T.Z. (2009). The neuro-symphony of stress. Nature. 10, 459-466.
Johnson, D.E., Guthrie, D., Smyke, A.T., Koga, S.F., Fox, N.A., Zeanah, C.H., et al. (2010).
Growth and associations between auxology, caregiving environment, and cognition in
socially deprived Romanian children randomized to foster vs ongoing institutional care.
Archives of pediatrics & adolescent medicine. 164, 507-516.
Judson, M.C., Bergman, M.Y., Campbell, D.B., Eagleson, K.L. and Levitt, P. (2009). Dynamic
gene and protein expression patterns of the autism-associated met receptor tyrosine
kinase in the developing mouse forebrain. The Journal of comparative neurology. 513,
511-531.
Judson, M.C., Eagleson, K.L., Wang, L. and Levitt, P. (2010). Evidence of cell-nonautonomous
changes in dendrite and dendritic spine morphology in the met-signaling-deficient mouse
forebrain. The Journal of comparative neurology. 518, 4463-4478.
Judson, M.C., Amaral, D.G. and Levitt, P. (2011a). Conserved subcortical and divergent cortical
expression of proteins encoded by orthologs of the autism risk gene MET. Cerebral
cortex. 21, 1613-1626.
Judson, M.C., Eagleson, K.L. and Levitt, P. (2011b). A new synaptic player leading to autism
risk: Met receptor tyrosine kinase. Journal of neurodevelopmental disorders. 3, 282-292.
Kanatsou, S., Ter Horst, J.P., Harris, A.P., Seckl, J.R., Krugers, J. and Joëls, M. (2016). Effects
of mineralocorticoid receptor overexpression on anxiety and memory after early life
stress in female mice. Frontiers in behavioral neuroscience.
152
Karst, H. and Joels, M. (2005). Corticosterone slowly enhances miniature excitatory postsynaptic
current amplitude in mice CA1 hippocampal cells. Journal of neurophysiology. 94, 3479-
3486.
Karst, H., Berger, S., Erdmann, G., Schutz, G. and Joels, M. (2010). Metaplasticity of amygdalar
responses to the stress hormone corticosterone. Proceedings of the National Academy of
Sciences. 107, 14449-14454.
Kessler, R.C., McLaughlin, K.A., Green, J.G., Gruber, M.J., Sampson, N.A., Zaslavsky, A.M., et
al. (2010). Childhood adversities and adult psychopathology in the WHO World Mental
Health Surveys. The British journal of psychiatry : the journal of mental science. 197,
378-385.
Kim, D.R., Bale, T.L. and Epperson, C.N. (2015). Prenatal programming of mental illness:
current understanding of relationship and mechanisms. Current psychiatry reports. 17, 5.
Kim, J.S. and Lee, S.H. (2016). Influence of interactions between genes and childhood trauma on
refractoriness in psychiatric disorders. Progress in neuro-psychopharmacology &
biological psychiatry.
Kim, Y. and Spruston, N. (2012). Target-specific output patterns are predicted by the distribution
of regular-spiking and bursting pyramidal neurons in the subiculum. Hippocampus. 22,
693-706.
Kinney, D.K., Miller, A.M., Crowley, D.J., Huang, E. and Gerber, E. (2008). Autism prevalence
following prenatal exposure to hurricanes and tropical storms in Louisiana. Journal of
autism and developmental disorders. 38, 481-488.
Kobayashi, T., Mitamura, T. and Nagasawa, H. (1997). Comparison of the spontaneous
behaviour of four strains of female mice with different physiological properties. In vivo.
11, 313-317.
Kohl, C., Wang, X.D., Grosse, J., Fournier, C., Harbich, D., Westerholz, S., et al. (2015).
Hippocampal neuroligin-2 links early-life stress with impaired social recognition and
increased aggression in adult mice. Psychoneuroendocrinology. 55, 128-143.
Kosten, T.A. and Kehoe, P. (2010). Immediate and enduring effects of neonatal isolation on
maternal behavior in rats. International journal of developmental neuroscience. 28, 53-
61.
Kurejova, M., Nattenmuller, U., Hildebrandt, U., Selvaraj, D., Stosser, S. and Kuner, R. (2010).
An improved behavioural assay demonstrates that ultrasound vocalizations constitute a
reliable indicator of chronic cancer pain and neuropathic pain. Molecular pain. 6, 18.
Lambert, N., Wermenbol, V., Pichon, B., Acosta, S., van den Ameele, J., Perazzolo, C., et al.
(2014). A familial heterozygous null mutation of MET in autism spectrum disorder.
Autism research. 7, 617-622.
153
Lang, A.J., Aarons, G.A., Gearity, J., Laffaye, C., Satz, L., Dresselhaus, T.R., et al. (2008).
Direct and indirect links between childhood maltreatment, posttraumatic stress disorder,
and women's health. Behavioral medicine. 33, 125-135.
Leon, M., Croskerry, P.G. and Smith, G.K. (1978). Thermal control of mother-young contact in
rats. Physiology & behavior. 21, 790-811.
