ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Pharmacy 284

Adolescent behavior

Links to early-life stress and alcohol in male andfemale rats

STINA LUNDBERG

ISSN 1651-6192ISBN 978-91-513-0893-7urn:nbn:se:uu:diva-405313

Dissertation presented at Uppsala University to be publicly examined in A1:107a,Biomedicinskt centrum (BMC), Husargatan 3, Uppsala, Friday, 24 April 2020 at 09:15 for thedegree of Doctor of Philosophy (Faculty of Pharmacy). The examination will be conductedin English. Faculty examiner: Assistant professor Heidi MB Lesscher (Utrecht University,Utrecht, The Netherlands).

AbstractLundberg, S. 2020. Adolescent behavior. Links to early-life stress and alcohol in male andfemale rats. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty ofPharmacy 284. 97 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0893-7.

Adolescence is an important developmental phase with large changes in behavior, physiologyand neurobiology, which transform an individual from immature child to independent adult. Dueto these changes, adolescence is a sensitive period for exposure to environmental factors suchas stress and drug exposure; it is also a common age of onset for alcohol consumption as wellas several psychiatric disorders. Despite these factors, less is known about this developmentalperiod than regarding adult individuals. Behavior is regulated by the central nervous system andcan be used as a lens to study these processes as well as for examination of associations betweenindividual differences, early-life stress and alcohol. The aim of this thesis was to experimentallyexamine adolescent behavior and its links to early-life stress and alcohol in adolescent maleand female rats. Different behavioral tests were used to profile adolescent animals togetherwith animal models of early-life stress, voluntary alcohol consumption and alcohol exposure.In addition, stress responsiveness after early-life stress and the impact of alcohol exposureon endogenous opioid peptide levels as well as blood alcohol concentrations were examined.The adolescent behavioral profile in the multivariate concentric square field™ (MCSF) wascharacterized and validated against the elevated plus maze and open field tests. The main findingwas subgroups based on individual variation that revealed three distinct behavioral types:Explorers, Shelter seekers and Main type animals. This pattern was replicated in an additional,independent cohort. Early-life stress, modelled by prolonged maternal separation, showedsmall effects on behavior in the MCSF and on social play behavior. However, an effect onstress responsiveness in males but not females subjected to prolonged maternal separation wasdiscovered. Predisposition for high alcohol consumption did not have a shared behavioral profileamong selectively bred rat lines. However, a subgroup of high drinking individuals in an outbredcohort showed behavioral similarities to one of the selectively bred lines. Alcohol exposureshowed small, but sex-dependent, effects on behavior and endogenous opioid peptide levels.Together, these studies provide new information about adolescent behavior and associationsto early-life stress and alcohol in males and females, relationships not extensively studied inadolescence.

Keywords: Adolescence, alcohol exposure, behavior, corticosterone, dynorphin, early life,endogenous opioid peptides, Met-enkephalin-Arg6Phe7 (MEAP), individual differences,multivariate concentric square field (MCSF), phenotyping, radioimmunoassay, selectivebreeding for alcohol consumption, voluntary alcohol intake

Stina Lundberg, Department of Pharmaceutical Biosciences, Box 591, Uppsala University,SE-75124 Uppsala, Sweden.

© Stina Lundberg 2020

ISSN 1651-6192ISBN 978-91-513-0893-7urn:nbn:se:uu:diva-405313 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-405313)

In the name of sacred knowledge,in the eyes of every god in every corner of the world,

I swear my allegiance to the library of Alexandria.I swear to protect the knowledge of this world

against all enemies, within and without.I swear to nurture and share such knowledge

with all who wish to learn. I swear to live, teach, preserve, study, fight and die in this cause.

– Sword and Pen, Rachel Caine

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Lundberg, S., Högman, C., Roman, E. (2019) Adolescent explor-

atory strategies and behavioral types in the multivariate concen-tric square field™ test. Frontiers in Behavioral Neuroscience, 13:41

II Lundberg, S., Nylander, I., Roman, E. (2020) Behavioral profil-ing in early adolescence and early adulthood of male Wistar rats after short and prolonged maternal separation. In press Frontiers in Behavioral Neuroscience 14:37

III Lundberg, S., Martinsson, M., Nylander, I., Roman, E. (2017) Altered corticosterone levels and social play behavior after pro-longed maternal separation in adolescent male but not female Wistar rats. Hormones and Behavior, 87:137–444

IV Lundberg, S., Roman, E., Bell, RL. (2020) Behavioral profiles of adolescent alcohol-preferring/non-preferring (P/NP) and high/ low alcohol-drinking (HAD/LAD) rats are dependent on line but not sex. In manuscript.

V Lundberg, S., Orodjlo, P., Sjöström, M., Nylander, I., Roman, E. (2020) Adolescents and alcohol – voluntary consumption or pas-sive exposure and links to behavior, endogenous opioids and blood alcohol levels in male and female rats. In manuscript.

Reprints were made with permission from the respective publishers.

Contents

Introduction ............................................................................................... 11Adolescence ......................................................................................... 11Behavior ............................................................................................... 12

Individual variation in behavior ...................................................... 13Stress .................................................................................................... 14

Early-life stress ................................................................................ 16Alcohol ................................................................................................. 16

Alcohol use disorders ...................................................................... 17Alcohol consumption in adolescence .............................................. 18

Methodology ............................................................................................. 19The rat as a model organism ................................................................ 19

Development of the rat .................................................................... 20Selective breeding ............................................................................ 22

Behavioral assessments ........................................................................ 23The multivariate concentric square field™ test ................................ 24The elevated plus maze and open field ............................................ 26The social play behavior test ........................................................... 27

Maternal separation .............................................................................. 28Voluntary alcohol consumption ........................................................... 30Alcohol exposure .................................................................................. 31Quantification of endogenous steroids and peptides ............................ 32

Sampling considerations .................................................................. 32Radioimmunoassay .......................................................................... 33

Multivariate data analysis ..................................................................... 34

Aims .......................................................................................................... 36

Methods..................................................................................................... 37Animals and housing ............................................................................ 38Behavioral tests .................................................................................... 39

The multivariate concentric square field™ test ................................ 39The elevated plus maze and open field ............................................ 41Social play behavior test .................................................................. 41

Stress procedures .................................................................................. 43Maternal separation ......................................................................... 43HPA axis reactivity test ................................................................... 43

Alcohol procedures .............................................................................. 44Voluntary alcohol consumption ....................................................... 44Alcohol exposure ............................................................................. 44

Blood and tissue analyses ..................................................................... 45Serum corticosterone radioimmunoassay ........................................ 45Endogenous opioid peptide radioimmunoassay .............................. 45Blood alcohol concentration analysis .............................................. 47

Statistical analysis ................................................................................ 47Parametrical statistics ...................................................................... 47Non-parametrical statistics .............................................................. 47Multivariate data analysis ................................................................ 48

Results and Discussion ............................................................................. 49Characterization of adolescent behavior in the MCSF ......................... 49

Sex-differences in adolescent behavior in the MCSF ..................... 49The adolescent behavioral types ...................................................... 50Repeated testing and/or effect of age in the MCSF ......................... 54The MCSF and the reference tests ................................................... 56

Links to early-life stress ....................................................................... 58Early adolescent HPA axis reactivity .............................................. 58Early adolescent behavioral profiles ................................................ 59Adolescent social play behavior ...................................................... 59Early adult behavioral profiles ........................................................ 60

Links to alcohol .................................................................................... 61Behavioral profiles after selective breeding for high or low alcohol consumption ..................................................................................... 61Adolescent behavior and later alcohol consumption ....................... 63Effect of alcohol exposure on adolescent behavioral profiles ......... 67Effect of alcohol exposure on endogenous opioid peptides ............ 68Blood alcohol concentrations after alcohol exposure ...................... 69

General discussion ................................................................................ 70Links to early-life stress .................................................................. 70Links to alcohol ............................................................................... 71

Conclusions ............................................................................................... 73

Populärvetenskaplig sammanfattning ....................................................... 74

Acknowledgements ................................................................................... 77

References ................................................................................................. 80

Abbreviations

ACTH Adrenocorticotropic hormone AFR Animal facility rearing ANOVA Analysis of variance AUDs Alcohol use disorders BAC Blood alcohol concentration BE Bridge entrance BNST Bed nucleus of the stria terminalis CNS Central nervous system CRH Corticotropin-releasing hormone CTRCI Central circle D Duration DCR Dark corner room DSM Diagnostic and Statistical Manual of Mental Disorders EPM Elevated plus maze F Frequency HAD High alcohol-drinking rats (selectively bred lines) HD, ID, LD High-, intermediate- or low drinking individuals (division

of outbred rats based on individual variation) HPA Hypothalamus-pituitary-adrenal HR/LR High/low responders to novelty ICD International Classification of Diseases and Related Health

Problems i.g. Intragastric i.p. Intraperitoneal ir Immunoreactive L Latency LAD Low alcohol-drinking rats (selectively bred lines) MCSF Multivariate concentric square field MEAP Met-enkephalin-Arg6Phe7 MS Maternal separation MS15 Daily 15 min maternal separation, postnatal day 1–21 MS360 Daily 360 min maternal separation, postnatal day 1–21 MVDA Multivariate data analysis NP Alcohol non-preferring rats (selectively bred line) Occ Occurrence OF Open field

P Alcohol-preferring rats (selectively bred line) PCA Principal component analysis PLS Partial least squares projection to latent structures PLS-DA PLS discriminant analysis PND Postnatal day PVN Paraventricular nucleus RIA Radioimmunoassay SAP Stretched attend posture SHRP Stress hypo-responsive period SPB Social play behavior SUDs Substance use disorders T0 Time-point 0 min after isolation in the

HPA axis reactivity test T120 Time-point 120 min after isolation in the

HPA axis reactivity test TOTACT Total activity TOTCORR Total corridor VTA Ventral tegmental area WHO World Health Organization

11

Introduction

Adolescent development has, in the last decades, gained more attention as a research subject. Large consortia have been initiated in this cause; examining the development of brain function in relation to individual variation, environ-mental and behavioral factors as well as development of psychiatric disorders (Schumann et al., 2010; Volkow et al., 2018). These studies, in human ado-lescents, are complemented by preclinical studies. Animal experiments pro-vide the means of precisely controlling exposure to environmental factors and detailed analysis of neurobiological consequences; studies that cannot be con-ducted in humans for ethical reasons. Both clinical and preclinical studies are needed to shed light on adolescent development and how it is affected by dif-ferent individual and environmental factors and how it, in turn, affects later health outcomes.

The objective of this thesis was to use preclinical studies to understand ad-olescent development and how it is affected by early-life stress and alcohol exposure as well as examining predictive factors for alcohol consumption. The focus was on examining adolescent behavior, including validation of a novel approach for behavioral profiling. Three different perspectives were applied: characterization of basal adolescent behavioral profiles and investigation of outcomes in adolescence after early-life stress and investigation of relation-ships to alcohol exposure and consumption.

Adolescence Adolescence is the developmental period between child- and adulthood, it is considered to encompass the second decade of life but has no clear boundaries. The World Health Organization (WHO, 2014) places adolescence from 10 to 19 years of age; however, recent arguments to extend the upper age limit to 24 years have been made on the basis of both physiological and social reasons (Sawyer et al., 2018). Generally, adolescence span from the onset of puberty, which normally falls between 8–13 years in girls and 9–14 years in boys (Khan, 2019) and continues until the individual is fully mature and independ-ent (Spear, 2000), which, when including brain maturation, stretches into the early twenties (Lebel and Deoni, 2018).

The development during adolescence involves large changes in behavior, physiology and neurobiology (van Duijvenvoorde et al., 2016; Lo Iacono and

12

Carola, 2018; Spear, 2000; Sturman and Moghaddam, 2011). Due to these extensive changes, adolescence is a vulnerable period for external influences such as stress and drug intake (Gray and Squeglia, 2018; Spear, 2018; Totten-ham and Galván, 2016). In addition, many psychiatric disorders, including mood and eating disorders, schizophrenia and substance use disorders (SUDs), first manifest during adolescence (Crews et al., 2016; Sturman and Moghaddam, 2011).

Adolescents across mammalian species show increased social behaviors, increased risk taking and more sensation- and novelty seeking than both younger and older individuals (Doremus-Fitzwater et al., 2010; van Duijvenvoorde et al., 2016; Sturman and Moghaddam, 2011). This is thought to facilitate the transition from immature juvenile to independent adult and aid in the search for new territory, food sources and mates (Yurgelun-Todd, 2007).

The brain alterations during adolescence are extensive; almost all neuro-transmitter systems as well as the white and gray matter are still developing (Casey et al., 2008; Crews et al., 2016; Spear, 2013; Sturman and Moghaddam, 2011). White matter volume increases steadily throughout and beyond adolescence signifying improved communication between different parts of the brain (Lebel and Deoni, 2018; Sturman and Moghaddam, 2011). The volume of cortical gray matter changes in a different pattern, forming an inverted U-shaped curve. During adolescence, the receptor and synaptic den-sity first increase to near a developmental maximum whereafter so called pruning takes place that reduces the levels to that of adults (Buwalda et al., 2011; Sturman and Moghaddam, 2011). Different parts of the brain mature in different rates, accounting for the cognitive and behavioral inconsistencies during adolescence. Imbalance in the risk-reward system make adolescents more impulsive and reckless in the search for sensation, reward and novelty, and difficulties to integrate emotional information can lead to poor decision making and an elevated emotional state (Crews et al., 2016; Doremus-Fitzwater et al., 2010; Ernst, 2014; Sturman and Moghaddam, 2011).

Behavior In 1973, Karl von Frisch, Konrad Lorenz and Nikolaas Tinbergen were awarded the Nobel Prize for their studies on patterns of individual and social behaviors (NobelPrize.org). They are considered to be the founders of the sci-entific field of ethology, the study of animal behavior, by their studies on in-sects, birds and fish but their theories have shown translational value across species and time; e.g. Tinbergen’s four questions are still used and commented upon until today (Bateson and Laland, 2013; Tinbergen, 1963).

The broadest definition of behavior is any activity of an animal’s effector organs, which includes both muscles and glands, and thus refers to everything

13

from running to salivation (Barnett, 2007). However, more generally, and of pertinence in this thesis, behavior is referring to the coordinated activity of skeletal muscles resulting in movement of the whole animal. A behavior is elicited due to a stimulus, internal or external, and interpreted by the central nervous system (CNS) that through motor neurons regulate muscle activity. Behavior, as an output of the CNS, can thus be used in neuroscience as a lens to study the events in the CNS that produce behavior (Thompson, 2001). But behavior is of value in several scientific disciplines, from psychology to vet-erinary medicine.

Behavior can roughly be divided into innate and learned behaviors. Innate behavior is often species-specific, stable over time and between individuals and, from first performance, the full behavioral complex is present and func-tional (Zha and Xu, 2015). Innate behavior often has a direct survival value for the individual or species, e.g. feeding, aggression and parental care, and is sometimes referred to as instinct or instinctive behaviors (Zha and Xu, 2015). Learned behaviors develop over time in the individual under the influence of the environment. As such, it is adaptive since it provides a way of changing the behavior of the individual depending on its previous experiences. One of the most basal types of learning is called classical or Pavlovian conditioning after its discoverer. It is the linking of two stimuli by their proximity in time, exemplified by Pavlov’s dogs where feeding was paired with the sound of a bell; eventually the sound of the bell alone elicited the same response as the feeding, in this case increased salivation (Kim, 2001; Pavlov, 2010). More complex learning often relies on operant conditioning, where a response (some type of behavior) of the individual is linked to a reinforcer or punish-ment which modulate the occurrence of the response (Fantino and Stolarz-Fantino, 2012).

The study of behavior has classically been divided in two schools; the “Eu-ropean” ethologists, studying innate and natural behaviors in the field or semi-natural settings, and the “North American” experimental behavioral psycholo-gists, working within laboratory settings and developed the operant or instru-mentalized behavioral approach, i.e. the operant, also called Skinner, box. The division is no longer as evident, and pertinent to this thesis is the ethoexperi-mental approach which combines these two traditions by striving to use bio-logically relevant laboratory tests and detailed attention to the animal behavior with the goal to uncover the relationship between behavior and underlying mechanisms (Blanchard and Blanchard, 1988; Pearson et al., 2017).

Individual variation in behavior Personality, defined as individual variation in behavior that is stable over time and between contexts, can be demonstrated in species across the breadth of animal taxa (Stamps and Groothuis, 2010a, 2010b). The module of such stable behavior is the behavioral trait, in contrast to behavioral state, which is the

14

behavior actually expressed in a certain context. In other words, an underlying trait modulates the probability of a certain state to be expressed. A common and well-established tool for describing human personality is the five-factor, or Big 5, model, which contains five broad domains: neuroticism, extraver-sion, openness, agreeableness and conscientiousness (Widiger et al., 2018). However, a plethora of assessments for evaluation the human personality and/or specific traits exists. They are based upon structured interviews or self-assessment questionnaires; to assess non-human animal personality, other ap-proaches are required (see Behavioral assessment below).

Behavioral traits or entire personality profiles can be associated with in-creased risk to develop certain disorders, e.g. high sensation seeking entails a vulnerability to develop SUDs (Belin et al., 2016; Blanchard et al., 2009; Zuckerman, 1996). There is therefore a value in identifying corresponding traits in animal models for experimental studies of additional biomarkers and disorder development. Several classifications of behavioral divisions have been developed and used in animal experiments. Individual variation in the response or reactivity to novelty have been used to divide rodents into high and low responders (the HR/LR model) (Piazza et al., 1989), where HR show similarities with high sensation seeking individuals (Belin et al., 2016; Blanchard et al., 2009). This trait dichotomy correlates to other behaviors and has been used as selection criterion for establishing selectively bred lines for HR and LR, respectively (Stead et al., 2006).

Another division of individual variation encompassing both behavioral and physiological measures is the concept of coping styles (de Boer et al., 2017; Coppens et al., 2010; Koolhaas et al., 2007), dividing rodents into proactive or reactive on the basis of their behavioral and physiological response to stress. The coping style spectrum has also been combined with emotionality to create a two-dimensional behavioral model (Koolhaas et al., 2007).

Stress Stress is the physiological response to a stressor, a stimulus interpreted as a real or perceived threat to the homeostasis of the individual (Selye, 1950). Everything from predator threat, tissue damage and starvation to social con-flict and even everyday activities like exercise, acts as stressors and elicit a stress response. This is the so-called flight-or-fight response, which in the short-term increases heart rate and blood pressure, focus senses and attention, and liberates stored energy reserves – anything that prepares the individual for dealing with the challenge at hand (Godoy et al., 2018; Goel et al., 2014).

The physiological stress response is mediated by two separate but inter-locked systems: the hypothalamus-pituitary-adrenal (HPA) axis (illustrated in Figure 1) and the sympathetic nervous system, with glucocorticoids and cate-

15

cholamines as their respective mediators. The HPA axis starts in the hypothal-amus, where neurons in the paraventricular nucleus (PVN) release corticotro-pin-releasing hormone (CRH) in a pulsatile, circadian cycle, which peaks dur-ing the first hours of the active phase (i.e. in the morning for diurnal, and in the evening for nocturnal animals). CRH is transported via the hypophyseal portal blood system to the anterior pituitary where it stimulates the release of adrenocorticotropic hormone (ACTH) into the circulation. ACTH acts on the adrenal cortex, mediating release of glucocorticoids (i.e. cortisol in humans and corticosterone in rodents) that exert a variety of functions on tissues throughout the body (Herman, 2009; Lightman, 2008; Tsigos and Chrousos, 2002).

Figure 1. The HPA axis in the rat.

The activity of the HPA axis is regulated by input from the limbic system and the brain stem to the PVN and is thereby responsive to a wide range of stress-ors, from emotional to physiological. Acute stress increases the secretion of CRH, which potentiates the glucocorticoid release (Lightman, 2008). Stress-ors also activate the sympathetic nervous system; noradrenergic projections from the brain stem stimulate release of noradrenaline and adrenaline from the adrenal medulla. At the central level, the HPA axis and the sympathetic nerv-ous system reciprocally activate each other (Tsigos and Chrousos, 2002), while the glucocorticoid effector of the HPA axis exerts negative feedback on CRH and ACTH secretion as well as on noradrenergic activity in the brain stem (Andrews et al., 2012; Tsigos and Chrousos, 2002). Additionally, CRH

16

and central catecholamines have central effects, coordinating the cognitive and behavioral stress response (Heinrichs and Koob, 2004).

Acute stress has an adaptive function by promoting survival in the face of acute challenges, the aforementioned adaptations in cardiovascular, percep-tive and metabolic functions works well for this purpose. However, if the stressor persists, is beyond the individual’s control or not enough recovery is allowed between stressful periods, the stress becomes chronic. Chronic stress is considered maladaptive, and is associated with an array of health effects, ranging from metabolic to psychiatric disorders. Disturbances in the HPA axis are also involved in many other disorders and the causal relationship between disorder and HPA axis dysregulation is often unclear. Up-regulation of HPA axis function has been associated with melancholic depression, anorexia ner-vosa, obsessive-compulsive disorder, diabetes mellitus, cancer and central obesity. Decreased HPA axis function on the other hand, has been connected to atypical depression, arthritis, posttraumatic stress disorder, chronic fatigue and chronic stress itself (Goel et al., 2014; de Kloet et al., 2005; Miller et al., 2007; Radley et al., 2015; Raison and Miller, 2003; Tsigos and Chrousos, 2002).