Levine, A., Worrell, T.R., Zimnisky, R. and Schmauss, C. (2012). Early life stress triggers
sustained changes in histone deacetylase expression and histone H4 modifications that
alter responsiveness to adolescent antidepressant treatment. Neurobiology of disease. 45,
488-498.
Levine, S. (2005). Developmental determinants of sensitivity and resistance to stress.
Psychoneuroendocrinology. 30, 939-946.
Li, J., Vestergaard, M., Obel, C., Christensen, J., Precht, D.H., Lu, M., et al. (2009). A
nationwide study on the risk of autism after prenatal stress exposure to maternal
bereavement. Pediatrics. 123, 1102-1107.
Li, J., Olsen, J., Vestergaard, M. and Obel, C. (2010). Attention-deficit/hyperactivity disorder in
the offspring following prenatal maternal bereavement: a nationwide follow-up study in
Denmark. European child & adolescent psychiatry. 19, 747-753.
Liao, X.M., Yang, X.D., Jia, J., Li, J.T., Xie, X.M., Su, Y.A., et al. (2014). Blockade of
corticotropin-releasing hormone receptor 1 attenuates early-life stress-induced synaptic
abnormalities in the neonatal hippocampus. Hippocampus. 24, 528-540.
Lim, C.S. and Walikonis, R.S. (2008). Hepatocyte growth factor and c-Met promote dendritic
maturation during hippocampal neuron differentiation via the Akt pathway. Cellular
signalling. 20, 825-835.
Lippmann, M., Bress, A., Nemeroff, C.B., Plotsky, P.M. and Monteggia, L.M. (2007). Long-
term behavioural and molecular alterations associated with maternal separation in rats.
The European journal of neuroscience. 25, 3091-3098.
Liu, D., Diorio, J., Tannenbaum, B., Caldji, C., Francis, D., Freedman, A., et al. (1997). Maternal
care, hippocampal glucocorticoid receptors, and hypothalamic-pituitary-adrenal
responses to stress. Science. 277, 1659-1662.
Liu, W.H., Chuang, H.L., Huang, Y.T., Wu, C.C., Chou, G.T., Wang, S., et al. (2016). Alteration
of behavior and monoamine levels attributable to Lactobacillus plantarum PS128 in
germ-free mice. Behavioural brain research. 298, 202-209.
Loi, M., Koricka, S., Lucassen, P.J. and Joëls, M. (2014). Age- and sex-dependent effects of
early life stress on hippocampal neurogenesis. Frontiers in endocrinology. 5, 13.
154
Lok, A., Bockting, C.L., Koeter, M.W., Snieder, H., Assies, J., Mocking, R.J., et al. (2013).
Interaction between the MTHFR C677T polymorphism and traumatic childhood events
predicts depression. Transl. psychiatry. 3, e288.
Lupien, S.J. and Lepage, M. (2001). Stress, memory, and the hippocampus: can't live with it,
can't live without it. Behavioural brain research. 127, 137-158.
Lyons, D.M., Parker, K.J. and Schatzberg, A.F. (2010). Animal models of early life stress:
Implications for understanding resilience. Developmental psychobiology. 52, 402-410.
Malter Cohen, M., Jing, D., Yang, R.R., Tottenham, N., Lee, F.S. and Casey, B.J. (2013a).
Early-life stress has persistent effects on amygdala function and development in mice and
humans. Proc Natl Acad Sci U S A. 110, 18274-18278.
Malter Cohen, M., Tottenham, N. and Casey, B.J. (2013b). Translational developmental studies
of stress on brain and behavior: implications for adolescent mental health and illness?
Neuroscience. 249, 53-62.
Maric, N.P., Dunjic, B., Stojiljkovic, D.J., Britvic, D. and Jasovic-Gasic, M. (2010). Prenatal
stress during the 1999 bombing associated with lower birth weight-a study of 3,815 births
from Belgrade. Archives of women's mental health. 13, 83-89.
McEwen, B.S., Biron, C.A., Brunson, K.W., Bulloch, K., Chambers, W.H., Dhabhar, F.S., et al.
(1997). The role of adrenocorticoids as modulators of immune function in health and
disease: neural, endocrine and immune interactions. Brain research reviews. 23, 79-133.
McEwen, B.S. and McEwen, C.A. (2015). Social, Psychological, and Physiological Reactions to
Stress. Emerging Trends in the Social and Behavioral Sciences. 1-15.
McLaughlin, K.A., Sheridan, M.A., Tibu, F., Fox, N.A., Zeanah, C.H. and Nelson, C.A., 3rd.
(2015a). Causal effects of the early caregiving environment on development of stress
response systems in children. Proceedings of the National Academy of Sciences. 112,
5637-5642.
McLaughlin, R.J., Verlezza, S., Gray, J.M., Hill, M.N. and Walker, C.D. (2015b). Inhibition of
anandamide hydrolysis dampens the neuroendocrine response to stress in neonatal rats
subjected to suboptimal rearing conditions. Stress. 1-11.
Mehta, M. and Schmauss, C. (2011). Strain-specific cognitive deficits in adult mice exposed to
early life stress. Behavioral neuroscience. 125, 29-36.
Millstein, R.A. and Holmes, A. (2007). Effects of repeated maternal separation on anxiety- and
depression-related phenotypes in different mouse strains. Neuroscience and
biobehavioral reviews. 31, 3-17.
Moffitt, T.E. (2013). Childhood exposure to violence and lifelong health: clinical intervention
science and stress-biology research join forces. Developmental Psychopathology. 25,
1619-1634.
155
Molet, J., Maras, P.M., Avishai-Eliner, S. and Baram, T.Z. (2014). Naturalistic rodent models of
chronic early-life stress. Developmental psychobiology. 56, 1675-1688.
Moriceau, S., Shionoya, K., Jakubs, K. and Sullivan, R.M. (2009). Early-life stress disrupts
attachment learning: the role of amygdala corticosterone, locus ceruleus corticotropin
releasing hormone, and olfactory bulb norepinephrine. The Journal of neuroscience. 29,
15745-15755.
Moriceau, S., Roth, T.L. and Sullivan, R.M. (2010). Rodent model of infant attachment learning
and stress. Developmental psychobiology. 52, 651-660.
Naninck, E.F., Hoeijmakers, L., Kakava-Georgiadou, N., Meesters, A., Lazic, S.E., Lucassen,
P.J., et al. (2015). Chronic early life stress alters developmental and adult neurogenesis
and impairs cognitive function in mice. Hippocampus. 25, 309-328.
Neilans, E.G., Holfoth, D.P., Radziwon, K.E., Portfors, C.V. and Dent, M.L. (2014).
Discrimination of ultrasonic vocalizations by CBA/CaJ mice (Mus musculus) is related to
spectrotemporal dissimilarity of vocalizations. PloS one. 9, e85405.
Niwa, M., Matsumoto, Y., Mouri, A., Ozaki, N. and Nabeshima, T. (2011). Vulnerability in early
life to changes in the rearing environment plays a crucial role in the aetiopathology of
psychiatric disorders. The international journal of neuropsychopharmacology. 14, 459-
477.
Noirot, E. (1966). Ultrasons et comportements maternels chez les petite rongeurs. Annales de la
Société royale zoologique de Belgique. 95, 47-56.
Noirot, E. (1972). Ultrasounds and maternal behavior in small rodents. Developmental
psychobiology. 5, 371-387.
Norman, R.E., Byambaa, M., De, R., Butchart, A., Scott, J. and Vos, T. (2012). The long-term
health consequences of child physical abuse, emotional abuse, and neglect: a systematic
review and meta-analysis. PLoS medicine. 9, e1001349.
O'Donnell, K., O'Connor, T.G. and Glover, V. (2009). Prenatal stress and neurodevelopment of
the child: focus on the HPA axis and role of the placenta. Developmental neuroscience.
31, 285-292.
O'Donnell, K.J., Glover, V., Barker, E.D. and O'Connor, T.G. (2014). The persisting effect of
maternal mood in pregnancy on childhood psychopathology. Development and
psychopathology. 26, 393-403.
Ognibene, E., Adriani, W., Macri, S. and Laviola, G. (2007). Neurobehavioural disorders in the
infant reeler mouse model: interaction of genetic vulnerability and consequences of
maternal separation. Behavioural brain research. 177, 142-149.
156
Okabe, S., Nagasawa, M., Kihara, T., Kato, M., Harada, T., Koshida, N., et al. (2013). Pup odor
and ultrasonic vocalizations synergistically stimulate maternal attention in mice.
Behavioral neuroscience. 127, 432-438.
Okaty, B.W., Freret, M.E., Rood, B.D., Brust, R.D., Hennessy, M.L., deBairos, D., et al. (2015).
Multi-Scale Molecular Deconstruction of the Serotonin Neuron System. Neuron. 88, 774-
791.
Ouyang, L., Fang, X., Mercy, J., Perou, R. and Grosse, S.D. (2008). Attention-
deficit/hyperactivity disorder symptoms and child maltreatment: a population-based
study. The Journal of pediatrics. 153, 851-856.
Pearson, R.M., Evans, J., Kounali, D., Lewis, G., Heron, J., Ramchandani, P.G., et al. (2013).
Maternal depression during pregnancy and the postnatal period: risks and possible
mechanisms for offspring depression at age 18 years. JAMA psychiatry. 70, 1312-1319.
Pedersen, C.A., Vadlamudi, S., Boccia, M.L. and Moy, S.S. (2011). Variations in Maternal
Behavior in C57BL/6J Mice: Behavioral Comparisons between Adult Offspring of High
and Low Pup-Licking Mothers. Frontiers in psychiatry. 2, 42.
Peng, Y., Huentelman, M., Smith, C. and Qiu, S. (2014). "MET receptor tyrosine kinase as an
autism genetic risk factor," in International review of neurobiology, ed. Konopka, G.
(Elsevier, Inc.), 135-165.
Pereira, M. and Ferreira, A. (2006). Demanding pups improve maternal behavioral impairments
in sensitized and haloperidol-treated lactating female rats. Behavioural brain research.
175, 139-148.