Early-life stress Early-life stress is a special case of chronic stress that, due to its developmen-tal timing, can have an imprinting function on the developing individual. Stud-ies in humans have linked adverse early experiences (e.g. physical, emotional or sexual abuse, neglect, parental loss or institution rearing) to increased risk for cardiovascular, metabolic and, in particular, psychiatric disorders (Bick and Nelson, 2016; Ehlert, 2013; Haller et al., 2014; Heim et al., 2010; Maniam et al., 2014; Nemeroff, 2016). The developing brain (Bick and Nelson, 2016; Nemeroff, 2016; Teicher et al., 2016) and HPA axis (Ehlert, 2013; Maniam et al., 2014) are established targets for negative effects of early-life stress but the mechanism linking early-life stress to these effects and later health outcomes are not completely understood. Neither are so-called sleeper effects, where the effects of early-life stress are not manifested until much later in development or even in adulthood (Bick and Nelson, 2016; Daskalakis et al., 2013). How the effects of early-life stress manifest during adolescence is one of the themes explored in this thesis.

Alcohol Alcohol (i.e. ethanol) is used globally and alcohol drinking is, in most cultures, socially accepted and often expected in certain settings (WHO, 2018). Alcohol is consumed for its euphoric and disinhibiting or sedative properties through its actions on the CNS, but alcohol has effects on most tissues in the body. In

17

the brain, alcohol has widespread effects on a variety of neurotransmitter and neuropeptide systems, both through inhibiting and potentiating mechanisms (Spanagel, 2009; Tabakoff and Hoffman, 2013; Vengeliene et al., 2008). Per-tinent to the reinforcing and addictive properties of alcohol is the converging effect on the brain’s reward system, which key feature is increased signaling in the mesolimbic dopamine system (Heilig et al., 2010; Söderpalm and Eric-son, 2011). Modulating the dopamine signaling and implicated in develop-ment of excessive alcohol intake and development of alcohol use disorders (AUDs) is the endogenous opioid system (Oswald and Wand, 2004; Palm and Nylander, 2016; Trigo et al., 2010).

Alcohol use disorders Most people consume alcohol in a controlled manner. However, some indi-viduals progress from controlled consumption, escalate their alcohol intake and develop AUDs. Exactly what factors contribute to the development of AUDs and why some individuals seem to be more sensitive than others are not fully known. The genetic risk component is estimated around 50% (Tawa et al., 2016; Verhulst et al., 2015), clearly emphasizing the importance of gene-environment interactions. To this end, a wide range of risk factors have been implicated, such as family history of AUDs, early alcohol debut, certain personality traits, psychiatric comorbidity and early-life adversity (Enoch, 2006; Oreland et al., 2017; Salvatore et al., 2015; Spanagel, 2009).

Two parallel systems are used to diagnose AUDs; the International Classi-fication of Diseases and Related Health Problems (ICD), 11th revision (WHO, 2019) and the Diagnostic and Statistical Manual of Mental Disorders (DSM), 5th edition (American Psychiatric Association, 2013). In ICD-11, disorders due to use of alcohol is the umbrella term used for any disorder relating to the use of alcohol, from episodes and patterns of harmful use of alcohol, alcohol intoxication and withdrawal, and lastly alcohol dependence, the addictive dis-order. In DSM-5, AUD is the sole diagnosis, with an included severity scale (mild, moderate, severe), approximately corresponding to ICD-11’s pattern of harmful alcohol use and alcohol dependence (Saunders et al., 2019). The cri-teria for AUD (DSM-5) and harmful alcohol use as well as alcohol depend-ence (ICD-11) are worded and divided differently but essentially describe the same symptoms: excessive alcohol consumption with loss of control over in-take; alcohol having a high priority despite negative effects on other aspects in life; and physiological symptoms such as craving, tolerance and withdrawal (Saunders et al., 2019). In this thesis, the term AUDs is used to generally de-scribe this combined symptomatology rather than either specific one.

18

Alcohol consumption in adolescence In Sweden, alcohol and other drug consumption among adolescents have been monitored in teenagers at age 15 (9th grade of elementary school) since 1971 and among teenagers 17 years old (2nd grade of secondary school) since 2004. For alcohol, the trend has been decreasing for both proportion of consumers and in estimated consumption since the beginning of the millennium, although the decline has halted in the last years (CAN, 2019). The same trend has been observed in other European and English-speaking countries (ESPAD Group, 2016; Pape et al., 2018; Vashishtha et al., 2019). Still, among Swedish ado-lescents aged 15 and 17, 42% and 69% respectively, have consumed alcohol during the last 12 months. The proportion of risk consumers, which in the survey includes both binge drinkers (consuming alcohol corresponding to at least one bottle of wine at one occasion) and high consumers (more than 14 or 9 standard drinks1/week for boys and girls, respectively), was 8% at age 15 and 21% at age 17. The proportion of girls among alcohol consumers was slightly higher in both age groups, while among risk consumers the proportion of boys was higher at 17 years of age whereas being the same among the gen-ders at age 15 (CAN, 2019).

Risk consumption of alcohol during adolescence and early debut is associ-ated with increased risk for development of AUDs later in life. In addition, it can also have long-lasting effects on the developing brain and lead to cognitive and behavioral deficits (Crews et al., 2016, 2019; Spear, 2018). In parallel to the normal development during adolescence, alcohol seems to have timing-dependent effects giving rise to different outcomes depending on when during adolescence it is consumed (Spear, 2015). There is however much that it is not known; for instance, the contribution of individual differences as predictive factors for high alcohol consumption, a relationship which is further explored in this thesis.

1 One standard drink contains 12 g of pure alcohol, corresponding to one bottle of beer (5%, 33 cl), one glass of wine (11–13%, 15 cl) or 4 cl spirits (40%).

19

Methodology

The rat as a model organism Model organisms are used in research to model something more or other than what the selected species is. It can be a model for animals or living organisms at large when studying basal biological processes or it can be a model for hu-mans studying our physiology or pathology. A research model can also refer to the specific experimental setup, such as genetic variant, environmental con-ditions etc., used together with a model organism to complete the model. When selecting a model system in relation to a research question it is im-portant to consider the validity of the research system. Validity entails the question of if and how an organism can be a model for something else; can a certain research question be answered with the use of a certain model system? When it comes to modeling human disorders, validity is often discussed in three perspectives: face, predictive and construct validity. Face validity is when the symptoms resemble the human condition; predictive validity is when treatments have effect in both humans and the model; and construct validity is when the underlying cause is homologous between humans and the model (Belzung and Lemoine, 2011; Spanagel, 2017; van der Staay, 2006; Stewart and Kalueff, 2015).

In biomedical research, the laboratory rat and mouse are the most common mammals used. They are domesticated variants of the Norwegian rat (Rattus norvegicus) and house mouse (Mus musculus) respectively and have been used in research for over a century. The mouse is nowadays dominating in numbers due to the large availability of genetical tools specific for this species. The rat, on the other hand, has been the classical choice for toxicological and physiological studies, and in neuroscience. However, complementary models in non-mammalian species such as zebrafish (Danio rerio) (Bambino and Chu, 2017; Demin et al., 2020; Fontana et al., 2018) have been on the rise of late. Rodents are common due to their ease of use; the small size and high reproductive rates make it possible to keep and breed many animals (Modlinska and Pisula, 2020). Moreover, rodents are relevant as models for human disorders as we share large genetic similarities (Gibbs et al., 2004; Modlinska and Pisula, 2020). In this thesis, outbred rats of the Wistar strain were used. Outbred rat strains are heterogenous, as opposed to inbred strains where each individual is genetically identical, which is achieved through sib-ling mating for more than twenty generations. In Paper I–III and V, outbred

20

Wistar rats were used for examining the research questions in a diverse popu-lation. In Paper IV, outbred Wistar rats were used as a standard for compari-sons with the selectively bred lines, which were the focus in the study.

In experimental, as well as in clinical, research male subjects are dominat-ing (Becker et al., 2016; Beery and Zucker, 2011; Palanza and Parmigiani, 2017). It has long been argued that females, due to their hormonal cycle, have higher intrinsic variability than males and that a higher number of subjects would have to be included to compensate for this. However, this does not hold true, and meta-analyses have actually shown that females are even less varia-ble than males in a number of outcomes (Becker and Koob, 2016; Becker et al., 2016, 2017). Additionally, it has often been assumed that findings would be the same in females and males but more and more areas are finding sex-dependent differences, and they are not always sex hormone-dependent (Abel and Rissman, 2012; Cahill, 2006; Carroll and Lynch, 2016; Cooke and Weath-ington, 2014; Sanchis-Segura and Becker, 2016; Simpson and Kelly, 2012). Furthermore, studies are often conducted on adult subjects leading to a greater shortage of studies concerning sex-dependent differences in adolescence. Spe-cific to this thesis are sex-dependent effects after early-life stress and relations to alcohol.

Development of the rat An overview of the development of the laboratory rat can be seen in Figure 2. The rat is an altricial animal; pups are born virtually blind and deaf with little ambulatory capacity and are thus entirely dependent on the care and nurture provided by the dam (Alberts, 2005). Pups do not only depend on the dam for their immediate survival but also for their development; it has been shown that maternal behaviors impact the pups’ physiological, neurobiological, social and behavioral development (Lai and Huang, 2011; Levine, 2002; Lucion and Bortolini, 2014). During most of the first two postnatal weeks, rat pups show a diminished ability to respond with increased corticosterone secretion to acute stressors, this is called the stress hypo-responsive period (SHRP) (Lai and Huang, 2011; Levine, 2002). The SHRP is proposed to provide protection during a developmental phase when the pup is sensitive to the effects of glu-cocorticoids.

Figure 2. Outline of the development of the laboratory rat from birth to adulthood. NB, newborn.

21

Around postnatal day (PND) 12–15, rat pups open their eyes and ambula-tion increase (Alberts, 2005). Before this point, pups show mostly dam-di-rected behaviors but as their sensory and mobility capacities increase there is an initiation of peer-directed behaviors (Auger and Olesen, 2009; Brown, 2005; Trezza et al., 2014). In rats, as in most mammalian species, juveniles and adolescents display a characteristic type of peer-directed behavior: social play behavior (SPB) (Panksepp et al., 1984). Playing is an essential part of normal social, cognitive and behavioral development (Panksepp et al., 1984; Trezza et al., 2010). Play deprivation, through social isolation during adoles-cence, has lasting social and behavioral effects, which are most likely based on neurobiological changes (Burke and Miczek, 2014; Buwalda et al., 2011; Trezza et al., 2014). Because of its developmental importance, social play is highly rewarding in adolescent animals, demonstrated in several studies using the same type of tests as with other reinforces such as food, sex or drugs of abuse (Trezza et al., 2010, 2014). In rats, the typical SPB is play fighting where the involved animals compete for access to the nape of each other. The frequency of SPB increases throughout the juvenile period and beginning of adolescence, it peaks around puberty and then fades to the low levels still pre-sent in adults (Auger and Olesen, 2009; Burke and Miczek, 2014).

Adolescence and puberty are two coinciding but different developmental processes occurring between the juvenile and adult stages of development. Puberty specifically encompasses the process of sexual maturation and can be clearly defined by hormonal and physiological events. The age at onset of pu-berty and the length of puberty is different for female and male rats. Female rats enter puberty around PND 30 and first ovulation occurs at approximately PND 40, thereafter estrus cycling is considered established. Male puberty is slower, with the first external signs visible around PND 40 but fertility is not reached until PND 60 when the first fully mature spermatozoa can be detected (Burke and Miczek, 2014; Schneider, 2008, 2013). Adolescence, on the other hand, refers to the behavioral and physiological maturation transforming a ju-venile individual to a mature adult (Doremus-Fitzwater et al., 2010; Schneider, 2008, 2013). The precise time span for adolescence in rodents is a debated question (Schneider, 2008, 2013; Sturman and Moghaddam, 2011). For rats, a conservative view places adolescence directly in conjunction with puberty (approximately PND 30–42) but a broader view encompasses the en-tire post-weaning period from PND 21 until 60–70 (Spear, 2015; Tirelli et al., 2003). The wider span is often subdivided into three phases; early adolescence (PND 21–30/34), mid-adolescence (PND 30/34–46) and late adolescence (PND 46–60/70) (Spear, 2015; Tirelli et al., 2003). Adulthood can be consid-ered reached after the animals have gone through puberty and the other be-havioral and physiological changes associated with adolescence, at approxi-mately PND 60–70.

22

Selective breeding Selective breeding is used to drive phenotypic diversity among domestic ani-mals and is the method behind everything from dog breeds to increased milk yields in dairy cows. In biomedical research, it is a valuable tool for studying inheritance and genetic contribution of phenotypes (Andersson, 2012). The relationship can both be studied retrospectively, as when using a dog breed with an, accidental, vulnerability to a certain disorder to examine the genetic cause (Switonski, 2014), and prospectively, selecting for the phenotype of in-terest in a laboratory setting.

Selective breeding has been used to create various lines of rats that can be used as models of human disorders as it creates an underlying predisposition for the phenotype selected for. It can as such be used to create behavioral state-models, such as the selective breeding on the HR/LR phenotype (Stead et al., 2006), which allows for studies of how the HR or LR trait develops in addition to its underlying genetics. Relevant for this thesis is bidirectional selective breeding for high or low voluntary alcohol intake and preference, which cre-ates pairs of rat lines with high or low underlying propensity for alcohol con-sumption. These lines can be used to study behavioral (Roman et al., 2012), neurobiological (Bell et al., 2012) and genetic (Bell et al., 2017; McBride et al., 2012) characteristics associated with each selection criteria. The high-al-cohol consuming lines can be used as models for AUDs and for evaluation of pharmaceutical candidates for treatment of AUDs (Bell et al., 2012, 2016, 2017).

Seven long-term, selective breeding programs have generated pairs of high/low alcohol-consuming lines around the world: the University of Chile bibulous/abstainer (UChB/UChA) rats (Quintanilla et al., 2006), the Finnish ALKO alcohol/non-alcohol (AA/ANA) rats (Sommer et al., 2006), the Sar-dinian alcohol-preferring/non-preferring (sP/sNP) rats (Colombo et al., 2006), the Warsaw high-/low-preferring (WHP/WLP) rats (Dyr and Kostowski, 2008), and the Indiana alcohol-preferring/non-preferring (P/NP) and high/low alcohol-drinking (HAD/LAD, replicate 1 and 2) rat lines (McBride et al., 2014). In this thesis, the three line pairs from Indiana (P/NP and HAD/LAD replicate 1 and 2) were used in Paper IV for assessment of adolescent behavior before contact with alcohol to examine behavioral correlates to each selection criteria.

The bidirectional selective breeding of the P/NP and HAD/LAD replicate rat lines is based on alcohol intake and preference in an alcohol drinking par-adigm with continuous access in the home cage with free choice between al-cohol (10% v/v) and water (Froehlich, 2010). The same selection criteria are used for the different lines; the high consumption criterion is defined as >5 g/kg alcohol consumed per day and a 2:1 preference for alcohol over water and the low consumption criterion is a daily alcohol intake <1.5 g/kg and low alcohol preference (0.2:1) (Froehlich, 2010). The separately named pairs of

23

lines are however derived using different foundation stocks; the P/NP rats were generated from outbred Wistar rats (Bell et al., 2006; Murphy et al., 2002), while the HAD/LAD replicate lines were generated from the N/NIH rats (Murphy et al., 2002), a heterogeneous foundation stock generated by crossing eight different inbred rat strains (Hansen and Spuhler, 1984; Li and Lumeng, 1984).

Behavioral assessments Assessment of personality in non-human animals rely on the assessment of behavior. But behavioral assessment, or behavioral tests, of laboratory ani-mals are used in biomedical research for many purposes, from evaluating how a certain genetic variant impacts the behavioral phenotype to evaluating phar-maceutical candidates (Hånell and Marklund, 2014). Broadly, behavioral tests can be divided into unconditioned and conditioned tests. Unconditioned tests rely on the expression of innate behaviors that are spontaneously elicited. Con-ditioned tests instead rely on learnt behavioral responses to specific stimuli and can be further divided into classical and instrumental conditioning with separate types of tests. In this thesis, only unconditioned behavioral tests are used and further discussion thus regards this type only, even though some ar-guments are valid for both unconditioned and conditioned behavioral tests.

The two general types of behavioral tests used in this thesis are exploration- and interaction-based tests. Exploration-based behavioral tests can be either forced, that is the animal is directly placed into a test arena, or voluntary, where a test arena is connected to the home cage and the animal can in its own time choose to enter and explore the novel arena. Forced exploratory tests are by far the most common of these types. Novelty, whether forced or voluntary, is considered to be a stressor (Koolhaas et al., 2011). As such, any behavior observed in the tests is thus stress-induced (when not using repeated testing or habituation). The exploratory drive governing exploration is modulated by the balances between approach/avoidance and neophilia/neophobia (Barnett, 2005). The level of locomotion or reactivity to novelty have been extensively studied and have been found to have several other behavioral, physiological and neurobiological correlates and the HR phenotype is often cited as a model for human sensation seeking (Blanchard et al., 2009). Interaction-based be-havioral tests measure the interaction between animals and depending on the specific design can quantify social behaviors from attack/defense, pup-dam interactions and social investigation and preference.

When using a behavioral test there are two important questions to ask be-fore starting: is the desired behavior measured in the test and is the behavior measured in a useful way? From an objective standpoint, it is apparent that the answer to both questions should be yes, but for some behavioral tests the pop-ularity seems to stem from convenience rather than theoretical reasons (Fonio

24

et al., 2012; Hånell and Marklund, 2014). Many commonly used tests only measure one behavior or type of behavior per test. If a wider behavioral char-acterization is desired, a battery of different tests is often used (Blanchard et al., 1998; Budde et al., 2013; Trent and Menard, 2013). A problem with this is that there might be carry-over effects between the tests so that the experi-ence gained in previous tests influence the performance in subsequent ones (McIlwain et al., 2001; Paylor et al., 2006). This problem can be avoided with the use of more diverse behavioral tests that give the animal a choice between different types of environments and thereby the ability to express different types of behaviors. One such test is the multivariate concentric square field™ (MCSF) test.

The multivariate concentric square field™ test The MCSF is a behavioral test based on forced exploration that rely on etho-experimental principles. The design is ethologically founded with a diverse, multivariate arena that emulates the diversity of the natural environment. In the MCSF arena, the animal has a free choice to explore zones of different qualities and thereby the chance to express a range of behaviors; the MCSF is thus considered to be unbiased with regard to mental condition (Meyerson et al., 2006). This allows for a more comprehensive characterization of the ani-mals’ native behaviors, referred to as a behavioral profile (Bikovski et al., 2020).

Figure 3. Schematic layout of the MCSF arena, the zones are: 1) center, 2a–c) corri-dors, 3) the dark corner room (DCR), 4) hurdle, 5) slope, 6) bridge entrance, 7) bridge and 8) central circle (CTRCI).

The MCSF (Figure 3) is a 100×100 cm square field arena enclosed by opaque walls on three sides and a transparent wall on the fourth. Internal walls sepa-rate the center (70×70 cm) from the peripheral corridors (15 cm wide). In the

25

center, a circular zone called the central circle (CTRCI, 25 cm in diameter) is included during analysis. Three corridors are accessible from the center through openings in the internal walls. Closed off in the end of one corridor is a covered shelter, called the dark corner room (DCR) and in the next corner is the hurdle, an elevated platform (10 cm above the floor) that is accessible from both neighboring corridors. In the hurdle is a hole-board with two holes mounted with a photocell to count the number of nose pokes into the holes. In the corridor along the side with the transparent outer wall is a stainless steel, wire-mesh construction (10 mm between bars). This section is divided into the slope, the inclined beginning of the construction connected to one of the cor-ridors; the bridge entrance (BE), first body length of the horizontal portion; and the bridge. The arena is illuminated from above and by a bridge lamp outside of the transparent wall to the following approximate levels (lux): cen-ter and corridors <25, DCR <1 and bridge ~500.

The MCSF has previously been validated in adult rats with focus on areas associated with risk and shelter (Meyerson et al., 2006). Open, elevated and illuminated areas are considered risk environments for rats, as these are areas where the animal would be more vulnerable to predation (Burn, 2008). In the MCSF, this is represented by the central circle and bridge. Enclosed and dark places are natural shelters for the rat, in the MCSF this is provided by the DCR. The quality of these zones has been confirmed in adult rats by demon-strating that food-restricted males hoard food pellets and lactating dams re-cover her pups from the bridge to the DCR, but food or pups placed in the DCR are not moved by the animal (Meyerson et al., 2006). Additionally, the center and corridors are considered as transit zones, associated with general- and exploratory activity. The elevated hurdle is considered as an exploratory incentive requiring additional effort to reach. The slope and bridge entrance, together with the so-called stretched attend posture (SAP), are associated with assessment of risk, whether the animal judges it to be safe to move from one zone to the next.

In adult rats, correlative analysis have identified associated parameters rep-resenting five different functional categories (i.e. risk taking, shelter seeking, general activity, exploratory activity and risk assessment) that are used in the interpretation of the result, together with a rank-order procedure called the trend analysis (Meyerson et al., 2013). This analysis creates compound varia-bles for the different behavioral categories by taking the correlated parameters for each category and ranking the individual animals’ values against each other. Thus, the animal with the highest score is given the highest rank and vice versa, for each of the correlated parameters. The rankings are then summed for each individual within the behavioral categories creating a rank sum, describing that animal’s relative performance in that behavioral category (Meyerson et al., 2013).