Plotsky, P.M., Thrivikraman, K.V., Nemeroff, C.B., Caldji, C., Sharma, S. and Meaney, M.J.
(2005). Long-term consequences of neonatal rearing on central corticotropin-releasing
factor systems in adult male rat offspring. Neuropsychopharmacology : official
publication of the American College of Neuropsychopharmacology. 30, 2192-2204.
Portfors, C.V. and Perkel, D.J. (2014). The role of ultrasonic vocalizations in mouse
communication. Current opinion in neurobiology. 28, 115-120.
Pryce, C. (2008). Postnatal ontogeny of expression of the corticosteroid receptor genes in
mammalian brains: Inter-species and intra-species differences. Brain Research Reviews.
57, 596-605.
Qiu, S., Anderson, C.T., Levitt, P. and Shepherd, G.M. (2011). Circuit-specific intracortical
hyperconnectivity in mice with deletion of the autism-associated Met receptor tyrosine
kinase. The Journal of neuroscience. 31, 5855-5864.
Qiu, S., Lu, Z. and Levitt, P. (2014). MET receptor tyrosine kinase controls dendritic
complexity, spine morphogenesis, and glutamatergic synapse maturation in the
hippocampus. The Journal of neuroscience. 34, 16166-16179.
157
Rai, D., Golding, J., Magnusson, C., Steer, C., Lewis, G. and Dalman, C. (2012). Prenatal and
early life exposure to stressful life events and risk of autism spectrum disorders:
population-based studies in Sweden and England. PloS one. 7, e38893.
Raineki, C., Moriceau, S. and Sullivan, R.M. (2010). Developing a neurobehavioral animal
model of infant attachment to an abusive caregiver. Biological psychiatry. 67, 1137-1145.
Raineki, C., Cortes, M.R., Belnoue, L. and Sullivan, R.M. (2012). Effects of early-life abuse
differ across development: infant social behavior deficits are followed by adolescent
depressive-like behaviors mediated by the amygdala. The Journal of neuroscience : the
official journal of the Society for Neuroscience. 32, 7758-7765.
Raineki, C., Sarro, E., Rincon-Cortes, M., Perry, R., Boggs, J., Holman, C.J., et al. (2015).
Paradoxical neurobehavioral rescue by memories of early-life abuse: the safety signal
value of odors learned during abusive attachment. Neuropsychopharmacology : official
publication of the American College of Neuropsychopharmacology. 40, 906-914.
Regev, L. and Baram, T.Z. (2014). Corticotropin releasing factor in neuroplasticity. Frontiers in
neuroendocrinology. 35, 171-179.
Réus, G.Z., Stringari, R.B., Ribeiro, K.F., Cipriano, A.L., Panizzutti, B.S., Stertz, L., et al.
(2011). Maternal deprivation induces depressive-like behaviour and alters neurotrophin
levels in the rat brain. Neurochemical research. 36, 460-466.
Rice, C.J., Sandman, C.A., Lenjavi, M.R. and Baram, T.Z. (2008). A novel mouse model for
acute and long-lasting consequences of early life stress. Endocrinology. 149, 4892-4900.
Roberts, E.M., English, P.B., Grether, J.K., Windham, G.C., Somberg, L. and Wolff, C. (2007).
Maternal residence near agricultural pesticide applications and autism spectrum disorders
among children in the California Central Valley. Environmental health perspectives. 115,
1482-1489.
Romano-Lopez, Q.F., Mendez-Diaz, M.D., Garcia, F.G.D., Regalado-Santiago, C.D., Ruiz
Contreras, A.E.D. and Prospero-Garcia, O.D. (2015). Maternal separation and early stress
cause long-lasting effects on dopaminergic and endocannabinergic systems and alters
dendritic morphology in the nucleus accumbens and frontal cortex in rats. Developmental
neurobiology.
Roth, T.L. and Sullivan, R.M. (2005). Memory of early maltreatment: neonatal behavioral and
neural correlates of maternal maltreatment within the context of classical conditioning.
Biological psychiatry. 57, 823-831.
Rudie, J.D., Hernandez, L.M., Brown, J.A., Beck-Pancer, D., Colich, N.L., Gorrindo, P., et al.
(2012). Autism-associated promoter variant in MET impacts functional and structural
brain networks. Neuron. 75, 904-915.
Sachs, B.D., Rodriguiz, R.M., Siesser, W.B., Kenan, A., Royer, E.L., Jacobsen, J.P., et al.
(2013). The effects of brain serotonin deficiency on behavioural disinhibition and
158
anxiety-like behaviour following mild early life stress. The international journal of
neuropsychopharmacology / official scientific journal of the Collegium Internationale
Neuropsychopharmacologicum. 16, 2081-2094.
Sanchez, M.M., Ladd, C.O. and Plotsky, P.M. (2001). Early adverse experience as a
developmental risk factor for later psychopathology: evidence from rodent and primate
models. Development and psychopathology. 13, 419-449.
Santarelli, S., Lesuis, S.L., Wang, X.D., Wagner, K.V., Hartmann, J., Labermaier, C., et al.
(2014). Evidence supporting the match/mismatch hypothesis of psychiatric disorders.
European neuropsychopharmacology : the journal of the European College of
Neuropsychopharmacology. 24, 907-918.
Sapolsky, R.M., Romero, L.M. and Munck, A.U. (2000). How do glucocorticoids influence
stress responses? Integrating permissive, suppressive, stimulatory, and preparative
actions. Endocrine Reviews. 21, 55-89.
Savignac, H.M., Dinan, T.G. and Cryan, J.F. (2011). Resistance to early-life stress in mice:
effects of genetic background and stress duration. Frontiers in behavioral neuroscience.
5, 13.
Schilling, E.A., Aseltine, R.H. and Gore, S. (2008). The impact of cumulative childhood
adversity on young adult mental health: measures, models, and interpretations. Social
science & medicine. 66, 1140-1151.
Schmidt, M.V., Wang, X.D. and Meijer, O.C. (2011). Early life stress paradigms in rodents:
potential animal models of depression? Psychopharmacology. 214, 131-140.
Semple, B.D., Blomgren, K., Gimlin, K., Ferriero, D.M. and Noble-Haeusslein, L.J. (2013).
Brain development in rodents and humans: Identifying benchmarks of maturation and
vulnerability to injury across species. Progress in neurobiology. 106-107, 1-16.
Shanks, N., Larocque, S. and Meaney, M.J. (1995). Neonatal endotoxin exposure alters the
development of the hypothalamic-pituitary-adrenal axis: early illness and later
responsivity to stress. The Journal of Neuroscience. 15, 376-384.
Shapero, B.G., Hamilton, J.L., Stange, J.P., Liu, R.T., Abramson, L.Y. and Alloy, L.B. (2015).
Moderate Childhood Stress Buffers Against Depressive Response to Proximal Stressors:
A Multi-Wave Prospective Study of Early Adolescents. Journal of abnormal child
psychology. 43, 1403-1413.
Sharma, S., Powers, A., Bradley, B. and Ressler, K.J. (2015). Gene x Environment Determinants
of Stress- and Anxiety-Related Disorders. Annual review of psychology. 67, 239-261.
Shepard, K.N., Lin, F.G., Zhao, C.L., Chong, K.K. and Liu, R.C. (2015). Behavioral relevance
helps untangle natural vocal categories in a specific subset of core auditory cortical
pyramidal neurons. The Journal of neuroscience : the official journal of the Society for
Neuroscience. 35, 2636-2645.
159
Shin, S.Y., Han, S.H., Woo, R.S., Jang, S.H. and Min, S.S. (2016). Adolescent mice show
anxiety- and aggressive-like behavior and the reduction of long-term potentiation in
mossy fiber-CA3 synapses after neonatal maternal separation. Neuroscience. 316, 221-
231.
Shonkoff, J.P. and Levitt, P. (2010). Neuroscience and the future of early childhood policy:
moving from why to what and how. Neuron. 67, 689-691.
Shonkoff, J.P., Garner, A.S., Siegel, B.S., Dobbins, M.I., Earls, M.F., McGuinn, L., et al. (2011).
The Lifelong Effects of Early Childhood Adversity and Toxic Stress. Pediatrics. 129,
e232-e246.
Shonkoff, J.P. (2012). Leveraging the biology of adversity to address the roots of disparities in
health and development. Proc. Natl. Acad. Sci. USA. 109, 17302-17307.
Sirotin, Y.B., Costa, M.E. and Laplagne, D.A. (2014). Rodent ultrasonic vocalizations are bound
to active sniffing behavior. Frontiers in behavioral neuroscience. 8, 399.
Smith, D.M. and Bulkin, D.A. (2014). The form and function of hippocampal context
representations. Neuroscience and biobehavioral reviews. 40, 52-61.
Sorge, R.E., Martin, L.J., Isbester, K.A., Sotocinal, S.G., Rosen, S., Tuttle, A.H., et al. (2014).
Olfactory exposure to males, including men, causes stress and related analgesia in
rodents. Nature methods. 11, 629-632.
Spinhoven, P., Penninx, B.W., Hickendorff, M., van Hemert, A.M., Bernstein, D.P. and Elzinga,
B.M. (2014). Childhood Trauma Questionnaire: factor structure, measurement
invariance, and validity across emotional disorders. Psychological assessment. 26, 717-
729.
Spratt, E.G., Nicholas, J.S., Brady, K.T., Carpenter, L.A., Hatcher, C.R., Meekins, K.A., et al.
(2012). Enhanced cortisol response to stress in children in autism. Journal of autism and
developmental disorders. 42, 75-81.
Stamoulis, C., Vanderwert, R.E., Zeanah, C.H., Fox, N.A. and Nelson, C.A. (2015). Early
Psychosocial Neglect Adversely Impacts Developmental Trajectories of Brain
Oscillations and Their Interactions. Journal of cognitive neuroscience. 27, 2512-2528.
Stein, M.B., Jang, K.L., Taylor, S., Vernon, P.A. and Livesley, W.J. (2002). Genetic and
environmental influences on trauma exposure and posttraumatic stress disorder
symptoms: a twin study. The American journal of psychiatry. 159, 1675-1681.