The MCSF has previously been used to study adolescent animals (Berardo et al., 2016; Palm et al., 2013; Wille-Bille et al., 2017, 2018), but no structured

26

comparison to other behavioral tests or other behavioral validation has been made previously to the work in Paper I in this thesis. Furthermore, the MCSF is used in this thesis to evaluate the behavioral profiles after early-life stress (Paper II) and for associations between the behavioral profiles and alcohol (Paper IV and V).

The elevated plus maze and open field The elevated plus maze (EPM) and open field (OF) tests are classical behav-ioral tests based on forced exploration of novel environments. These tests, to-gether with additional ones such as the light/dark box, generally provide two choices for the subject; usually to enter or not to enter a novel and/or risky environment. These tests can therefore be grouped together as univariate be-havioral tests, since they measure behavior on one scale. Although, the per-formance often needs to be interpreted with the underlying locomotor activity in mind.

The EPM is based on the approach/avoidance balance between exploring the open and elevated parts of the arena or not. It is derived from a study by Montgomery (1955), describing the balance between exploratory and fear-re-lated drives when rats were exposed to a novel stimulus in the form of open or enclosed maze alleys. Today, the arena is usually, as its name suggests, plus-sign shaped with two opposing arms surrounded by walls (‘closed’ arms) and two ‘open’ arms without walls. A variant with a circular runway (elevated zero maze) has also been designed, eliminating the ambiguous central zone were the arms meet in the classical cross shape (Braun et al., 2011; Shepherd et al., 1994). The EPM was popularized as a test for measuring anxiety in the 1980’s after work by Pellow and co-workers (1985). Stressful procedures or anxiogenic drugs were shown to decrease the time spent on the open arms while anxiolytic drugs increased the time spent on the open arms; rats also had an innate preference for the closed over the open arms (Pellow et al., 1985), as was also shown by Montgomery (1955). The question of anthropomorphiz-ing rat behavior has influenced the interpretation of the behavior in the EPM, from measuring anxiety, the designation has changed to anxiety-like behavior and now the denomination of ‘disorder-like behavior’ is also being questioned (Stanford, 2017). In the present thesis, the EPM is used as a reference test to the MCSF in Paper I with focus on the approach/avoidance balance and how the animal chooses to distribute its time between the sheltered and open (i.e. risky) parts of the arena.

The OF was first devised as a test to assess emotionality by measuring the defecation and urination evoked by exposure to a novel environment, in this case a large open arena (Hall, 1934). Currently, the OF arena vary considera-bly in size and shape between studies, and testing can be performed under differing illumination levels. The OF can be used to measure general locomo-tion and center versus thigmotactic (‘wall-hugging’) activity; center activity

27

can be interpreted as risk taking while thigmotaxis is risk-avoiding. Increasing illumination reduces the central activity (Bikovski et al., 2020; Momeni et al., 2014) while testing under red- or no-light conditions focus on locomotion (Bikovski et al., 2020) and is sometimes referred to as a locobox. The OF is, similarly to the EPM, subject to argumentation of what is actually measured in the test (exemplified in the still relevant commentary exchange in the March issue of Journal of Psychopharmacology in 2007 (Blizard et al., 2007; Overstreet, 2007; Rodgers, 2007; Stanford, 2007b, 2007a)). In the present the-sis, the OF is used as a reference test to the MCSF in Paper I with focus on locomotion and the distribution of central (i.e. risky) versus thigmotactic ac-tivity. In addition to the classical OF, the test is also used together with a so-called start box attached to the outside of the arena, which allows for assess-ment of emerge latency from the shelter (i.e. the start box) into the open area and shelter-seeking behavior in general.

As stated above (under Behavioral assessment), testing in these types of tests, which entail forced exposure to novelty, is a stressor for the tested ani-mal and it is important to remember that the behavior is thus stress-induced. Testing time have an impact on this; in the EPM, short testing times are often used (5 min typically) which focus on the initial response to the novel arena and the observed behavior is thus to a higher degree stress-induced. Testing times in the OF are more varied from correspondingly short, focusing on the initial response as in the EPM, to times 30–120 minutes more focused on the habituation response, locomotion and exploration. The testing time used for the EPM and OF in the present thesis (Paper I) was 20 minutes from first in-troduction to the respective arena. The observations thus include both the ini-tial response and the habituation and subsequent full exploration of the arenas by all animals.

The social play behavior test The SPB test is an interaction-based behavioral test specifically targeting ju-venile and adolescent social interactions. As stated above (under Development of the rat), SPB is essential for normal development of the rat (Panksepp et al., 1984; Trezza et al., 2010). In the rat, a play bout is initiated by a behavior called a pounce where one animal leaps and tries to make contact with the other’s nape and nuzzle it with its snout (Pellis et al., 1997; Trezza et al., 2010). The animal being pounced upon can avoid contact in several ways but the most common is to roll over to a supine position with the initiating animal standing over it, a behavior called pinning. Pinning can also be the outcome of boxing and wrestling, actions when two animals stand on their hind legs and paw at each other or when they tumble and roll about (Auger and Olesen, 2009). Pouncing and pinning are considered to be more objective measures of play, and these behaviors correlate well to other play behaviors (van Hasselt et al., 2012; van Kerkhof et al., 2013; Panksepp et al., 1984).

28

SPB can be studied naturalistically, with observations carried out in the home cage with animals normally housed together in their established social hierarchy. This method allows for the study of play partner choice, true play prevalence and questions regarding social structures in adolescent animals (Meaney et al., 1982; Parent and Meaney, 2008). However, observations car-ried out in the home cage require prolonged observations, spanning over sev-eral days to obtain sufficient numbers of observations. To get an estimation of the level and composition of SPB in different experimental groups, observa-tion of selected play pairs after a period of isolation is more often used (van Hasselt et al., 2012; van Kerkhof et al., 2013; Niesink and Van Ree, 1989; Vanderschuren et al., 1995). Isolation increases the frequency of expressed SPB (Niesink and Van Ree, 1989), allowing for shorter observation times. When using this method, it is important to minimize environmental and social influences that can disturb the play session or create an unbalanced play pair. Therefore, testing is performed in an arena that the animals are habituated to under low light conditions, as novel environment and bright lighting decrease the frequency of play (Vanderschuren et al., 1995). Furthermore, the animals are partnered according to weight to unfamiliar partners as this creates play pairs without preexisting social hierarchies (Panksepp et al., 1984).

In the present thesis, the SPB test is used as a reference test to the MCSF in Paper I for comparison of exploratory and social behaviors among adoles-cent rats. In Paper III, the SPB test is used to evaluate the effects of early-life stress on adolescent social behavior.

Maternal separation Several preclinical models of early-life stress have been developed to study the causal relationship between early-life stress and later outcomes. These models mostly rely on one of two basic concepts: withholding maternal care by separating the offspring from the caretaker or applying factors that decrease the caretaker’s quality of care towards the offspring (Knop et al., 2017; Ladd et al., 2000; Maccari et al., 2014; Nylander and Roman, 2013; Pryce and Feldon, 2003). Methods separating caretaker and offspring vary from com-plete separation (e.g. artificial- or peer-rearing) to repeated separations (ma-ternal separation, MS) or single 24-h separation at a defined time-point during the postnatal period (maternal deprivation). Decreasing the quality of care can be done by stressing the caretaker directly or by limiting access to nesting and bedding material making it impossible for the caretaker to provided sufficient care to the offspring (Knop et al., 2017; Nylander and Roman, 2013).

MS is not a single model, but rather an umbrella term for manipulations of the early environment of rodent pups by removing them from the dam on re-peated occasions during the postnatal period. However, MS was first not de-vised as a model for early-life stress, instead it was noted that early handling

29

of rat pups, and thus removal from the dam, was connected with positive phys-iological and behavioral effects later in life (Levine and Lewis, 1959; Levine et al., 1957). This effect is attributed to a simulation of the natural condition, as rat dams leave the nest at regular, but short, intervals in natural settings (Fleming and Rosenblatt, 1974; Jans and Woodside, 1990). This effect spurred the development of a rearing condition with the opposite effect, which resulted in the prolonged MS condition. However, it is notable that the consequences of prolonged MS are highly variable (Lehmann and Feldon, 2000; Macrì and Würbel, 2006; Nylander and Roman, 2013; Pryce and Feldon, 2003).

As previously stated, MS is not a standardized procedure and the protocol varies a lot between studies (Lehmann and Feldon, 2000; Nylander and Roman, 2013; Roman and Nylander, 2005). During separation, the pups can be individually separated or kept together in the litter, with very different out-come on the offspring (Nylander and Roman, 2012, 2013). Additionally, the length of separation varies substantially for prolonged MS, where 180–360 minutes is the common interval. Short MS, on the other hand, is typically 15 minutes. Likewise, the number of days the separations are carried out differs between studies. Separations are most often initiated during one of the first PNDs but cessation can occur between PND 10–21.

Furthermore, different control groups are used; some studies focus on com-paring short and prolonged MS while additional groups controlling for han-dling and/or separation conditions are often added. Animal facility rearing (AFR) produces conventionally reared laboratory rats and is used as a control for the MS procedure as a whole. A group were pups are handled daily, like the MS groups, but only briefly removed from the home cage can be used to control for separation, this group can be referred to as handled or MS0. How-ever, handled is sometimes used interchangeably with any short MS condition. A non-handled group can be added to control for handling, this group is reared with minimal human contact and therefore, often without cage cleaning (Lehmann and Feldon, 2000; Nylander and Roman, 2013; Pryce and Feldon, 2003). Possibly due to their limited exposure to external stimuli during rear-ing, non-handled animals exhibit aberrant behavior (Pryce et al., 2001) and neurobiology (Nylander and Roman, 2012), and their suitability as controls has been questioned (Nylander and Roman, 2013). The well-used AFR con-dition has also received criticism from an ethological perspective, considering what actually is a natural rearing condition for the rat, and by extension this raises the question of how ‘normal’ a standard laboratory rat is (Nylander and Roman, 2012).

In the present thesis, MS was used to study the effect of early-life stress on behavioral development through early adolescence to early adulthood (Pa-per II) and on adolescent HPA axis reactivity and social behavior (Paper III). In Paper II, prolonged MS was studied together with short MS and AFR con-trols; short MS controls for the included procedures (i.e. handling and separa-tion) and AFR was included as a control for behavioral development after

30

conventional rearing. In Paper III, prolonged MS was studied together with short MS controls only, as the more invasive handling and testing procedures were assumed to diversely affect AFR animals not as habituated to experi-menter contact as the daily handled MS animals of either separation length. By using only MS animals any detected differences would be due to the sep-aration duration only.

Voluntary alcohol consumption Voluntary alcohol consumption models in rats have high face and construct validity to human alcohol consumption (Becker, 2012; Carnicella et al., 2014; Crabbe, 2014; Sanchis-Segura and Spanagel, 2006). When given access to al-cohol, rats will voluntary consume it (Meisch, 2001; Sanchis-Segura and Spanagel, 2006), albeit to highly variable individual levels in both adolescents (Spoelder et al., 2015) and adults (Momeni and Roman, 2014). Most of the voluntary alcohol consumption models rely on non-operant intake, i.e. the rats have access to alcohol without any previous actions required, often trough ac-cess in the home cage. Key in these models is the voluntary consumption, the animal can choose between alcohol and water when drinking. Voluntary con-sumption in outbred rats is usually used to study acquisition and habitual al-cohol intake and not AUDs, as the intake levels in these models are moderate and do not induce intoxication. To model excessive intake and AUDs, other methods with initial forced intake or another choice of animal is needed (Bell et al., 2014; McBride et al., 2014; Sanchis-Segura and Spanagel, 2006; Vendruscolo and Roberts, 2014).

Access to alcohol in the home cage can be given according to different schedules: continuous, intermittent and restricted. In continuous access sched-ules, alcohol is always available, intermittent models have access scheduled to every other day and restricted scheduling limits the intake to a few hours every day as opposed to the others, which gives full 24 h access (Becker, 2012; Crabbe et al., 2011). Intermittent scheduled access has been shown to lead to higher alcohol intake than continuous access (Carnicella et al., 2014; Simms et al., 2008; Wise, 1973) and restricted access is often used when assessing timed effects of alcohol intake. In the lab, a modified intermittent schedule has been devised with alcohol access three consecutive days per week and four days of only water drinking in-between the access days (Momeni and Roman, 2014; Palm and Nylander, 2014a), to simulate a common human drinking pat-tern with alcohol intake limited to the weekends.

Another variable, besides the timing of alcohol access, is the alcohol con-centration and number of alcohol bottles (Sanchis-Segura and Spanagel, 2006). Low alcohol concentrations (<6%) have a sweet taste and are often preferred over higher concentrations (>10%) (Palm et al., 2011a; Sanchis-Segura and Spanagel, 2006). Variations in preference for different alcohol

31

concentrations can be studied in paradigms using more than one alcohol con-centration, enabling studies on individual variation (Palm et al., 2011a). An-other common paradigm is the intermittent two-bottle choice, where the ani-mal has a free-choice between one alcohol concentration (typically 20%) and water (Carnicella et al., 2014). Animals readily acquire and maintain or even escalate their alcohol intake in this paradigm, without any initiation proce-dures (e.g. alcohol sweetening, water/food restriction or priming through forced exposure) (Carnicella et al., 2014).

In the present thesis, voluntary alcohol consumption was assessed in a two-bottle, free-choice paradigm (20% alcohol) on the modified intermittent schedule to ascertain the propensity of alcohol intake and preference (Pa-per V).

Alcohol exposure Alcohol exposure, as opposed to voluntary consumption, is used to evaluate the effect of a set dose of alcohol. As indicated above (under Voluntary alco-hol consumption), the individual level of alcohol consumption varies substan-tially and makes studies of the pharmacological effects of alcohol difficult.

Alcohol exposure can be administered through a variety of means and de-livery sights. An administration route common as pretreatment to increase consumption before home cage drinking or operant self-administration of al-cohol is vapor exposure (Vendruscolo and Roberts, 2014). Inhalation does however not resemble human alcohol consumption and for studies of discrete doses it is not generally used. What does resemble human consumption is in-tragastric (i.g.) administration, either through gavage or permanent cannula-tion. As cannulation requires surgery with associated recovery, it is not easily used during the narrow time window of adolescence nor is it justifiable for few administration occasions.

Different types of parenteral injections are a common means for various drug exposures. Intravenous injection allows for rapid delivery and onset of effects, differing significantly from oral administrations. Intraperitoneal (i.p.) injection on the other hand, partly resembles oral administration in the absorp-tion pathway, excluding passage over the gut wall, whereafter the absorption to the systemic circulation is the same for i.g. and i.p. administrations: transport via the mesenteric vasculature, through the portal vein and liver to the heart (Cederbaum, 2012; Claassen, 1994). Although, some of the absorp-tion after i.p. administration occur through non-hepatic pathways (Claassen, 1994). I.p. injection confers one dose-stabilizing property over i.g. administra-tion, alcohol absorption from the gut depend on the feeding state of the animal (Cederbaum, 2012) giving rise to variation in freely fed animals. This can be

32

abated by food deprivation before i.g. administration but this gives rise to be-havioral and physiological adjustments confounding other measurements (Rowland, 2007).

In the present thesis, alcohol exposure through i.p. injection was used to study the impact of alcohol on behavior and endogenous opioid peptide levels (Paper V). Furthermore, blood alcohol concentrations (BACs) were compared following alcohol exposure through i.p. or i.g. administration (Paper V).

Quantification of endogenous steroids and peptides Sampling considerations Corticosterone assessment In the acute stress response, corticosterone is rapidly released and peaks within 30 minutes. When the stressor disappears or the animal habituates, the corti-costerone levels decline and return to normal within 60–120 minutes (Spencer and Deak, 2016). To capture this time-scale, blood samples are dominating; corticosterone is readily measurable from blood after serum or plasma sepa-ration. There are different methods for blood collection; many studies use end-point collection by decapitation or terminal exsanguination. With this method, repeated sampling in the same animal is not possible and separate groups of animals are needed to capture the different phases of the HPA axis response. Therefore, end-point blood collection is not an option if individual develop-ment over time is of interest. Another method is automated blood sampling through an implanted venous cannula. This method allows for repeated, fre-quent sampling as minute samples can be collected without disturbing the an-imal and it generates measures with high time-resolution (Clark et al., 1986; Windle et al., 1998). However, this method is difficult to use in juvenile and adolescent animals, as cannulation is a surgical procedure that requires several days of recovery before sampling can be initiated, which can be enough to miss the developmental window of interest. Due to these considerations, a blood sampling method with repeated peripheral sampling was chosen in Pa-per III in the present thesis. This method allowed for repeated sampling of the same individual with enough frequency to capture the phases of the HPA axis response, without age-constrains in the adolescent animals.

Peptide assessments Peptides are short chains of amino acids and are common as hormones and neurotransmitters. The class of peptides analyzed in this thesis is the endoge-nous opioid peptides that are present throughout the brain (Le Merrer et al., 2009), are important in the regulation of for example stress and mood and are important players in psychiatric disorders including SUDs (Bodnar, 2020). In

33

this thesis, two of the opioid peptides are measured in Paper V: Met-enkeph-alin-Arg6Phe7 (MEAP), a heptapeptide of the enkephalin family and dy-norphin B, a 13 amino acid long peptide of the dynorphin family.

The issue with measuring peptides is their rapid turn-over rate, which con-tinues postmortem in fresh tissue. To acquire representative measures of in vivo levels this process must be inactivated (Hallberg, 2015; Hallberg and Nyberg, 2003). The common practice for brain preservation is dissection on ice followed by low temperature freezing, typically on dry ice. However, this still allows for ample time for degradation to take place depending on the level of dissection (Segerström et al., 2016). In addition, thawing and/or reheating of freshly frozen tissues allows for further degradation as enzymes are reac-tivated (Segerström et al., 2016). Stabilization through fixation chemicals such as formaldehyde can be used on whole tissue but the penetration time does not lend it for fixation for measurement of peptides. Fixation chemicals can also be used as part of the euthanization process and perfused throughout the animal’s circulation system allowing for faster penetration into tissues (Lamberts and Goldsmith, 1986), this does however require anesthetization which affects the levels of neurotransmitters. Thermal treatment to stop enzy-matic activity through denaturing of the proteins is a method that does not require addition of any chemicals that can affect subsequent analysis. Both microwave irradiation (Mathè et al., 1990; Nylander et al., 1997) and conduc-tive heat transfer (Segerström et al., 2016) have been used to stabilize brain samples for analysis of neuropeptides. In Paper V in the present thesis, sam-ples were first rapidly frozen due to practical reasons and then stabilized with the Stabilizor T1 (Figure 4A) using conductive heat transfer before dissection. Before analysis, the samples are homogenized and purified by cation ex-change chromatography (Figure 4B) to separate similar peptides prone to cross-reactivity in the immunoassay.

Radioimmunoassay Immunoassays can be used to quantify any number of endogenous or exoge-nous substances in almost any tissue. These assays rely on the binding of an antibody to its antigen and the formation of an antigen-antibody complex. In a competitive immunoassay (Figure 4C), the unknown concentration of anti-gen in the sample competes with a known concentration of labeled antigen for binding to the antibodies. With increasing concentration of the substance of interest in the sample, the labeled antigen will be increasingly displaced from the antibodies in favor of the unlabeled antigen. After separation of the un-bound antigens from the antigen-antibody complex, the amount of labeled an-tigen in complex with the antibody can be measured and will have an inverse relationship with the concentration of antigen in the sample of interest. The type of labeling determines the immunoassay type and the method of detec-tion. Radioactive labeling is the basis in radioimmunoassay (RIA) where the

34

radioactivity can be measured in a β- or γ-counter depending on the radioiso-tope used (e.g. 3H or 125I). Antigen conjugation to an enzyme and measurement of the enzyme activity is used in enzyme-linked immunosorbent assays (ELISA).

Figure 4. The work flow for sample preparation and analysis of endogenous opioid peptides. A) Tissue stabilization with the Stabilizor T1 with Maintainor tissue card, B) sample purification through cation exchange chromatography on Sephadex col-umns and C) competitive RIA with antigen-antibody complex formation. Modified with permission from Linnea Granholm (2018).

In this thesis, radioimmunoassays with 125I-labled corticosterone or opioid peptides were used to quantify the amount of immunoreactive (ir) corti-costerone in serum (Paper III) or ir levels of endogenous opioid peptides in various brain regions (Paper V). From here on, ir levels, as measured by RIA, will only be named levels.

Multivariate data analysis Behavioral testing in the MCSF generates many parameters from the same test session, resulting in “short and wide” data tables, i.e. multivariate data, like-wise reflected in the name of the MCSF. Behavioral studies have, through novel approaches and automated detection, begun to enter the field of multi-variate data analysis (MVDA). This has otherwise in biology been dominated by other fields e.g. pharmaceutics (Rajalahti and Kvalheim, 2011) and omics research (Csala and Zwinderman, 2019).

MVDA allows for reduction of the dimensionality of the available data, allowing for additional extraction of information as a complement to classical statistics, as patterns and relationships can be more easily identified (Shah et al., 2016). In this thesis, three related types of MVDA were used: principal component analysis (PCA), partial least squares projections to latent struc-tures (PLS) analysis, and PLS-discriminant analysis (PLS-DA). A PCA pro-vides an overview and summary of the data as it examines the relationship

35

among all entered parameters (X:s). A PLS analysis is used to provide an over-view of the relationship between blocks of data (X:s and Y:s). PLS-DA uses a given classification (a Y) and relate the parameters (X:s) in relation to the clas-sification, facilitating the exploration of class-dependent differences.