Stern, J.M. (1997). Offspring-induced nurturance: animal-human parallels. Developmental
psychobiology. 31, 19-37.
Stern, J.M. and Azzara, A.V. (2002). Thermal control of mother-young contact revisited:
hyperthermic rats nurse normally. Physiology & behavior. 77, 11-18.
160
Stern, J.M. and Keer, S.E. (2002). Acute hunger of rat pups elicits increased kyphotic nursing
and shorter intervals between nursing bouts: implications for changes in nursing with
time postpartum. Journal of comparative psychology. 116, 83-92.
Strange, B.A., Witter, M.P., Lein, E.S. and Moser, E.I. (2014). Functional organization of the
hippocampal longitudinal axis. Nature reviews. Neuroscience. 15, 655-669.
Stumpf, W.E., Heiss, C., Sar, M., Duncan, G.E. and Craver, C. (1989). Dexamethasone and
corticosterone receptor sites. Differential topographic distribution in rat hippocampus
revealed by high resolution autoradiography. Histochemistry. 92, 201-210.
Suderman, M., Borghol, N., Pappas, J.J., Pinto Pereira, S.M., Pembrey, M., Hertzman, C., et al.
(2014). Childhood abuse is associated with methylation of multiple loci in adult DNA.
BMC medical genomics. 7, 13.
Sullivan, P.F., Neale, M.C. and Kendler, K.S. (2000). Genetic epidemiology of major
depression: review and meta-analysis. The American journal of psychiatry. 157, 1552-
1562.
Thompson, B.L. and Levitt, P. (2015). Complete or partial reduction of the Met receptor tyrosine
kinase in distinct circuits differentially impacts mouse behavior. Journal of
neurodevelopmental disorders. 7, 35.
Troller-Renfree, S., McDermott, J.M., Nelson, C.A., Zeanah, C.H. and Fox, N.A. (2015). The
effects of early foster care intervention on attention biases in previously institutionalized
children in Romania. Developmental science. 18, 713-722.
Tsuda, M.C., Yamaguchi, N. and Ogawa, S. (2011). Early life stress disrupts peripubertal
development of aggression in male mice. Neuroreport. 22, 259-263.
Tsuda, M.C. and Ogawa, S. (2012). Long-lasting consequences of neonatal maternal separation
on social behaviors in ovariectomized female mice. PloS one. 7, e33028.
Tsuda, M.C., Yamaguchi, N., Nakata, M. and Ogawa, S. (2014). Modification of female and
male social behaviors in estrogen receptor beta knockout mice by neonatal maternal
separation. Frontiers in neuroscience. 8, 274.
Tsuzuki, H., Maekawa, M., Konno, R. and Hori, Y. (2012). Functional roles of endogenous D-
serine in pain-induced ultrasonic vocalization. Neuroreport. 23, 937-941.
Turner, H.A., Finkelhor, D. and Ormrod, R. (2006). The effect of lifetime victimization on the
mental health of children and adolescents. Social science & medicine. 62, 13-27.
Van den Bergh, B.R. and Marcoen, A. (2004). High antenatal maternal anxiety is related to
ADHD symptoms, externalizing problems, and anxiety in 8- and 9-year-olds. Child
development. 75, 1085-1097.
161
Van den Bergh, B.R., Mennes, M., Oosterlaan, J., Stevens, V., Stiers, P., Marcoen, A., et al.
(2005). High antenatal maternal anxiety is related to impulsivity during performance on
cognitive tasks in 14- and 15-year-olds. Neuroscience and biobehavioral reviews. 29,
259-269.
van der Kooij, M.A., Grosse, J., Zanoletti, O., Papilloud, A. and Sandi, C. (2015). The effects of
stress during early postnatal periods on behavior and hippocampal neuroplasticity
markers in adult male mice. Neuroscience. 311, 508-518.
van Hasselt, F.N., Boudewijns, Z.S., van der Knaap, N.J., Krugers, H.J. and Joels, M. (2012).
Maternal care received by individual pups correlates with adult CA1 dendritic
morphology and synaptic plasticity in a sex-dependent manner. Journal of
neuroendocrinology. 24, 331-340.
van Heerden, J.H., Russell, V., Korff, A., Stein, D.J. and Illing, N. (2010). Evaluating the
behavioural consequences of early maternal separation in adult C57BL/6 mice; the
importance of time. Behavioural brain research. 207, 332-342.
Veenema, A.H., Bredewold, R. and Neumann, I.D. (2007). Opposite effects of maternal
separation on intermale and maternal aggression in C57BL/6 mice: link to hypothalamic
vasopressin and oxytocin immunoreactivity. Psychoneuroendocrinology. 32, 437-450.
Venerosi, A., Cirulli, F., Capone, F. and Alleva, E. (2003). Prolonged perinatal AZT
administration and early maternal separation: effects on social and emotional behaviour
of periadolescent mice. Pharmacology, biochemistry, and behavior. 74, 671-681.
Voineagu, I., Wang, X., Johnston, P., Lowe, J.K., Tian, Y., Horvath, S., et al. (2011).