All three analyses rely on projection and the creation of components repre-senting dimensions of variability in the data. Each component is described by R2X, Q2 and, if Y is used, R2Y values. R2 is the amount of variation in the data (X or Y) that is explained by the component; the first component always has the highest R2 as the components are created hierarchical. The first component is selected by maximizing the variability explained and subsequent compo-nents are orthogonal to the previous. A model can contain any number of com-ponents (up to the number of parameters), but retention of a component in the inspection stage is balanced by gains or losses in Q2 as the number of compo-nents grow. Q2 estimates the predictive capacity of the component by cross-validation: data (both parameters and individuals) are serially removed and how well the remaining data can predict the removed data is calculated.

In the present thesis, for each model, two-dimensional score and loading plots are generated from the first two components. The score plot shows the relationship between the individuals as it summarizes the variables for each individual in each component to a t-score. When the components are plotted against each other it enables interpretation of the relatedness between individ-uals and identification of groups and outliers. The loading plot shows the op-posite, i.e., summarizes all individuals for each parameter to a p-loading for PCAs and w*c-loadings for both types of PLS analyses and enables interpre-tation of the relatedness among parameters.

36

Aims

The overall aim was to examine adolescent behavior and its links to early-life stress and alcohol in adolescent male and female rats. The specific aims were:

• To explore the general behavioral profile of adolescent males and to vali-

date adolescent behavior in the MCSF in comparison to other behavioral tests (Paper I)

• To evaluate the behavioral profile in early adolescence and early adult-hood after early-life stress in males (Paper II)

• To study the stress response and social play behavior in adolescence after early-life stress (Paper III)

• To examine the link between genetic propensity for high or low alcohol consumption and the adolescent behavioral profile before first contact with alcohol (Paper IV)

• To investigate the association between the adolescent behavioral profile and later voluntary alcohol consumption (Paper V)

• To evaluate the behavioral and neurobiological effects of acute alcohol

exposure during adolescence (Paper V)

37

Methods

An overview of the experiments is presented in Figure 5.

Figure 5. A) Developmental phases of the rat (compare with Figure 2) and B–F) cor-responding experimental outlines in Paper I–V. Gray bars denote time spent at the animal vendor, orange indicate stress procedures and red indicate alcohol proce-dures. NB, newborn.

38

In Paper I, animals were tested in either the MCSF, EPM, OF with or without start box or SPB test at approximately 5 weeks of age (behavior 1 in Figure 5B) and then at 6 weeks of age in either the MCSF or SPB test (behavior 2 in Figure 5B). In Paper II (Figure 5C), animals were subjected to MS during the three first postnatal weeks and then tested in the MCSF at approximately 4 and 9 weeks of age (PND 26–31 and 61–65, respectively). In Paper III (Figure 5D), animals were also subjected to MS as in Paper II and then assessed in an HPA axis reactivity test when 3 weeks old and in the SPB test at 4 weeks of age (PND 25/26 and 33–35, respectively).

In Paper IV (Figure 5E), subjects were tested in the MCSF when 4 weeks old (PND 30–35). In Paper V (Figure 5F), subjects were tested in the MCSF at approximately 6 weeks of age, with and without exposure to alcohol. The non-treated group went on to voluntarily consume alcohol throughout late ad-olescence and into adulthood (between 7 and 14 weeks of age). The alcohol-exposed group received one additional alcohol exposure at approximately 7 weeks of age when tissues were collected for analysis. A third group were exposed to alcohol at 4 weeks of age and sampled for determination of BACs.

Animals and housing Animals were either delivered post-weaning from a vendor (Paper I, IV and V), born at the animal facility of dams delivered from a vendor at gestational day 14–16 (Paper II and III) or bred in-house as part of a selective breeding program (Paper IV). The sex and strain or line of the included adolescent rats, the experimental groups and subgroups based on individual variation used in the statistical analysis are shown in Table 1. In this thesis, when describing a result as obtained in outbred rats, it is henceforth referring to Wistars as in-cluded in the papers in the thesis.

Table 1. Overview of the sex, rat strain or line and groups used in this thesis. Exper-imental groups were created by varying the experimental conditions for the animals, while groups based on individual variation were derived during the data analysis.

Paper Sex Strain/Line Experimental group Groups based on individual variation

I Behavioral ty pe II

♂ Wistar (R ccHan)♂ Wistar (RccHan) MS360/MS15/AFR Behavioral type

III ♂♀ Wistar (RccHan, MS360/MS15

IV

♂♀

HanTac) P/NP HAD1/LAD1 HAD2/LAD2 Wistar (RccHan)

V: Exp. 1 ♂♀ Wistar (RccHan) Alcohol intake Exp. 2 ♂♀ Wistar (RccHan) 0.0/0.5/1.0 g alcohol i.p. Exp. 3 ♂♀ Wistar (RccHan) 0.5/1.0/1.5 g alcohol i.p./i.g.

39

Animals born in-house were weaned on PND 22 (Paper II and III) or 22–27 (Paper IV), post-weaning rats were delivered approximately 3 weeks old (Pa-per I and V). Animals were housed in same-sex and, when possible, same-lit-ter groups of 2–5 animals in cages type IV (59×38×20 cm, Paper I–III and V) or shoe-box cages (22×44×20 cm, Paper IV), with ad libitum access to food and water. In Paper I, II, IV and V, the animals were transferred to a reversed light/dark cycle in conjunction with weaning or delivery to facilitate testing under the dark phase of the light/dark cycle. In Paper III, due to the early HPA axis reactivity testing, the experiments were carried out under a normal light/dark cycle with testing during the light phase. Transportation and/or light/dark cycle reversal was followed by an acclimatization period where af-ter experimental procedures were initiated. Animals were identified by either ear pouncing (Paper I–III and V) or tail markings (Paper IV).

All animal experimental protocols were approved by the relevant ethical committee (Paper I–III and V: the Uppsala Animal Ethical Committee; Pa-per IV: The Institutional Animal Care and Use Committee of the Indiana Uni-versity School of Medicine), in accordance with the appropriate laws and guidelines concerning work with laboratory animals.

Behavioral tests Prior to behavioral testing, the animals were habituated to handling and trans-portation. Testing was carried out under the same environmental conditions as in the animal room, in a room adjacent to the animal room (Paper I–III and V) or adjacent to a holding room where the animals to be tested were transported to in their home cage at least 2 h before testing was initiated (Paper IV). On each testing day (except for the SPB test), preceding the experimental rats, an age-matched out-of-test rat was first allowed to explore the behavioral arena to avoid any first-in-line effects.

The multivariate concentric square field™ test Animals were tested in the MCSF test at different ages across adolescence and early adulthood (Paper I, II, IV and V, see Figure 5 for details). The arena is described in detail under Methodology in this thesis and elsewhere (Meyerson et al., 2006; Roman and Colombo, 2009). Briefly, the arena (Figure 3) is sub-divided into zones of different quality and the animal is allowed to freely ex-plore the arena.

The animals were tested during the dark phase of the light/dark cycle, in mixed order to avoid time- and order bias. If males and females were part of the same batch, males were always tested before females on any given day. In all instances, except in Paper IV, animals in the same cage were not tested on the same day. In Paper IV, since testing required transportation to a holding

40

room, the animals in a cage were tested on the same day but never directly after one another.

Each animal to be tested was transferred one by one from the animal or holding room to the testing room in a transportation bucket or cage and placed in the center facing the wall separating the center and bridge. Recording started immediately and lasted for 20 minutes. After each test session, the arena was wiped off with a 10% ethanol solution and was allowed to dry com-pletely before the next session.

The test sessions were recorded from above by video camera. After each trial the number of fecal boli, urinations and nose pokes by photocell counts (except in Paper IV where no photocell was used) were noted. EthoVision XT (Noldus Inc., Wageningen, the Netherlands) was used for manual scoring of zone measures, rearing, grooming and SAPs, and automatic tracking of total distance and mean velocity (Paper I, IV and V (automatic tracking was not used in Paper IV)). In Paper II, Score 3.3 (Soldis, Uppsala, Sweden) was used to manually score zone measures, EthoVision 2.3 was used for automatic tracking of total distance and mean velocity and direct observation was used to score number of rearing, grooming and SAPs. A zone transition was only counted when both hind legs crossed into the new section. For description of the primary behavioral parameters, see Table 2. In addition, the following pa-rameters were derived: latency to leave the center (L leave); the sum of all frequencies (total activity, TOTACT); frequency and duration spent in the cor-ridors (F and D TOTCORR); occurrence (Occ, only reported when a zone or behavior was not visited or performed by all animals) for zone visits, rearing, grooming, SAPs, nose pokes, urinations and fecal boli; and duration per visit (D/F), percental frequency (%F) and percental duration (%D) for all zones.

Table 2. Parameters recorded in the MCSF, EPM and OF test. A latency score re-quired a frequency minimum of 1.

Parameters Description

Rearing The animal lifts both fore paws into the air or onto a wall Grooming The animal grooms itself; face-, body- or anogenital wash Stretched attend posture (SAP) The animal elongates its body from one zone into another Latency (L [s]) The time until the first visit to a zone Frequency (F) The number of visits to a zone Duration (D [s]) The time spent in a zone Distance [cm] The total distance travelled Velocity [cm/s] The mean velocity MCSF specific Nose pokes The animal dips its head into the hole-board in the hurdle

41

The elevated plus maze and open field The EPM and OF were used as reference tests as part of the validation of ad-olescent behavior in the MCSF (Paper I). Each animal was tested in one of the complementary tests at 5 weeks of age.

The EPM consists of two crossed platforms (90×10 cm) forming a four-armed cross. Two opposite arms were enclosed by walls 40 cm high, creating two closed and two open arms used as zones in the analysis. The central square where the arms meet was considered its own zone. The entire arena was ele-vated 51 cm of the floor. The illumination in the arena was 100 lux, measured in the central square.

The OF arena is circular (90 cm in diameter) with wire-mech floor and a 35 cm high outer wall. The arena was used both with and without the use of a start box (25×25×25 cm) attached to the outside of the arena, accessible through a door in the wall. During analysis, the open field is divided into three zones by concentric circles: the center (30 cm in diameter), the inner circle (15 cm wide) and the outer circle (15 cm wide). The illumination in the arena was 100 lux, measured in the center.

The animals were tested during the dark phase of the light/dark cycle, in mixed order to avoid time- and order bias. Animals in the same cage were not tested on the same day. Each animal to be tested was transferred one by one from the animal room to the testing room in a transportation bucket and placed in the central square facing an open arm (EPM), in the outer circle facing the wall (OF) or in the start box (OF with start box). Recording started immedi-ately and lasted for 20 minutes. After each test session, the arena was wiped off with a 10% ethanol solution and was allowed to dry completely before the next session.

The test sessions were observed and recorded from above by video camera. After each trial the number of fecal boli in both the EPM and OF, and urina-tions in the EPM were noted. EthoVision XT (Noldus Inc., Wageningen, the Netherlands) was used for manual scoring of rearing, grooming and SAPs, and automatic tracking was used for acquisition of zone measures, total distance and mean velocity. A zone transition was defined as the tail body point moving into the new section. For description of the primary behavioral parameters, see Table 2. In addition, the following parameters were derived: the sum of all frequencies (total activity, TOTACT) and duration per visit (D/F), percental frequency (%F) and percental duration (%D) for all zones.

Social play behavior test SPB was assessed in Paper I, as part of the validation of adolescent behavior in the MCSF with animals tested at 5 or 6 weeks of age and in Paper III, to explore the effect of MS on adolescent social behavior with animals tested at 4 weeks of age.

42

Testing was carried out in a square arena (50×50 cm) with a 25 cm high outer wall; the arena floor was covered with approximately 2 cm of wood-chip bedding. The arena was illuminated from above to a level of approximately 10 lux.

Before testing, the animals were habituated to the arena on two consecutive days. The animals were allowed to freely explore the arena either alone for 5 minutes (Paper I) or together with its cage mates for 10 minutes (Paper III) each day. Play pairs consisted of unfamiliar individuals, in Paper I pairing was performed with respect to body weight (<15% body weight difference); in Pa-per III pairing was performed with respect to rearing condition, vendor, sex and body weight (5% median body weight difference, min <1%, max 23%).

On the day of testing, animals were isolated for 3.5 h in cages type III (42.5×26.5×18 cm) with access to food and water. The animals remained in the animal room and thus had vocal, visual and olfactory contact with others whilst being deprived of physical social contact. The play partners were marked on the tip or base of the tail with a black, xylene-free marker pen, for later individual-based behavioral analysis. At the end of isolation, the play partners where brought, in their isolation cages, to the testing room and re-leased into opposite ends of the arena. Recording started immediately and lasted for 15 minutes. Between each test, the wood-chip bedding was replaced and the arena wiped off with a 10% ethanol solution, the arena was allowed to dry completely before fresh bedding was added.

The test sessions were observed and recorded from above by video camera; EthoVision XT (Noldus Inc., Wageningen, the Netherlands) was used for manual scoring of social behaviors, see Table 3 for description of the behav-iors and parameters. In Paper I, both individual play parameter and the quality of contact in the pair were assessed. In Paper III, only individual play param-eters were assessed.

Table 3. Ethogram over behaviors scored in the SPB test. Individual behaviors were assessed with respect to which animal initiated the behavior, giving an individual score. Quality of contact was assessed for the pair. In Paper I, latency [s] to first ex-pression, frequency and duration [s] were assessed for all behaviors. In Paper III, only latency and frequency for pouncing and pinning were assessed. A latency score required a frequency minimum of 1.

Category Behavior Description

Individual behaviors Pouncing When the animal tries or succeeds to nuzzle the nape of its play partner with its snout

Pinning When the animal stands over its play partner, which is in a supine position

Climbing When the animal climb over its play partner Quality of contact No physical contact The animals are not in immediate contact Contact behavior The animals are touching or sniffing each other Vigorous play The animals are engaged in play behaviors

43

Stress procedures

Maternal separation MS was used to simulate early-life stress in rat pups by removal of the litter from the dam for 360 minutes daily (MS360) during the full pre-weaning pe-riod (PND 1–21). Control rearing conditions used were short MS (15 min daily separation, MS15) and AFR (Paper II) or only MS15 (Paper III).

Starting at gestational day 21, dams were checked twice daily for parturi-tion, in the morning and late afternoon. Day of birth (PND 0) was considered from one afternoon check to the next, i.e. litters were considered born on the same day when born after the afternoon check on one day but before the same check the following day. On PND 0, pups were sexed and cross-fostered, within the same supplier when applicable, into litters of 9–10 pups (5–6 males and 4–5 females per litter). The litters were randomly assigned (with respect to vendor when relevant) to one of three rearing conditions.

The separations occurred once daily on PND 1–21, during the light period of the light/dark cycle, starting from 09:00. First, the dam was removed from the home cage and placed in a holding cage (type II; 22×16×14 cm), thereafter the litter was placed together in a cage type II with wood-chip bedding and moved to a heating cabinet (29±1.0°C) in an adjacent room to avoid hypother-mia. The MS15 dams spent the separations in the holding cage, the MS360 dams were returned to the home cage during the separation but were once again removed before the return of the litter. The litter was returned to the empty home cage and placed together in the nest, thereafter the dam was re-turned.

Home cages were changed twice during the pre-weaning phase, on PND 7 and 16. MS litters were weighed seven times, on PNDs 0/1, 3/4, 7, 10, 13/14, 16/17 and 20. The animals in the AFR group were left undisturbed except for the cage changes, at which time they were also weighed.

HPA axis reactivity test Basal, challenged and recovery levels of corticosterone were assessed on PND 25/26 in early adolescence in males and females after MS (Paper III). Blood samples were collected immediately before and after 30 minutes (T0) of iso-lation in cages type II and after 120 minutes (T120) of recovery in the home cage.

Testing was carried out in a room adjacent to the animal room, under the same environmental conditions. Blood samples were collected from a tail cut (Microvette serum 100 μl, Sarstedt AG & Co, Nümbrecht, Germany), with basal samples collected between 09:00–09:25 for females and 10:15–10:40 for males. All samples were collected during the light phase of the light/dark

44

cycle. The blood samples were allowed to coagulate in room temperature for 20–30 minutes and then refrigerated. Serum was collected after centrifugation (10 min, 3000 g, 4 °C) and stored at -80 °C until analysis.

Alcohol procedures

Voluntary alcohol consumption In Paper V, experiment 1, the animals received access to alcohol for 8 weeks from 7 weeks of age. Alcohol was available in a modified intermittent sched-ule with alcohol access three consecutive days per week (Tuesday, Wednes-day, Thursday), followed by four days of water only (Momeni and Roman, 2014; Palm and Nylander, 2014a), resulting in a total of 24 alcohol access sessions á 24 h. Alcohol access was initiated during the dark phase and the water and alcohol solutions were changed before access each day. The place-ment of the bottles was rotated to avoid position bias.

Alcohol was given in a two-bottle choice paradigm, with free choice be-tween water and alcohol (20% alcohol diluted from 96% (Solveco AB, Rosersberg, Sweden) in tap water [v/v]). The amount of fluid consumed was measured every access day by weighing the bottles. The animals were weighed twice a week, the day before the first alcohol access day and after removal of alcohol on the last day of access. The total alcohol intake [g/kg], alcohol preference [%] and water as well as total fluid intake [g/kg] were cal-culated for each access day and analyzed as averages for the three days in a week.

During alcohol access, a cage divider was placed in the cage (Palm, 2014; Palm and Nylander, 2014b; Scott et al., 2020), separating the pair-housed an-imals. The divider was made of transparent acrylic and wire mesh and was used to facilitate the acquisition of individual measurement of fluid intake dur-ing the alcohol access days while allowing for tactile, auditory and olfactory contact between the cage mates. The divider was inserted into the cage directly before presentation of alcohol on the first access day and removed together with the alcohol on the last access day.

Alcohol exposure In Paper V, experiments 2 and 3, animals were exposed to various alcohol doses through intraperitoneal (i.p.) or intragastric (i.g.) administration.

In experiment 2, adolescent male and female rats were administered 0.0, 0.5 or 1.0 g/kg alcohol i.p. (15% alcohol diluted from 96% (Solveco AB, Rosersberg, Sweden) in saline (B. Braun AG, Melsungen, Germany) [v/v]) at two time-points: 5 minutes before testing in the MCSF at 6 weeks of age and

45

30 minutes before euthanization one week later. Animals were euthanized by decapitation and the brains were quickly removed, flash-frozen in isopentane and stored at -80 °C until analysis.

In Paper V, experiment 3, adolescent male and female rats at 4 weeks of age were administered 0.5, 1.0 or 1.5 g/kg alcohol i.p. (20% alcohol diluted from 96% (Solveco AB, Rosersberg, Sweden) in saline (B. Braun AG, Mel-sungen, Germany) [v/v]) or i.g. (20% alcohol diluted from 96% (Solveco AB, Rosersberg, Sweden) in tap water [v/v]). 30- and 120-minutes post-treatment, blood samples were collected from a hind paw vein (CapiJect T-MLH 500 μl, Terumo Medical Corp., Elkton, USA). The samples were stored on ice until centrifugation (10 min, 570 g, 4 °C), plasma was collected and stored at -80°C until analysis.

Blood and tissue analyses Serum corticosterone radioimmunoassay Corticosterone was measured in serum samples collected after the HPA axis reactivity test in adolescent males and females previously subjected to MS (Paper III).

Serum samples were thawed and analyzed with the commercial Im-muChem™ Double Antibody Corticosterone 125I RIA kit for rats and mice (MP Biomedicals LLC, Orangeburg, NY, USA), in accordance with the in-cluded protocol, with one additional standard (12.5 ng/ml). All samples were analyzed in duplicate. According to the included protocol, the intra-assay var-iation was 4.4–10.3% and the inter-assay variation 6.5–7.2%. The corti-costerone antiserum had 100% reactivity with corticosterone, cross-reactivity to other steroids were 0.34% to deoxycorticosterone, 0.10% to testosterone, 0.05% to cortisol and <0.05% to other tested steroids.

Endogenous opioid peptide radioimmunoassay Two types of endogenous opioid peptides, MEAP and dynorphin B, were measured in six different brain regions (Figure 6) after alcohol exposure through i.p. injection in adolescent male and female rats (Paper V). The col-lected brains were stabilized and dissected, the dissected tissue samples were homogenized and purified before the peptide levels were measured with RIA.

Whole brains were moved to -20°C over night before heat-stabilization with the Stabilizor T1 (Denator AB, Uppsala, Sweden) on the “frozen struc-ture preserve” followed by “fresh structure preserve” mode (Granholm et al., 2018; Segerström et al., 2016). After stabilization, the hypothalamus was re-moved with a pair of forceps, the brains were placed in a rat brain matrix (ASI Instruments Inc., Warren, MI, USA) and dissected using razor blades (coronal

46

sections, 1 mm slots). The nucleus accumbens, bed nucleus of the stria termi-nalis (BNST), striatum (caudate putamen), amygdala and ventral tegmental area (VTA) were dissected from the sections with the guidance of a rat brain atlas (Paxinos and Watson, 2007). Stabilized tissue samples were stored at -20°C.

Figure 6. The brain regions used in analysis of endogenous opioid peptides after al-cohol exposure in Paper V, sagittal view.