Transcriptomic analysis of autistic brain reveals convergent molecular pathology. Nature.
474, 380-384.
Volk, H.E., Kerin, T., Lurmann, F., Hertz-Picciotto, I., McConnell, R. and Campbell, D.B.
(2014). Autism spectrum disorder: interaction of air pollution with the MET receptor
tyrosine kinase gene. Epidemiology. 25, 44-47.
Wadhwa, P.D., Sandman, C.A., Porto, M., Dunkel-Schetter, C. and Garite, T.J. (1993). The
association between prenatal stress and infant birth weight and gestational age at birth: a
prospective investigation. Am J Obstet Gynecol. 169, 858-865.
Wang, L., Jiao, J. and Dulawa, S.C. (2011a). Infant maternal separation impairs adult cognitive
performance in BALB/cJ mice. Psychopharmacology (Berl). 216, 207-218.
Wang, X.D., Rammes, G., Kraev, I., Wolf, M., Liebl, C., Scharf, S.H., et al. (2011b). Forebrain
CRF(1) modulates early-life stress-programmed cognitive deficits. The Journal of
neuroscience. 31, 13625-13634.
Wang, X.D., Labermaier, C., Holsboer, F., Wurst, W., Deussing, J.M., Muller, M.B., et al.
(2012). Early-life stress-induced anxiety-related behavior in adult mice partially requires
162
forebrain corticotropin-releasing hormone receptor 1. The European journal of
neuroscience. 36, 2360-2367.
Wang, X.D., Su, Y.A., Wagner, K.V., Avrabos, C., Scharf, S.H., Hartmann, J., et al. (2013).
Nectin-3 links CRHR1 signaling to stress-induced memory deficits and spine loss. Nature
neuroscience. 16, 706-713.
Ward, A.J. (1990). A comparison and analysis of the presence of family problems during
pregnancy of mothers of "autistic" children and mothers of normal children. Child
psychiatry and human development. 20, 279-288.
Weinstock, M. (2005). The potential influence of maternal stress hormones on development and
mental health of the offspring. Brain, behavior, and immunity. 19, 296-308.
White, N.R., Adox, R., Reddy, A. and Barfield, R.J. (1992). Regulation of rat maternal behavior
by broadband pup vocalizations. Behavioral and neural biology. 58, 131-137.
Whitnall, M.H. (1993). Regulation of the hypothalamic corticotropin-releasing hormone
neurosecretory system. Progress in neurobiology. 40, 573-629.
Widom, C.S. (1999). Posttraumatic stress disorder in abused and neglected children grown up.
The American journal of psychiatry. 156, 1223-1229.
Wiley, J.W., Higgins, G.A. and Athey, B.D. (2016). Stress and glucocorticoid receptor
transcriptional programming in time and space: Implications for the brain-gut axis.
Neurogastroenterology and motility. 28, 12-25.
Williams, W.O., Riskin, D.K. and Mott, A.K. (2008). Ultrasonic sound as an indicator of acute
pain in laboratory mice. Journal of the American Association for Laboratory Animal
Science : JAALAS. 47, 8-10.
Windham, G.C., Zhang, L., Gunier, R., Croen, L.A. and Grether, J.K. (2006). Autism spectrum
disorders in relation to distribution of hazardous air pollutants in the san francisco bay
area. Environmental health perspectives. 114, 1438-1444.
Windle, R.J., Wood, S.A., Kershaw, Y.M., Lightman, S.L., Ingram, C.D. and Harbuz, M.S.
(2008). Increased Corticosterone Pulse Frequency During Adjuvant-Induced Arthritis and
its Relationship to Alterations in Stress Responsiveness. Journal of neuroendocrinology.
13, 905-911.
Wolterink-Donselaar, I.G., Meerding, J.M. and Fernandes, C. (2009). A method for gender
determination in newborn dark pigmented mice. Lab animal. 38, 35-38.
Woodside, B., Pelchat, R. and Leon, M. (1980). Acute elevation of the heat load of mother rats
curtails maternal nest bouts. Journal of comparative and physiological psychology. 94,
61-68.
163
Xie, L., Korkmaz, K.S., Braun, K. and Bock, J. (2013). Early life stress-induced histone
acetylations correlate with activation of the synaptic plasticity genes Arc and Egr1 in the
mouse hippocampus. Journal of neurochemistry. 125, 457-464.
Yang, X.D., Liao, X.M., Uribe-Marino, A., Liu, R., Xie, X.M., Jia, J., et al. (2015). Stress during
a critical postnatal period induces region-specific structural abnormalities and
dysfunction of the prefrontal cortex via CRF1. Neuropsychopharmacology : official
publication of the American College of Neuropsychopharmacology. 40, 1203-1215.
Yuen, E.Y., Liu, W., Karatsoreos, I.N., Feng, J., McEwen, B.S. and Yan, Z. (2009). Acute stress
enhances glutamatergic transmission in prefrontal cortex and facilitates working memory.
Proceedings of the National Academy of Sciences. 106, 14075-14079.