The tissue samples were thawed in 95°C 1M acetic acid and homogenized by sonication (Branson Sonifier, Danbury, CT, USA). The homogenates were centrifuged (15 min, 15200 g, 4°C) and the supernatants were purified by cat-ion exchange chromatography on conditioned Sephadex gel columns. The peptides were eluted with pyridine and formic acid buffers of increasing ion strengths (Christensson-Nylander et al., 1985; Segerström et al., 2016). The eluates were dried by vacuum centrifugation and stored at -20°C.

The opioid peptide RIAs were performed as previously described for MEAP (Nylander et al., 1995a) and dynorphin B (Nylander et al., 1997). All samples were analyzed in duplicate. Prior to analysis of MEAP, samples were subjected to oxidization by treatment with 1M acetic acid and 30% hydrogen peroxide. Antisera for the two peptides were generated in rabbits. The MEAP antiserum (90:3D II) was used at a final dilution of 1:140000. The cross-reac-tivity with Met-enkephalin, Met-enkephalin-Arg6, Met-enkephalin-Arg6Gly7Leu8, Leu-enkephalin and dynorphin A(1–6) were <0.1%. The dy-norphin B antiserum (113+) was used at a final dilution of 1:500000. The cross-reactivity with dynorphin B 29 was 1% and 100% with big dynorphin (dynorphin 32). Antibody-bound peptides were separated from unbound pep-tide by adding charcoal suspension in the MEAP RIA and goat-anti-rabbit-IgG and normal rabbit serum in the dynorphin B RIA.

47

Blood alcohol concentration analysis Plasma samples were brought to room temperature and BACs were measured using the Analox AM1 analyzer (Analox Instruments Ltd., Stourbridge, United Kingdom), in accordance with the manufacturer’s instructions. The limit of detection was stated as 4.5 mg/dl and the analysis volume was 5 μl. Samples were analyzed in triplicates whenever possible. Electrode sensitivity was tested with known standards every ten samples and variation was under 5% throughout measurements.

Statistical analysis Classical statistical analysis was carried out in Statistica (TIBCO Software Inc., Tulsa, USA) or R (R Core Team, 2020). First, normality was examined with Shapiro Wilk’s test which directed the choice of further statistical tests. P-values were used to determine statistical significance in three grades; <0.05, <0.01 and <0.001 unless otherwise stated. Categorical data were analyzed with Pearson or Maximum-Likelihood Chi2-test. Effect sizes of significant differences were calculated according to Fritz et al. (2012).

Parametrical statistics Body weights in Paper IV and V, and peptide levels in Paper V were normally distributed and thus analyzed with parametrical statistics. Single point measures were analyzed with factorial analysis of variance (ANOVA) and longitudinal data sets were analyzed with repeated measures ANOVA. Signif-icant effects and interactions were further examined with post hoc Tukey HSD test.

Non-parametrical statistics All behavioral data (Paper I–V), corticosterone measurements (Paper III), BACs (Paper V) and body weights in Paper I and III showed largely non-normal distribution and were thus analyzed with non-parametrical statistics.

Single point measures were analyzed with Mann-Whitney U-test when comparing two initial groups or with Kruskal-Wallis ANOVA by ranks with post hoc Mann-Whitney U-test when comparing more than two initial groups. Correlations were examined with Spearman rank order test.

Longitudinal data sets were analyzed with the R package nparLD (Noguchi et al., 2012), facilitating analysis of main effects and interactions in factorial designs. Significant effects and interactions were further examined with post hoc Mann-Whitney U-test (between subject-dependent) and/or Wilcoxon’s matched pairs test or the nparLD package (within subject-dependent).

48

Multivariate data analysis MVDA was carried out in SIMCA (Sartorius Stedim Data Analytics AB, Umeå, Sweden). Models were generated with the autofit option, the created components were inspected and excluded if the eigenvalue was below 2.0 or if the Q2 had a large negative value. Latencies, occurrences and percental du-ration and frequency from the behavioral tests were not included in the models and additional parameters were excluded when advised by the software.

49

Results and Discussion

Characterization of adolescent behavior in the MCSF Four aspects of general adolescent behavior in the MCSF were covered in this thesis: 1) sex-differences in the behavioral profile (Paper IV and V), 2) indi-vidual variation in the behavioral profile and identification of the adolescent behavioral types (Paper I and II), 3) repeated testing and/or effect of age at testing (Paper I and II) and 4) relationship to other behavioral tests (Paper I).

Sex-differences in adolescent behavior in the MCSF Sex-differences in adolescent behavior in the MCSF were limited in untreated animals. In Paper V, in outbred males and females tested without any preced-ing treatment, females had lower average visit time and higher velocity in the center, lower duration spent in the corridors, lower percental frequency of vis-its to the slope and performed more rearings than males. In Paper IV, where outbred animals were analyzed together with selectively bred lines with line/strain as a co-variate, the effect of sex was also small. However, when animals were pretreated with an i.p. injection of alcohol or saline as control (Paper V), notable sex and sex×treatment effects were discovered (see further details in Effect of alcohol exposure on adolescent behavioral profiles).

The discrepancy between treated and untreated cohorts may lie in a sex-differentiated response to the i.p. administration itself. This is a probable ex-planation as the results in the untreated cohorts in Paper IV and V are in line with the few previous studies examining sex-differences in the MCSF. Small to modest sex-differences have been reported in both adolescent (Fernández et al., 2020) and adult (Lundberg et al., 2018; Meyerson et al., 2006) animals. Among adolescent rats short-term selectively bred (three generations) for high or low alcohol consumption, higher scores of risk-associated parameters and lower shelter-associated parameters were reported in female relative male an-imals using a modified MCSF (Fernández et al., 2020). In adults, higher gen-eral and exploratory activity has been reported in female compared to male Wistar rats (Lundberg et al., 2018), while another study described no marked difference between male and female Sprague-Dawley rats (Meyerson et al., 2006).

50

The adolescent behavioral types The discovery of the adolescent behavioral types in the behavioral profile from the MCSF was the main finding in Paper I and independently replicated in Paper II. The identification process of the behavioral types and the devel-opment of a rank-order procedure, named the type scores, is presented below.

In Paper I, the adolescent performance in the MCSF was compared with the functional behavioral categories in the trend analysis, which is used to in-terpret adult behavior in the MCSF (Meyerson et al., 2013). The ranked pa-rameters included in the trend analysis was used in a PCA of the adolescent individuals in Paper I. Parameters in different functional categories showed considerable overlap, especially parameters related to risk-taking behavior on the bridge and risk assessment parameters as well as the parameters in the general and exploratory activity categories. This indicates that the trend anal-ysis is not directly applicable on adolescent behavior in the MCSF. That the risk taking and risk assessment categories does not separate in adolescence could be due to the generally higher levels of risk taking and impulsivity in adolescence and higher novelty seeking may drive the overlap between gen-eral and exploratory activity (Doremus-Fitzwater et al., 2010; Spear, 2000; Sturman and Moghaddam, 2011).

Instead of a relationship to the trend analysis categories, subgroups relating to individual differences in the parameters could be seen in the score plot of the aforementioned PCA (Figure 7A). These clusters were chosen for further analysis; 62 individuals received a classification while eight intermediate in-dividuals were left without. The classification was used in a PLS-DA of all MCSF parameters, to examine the persistence of the classification and which parameters were important for the different groups (Figure 7B–C). The PLS-DA generated one significant component, so while the groups were distinct in the score plot of this analysis (Figure 7B) and in a PCA of the same data, the classification reflects a mostly one-dimensional, but multivariate, difference between the groups. The parameters important for the distinction between the groups were identified in the PLS-DA loading plot (Figure 7C) and verified by inspection of their regression coefficients. The classification as a whole was named behavioral types, with the subgroups Explorers, Shelter seekers and Main types. The types’ characteristics and important correlated parame-ters are summarized in Table 4. The outlying behavioral types, i.e. the Ex-plorer and Shelter seeking type, each comprised approximately 20% of the total cohort, that is 13 out of 70 individuals each.

51

Figure 7. Identifying and characterizing the behavioral types. A) Individuals scores from the PCA (n=70, 3 components, R2X=0.558, Q2=0.333) of ranked parameters from the trend analysis. B) Individual scores and C) parameter loadings from the PLS-DA (n=62, one significant component: R2X=0.359, R2Y=0.343, Q2=0.277; a second component was added to obtain a 2-dimentional model for visualization: R2X=0.080, R2Y=0.143, Q2=-0.046) of the MCSF parameters (X) and behavioral type as class (Y).

52

Table 4. Characteristics of the behavioral types discovered in Paper I and their as-sociated regression coefficients used to create the type scores in Paper II.

Main type Explorers Shelter seekers

Characteristics ↔ Activity

↑ Activity ↓ Risk and activity ↑ Shelter seeking

Significant regression coefficients

+ D slope - D/F DCR

+ Velocity, distance - D/F center

+ D DCR, D/F DCR, nose poke - Total activity, velocity, dis-tance, D center, D bridge, D bridge entrance

Classical statistical analysis comparing the three behavioral types further con-firmed the widespread difference among them: 46 out of 71 parameters had a significant difference between at least two of the types, whereof 16 parameters differed among all three types. As indicated in the MVDA, Explorers were somewhat more similar to the Main type than the Shelter seekers were and the difference between Explorers and Shelter seekers were all-encompassing. The difference in movement and overall performance of the behavioral types in the MCSF is illustrated in Figure 8, where the automatic tracking in Etho-Vision XT for random samples from each behavioral type is shown.

Figure 8. Schematic layout of the MCSF (compare with Figure 3) overlaid with tracks of movement (top row) and heatmap of time spent (bottom row) in three ran-domly selected individuals from Paper I, one from each of the three behavioral types: main type (left), explorer (middle) and shelter seeker (right).

To proceduralize the identification of the behavioral types and to create a tool for quantification that could be used to assess the type variability in smaller cohorts, a rank-order procedure was created. The procedure was influenced by the trend analysis; used for interpretation of adult performance in the

53

MCSF (Meyerson et al., 2013). The parameters selected were the ones having significant regression coefficients in the PLS-DA (Table 4). Rankings were made, giving the individual with the lowest parameter value rank 1, the next rank 2 and so forth, to the individual with the highest parameter value receiv-ing rank n (n=number of ranked individuals). Parameters with a negative re-gression coefficient were reversed and the ranks for a type were then summed. Each individual thus received two rank sums that were named after the corre-sponding behavioral type, i.e. exploration and shelter seeking, and the whole procedure was called type score. When the procedure was retrospectively ap-plied to the data in Paper I, the 20% of individuals with highest exploration score encompassed 77% of the Explorers (10 of 13 individuals) and no Shelter seekers. The top 20% of individuals with highest shelter seeking score in-cluded 85% of the Shelter seekers (11 of 13 individuals) and none of the Ex-plorers.

Figure 9. Replication of the adolescent behavioral separation into behavioral types. A) Individual scores and B) parameter loadings in the PCA (n=71, 2 components, R2X=0.439, Q2=0.314) of the MCSF parameters from the male early-adolescent rats in Paper II.

54

Initial analysis of the data in Paper II indicated a similar division based on individual variation as in Paper I. The type score procedure was used to iden-tify the highest scoring individuals of each type and these individuals guided the identification of the behavioral types in the first MCSF trial in Paper II. However, in this younger cohort (4 weeks old as opposed to 5 and 6 weeks old in Paper I) the 20% division did not correspond to the clusters seen in the PCA score plot (Figure 9A). Instead, the cluster of Shelter seekers was bigger (28%, 20 out of 71 individuals) and the number of Explorers was smaller (13%, 9 out of 71 individuals) than in Paper I. This was not entirely unex-pected as exploratory behavior have been shown to increase throughout ado-lescence (Lynn and Brown, 2009) and higher shelter seeking is consistent with theories of behavioral development (Doremus-Fitzwater et al., 2010; Sturman and Moghaddam, 2011; Yurgelun-Todd, 2007). As in Paper I, classical statis-tical analysis of the MCSF parameters by behavioral type showed widespread differences; 46 of 75 parameters showed some type-dependent difference and approximately half of those had differences among all three types.

The differences between the behavioral types were also examined in the second trial at 9 weeks of age in Paper II, when the animals had entered early adulthood. The differences were not as pronounced at this age as when iden-tified in adolescence; but notably, the shelter seeking behavioral profile was still identifiable while the differences between Explorers and Main types had virtually disappeared. Another noteworthy finding was that the Shelter seekers differed in all the trend analysis categories except precisely shelter seeking. In the trend analysis, shelter seeking only includes DCR-related parameters while the behavioral type encompasses both increased shelter-seeking behav-ior as well as decreased activity and risk-taking behavior. This finding pro-vides further information about the behavioral types and how they develop and relate to behavior in adulthood.

Repeated testing and/or effect of age in the MCSF Two cases of repeated testing in the MCSF in male animals were included in this thesis. In Paper I, retesting was done with one-week inter-trial interval at 5 and 6 weeks of age and in Paper II, testing was done at 4 and 9 weeks of age, an inter-trial interval of five weeks.

Testing with one week between trials resulted in differences among 20 out of 68 parameters including increased total activity, driven by increased num-ber of visits to the center, corridors and DCR, while the frequency to the bridge decreased. The duration on the bridge also decreased. Retesting also shortened the latency to leave the center after the start of the trial and the latencies to first visit to all zones except the central circle, DCR and bridge entrance. Nei-ther the distance moved, mean velocity nor number of rearings were affected by retesting.

55

Retesting with five weeks between trials, and an earlier first test, had a more pronounced effect on the behavior in the MCSF. In the second trial, 44 of 75 parameters differed compared to the first trial, including total activity, distance moved, mean velocity and rearing, which all increased, as did the proportion of animals visiting all zones. The latency to leave the center and latency to the individual zones (except the central circle) decreased in trial 2 relative trial 1. The duration spent in the corridors and hurdle increased, the time spent in the DCR decreased and the duration in the other zones were unchanged between the trials.

That the differences were more numerous when retesting with five compared to one week inter-trial interval indicate an involvement of development over the course of adolescence in Paper II. This is further supported by comparisons of the animals in Paper I tested for the first time in the MCSF at the two ages also used for retesting (one-week interval), the differences in the retested cohort were more distinct than the developmental effect. The effect of age was further ex-amined in Paper I, as the adolescent cohort was compared to a literature cohort aged between 10 and 11 weeks. The adolescent and adult cohorts could be clearly separated in a PLS-DA and the model generated predicted ages that dif-fered significantly between the adolescent and adult cohorts.

A hypothesized model of development of activity in the MCSF has been developed based on the included data on retesting and development discussed above and is presented in Figure 10.

Figure 10. Hypothesized development of activity in the MCSF, based upon data from Paper I and II in outbred males. Note that the scale of activity is arbitrary and generalized, it does not reflect actual fold-differences at different ages.

There are previous studies on retesting in the MCSF in adult males with inter-trial intervals of 1–6 weeks, which demonstrate some memory retention of the experience in the MCSF between trials and similar findings have been seen across studies in different strains and lines (Karlsson et al., 2009; Magara et al., 2015; Meyerson et al., 2006; Roman and Colombo, 2009; Roman et al.,

56

2007). The main findings in adults when retesting have been lower risk-taking behavior on the bridge in most (Magara et al., 2015; Meyerson et al., 2006; Roman et al., 2007) but not all studies (Roman and Colombo, 2009) which corresponds to the finding in Paper I. Contrary to the increased activity seen when retesting in both Paper I and II, retesting in adult male subjects consist-ently decrease activity (Magara et al., 2015; Meyerson et al., 2006; Roman and Colombo, 2009; Roman et al., 2007).

The MCSF and the reference tests In Paper I, adolescent males were tested in the MCSF and one additional be-havioral test: the EPM, two OF variants or the SPB test, at 5 and 6 weeks of age. The performance in the MCSF was analyzed compared to the other ex-ploration-based tests and in relation to the SPB separately.

The relationship between the MCSF parameters (X) and parameters from the EPM and OF tests (Y) was analyzed in a PLS analysis; the cumulative R2 values indicated high coverage of the variability in the data (R2X=0.639, R2Y=0.520, n=36, 4 components) while the Q2 was poor (-0.087) due to the high amount of missing data in Y since each individual only had data from one of the complementary behavioral tests. Despite this, the model showed a good correspondence between the X and Y parameters, as indicated by the relation-ship between the latent variables of the X and Y spaces (t[1] and u[1], respec-tively, Figure 11). Furthermore, the behavioral types were still identifiable in the analysis score plot and in the relationship between the latent variables (Fig-ure 11).

Figure 11. Correlation between the latent variables t[1] and u[1] representing the in-ternal relationship between X (MCSF parameters) and Y (EPM and OF parameters) (n=36). p<0.001, r=0.57, Spearman rank order correlation.

57

The PLS-DA guided selection of parameters for conventional correlative anal-ysis. Duration per visit in the center, DCR, corridors and hurdle were not cor-related to the parameters of the closed arm in the EPM nor the start box in the OF with start box. Likewise, no correlation was found between the bridge and open arm parameters despite the similarity in design and interpretation of these zones. However, the frequency and duration on the bridge in the MCSF showed positive correlations to the total distance travelled and number of rear-ings in the EPM. Activity parameters from the OF without start box showed a relationship to parameters from the center in the MCSF, especially the dura-tion spent in the center, which correlated to total activity, distance, velocity and rearing in the OF. Finally, the consistency of the different activity param-eters and number of fecal boli was examined between the MCSF and the EPM or OF tests. The total activity, distance travelled, and number of fecal boli did not correlate between the MCSF and the other tests. However, velocity in the MCSF and the OF showed a positive correlation, as did rearing in the MCSF and OF, and MCSF and EPM.

That rearing, a species-typical behavior, correlated well between the tests is in line with previous studies; in adult rats rearing in the OF and elevated zero maze were correlated (Blokland et al., 2012) and in adult mice rearing showed positive correlations between the EPM, OF and light/dark box tests (O’Leary et al., 2013). However, locomotion has also been reported to corre-late between tests in adult mice (O’Leary et al., 2013) and adolescent rats (Acevedo et al., 2014) which was not seen here. This could be due to method-ological differences (e.g., different testing times and/or inter-test intervals) or alternatively, that these behaviors reflect different behavioral qualities in the MCSF compared to the reference tests, due to the varied design of the MCSF arena. Taken together, it seems that the behavior in the MCSF corresponds to the behavior in the other exploratory tests but not in a linear way, i.e., the same measurement in the different tests does not correlate with each other directly. This highlights the differences arising when the animal is given more choices to express different behaviors and the importance of basing behavioral inter-pretations in the MCSF on several parameters.

In contrast to the relationship between the MCSF and the other exploration-based tests, the covariance between behavior in the MCSF and behavior in the SPB test was low. This is, however, not entirely surprising as environmental exploration and social behaviors are often described as opposing entities, and in adolescent rats a novel environment even inhibits the expression of SPB (Vanderschuren et al., 1995). Additionally, social play is very reciprocal and the performance in the SPB test is thus a product depending equally on both play partners whereas the MCSF performance is driven by the internal moti-vation to explore the environment of each individual. To balance these factors and successfully detangle any relationship between exploration and social be-haviors is a methodological challenge, but any overt relationship between ex-ploration in the MCSF and SPB in adolescence can be excluded.

58

Links to early-life stress The effects of early-life stress, modeled by prolonged MS, were studied in Paper II and III. The main results are presented below in ontological order.

Early adolescent HPA axis reactivity HPA axis reactivity was assessed in males and females at 3 weeks of age (Pa-per III). Females had generally higher corticosterone than males and further analysis was made with males and females separately. Among adolescent males (Figure 12A), MS360 males had higher corticosterone levels at baseline and at T120 than MS15 males. Over time, both MS15 and MS360 males in-creased their corticosterone levels in response to challenge (T0) and decreased after recovery (T120). However, only MS15 males decreased to a level below baseline while corticosterone levels at T120 in MS360 males were comparable to their baseline levels.

Among adolescent females (Figure 12B), analysis did not show any effect or interaction with rearing condition. Post hoc analysis in the whole female cohort, revealed a normal corticosterone response with increased levels di-rectly after challenge (T0) and a decrease after recovery (T120) to levels be-low baseline.

Figure 12. Adolescent HPA axis reactivity in MS15 and MS360 A) males (n=11 and 16, respectively) and B) females (n=14 and 11, respectively) in Paper III. Samples collected immediately before (baseline) and after (T0) 30 min isolation, and after re-covery in the home cage (T120). Data presented as median with upper and lower quartiles. *p<0.05 compared to MS15 males (post hoc Mann-Whitney U test); BBB p<0.001 compared to baseline, 000 <0.001 compared to T0 (post hoc nparLD).

Higher HPA axis activity in adult females relative to males is well established (Goel et al., 2014; Sze and Brunton, 2019). However, the HPA axis was not fully developed at the age of testing in Paper III (Brown and Spencer, 2013;

59

Burke and Miczek, 2014), but sex-dependent differences have been docu-mented throughout development (Goel et al., 2014; Panagiotakopoulos and Neigh, 2014; Patchev et al., 1999; Romeo et al., 2016). That prolonged MS has no influence on adolescent female HPA axis reactivity, with the use of repeated sampling in the same individual, has not been demonstrated previ-ously. Unchanged HPA axis reactivity has since been reported in adult females comparing prolonged MS and AFR (Gehrand et al., 2018). The rearing condi-tion-dependent effect seen in the adolescent males, is in line with a previous study examining HPA axis activity in adolescent males subjected to prolonged MS compared to AFR (Veenema and Neumann, 2009). However, it is in con-trast to studies in adult males were baseline corticosterone is consistently not affected by rearing condition (Biggio et al., 2014; Gehrand et al., 2018; Plotsky and Meaney, 1993; Pryce et al., 2001; Roman et al., 2006), although exemptions from this have been reported (Biggio et al., 2018; Marais et al., 2008; Odeon et al., 2017).

Early adolescent behavioral profiles The first MCSF trial in Paper II at 4 weeks of age, revealed only a few differ-ences depending on rearing condition in male offspring. MS360 and MS15 males behaved differently in the center zone of the MCSF (lower duration, duration per visit, and percental duration and frequency) compared to AFR males. MS360 males had higher duration per visit on the bridge than both MS15 and AFR males, and had lower distance travelled than AFR males. No difference in the type scores were observed among the rearing conditions.

The behavioral effects of both short and prolonged MS, as assessed in the MCSF, seem to be limited compared to AFR in male rats when assessed in early adolescence, this is in line with what has previously been reported (Palm et al., 2013). The few differences detected were enriched among parameters from the center zone where MS males, independent of separation length, had decreased activity. This can be due to higher avoidance of open areas in MS males or, as there were no differences among central circle parameters, de-creased thigmotactic behavior in MS males compared to AFR males. The avoidance theory can be partially supported by a study assessing behavior with the EPM shortly after weaning which found increased latency to first visit the open arms and fewer visits to the open arms in MS360 males compared to AFR while no differences were seen relative MS15 males (Ploj et al., 2002).

Adolescent social play behavior In the SPB test at 4 weeks of age in Paper III, male offspring had overall higher frequencies of pouncing and pinning than female offspring. A rearing condi-tion-dependent effect was found for number of pinning actions in male off-spring, where MS360 males had higher frequency than MS15 males and

60

MS360 females. In the number of pouncing actions, MS360 males had higher frequency than MS360 females only. No sex- or rearing condition-dependent differences were detected in the latency to first performed pounce or pin.

Higher SPB activity in males than in females is consistent with the litera-ture, although somewhat dependent on the method of testing (for review, see Auger and Olesen, 2009) and choice of animal strain (Himmler et al., 2014; Northcutt and Nwankwo, 2018). Previous studies examining SPB after pro-longed MS have found differing results. Two studies reported only minor dif-ferences in SPB after prolonged MS with no effect on the number of pouncing actions, the probability to respond with a pin or number of pins in males sub-jected to short or prolonged MS compared to AFR (Arnold and Siviy, 2002), and in males and females subjected to prolonged MS compared to AFR (Zimmerberg and Sageser, 2011). Another study reported sex-dependent changes in SPB after prolonged MS compared to AFR; MS males had de-creased number of pounces while MS females had decreased number of pins (Muhammad and Kolb, 2011). In an adolescent resident/intruder paradigm, males subjected to prolonged MS performed more pounces and were subjected to fewer pinning actions than AFR males, while the number of initiated pins did not differ (Veenema and Neumann, 2009). That the results vary between studies can probably be attributed to the high methodological variability that exists in studies examining SPB. When studying adolescent animals, the age at testing is likely to have a major influence on the obtained results, as the rate and composition of SPB changes throughout the juvenile and adolescent pe-riod (Auger and Olesen, 2009).

The increase in pinning among MS360 males, seen in this study, is contra-dictory in light of the elevated corticosterone levels, as measured in the HPA axis reactivity test. Studies using other means of corticosterone elevation, e.g. neonatal corticosterone administration (Meaney et al., 1982) or acute stress (Vanderschuren et al., 1995), have been shown to reduce the amount of SPB in adolescence.

Early adult behavioral profiles The second MCSF trial in Paper II at 9 weeks of age, revealed even fewer differences than the first trial. MS15 males had lower duration per visit on the slope than MS360 and AFR males, MS360 males performed more SAPs than MS15 males and a higher proportion of MS360 males urinated in the arena compared to AFR males.

The behavioral effects of both short and prolonged MS seem to be limited compared to AFR in male rats when assessed in early adulthood. The few dif-ferences detected were among parameters associated with risk assessment, which previously has been seen to be affected by prolonged maternal separa-tion in adult males (Roman et al., 2006).

61

Links to alcohol Behavioral and neurobiological links to alcohol were studied in Paper IV and V with three perspectives: 1) the effect of selective breeding for alcohol consumption, 2) voluntary alcohol consumption and 3) the effect of acute ex-posure to alcohol.

Behavioral profiles after selective breeding for high or low alcohol consumption The adolescent, alcohol-naïve behavioral profile was assessed at 4 weeks of age among lines selectively bred for high or low alcohol consumption (Paper IV). The effect of sex on the behavioral profile was low and further analysis was performed with sex collapsed. The effect of line within each pair and within each selection criterion varied substantially (Figure 13).

Comparing P and NP rats (Figure 13A), higher frequency of visits to the corridors, hurdle, slope, bridge entrance and bridge were seen in P relative NP rats, which also led to a higher total activity in P relative NP rats. P rats also had shorter total and average duration in the corridors and DCR, while spend-ing more time on the bridge with a shorter average visit time than NP rats. On the other hand, HAD1 and LAD1 animals (Figure 13A) had highly similar behavioral profiles as differences were only evident for a few parameters. HAD2 and LAD2 animals (Figure 13A) instead showed differences mostly among center and central circle parameters. HAD2 rats had lower activity in the central circle than LAD2 rats and had lower frequency of visits to the cen-ter while having higher average visit time there relative LAD2 rats.

When comparing the high alcohol-consuming lines (Figure 13B), P rats differed markedly from the two HAD replicate lines by having higher fre-quency and lower average visit time to the majority of the zones, indicating higher activity in P than HAD1 and HAD2 rats. P rats were more active in the central circle than either HAD line which, together with differences in the bridge parameters, indicate increased risk-associated behavior in P relative HAD rats. The differences between the HAD lines were not as pronounced but more marked than in previous studies (Hwang et al., 2004; Overstreet et al., 1997; Rodd et al., 2004; Viglinskaya et al., 1995). Overall, P rats had the highest activity and risk-associated behavior of the high consuming lines while HAD2 rats had the lowest activity and high shelter-seeking behavior.

Among the low alcohol-consuming lines (Figure 13C), the differences were not as prominent as among the high consuming lines. The two LAD replicate lines only showed a few differences while they differed compared to the NP rats. The frequency of visits to the center, central circle, corridors, DCR and hurdle were higher in NP rats relative both LAD1 and LAD2 rats, leading to a higher total activity among NP rats. These and additional differences indi-cate higher activity and exploration in NP rats relative the LAD replicate lines.

62

Figure 13. Heatmap displaying effects sizes for significant differences comparing A) the high and low alcohol-consuming lines within each selectively bred pair, B) the three high consuming lines and C) the three low consuming lines in Paper IV. P n=40; NP n=23; HAD1 n=37; LAD1 n=23; HAD2 n=40; LAD2 n=24. n.s., non-significant.

The results indicate no common behavioral profile for neither the high nor low alcohol-consumption selection criterion in adolescent animals. That selec-tively bred lines differ in their behavior has previously been reported (Overstreet et al., 1997; Roman et al., 2012) but the precise findings herein differ from previous reports that mostly have studied adult animals. The dif-fering findings may thus be age-dependent and it is possible that different be-havioral phenotypes are expressed at different time-points during develop-ment.

63

Adult P rats have been reported to have both increased anxiety-like behav-ior (Hwang et al., 2004; Pandey et al., 2005; Stewart et al., 1993; Zhang et al., 2010) and increased locomotion or novelty-seeking behavior (Badishtov et al., 1995; McKinzie et al., 2002; Nowak et al., 2000; Roman et al., 2012). The current findings in adolescent animals settles with the latter interpretation but an explanation unifying these disperse findings is lacking. The HAD replicate lines have previously been found to be similar both within and between the replicates (Hwang et al., 2004; Overstreet et al., 1997; Rodd et al., 2004; Viglinskaya et al., 1995), which match the current findings in replicate 1 and between the LAD lines. However, the HAD2 rats differed from this with marked low activity and high shelter-seeking behavior in the MCSF. This may, as previously mentioned, be an age-dependent effect that has not previ-ously been discovered or it might be a newly arisen phenotype in the sepa-rately kept breeding stocks.

Adolescent behavior and later alcohol consumption In Paper V, one of the relationships of interest was behavioral characteristics preceding different levels of voluntary alcohol consumption in a cohort of out-bred male and female rats. As such, even though behavioral testing preceded assessment of alcohol consumption, the drinking behavior is described first followed by its links to the alcohol-naïve behavioral profile. The behavioral profiles were assessed in the MCSF at 6 weeks of age, followed by eight weeks of voluntary alcohol intake between 7 and 14 weeks of age.

Voluntary alcohol consumption in late adolescence into adulthood Females had higher intake of alcohol than males the last three weeks of the experiment, after approximately 12 weeks of age. When examining the devel-opment of alcohol intake over the eight weeks of access, females were largely stable in their alcohol intake as only a decrease from week 1 to 2 was detected. Among the males, the intake was stable the first five weeks, then the intake decreased in week 6–8 in relation to week 4.

A division based on overall alcohol intake was made by ranking, separate by sex, the average daily intake for each week and summing the weekly rank into a cumulative rank (Spoelder et al., 2015). A tertiary split then divided the individuals into high-, intermediate- and low alcohol drinkers (HD, ID and LD respectively; HD and LD n=7/sex/group, ID n=6/sex).

The alcohol intake in males divided into drinking subgroups is shown in Figure 14A. Male HD rats had higher alcohol intake than LD males all weeks except week 1, and ID males had higher alcohol intake than LD males weeks 1, 2 and 4. There was no difference between HD and ID males at any week. The intake was stable over the weeks in males in all three drinking subgroups when relating week 1 and 4 to all subsequent weeks, except for ID and LD males at weeks 6 where the intake was lower than in week 4.

64

The alcohol intake in females divided into drinking subgroups is shown in Figure 14B. Female HD rats had higher alcohol intake than LD females at weeks 2, 3 and 6–8; ID females had higher alcohol intake week 6 and 8 than LD females. There was no difference between HD and ID females at any week. The intake was stable over the weeks in females in all three drinking subgroups when relating week 1 and 4 to all subsequent weeks, except for ID females at week 2 where the intake was lower than in week 1.

Figure 14. Alcohol intake in A) males and B) females from late adolescence into adulthood divided into HD, ID and LD subgroups in Paper V (n=6–7/group/sex). Data presented as medians with upper and lower quartiles. *p<0.05, **<0.01 com-pared to LD, °<0.05 compared to males within the same group (post hoc Mann-Whitney U test); @<0.05 compared to week 1 within the same group, §<0.05 com-pared to week 4 within the same group (post hoc Wilcoxon’s test).

Individual variation in alcohol intake and separation into high and low drink-ing subgroups have previously been demonstrated both in adolescent (Fernán-dez et al., 2020; Labots et al., 2018) and adult rats (Labots et al., 2018; Momeni and Roman, 2014; Spoelder et al., 2015). This procedure has been shown to be useful for studies of initiation of, and adaptation to, excessive alcohol consumption (Carnicella et al., 2014). The intake levels in the HD group hovers around 5 g/kg, although somewhat lower for males than females, which is similar to what has been used for selection of high alcohol consumers in selective breeding programs (McBride et al., 2014). This indicates that this group could be interesting for further studies of alcohol consumption in ado-lescence.

65

Higher alcohol intake in adolescents relative adults (Spear, 2018) and in adult females relative males (Becker and Koob, 2016) is well documented in the literature even though exceptions to this have also been reported (Labots et al., 2018 (higher intake in adulthood relative adolescence); Schramm-Sapyta et al., 2014 (no sex difference in intake)). Compared to previous stud-ies using the modified intermittent alcohol access paradigm in adults, the lev-els of alcohol intake in adolescence in Paper V seem slightly higher than what have been reported for adult males (Momeni and Roman, 2014; Scott et al., 2020) and females (Lundberg et al., 2017; Scott et al., 2020). Mixed results are reported regarding when sex-differences in alcohol intake emerge, some suggest that the difference is present already during adolescence (Walker et al., 2008), other that it emerges first in late adolescence (Marco et al., 2017) or adulthood (Bell et al., 2011) and even that the relationship is reversed in adolescence relative what is seen in adulthood (Vetter-O’Hagen et al., 2009). In the present results, there seem to be a synergistic effect between the emer-gence of sex- and age-dependent differences; as the male intake declines from week 6, the sex difference with higher intake in females appears. An earlier or more sustained decline in males than in females in the gradual shift from ad-olescent to adult-like alcohol intake levels may thus be the mechanism behind sex-dependent difference emerging with adulthood, at least with the current modified intermittent alcohol access paradigm.

Behavioral profiles preceding voluntary alcohol consumption Overall analysis of the behavior in the MCSF indicated a limited influence of sex with a moderate effect of drinking group. Further results are thus discussed with sex collapsed. The type score exploration, but not shelter seeking, showed a relationship with level of alcohol consumption (Figure 15A); explo-ration was higher in HD than in ID animals while LD animals did not differ from the other groups. Among the behavioral parameters, HD animals had a higher total activity than ID animas, driven by higher number of visits to the center, corridors, hurdle and slope in HD relative ID animals. HD animals also had longer distance moved and higher average velocity in the arena as well as higher number of rearings than ID animals. Moreover, HD animals had longer distance moved and higher average velocity in the center, spent more time in the hurdle and performed more nose pokes than the ID animals. Relative to LD animals, HD rats performed more rearings and visits to the center, and visited the DCR fewer times percentage-wise. Lastly, LD rats visited the cor-ridors more times percentage-wise than ID animals, and had lower average visit time on the bridge.

Together, these differences indicate that a behavioral profile with high ac-tivity and exploration can predict which animals will consume more alcohol on a voluntary basis. There were more differences comparing the HD to the ID group, while fewer differences were seen in relation to the LD animals.

66

Previous studies examining adult behavior in the MCSF and links to vol-untary alcohol intake have found high risk-assessment behavior (Momeni et al., 2014) and high thigmotactic behavior (Momeni and Roman, 2014) char-acterizing high alcohol consuming males. The differing behavioral correlates in adolescent and adult animals are likely due to age-dependent differences in the performance in the MCSF as discussed above (see Repeated testing and/or effect of age in the MCSF). The present results imply that neither high risk-taking behavior nor high emotionality, approximated by high shelter-seeking behavior in the MCSF, characterize high alcohol consumption in adolescent males and females. These are the typically identified behavioral correlates to high alcohol consumption (Spanagel, 2009) and may indicate unexplored be-havioral differences between adolescent and adult rats. A hypothesis is that high activity, and especially exploration, during adolescence could be viewed as an adolescent risk-associated behavior that is associated with high volun-tary alcohol intake as seen in the HD animals.

Figure 15. Type scores from the MCSF and links to alcohol. Mid-adolescent males and females divided by A) their later voluntary alcohol intake (HD, ID or LD, n=6–7/group/sex) and B) alcohol dose (1.0, 0.5 or 0.0 g/kg i.p., n=12/dose/sex) in Paper V. Data presented as medians with upper and lower quartiles. +p<0.05 comparing HD and ID animals with sex collapsed, °<0.05 comparing males and females within the same dose (post hoc Mann-Whitney U test).

67

Effect of alcohol exposure on adolescent behavioral profiles In Paper V, the second relationship of interest was behavioral effects of alco-hol exposure in adolescent males and females. The behavioral profiles were assessed in the MCSF after an i.p. injection of 0.0, 0.5 or 1.0 g/kg alcohol at 6 weeks of age.

The omnibus analysis of the parameters of the MCSF indicated an overall sex-dependent effect on the behavioral response after acute alcohol exposure as there were notable effects of sex and sex×dose while the influence of alco-hol dose alone was limited. Among the type scores (Figure 15B), an effect of sex was seen for exploration where females had higher scores. For shelter seeking, there were effects on the sex and sex×dose levels; females had gen-erally lower scores driven by a lower shelter seeking score in the control fe-males given 0.0 g/kg alcohol i.p. relative control males.

Among the behavioral parameters, females given the highest dose (1.0 g/kg) spent less time, both in seconds and percentage-wise, in the corridors, center and hurdle than the 0.5 and 0.0 g/kg dose groups. At the highest dose, females also had lower average visit time to the corridors and hurdle than fe-males in the low dose group and controls. Additional effects of dose among the females were a higher proportion of high dose females visiting the central circle compared to the low dose females and fewer rearings in the high dose compared to the control dose females. In males, no effect of dose on the zone durations was found. Males given the highest alcohol dose had, in parallel to the females, lower average duration per visit to the corridors and hurdle than control males, and for the average duration in the corridors also relative to males given 0.5 g/kg alcohol.

Sex-dependent differences were enriched with lowering alcohol dose and were thus most pronounced between control dose males and females. At the control dose (0.0 g/kg), females and males differed in 15 additional behavioral parameters apart from the shelter seeking score; this included higher total ac-tivity, proportion visiting the central circle, number of rearings and visits to the central circle, bridge entrance and bridge in control females relative con-trol males. At the low dose (0.5 g/kg), females visited the central circle more, had longer duration in the hurdle and a lower proportion defecated in the arena than in the low dose males. Lastly, at the highest dose (1.0 g/kg) the proportion of females visiting the central circle was higher and the proportion defecating was lower than in the high dose males.

Together, these differences indicate that females have a moderate effect of acute alcohol exposure up to 1.0 g/kg, reacting with decreased time spent in zones associated with exploration, while males show a limited effect of alco-hol exposure at the studied doses. Sex-dependent differences were most pro-nounced in the control group where females had higher activity and risk-asso-ciated behavior compared to males.

68

The overall small effect of acute alcohol exposure on the behavioral profile was surprising. The doses used have previously been shown to have readily detectable effects on adult behavior in the MCSF (Karlsson and Roman, 2016) and effects on social (Varlinskaya and Spear, 2002) and anxiety-like behavior (Acevedo et al., 2014) in adolescence. As mentioned above (see Sex-differ-ences in adolescent behavior in the MCSF), one hypothesis is that there is a sex-dependent effect of the i.p. injection itself, this may obscure the alcohol-induced effects, and the fact that that the sex differences are enriched in the 0.0 g/kg dose indicates this.

Effect of alcohol exposure on endogenous opioid peptides Furthermore, in Paper V, the acute effect of alcohol exposure on the levels of two endogenous opioid peptides were assessed in males and females. The pep-tide levels were assessed after an i.p. injection of 0.0, 0.5 or 1.0 g/kg alcohol at 7 weeks of age. A limited impact on the endogenous opioid system was revealed. Only the level of MEAP in the amygdala showed an alcohol-depend-ent effect (Figure 16A) which was independent of sex. In animals exposed to 1.0 g/kg alcohol, the MEAP level was lower than in animals administered 0.5 g/kg alcohol. Neither of the alcohol doses differed from the saline control.

Sex-dependent differences were found in MEAP levels (Figure 16B) in the hypothalamus, nucleus accumbens and VTA, where females had consistently higher levels than males. Contrary, in the BNST, the level of dynorphin B was lower in females than in males (Figure 16C).

Figure 16. Levels of endogenous opioid peptides after alcohol exposure in Paper V. Levels analyzed by alcohol exposure×sex, result presented on the level of significant effect. A) MEAP by alcohol exposure (n=22/dose), B) MEAP by sex and C) dy-norphin B by sex (n=34–36/sex). Data presented as median with 95% confidence in-terval. °p<0.05 comparing 1.0 g/kg and 0.5 g/kg; *<0.05, **<0.01 comparing males and females (post hoc Tukey HSD test). DynB, dynorphin B; HT, hypothalamus; NAc, nucleus accumbens.

69

Previous studies in adolescent animals investigating alcohol-induced effects following voluntary intake (Palm et al., 2013) and social isolation during al-cohol intake (Palm and Nylander, 2014b) have found more pronounced alco-hol-induced effects on endogenous opioid peptide levels. Specifically, the de-crease in MEAP levels in the amygdala at the high (1.0 g/kg) compared to the low (0.5 g/kg) alcohol dose is the opposite of what have been detected after voluntary alcohol consumption in adolescent (Palm and Nylander, 2014b; Palm et al., 2013) and adult (Gustafsson et al., 2007) male rats. This discrep-ancy may be explained by the different routes of administration (Palm and Nylander, 2016); acute administration in the present study compared to vol-untary alcohol intake in previous studies. Long-term voluntary alcohol intake may alter the MEAP synthesis and thereby have more long-term effects on the neurochemical balance (Palm and Nylander, 2016; Trigo et al., 2010). The acute exposure, on the other hand, may trigger a release of MEAP that is rap-idly metabolized, leading to the detected lower levels after 1.0 g/kg compared to 0.5 g/kg in the amygdala. However, different studies disagree on the time-course for enkephalin release after alcohol exposure and it seems to be both dose- and region-dependent (Jarjour et al., 2009; Méndez et al., 2010).