Zoicas, I. and Neumann, I.D. (2016). Maternal separation facilitates extinction of social fear in
adult male mice. Behavioural brain research. 297, 323-328.
Abstract (if available)
Abstract
Evidence from clinical studies has demonstrated a significant relation between early adverse experiences, noted as early-life stress (ELS), and later risk for emotional, cognitive, and physical health problems. However, many mental health disorders that have been associated with ELS have a significant heritability component that is currently unexplained by results from genome-wide association studies (GWAS). Interactions between multiple genetic and environmental factors are thought to be partly responsible for this ‘missing heritability’. The MET gene is a pertinent candidate that may interact with ELS to alter brain structure and behavior, as altered expression of MET receptor tyrosine kinase protein in mice has shown to affect synaptogenesis, neuronal growth and connectivity during early development, and behavior in adult mice. Moreover, a common variant (rs1858830 ‘C’ allele) in the promoter region has been associated with increased risk of autism spectrum disorder (ASD), as well as altered structural and functional outcomes in brain imaging studies, which validates research into the role of MET in brain development. ❧ To study the question whether Met and ELS interact to have enduring effects on neuronal morphology and behavior in mice, we used a limited bedding and nesting paradigm that previously has been shown to induce a stress response in mouse pups, and result in lasting effects on mouse brain structure and function. However, the mechanism by which this paradigm induces ELS in mice is not completely clear. Extended temporal behavioral analyses of nest entry behavior by dams and vocalizations by pups suggest that pups in ELS conditions experienced aversive and painful stimuli due to altered nest entry behavior by the dam. The nest entry behaviors by the dams due to limited resources were transient and not transmitted to the next generation of females. ❧ ELS was induced with this limited bedding paradigm in both wild-type pups and Met+/- pups, which express 50% of normal MET receptor tyrosine kinase protein in the central nervous system, and morphological characteristics were assessed in young adult brains. Analyses of pyramidal neurons that are part of a network regulating complex behaviors that are affected by ELS, showed that dendritic arbor complexity of neurons that project from the ventral CA1 region of the hippocampus to basolateral amygdala was reduced in wild-type mice that had experienced ELS, as well as in Met+/- mice raised in control conditions. However, ELS had the opposite effect in Met+/- mice as in wild-type mice, as arbor complexity in these mice was increased compared to non-stressed Met+/- mice, possibly due to the precocious maturational state of neurons as a result of a reduction in MET protein. The effect on dendritic morphology appears to be selective, as Met+/-, ELS, or the combination of these genetic and environmental factors did not alter spine characteristics in these specific neurons. ❧ Behavioral analyses of wild-type and Met+/- mice showed that Met genotype reduced anxiety-like behavior and impaired contextual fear memory, while fear acquisition and extinction were unaffected. Social behavior was impaired by ELS, specifically in Met+/- mice. These results suggest that the combination of these genetic and environmental factors results in an increased number of affected behavioral domains, and that Met+/- mice may be more sensitive to ELS. ❧ In summary, the outcomes of this research project show that gene-environment interactions between Met expression and ELS affect neuronal structure and behavior in mice. This body of work creates a foundation for future research on the mechanism of this interaction. Furthermore, the outcomes provide the first justification to study an interaction between MET rs1858830 genotype and early adverse experiences in the human population, and its effects on severity and risk of mental health disorders.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The impact of genetic pleiotropy on heterogeneity in the developing forebrain and hindbrain
PDF
Estrogen, progesterone and BDNF interactions: roles in neuroprotection
PDF
The role of inflammation in mediating effects of obesity on Alzheimer's disease
PDF
Uncovering cell type-specific roles of proteins involved in glutamatergic synapse regulation
PDF
Experience-dependent neuroplasticity of the dorsal striatum and prefrontal cortex in the MPTP-lesioned mouse model of Parkinson’s disease
PDF
Uncovering the protective role of protein glycosylation in Parkinson's disease utilizing protein semi-synthesis
Asset Metadata
Creator
Heun-Johnson, Hanke
(author)
Core Title
Gene-environment interactions in neurodevelopment
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Neuroscience
Publication Date
08/03/2016
Defense Date
03/16/2016
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
amygdala,anxiety,dendrite,early-life stress,fear memory,G x E,hippocampus,maternal behavior,MET receptor tyrosine kinase,morphology,OAI-PMH Harvest,social interaction,spine,ultrasonic vocalizations
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Wood, Ruth I. (
committee chair
), Levitt, Pat (
committee member
), McGee, Aaron W. (
committee member
), Stiles, Bangyan L. (
committee member
)
Creator Email
hankeheun@gmail.com,heunjohn@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-294826
Unique identifier
UC11279564
Identifier
etd-HeunJohnso-4730.pdf (filename),usctheses-c40-294826 (legacy record id)
Legacy Identifier
etd-HeunJohnso-4730.pdf
Dmrecord
294826
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Heun-Johnson, Hanke
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
amygdala
anxiety
dendrite
early-life stress
fear memory
G x E
hippocampus
maternal behavior
MET receptor tyrosine kinase
morphology
social interaction
ultrasonic vocalizations