The detected sex-differences in MEAP and dynorphin B levels are novel findings. Both these classes of endogenous opioid peptides are involved in emotional behaviors: enkephalins are associated with positive, reinforcing ef-fects and dynorphins with negative, aversive effects when released in the lim-bic system (Palm and Nylander, 2016). This could further explain the lack of alcohol-induced behavioral findings in males and the above hypothesis of a sex-dependent effect of the i.p. injection itself as the reduced MEAP and in-creased dynorphin B levels in the males could suggest a greater negative and aversive response of the i.p. injection in the males than in the females. Further supporting this hypothesis is the reported decrease in Met- and Leu-enkepha-lin and dynorphin A levels in saline-injected (subcutaneous) compared with untreated adult male rats (Nylander et al., 1995b).

Blood alcohol concentrations after alcohol exposure Lastly, in Paper V, the influence of sex and administration route on BACs after alcohol exposure was examined. BACs were assessed 30 and 120 minutes after alcohol exposure through i.g. or i.p. administration of 0.5, 1.0 or 1.5 g/kg in males and females at 4 weeks of age.

Thirty minutes after alcohol administration, the BACs in the animals that received alcohol i.g. were lower than the i.p. group in all doses in males and females, except in females given 0.5 g/kg alcohol. After 120 minutes, the males receiving alcohol i.g. did not differ from their counterparts in the i.p. group. On the other hand, females given 1.5 g/kg alcohol i.g. had lower BACs after 120 minutes than the same dose i.p., while the two lower doses of alcohol resulted in higher BAC levels at 120 minutes after i.g. than i.p. administration.

70

Males showed a clear relationship between dose and BAC after both i.g. and i.p. administration, as did the females receiving alcohol through i.p. injec-tion. However, females receiving alcohol i.g. showed a blunted dose-depend-ent relationship as the two higher doses did not differ, the reason behind this is unclear but some undetected methodological error cannot be excluded. This also influenced the analysis of sex-dependent differences in BAC after i.g. administration, resulting in a more pronounced difference between males and females after i.p. injection. Lower BACs were detected in females than in males when measured 30 minutes after i.p. administration in all doses.

Information on BACs after administration of low to moderate doses is lim-ited and these results provide new information on route- and sex-dependent differences among adolescent male and female rats. The lower BACs after i.g. administration may be due to small differences in absorption between i.g. and i.p. administration leading to temporally shifted dose-response curves. In adults, peak-BACs have been reached after 30–40 minutes after both i.g. and i.p. administration (Crippens et al., 1999; Robinson et al., 2002) suggesting detection of near maximum BACs in the current result. However, BACs in adult females have been shown to peak up to 20 minutes earlier than in males after i.g. administration (Robinson et al., 2002). If this is the same in adoles-cent animals and if it is due to differences shared by absorption after both i.g. and i.p. administration, this could explain the sex-differences seen at 30 minutes after administration.

General discussion

Links to early-life stress The behavioral effects of MS vary considerably between studies using a wide variety of behavioral tests assessing exploration and locomotion in, primarily, male offspring in both adolescence and adulthood. Prolonged MS has been associated with as disparate behavioral characteristics as impulsiveness/hy-peractivity (Brake et al., 2004; Colorado et al., 2006) and high anxiety- and/or depression-like behavior (Jin et al., 2018; Kalinichev et al., 2002; Muhammad and Kolb, 2011), but no or small behavioral effects have also been reported (Biggio et al., 2018; Farkas et al., 2009; Marmendal et al., 2004; Shalev and Kafkafi, 2002). Findings also vary when SPB have been assessed in adoles-cence; subtle sex-specific effects akin to what was seen in Paper III have been reported, although the exact differences differed (Arnold and Siviy, 2002; Zimmerberg and Sageser, 2011). Additionally, both decreased (Muhammad and Kolb, 2011) and increased (Veenema and Neumann, 2009) levels of SPB have been reported in male offspring after prolonged MS.

71

These discrepancies are likely due to a combination of several factors: the different behavioral tests, and variants of them, used (Hånell and Marklund, 2014), differences in the age when outcomes were assessed, especially in ad-olescence when behavior changes rapidly (Auger and Olesen, 2009; Sturman and Moghaddam, 2011) and methodological differences in the MS protocols (Lehmann and Feldon, 2000; Nylander and Roman, 2012, 2013; Pryce and Feldon, 2003; Vetulani, 2013). However, in a series of experiments using a consistent MS paradigm with separations throughout the pre-weaning period, as used in this thesis, minor basal behavioral effects are a consistent finding in male offspring, independent of testing age and behavioral test (Lundberg et al., 2017; Palm et al., 2013; Ploj et al., 2002; Roman et al., 2006). Although, differences in alcohol consumption (Comasco et al., 2015; Daoura and Nylander, 2011; Daoura et al., 2011; Gustafsson and Nylander, 2006; Ploj et al., 2003) and neurobiology (Bendre et al., 2015; Comasco et al., 2015; Granholm et al., 2017; Gustafsson et al., 2007, 2008; Palm and Nylander, 2014b; Palm et al., 2013) have been reported using the same paradigm. This is in line with the working hypothesis that this type of prolonged MS (daily 360-minute separations throughout the pre-weaning period) leads to a pheno-type predisposed to develop stress-related disorders but not overt changes in physiology and behavior without additional provocations in male offspring (Nylander and Roman, 2012; Palm et al., 2013). This is supported by findings of dysregulated HPA axis reactivity, in both young (Paper III) and adult (Roman et al., 2006) MS360 males. There also seems to be a developmental aspect of the dysregulation of the HPA axis; shortly after weaning, MS360 males have increased baseline and recovery levels of corticosterone (Paper III), while in adulthood, the stress-induced corticosterone levels are blunted (Roman et al., 2006).

Why females seem to be less affected by prolonged MS is not fully under-stood and possible explanations range from sex differences in the dam-infant relationship, neuronal or endocrine development or in the neurochemical re-sponse to the separation itself, or a combination thereof. One hypothesis is that females have increased buffering capabilities compared to males and that an even larger perturbation is needed to unmask the effects in females than in males after prolonged MS (Lundberg et al., 2017).

Links to alcohol In Paper IV and V, the alcohol-naïve behavioral profile was assessed in ani-mals selectively bred for high or low alcohol consumption and in outbred an-imals later given access to alcohol on a voluntary basis. A similar behavioral tendency of high activity and exploration was shared between P rats and the HD group in the outbred cohort, although P rats also exhibited increased risk-associated behavior. This indicates one common phenotype for high voluntary

72

alcohol consumption, albeit not the only one as indicated by the differing be-havioral characteristics of the HAD replicate lines in comparison. Which phe-notype associates with high voluntary alcohol intake varies among outbred animals and has been shown to be both strain- and supplier-dependent (Cailhol and Mormède, 2001; Goepfrich et al., 2013; Momeni et al., 2015; Palm et al., 2011a, 2011b; Tjernström and Roman, 2018; Wood et al., 2016). Addition-ally, it is also possible to reach different conclusions using the same strain and vendor as outbred animals by design are heterogeneous and the variability from batch to batch is often unknown. This further emphasizes the variability in etiology of development of excessive alcohol consumption and AUDs.

Overall in Paper V, the effect of alcohol exposure on behavior and endog-enous opioid peptides was small. This can be due to lower sensitivity in ado-lescents than adults to many of the intoxicating and impairing effects of alco-hol (Spear and Swartzwelder, 2014). However, the lower dose range was se-lected as the possible stimulatory and/or anxiolytic effects of alcohol at lower doses were of greater interest than the general sedative effects of higher alco-hol doses, especially for the behavioral assessment in the MCSF. The dose range was also supported by a previous study of adult males in the MCSF (Karlsson and Roman, 2016) and studies on other behavioral effects in ado-lescents (Acevedo et al., 2014; Varlinskaya and Spear, 2002).

The possible sex-dependent effect of the i.p. injection by itself have been discussed in different perspectives in other sections of this thesis. However, that there was only a very limited effect of alcohol on the analyzed MEAP and dynorphin B levels may indicate that the administered doses did not generate a pharmacologically relevant effect, even though an i.p. injection of 1.0 g/kg produced BACs around the criterion for an alcohol binge (>80 mg/dl) (NIAAA National Advisory Council, 2004) in the younger cohort used to as-sess BACs. Although, the sex-dependent differences found regarding the en-dogenous opioid peptides may support an injection-induced effect. The higher dynorphin B levels in the BNST in males relative females may indicate acti-vation of a dynorphin circuitry from the BNST to the amygdala proposed to mediate anxiety-like behavior (Poulin et al., 2009), this could be driving the detected sex-dependent differences. That this effect is mediated by the i.p. in-jection per se and not by the 30 minutes of individual housing post-injection is supported by findings by Granholm et al. (2015), where neither MEAP nor dynorphin B were affected by single housing for 30 minutes in males in early adolescence.

73

Conclusions

These studies provide new information about adolescent behavior and associ-ations to early-life stress and alcohol in males and females as well as outcomes on hormone and brain physiology.

• Paper I and II provide characterization of adolescent behavior and de-

scribe the adolescent behavioral types in the MCSF test. Outlying from the Main type of individuals are types called Explorers, characterized by high activity, and Shelter seekers, characterized by high shelter-seeking as well as low activity and risk-taking behavior. These types provide a tool for further interpretation of adolescent behavior through the rank-or-der procedure type scores. Furthermore, Paper I and II provide new knowledge about effects of age and development on the performance in the MCSF.

• Paper II replicates the finding of minor effects on the behavioral profiles in early adolescence and early adulthood after short and prolonged MS.

• Paper III is, to our knowledge, the first to map individual HPA axis reac-

tivity and peak SPB in adolescent male and female offspring subjected to short versus prolonged MS. The results provide early detectable signs of the systemic impact of prolonged MS and emphasize the importance of sex and development on effects after early-life stress.

• Paper IV corroborates heterogeneity in background for development of

AUDs by not finding a common behavioral ‘pathway’ linked to selective breeding for high alcohol consumption in adolescent rats. Again, the ef-fect of development is emphasized as the findings in adolescents differed from previous results in adults.

• Paper V provides new information about sex-dependent effects on the re-

lationship between alcohol, behavior and neurobiology in adolescence.

Together, these findings emphasize the importance of studying these research questions from a broader perspective on the aspects of age, development and sex as previously unknown effects were uncovered when examining links to early-life stress and alcohol in adolescent males and females.

74

Populärvetenskaplig sammanfattning

Ungdomsperioden är den utvecklingsperiod som infaller mellan barndom och vuxen ålder och utmärks av de förändringar som sker i hela kroppen. Föränd-ringarna som sker är menade att hjälpa individen utvecklas från ett barn, som är beroende av sina föräldrar, till en självständig individ. Under den här peri-oden, samt in i ung vuxen ålder, utvecklas och mognar hjärnan. Vi blir bättre på abstrakt tänkande, planering och bedömning av konsekvenser. Beteendet påverkas också av förändringarna i hjärnan och karakteriseras av ökat riskta-gande, socialt och sensationssökande beteende vilket är typiskt för ungdoms-perioden hos både människor och övriga däggdjur.

På grund av de stora förändringar som sker under ungdomsperioden är in-dividen också mer känslig för yttre omständigheter, så som stress och drogin-tag, vilka kan ge långvariga följder då de kan påverka utvecklingen. Ung-domsperioden är en vanlig debutålder för psykiatriska sjukdomar som ångest, depression, ätstörningar och schizofreni. Även drog- och alkoholintag debu-terar vanligtvis under ungdomsperioden.

I denna avhandling studeras beteendet under ungdomsperioden samt hur tidig stress och alkohol påverkar beteendet. De här faktorerna, och många andra, är inte studerade i lika stor utsträckning under ungdomsperioden som hos fullvuxna individer. Detta trots att ungdomsperioden är särskilt känslig med tanke på utvecklingen som sker.

Dessa frågeställningar kan undersökas på flera olika sätt. Djurexperimen-tella modeller är ett alternativ för att ha detaljerad kontroll över de yttre om-ständigheterna samtidigt som evolutionärt konserverade system möjliggör kartläggning av konsekvenser och mekanismer som kan vägleda bekräftande studier i människa. I denna avhandling har råttor använts, för vilka det finns modeller för att studera bland annat beteende, tidig stress och uppväxtmiljö, frivilligt alkoholintag och alkoholexponering. I dessa modeller är det även möjligt att mäta olika hormoner och signalsubstanser, både i blod och direkt i hjärnan. Av intresse i den här avhandlingen är stressystemet och dess hormo-ner samt de kroppsegna opioiderna som bland annat reglerar aktiviteten i hjär-nans belöningssystem.

Hos råttor pågår ungdomsperioden ungefär från tre till tio veckors ålder. Den här avhandlingen visar att beteendeprofilen under den perioden skiljer sig avsevärt från äldre individers. Skillnaden mellan individer under ungdomspe-rioden visade sig vara stor med tre tydliga grupperingar. En grupp uppvisade en beteendeprofil med hög aktivitet medan en andra, motsatt grupp uppvisade

75

låg aktivitet, mycket skyddssökande och lågt risktagande beteende. Båda de här grupperna skilde sig avsevärt från varandra och från den största gruppe-ringen av individer vars beteendeprofil var mellanliggande de andra två.

Tidig stress är förenat med ökad risk för en mängd olika sjukdomar, psyki-atriska som kroppsliga, senare i livet. Hos råttor är tidig stress något som på-verkar ungarna innan tre veckors ålder, då de fortfarande är beroende av mam-man. Genom att separera dem från mamman under timslånga perioder varje dag kan uppväxtmiljön påverkas för att utgöra en stress för ungarna. I den här avhandlingen studerades utvecklingen av beteendeprofilen, lekbeteendet och stressystemet under ungdomsperioden efter tidig stress. Effekterna på beteen-det var begränsade medan en påverkan på stressystemet hos hanar men inte honor hittades. Ett förändrat stressystem ses vid många sjukdomar och för-ändringen som hittades här kan vara ett tidigt tecken på utveckling mot eller en riskfaktor för någon av dessa sjukdomar. Honor har även tidigare visats vara okänsliga för den här typen av tidig stress; vad som skiljer honorna från hanarna och gör dem motståndskraftiga är inte känt.

I avhandlingen studerades sambandet mellan beteende under ungdomsål-dern och alkohol i tre olika perspektiv: effekten av anlag för högt alkoholintag, frivilligt alkoholintag i normalpopulationen och effekten av alkoholexpone-ring. Risken för att utveckla alkoholberoende är jämnt fördelad mellan arv och miljö, ingen av delarna är dock klarlagd vilket försvårar studier av specifika riskfaktorer. En modell för att studera effekterna av arv är upprättandet av selektivt avlade linjer av råttor där individer som frivilligt dricker mycket eller lite alkohol väljs för avel i flera generationer. På så sätt fås linjer av djur med hög eller låg genetisk benägenhet för alkoholkonsumtion. I en vanlig råttstam finns det en stor variation i hur mycket alkohol varje individ dricker, den va-riationen användes för att studera samband till beteendet innan alkoholdebut. Slutligen, användes fasta doser av alkohol för att studera hur beteendet påver-kas och hur hjärnan svarar på alkohol. Dos-responsförhållanden mellan alko-hol och beteende, kroppsegna opioider eller blodalkoholnivåer undersöktes.

Ingen entydig beteendeprofil hittades för råttor med nedärvt anlag för högt alkoholintag. Istället fanns olika beteendemässiga skillnader bland de olika avelslinjerna. Detta liknar vad som återfinns kliniskt, olika underliggande arvs- och miljöfaktorer kan leda fram till samma kliniska bild av hög alkohol-konsumtion. I normalpopulationen karakteriserades de individer, både honor och hanar, med högst alkoholintag av en beteendeprofil med hög aktivitet och högt utforskande beteende. Den beteendeprofilen har vissa liknelser med pro-filen hos en av linjerna som är selektivt avlad för högt alkoholintag. Alkohol-exponering vid de låga till medelhöga doser som användes i den här avhand-lingen visade sig ha små effekter på beteende och kroppsegna opioider, och vissa av effekterna var könsberoende. Honor, men inte hanar, uppvisade en beteendeprofil med sänkt utforskande beteende efter exponering av alkohol. Alkoholexponeringen hade begränsad effekt på de kroppsegna opioidnivåerna i hjärnan, dock fanns det flera, tidigare okända, skillnader i nivåerna mellan

76

honor och hanar. Även dos-responsförhållandet för blodalkoholnivåer visade sig vara könsberoende.

Sammantaget bidrar resultaten i den här avhandlingen till ökad kunskap om beteendet under ungdomsperioden hos råttor samt ålders-, utvecklings- och könsberoende skillnader i effekterna efter tidig stress och för olika aspekter av alkohol. Detta är omständigheter som är viktiga att ta hänsyn till för att iden-tifiera riskfaktorer för att utveckla stress- och alkoholrelaterad problematik samt i förebyggande och behandlande arbete.

77

Acknowledgements

The studies included in this thesis were carried out in the Neuropharmacology, Addiction and Behavior research group at the Department of Pharmaceutical Biosciences, Uppsala University, Sweden, with support from the Uppsala University Behavioral Facility (UUBF), Faculty of Medicine and Pharmacy, Uppsala University. The studies were supported by grants from the Swedish Research Council, the Alcohol Research Council of the Swedish Alcohol Re-tail Monopoly, the Facias Foundation, the Gahlin Foundation and the National Institute of Health/National Institute on Alcohol Abused and Alcoholism.

The visit to the Department of Psychiatry, Indiana School of Medicine, Indi-ana University–Purdue University Indianapolis, USA resulting in collection of the data in Paper IV, was supported by Bertil Göranssons resestipendium för unga alkoholforskare from the Swedish Alcohol Retail Monopoly as well as Rönnows and CR Leffmans travel stipends from the Disciplinary Domain of Medicine and Pharmacy, Uppsala University. Travel stipends from Apotekarsocieteten (Apotekare CD Carlssons stiftelse, IFs stiftelse inklusive Källrotstipendiet and Elisabeth och Alfred Ahlqvist stiftelse) funded confer-ences and international courses.

First and foremost, to my super supervisor Erika. Thank you for your advice, your patience and your guidance. You’ve allowed me free range to explore the world of science, both deep dives and long flights. Thank you for believing in me from the start and all the way through! Better praise than saying “I would do it all over again” can’t be given and that’s 100% true!

My co-supervisor Ingrid, to have your support and balance have been so val-uable. Even when running between meetings you make the time to share your knowledge and come with comments and news on our research and the rest of the university world.

Thanks to all past and present members of the research group; small and sweet, NAB is truly a family. Nikita, it has been a pleasure to share an office with you the last half of my PhD. We’ve shared anything from work to life’s weird indulgences of gin, cats and tattoos. I’m amazed how you power through your long, long studies; good things will come for those who wait (and work hard)! Sara P, you were my first roommate and taught me all the basics of being a

78

PhD student: work hard, say no when you need to and take your vacation! Shima and Linnea, the power duo of laughter, thank you for all the help throughout the years. Joep, our newest addition, you bring new perspectives to the group and additional love for gin! Teaching is a big part of this family and making sure it runs as smooth as possible is Anne-Lie, Fredrik, Viktoria and Lena B. Through thick and thin you make it possible! To Anne-Lie, thank you for our talks about this and that, your door has always been open. To all other new and old PhD students sharing the course labs: Jonathan, Anneli, Mikaela, Frida, Oly, Ida and brand-new Kimia and Erica. We make it through together, and with help of Fredrik of course! To Carin, keeper of our rules and measures, thank you for the help with the RIAs. The newest addition, Ottil, soon I’ll come dance with you! A huge thanks to my rodent whisperers, Marita and Lova, anything you don’t know about rats isn’t worth knowing. Lova, our Baloo, you are truly missed in the corridor!

Without UUBF no behavioral studies and without behavioral studies no thesis. A big thanks to Åsa KG, Thomas and Sara EL at the platform. Sharing knowledge, fixing unresponsive computers and anything else, you’re always a great help, support and company!

The work in this thesis wouldn’t have been possible without Bengt Meyerson, the creator of the MCSF. Not here to see me through, though I hope I’ve made you proud!

A big thanks to Rick Bell for granting me the opportunity to work in his lab at IUPUI in Indianapolis. I learned and grew so much during the visit there! Thanks to the lovely people at the lab: Erin, Caron, Tom and Jason, you made my work easy and fun. And to all the others that made my work and stay abroad one of the best experiences of my life! The nights at brick house duel-ing pianos will forever be remembered.

To my co-authors here back home. Cecilia, without you no Paper I and then this whole experience would have been completely different. My, my partner-in-crime, all the work we put in and all the fun we had along the way. Parisa, the last minute, hardworking hero. Everything will turn out fine, I promise! Martina, the turns data take, thank you for the work you put in all the way back.

Sometime the time is not enough, and then I have been lucky to have a couple of heroic students come along with extra hands. Nelson, I can’t thank you enough for the hours you put down on the recordings for Paper IV. Helene, even though your work didn’t make it to the thesis, you did amazingly well and that work still continues. Matilda, my first student, invaluable, curious and brave. You kept track of my progress over the years and see, I made it!

79

To all the ones working to make work easier for everyone else. Thank you to the stars in the animal facility, you are amazing and all this work would not have been possible without you. Thank you to the hard-working technical and administrative staff at the department, any issue big or small, posters, bills, packages, contracts and anything else is no problem for you! The same goes for the workshop staff at BMC! You built all the MCSF arenas for us and helped ship one across the Atlantic. It’s not easy finding packaging for an arena a meter squared, so you built a custom-made freight create to hold it!

Obviously, I can’t keep to just research and teaching at work. A by-election somehow turned into four years in the board of FDR. Thanks to all my fellow board members: Ilse, Anders, Henrik J, Annelie, Rebecka, Sara R, Mikaela, Rekha, Sara MH, Albin, Karin, Jonas, Andrea, Sofia Z, Oliver, Sohan and Henrik BN. With you I have learnt so much, been angry, happy and anything in-between. Keep up the good work!

To all PhD students and others I’ve had the pleasure working with in some form or another over the years, this experience wouldn’t have been the same without you: Erik N, Anna, Erika B, Alfhild, Mathias, Erik B, Lena N, Oskar, Lisa, Patrik, Svante, Arshi, Laura, Johanna, Erica, Maria, Megha, the list goes on and on…

To Mikaela, you’re truly an inspiration, I don’t know how you do it all! To Milon, thank you for everything, I really can’t be more specific than that!

For much-needed escape during these years, thanks to all my lovely board game, roleplay and LARP friends. Both in and out of the games, I can trust that you will be there.

To my family, your love and support makes me who I am. Mom and dad, you taught me that everything is possible. Mom, you made the teacher part of me and dad, from you I get my stubbornness. My sisters, Ida and Åsa, you’re always there whether living one block away or across the country, three is truly a magic number! Tommy, your caring and your daring drives me to be more kind and courageous. I love you, in all possible and impossible ways <5

To each and every one of you, thank you so very much, I couldn’t have done this without you!

With love, Stina

80

References

Abel, J.L., and Rissman, E.F. (2012). Location, location, location: Genetic regulation of neural sex differences. Rev. Endocr. Metab. Disord. 13, 151–161.

Acevedo, M.B., Nizhnikov, M.E., Molina, J.C., and Pautassi, R.M. (2014). Relation-ship between ethanol-induced activity and anxiolysis in the open field, elevated plus maze, light-dark box, and ethanol intake in adolescent rats. Behav. Brain Res. 265, 203–215.

Alberts, J.R. (2005). Infancy. In The Behavior of the Laboratory Rat: A Handbook with Tests, (Oxford university press), pp. 266–277.

American Psychiatric Association (2013). Diagnostic and Statistical Manual of Men-tal Disorders (DSM-5®) (American Psychiatric Pub).

Andersson, L. (2012). How selective sweeps in domestic animals provide new insight into biological mechanisms. J. Intern. Med. 271, 1–14.

Andrews, M.H., Wood, S.A., Windle, R.J., Lightman, S.L., and Ingram, C.D. (2012). Acute glucocorticoid administration rapidly suppresses basal and stress-induced hypothalamo-pituitary-adrenal axis activity. Endocrinology 153, 200–211.

Arnold, J.L., and Siviy, S.M. (2002). Effects of neonatal handling and maternal sepa-ration on rough-and-tumble play in the rat. Dev. Psychobiol. 41, 205–215.

Auger, A.P., and Olesen, K.M. (2009). Brain sex differences and the organisation of juvenile social play behaviour. J. Neuroendocrinol. 21, 519–525.

Badishtov, B.A., Overstreet, D.H., Kashevskaya, O.P., Viglinskaya, I.V., Kampov-Polevoy, A.B., Seredenin, S.B., and Halikas, J.A. (1995). To drink or not to drink: Open field behavior in alcohol-preferring and-nonpreferring rat strains. Physiol. Behav. 57, 585–589.

Bambino, K., and Chu, J. (2017). Zebrafish in Toxicology and Environmental Health. In Current Topics in Developmental Biology, K.C. Sadler, ed. (Academic Press), pp. 331–367.

Barnett, S.A. (2005). Ecology. In The Behavior of the Laboratory Rat: A Handbook with Tests, I.Q. Whishaw, and B. Kolb, eds. (Oxford University Press), pp. 15–24.

Barnett, S.A. (2007). The Rat: A Study in Behavior (Transaction Publishers). Bateson, P., and Laland, K.N. (2013). Tinbergen’s four questions: an appreciation and

an update. Trends Ecol. Evol. 28, 712–718. Becker, H.C. (2012). Animal Models of Excessive Alcohol Consumption in Rodents.

In Behavioral Neurobiology of Alcohol Addiction, W.H. Sommer, and R. Spanagel, eds. (Springer Berlin Heidelberg), pp. 355–377.

Becker, J.B., and Koob, G.F. (2016). Sex Differences in Animal Models: Focus on Addiction. Pharmacol. Rev. 68, 242–263.

Becker, J.B., Prendergast, B.J., and Liang, J.W. (2016). Female rats are not more var-iable than male rats: a meta-analysis of neuroscience studies. Biol. Sex Differ. 7.

Becker, J.B., McClellan, M.L., and Reed, B.G. (2017). Sex differences, gender and addiction. J. Neurosci. Res. 95, 136–147.

81

Beery, A.K., and Zucker, I. (2011). Sex bias in neuroscience and biomedical research. Neurosci. Biobehav. Rev. 35, 565–572.

Belin, D., Belin-Rauscent, A., Everitt, B.J., and Dalley, J.W. (2016). In search of pre-dictive endophenotypes in addiction: insights from preclinical research. Genes Brain Behav. 15, 74–88.

Bell, R.L., Rodd, Z.A., Lumeng, L., Murphy, J.M., and McBride, W.J. (2006). The alcohol-preferring P rat and animal models of excessive alcohol drinking. Addict. Biol. 11, 270–288.

Bell, R.L., Rodd, Z.A., Smith, R.J., Toalston, J.E., Franklin, K.M., and McBride, W.J. (2011). Modeling binge-like ethanol drinking by peri-adolescent and adult P rats. Pharmacol. Biochem. Behav. 100, 90–97.

Bell, R.L., Sable, H.J.K., Colombo, G., Hyytia, P., Rodd, Z.A., and Lumeng, L. (2012). Animal models for medications development targeting alcohol abuse us-ing selectively bred rat lines: Neurobiological and pharmacological validity. Phar-macol. Biochem. Behav. 103, 119–155.

Bell, R.L., Rodd, Z.A., Engleman, E.A., Toalston, J.E., and McBride, W.J. (2014). Scheduled access alcohol drinking by alcohol-preferring (P) and high-alcohol-drinking (HAD) rats: Modeling adolescent and adult binge-like drinking. Alcohol 48, 225–234.

Bell, R.L., Hauser, S., Rodd, Z.A., Liang, T., Sari, Y., McClintick, J., Rahman, S., and Engleman, E.A. (2016). A Genetic Animal Model of Alcoholism for Screen-ing Medications to Treat Addiction. In International Review of Neurobiology, R.L. Bell, and S. Rahman, eds. (Academic Press), pp. 179–261.

Bell, R.L., Hauser, S.R., Liang, T., Sari, Y., Maldonado-Devincci, A., and Rodd, Z.A. (2017). Rat animal models for screening medications to treat alcohol use disor-ders. Neuropharmacology 122, 201–243.

Belzung, C., and Lemoine, M. (2011). Criteria of validity for animal models of psy-chiatric disorders: focus on anxiety disorders and depression. Biol. Mood Anxiety Disord. 1, 9.

Bendre, M., Comasco, E., Nylander, I., and Nilsson, K.W. (2015). Effect of voluntary alcohol consumption on Maoa expression in the mesocorticolimbic brain of adult male rats previously exposed to prolonged maternal separation. Transl. Psychiatry 5, e690.

Berardo, L.R., Fabio, M.C., and Pautassi, R.M. (2016). Post-weaning Environmental Enrichment, But Not Chronic Maternal Isolation, Enhanced Ethanol Intake during Periadolescence and Early Adulthood. Front. Behav. Neurosci. 10.

Bick, J., and Nelson, C.A. (2016). Early Adverse Experiences and the Developing Brain. Neuropsychopharmacol. Off. Publ. Am. Coll. Neuropsychopharmacol. 41, 177–196.

Biggio, F., Pisu, M.G., Garau, A., Boero, G., Locci, V., Mostallino, M.C., Olla, P., Utzeri, C., and Serra, M. (2014). Maternal separation attenuates the effect of ad-olescent social isolation on HPA axis responsiveness in adult rats. Eur. Neuro-psychopharmacol. J. Eur. Coll. Neuropsychopharmacol.

Biggio, F., Talani, G., Locci, V., Pisu, M.G., Boero, G., Ciarlo, B., Grayson, D.R., and Serra, M. (2018). Low doses of prenatal ethanol exposure and maternal sep-aration alter HPA axis function and ethanol consumption in adult male rats. Neu-ropharmacology 131, 271–281.

Bikovski, L., Robinson, L., Konradsson-Geuken, A., Kullander, K., Viereckel, T., Winberg, S., Roman, E., and Tsoory, M. (2020). Lessons, insights and newly de-veloped tools emerging from behavioral phenotyping core facilities. J. Neurosci. Methods 334, 108597.

82

Blanchard, D.C., and Blanchard, R.J. (1988). Ethoexperimental Approaches to the Biology of Emotion. Annu. Rev. Psychol. 39, 43–68.

Blanchard, M.M., Mendelsohn, D., and Stamp, J.A. (2009). The HR/LR model: Fur-ther evidence as an animal model of sensation seeking. Neurosci. Biobehav. Rev. 33, 1145–1154.

Blanchard, R.J., Hebert, M.A., Ferrari, P., Palanza, P., Figueira, R., Blanchard, D.C., and Parmigiani, S. (1998). Defensive behaviors in wild and laboratory (Swiss) mice: the mouse defense test battery. Physiol. Behav. 65, 201–209.

Blizard, D.A., Takahashi, A., Galsworthy, M.J., Martin, B., and Koide, T. (2007). Test standardization in behavioural neuroscience: a response to Stanford. J. Psycho-pharmacol. (Oxf.) 21, 136–139.

Blokland, A., ten Oever, S., van Gorp, D., van Draanen, M., Schmidt, T., Nguyen, E., Krugliak, A., Napoletano, A., Keuter, S., and Klinkenberg, I. (2012). The use of a test battery assessing affective behavior in rats: Order effects. Behav. Brain Res. 228, 16–21.

Bodnar, R.J. (2020). Endogenous opiates and behavior: 2017. Peptides 124, 170223. de Boer, S.F., Buwalda, B., and Koolhaas, J.M. (2017). Untangling the neurobiology

of coping styles in rodents: Towards neural mechanisms underlying individual differences in disease susceptibility. Neurosci. Biobehav. Rev. 74, 401–422.

Brake, W.G., Zhang, T.Y., Diorio, J., Meaney, M.J., and Gratton, A. (2004). Influence of early postnatal rearing conditions on mesocorticolimbic dopamine and behav-ioural responses to psychostimulants and stressors in adult rats. Eur. J. Neurosci. 19, 1863–1874.

Braun, A.A., Skelton, M.R., Vorhees, C.V., and Williams, M.T. (2011). Comparison of the elevated plus and elevated zero mazes in treated and untreated male Spra-gue–Dawley rats: Effects of anxiolytic and anxiogenic agents. Pharmacol. Bio-chem. Behav. 97, 406–415.

Brown, R.W. (2005). Adolescence. In The Behavior of the Laboratory Rat: A Hand-book with Tests, (Oxford university press), pp. 278–286.

Brown, G.R., and Spencer, K.A. (2013). Steroid hormones, stress and the adolescent brain: A comparative perspective. Neuroscience 249, 115–128.

Budde, M.D., Shah, A., McCrea, M., Cullinan, W.E., Pintar, F., and Stemper, B.D. (2013). Primary Blast Traumatic Brain Injury in the Rat: Relating Diffusion Ten-sor Imaging and Behavior. Front. Neurol. 4.

Burke, A.R., and Miczek, K.A. (2014). Stress in adolescence and drugs of abuse in rodent models: Role of dopamine, CRF, and HPA axis. Psychopharmacology (Berl.) 231, 1557–1580.

Burn, C.C. (2008). What is it like to be a rat? Rat sensory perception and its implica-tions for experimental design and rat welfare. Appl. Anim. Behav. Sci. 112, 1–32.

Buwalda, B., Geerdink, M., Vidal, J., and Koolhaas, J.M. (2011). Social behavior and social stress in adolescence: a focus on animal models. Neurosci. Biobehav. Rev. 35, 1713–1721.

Cahill, L. (2006). Why sex matters for neuroscience. Nat. Rev. Neurosci. 7, 477–484. Cailhol, S., and Mormède, P. (2001). Sex and strain differences in ethanol drinking:

effects of gonadectomy. Alcohol. Clin. Exp. Res. 25, 594–599. CAN (2019). Skolelevers drogvanor 2019 (Centralförbundet för alkohol- och narko-

tikaupplysning). Carnicella, S., Ron, D., and Barak, S. (2014). Intermittent ethanol access schedule in

rats as a preclinical model of alcohol abuse. Alcohol 48, 243–252. Carroll, M.E., and Lynch, W.J. (2016). How to study sex differences in addiction us-

ing animal models. Addict. Biol. 21, 1007–1029.

83

Casey, B.J., Jones, R.M., and Hare, T.A. (2008). The Adolescent Brain. Ann. N. Y. Acad. Sci. 1124, 111–126.

Cederbaum, A.I. (2012). Alcohol metabolism. Clin. Liver Dis. 16, 667–685. Christensson-Nylander, I., Nyberg, F., Ragnarsson, U., and Terenius, L. (1985). A

general procedure for analysis of proenkephalin B derived opioid peptides. Regul. Pept. 11, 65–76.

Claassen, V. (1994). Intraperitoneal Drug Administration. In Techniques in the Be-havioral and Neural Sciences, (Elsevier), pp. 46–58.

Clark, R.G., Chambers, G., Lewin, J., and Robinson, I.C. a. F. (1986). Automated repetitive microsampling of blood: growth hormone profiles in conscious male rats. J. Endocrinol. 111, 27–35.

Colombo, G., Lobina, C., Carai, M.A.M., and Gessa, G.L. (2006). Phenotypic char-acterization of genetically selected Sardinian alcohol-preferring (sP) and -non-preferring (sNP) rats. Addict. Biol. 11, 324–338.

Colorado, R.A., Shumake, J., Conejo, N.M., Gonzalez-Pardo, H., and Gonzalez-Lima, F. (2006). Effects of maternal separation, early handling, and standard facility rearing on orienting and impulsive behavior of adolescent rats. Behav. Processes 71, 51–58.

Comasco, E., Todkar, A., Granholm, L., Nilsson, K.W., and Nylander, I. (2015). Αl-pha 2a-Adrenoceptor Gene Expression and Early Life Stress-Mediated Propen-sity to Alcohol Drinking in Outbred Rats. Int. J. Environ. Res. Public. Health 12, 7154–7171.

Cooke, B.M., and Weathington, J.M. (2014). Human and animal research into sex-specific effects of child abuse. Horm. Behav. 65, 416–426.

Coppens, C.M., de Boer, S.F., and Koolhaas, J.M. (2010). Coping styles and behav-ioural flexibility: towards underlying mechanisms. Philos. Trans. R. Soc. B Biol. Sci. 365, 4021–4028.

Crabbe, J.C. (2014). Use of animal models of alcohol-related behavior. In Handbook of Clinical Neurology, E.V. Sullivan, and A. Pfefferbaum, eds. (Elsevier), pp. 71–86.

Crabbe, J.C., Harris, R.A., and Koob, G.F. (2011). Preclinical studies of alcohol binge drinking. Ann. N. Y. Acad. Sci. 1216, 24–40.

Crews, F.T., Vetreno, R.P., Broadwater, M.A., and Robinson, D.L. (2016). Adoles-cent Alcohol Exposure Persistently Impacts Adult Neurobiology and Behavior. Pharmacol. Rev. 68, 1074–1109.

Crews, F.T., Robinson, D.L., Chandler, L.J., Ehlers, C.L., Mulholland, P.J., Pandey, S.C., Rodd, Z.A., Spear, L.P., Swartzwelder, H.S., and Vetreno, R.P. (2019). Mechanisms of Persistent Neurobiological Changes Following Adolescent Alco-hol Exposure: NADIA Consortium Findings. Alcohol. Clin. Exp. Res. 43, 1806–1822.

Crippens, D., White, M.L., George, M.A., Jaworski, J.N., Brunner, L.J., Lancaster, F.E., and Gonzales, R.A. (1999). Gender Differences in Blood Levels, But Not Brain Levels, of Ethanol in Rats. Alcohol. Clin. Exp. Res. 23, 414–420.

Csala, A., and Zwinderman, A.H. (2019). Multivariate Statistical Methods for High-Dimensional Multiset Omics Data Analysis. In Computational Biology, H. Husi, ed. (Brisbane (AU): Codon Publications), p.

Daoura, L., and Nylander, I. (2011). The response to naltrexone in ethanol-drinking rats depends on early environmental experiences. Pharmacol. Biochem. Behav. 99, 626–633.

Daoura, L., Haaker, J., and Nylander, I. (2011). Early environmental factors differen-tially affect voluntary ethanol consumption in adolescent and adult male rats. Al-cohol. Clin. Exp. Res. 35, 506–515.

84

Daskalakis, N.P., Bagot, R.C., Parker, K.J., Vinkers, C.H., and de Kloet, E.R. (2013). The three-hit concept of vulnerability and resilience: Toward understanding ad-aptation to early-life adversity outcome. Psychoneuroendocrinology 38, 1858–1873.

Demin, K.A., Lakstygal, A.M., Volgin, A.D., de Abreu, M.S., Genario, R., Alpyshov, E.T., Serikuly, N., Wang, D., Wang, J., Yan, D., et al. (2020). Cross-species Anal-yses of Intra-species Behavioral Differences in Mammals and Fish. Neuroscience S0306452219308905.

Doremus-Fitzwater, T.L., Varlinskaya, E.I., and Spear, L.P. (2010). Motivational sys-tems in adolescence: possible implications for age differences in substance abuse and other risk-taking behaviors. Brain Cogn. 72, 114–123.

van Duijvenvoorde, A.C.K., Peters, S., Braams, B.R., and Crone, E.A. (2016). What motivates adolescents? Neural responses to rewards and their influence on ado-lescents’ risk taking, learning, and cognitive control. Neurosci. Biobehav. Rev. 70, 135–147.

Dyr, W., and Kostowski, W. (2008). Warsaw high-preferring (WHP) and Warsaw low-preferring (WLP) lines of rats selectively bred for high and low voluntary ethanol intake: Preliminary phenotypic characterization. Alcohol 42, 161–170.

Ehlert, U. (2013). Enduring psychobiological effects of childhood adversity. Psycho-neuroendocrinology 38, 1850–1857.

Enoch, M.-A. (2006). Genetic and Environmental Influences on the Development of Alcoholism. Ann. N. Y. Acad. Sci. 1094, 193–201.

Ernst, M. (2014). The triadic model perspective for the study of adolescent motivated behavior. Brain Cogn. 89, 104–111.

ESPAD Group (2016). ESPAD Report 2015. Results from the European School Sur-vey Project on Alcohol and other drugs.

Fantino, E., and Stolarz-Fantino, S. (2012). Operant Conditioning. In Encyclopedia of Human Behavior (Second Edition), V.S. Ramachandran, ed. (San Diego: Aca-demic Press), pp. 749–756.

Farkas, J., Reglodi, D., Gaszner, B., Szogyi, D., Horvath, G., Lubics, A., Tamas, A., Frank, F., Besirevic, D., and Kiss, P. (2009). Effects of maternal separation on the neurobehavioral development of newborn Wistar rats. Brain Res. Bull. 79, 208–214.

Fernández, M.S., Bellia, F., Ferreyra, A., Chiner, F., Jiménez García, A.M., D’Adda-rio, C., and Pautassi, R.M. (2020). Short-term selection for high and low ethanol intake during adolescence exerts lingering effects in stress-induced ethanol drink-ing and yields an anxiety-prone phenotype. Behav. Brain Res. 380, 112445.

Fleming, A.S., and Rosenblatt, J.S. (1974). Maternal behavior in the virgin and lac-tating rat. J. Comp. Physiol. Psychol. 86, 957–972.

Fonio, E., Golani, I., and Benjamini, Y. (2012). Measuring behavior of animal models: faults and remedies. Nat. Methods 9, 1167–1170.

Fontana, B.D., Mezzomo, N.J., Kalueff, A.V., and Rosemberg, D.B. (2018). The de-veloping utility of zebrafish models of neurological and neuropsychiatric disor-ders: A critical review. Exp. Neurol. 299, 157–171.

Fritz, C.O., Morris, P.E., and Richler, J.J. (2012). Effect size estimates: Current use, calculations, and interpretation. J. Exp. Psychol. Gen. 141, 2–18.

Froehlich, J.C. (2010). What Aspects of Human Alcohol Use Disorders Can Be Mod-eled Using Selectively Bred Rat Lines? Subst. Use Misuse 45, 1727–1741.

Gehrand, A.L., Hoeynck, B., Jablonski, M., Leonovicz, C., Cullinan, W.E., and Raff, H. (2018). Programming of the Adult HPA Axis After Neonatal Separation and Environmental Stress in Male and Female Rats. Endocrinology 159, 2777–2789.

85

Gibbs, R.A., Weinstock, G.M., Metzker, M.L., Muzny, D.M., Sodergren, E.J., Scherer, S., Scott, G., Steffen, D., Worley, K.C., Burch, P.E., et al. (2004). Ge-nome sequence of the Brown Norway rat yields insights into mammalian evolu-tion. Nature 428, 493–521.

Godoy, L.D., Rossignoli, M.T., Delfino-Pereira, P., Garcia-Cairasco, N., and de Lima Umeoka, E.H. (2018). A Comprehensive Overview on Stress Neurobiology: Basic Concepts and Clinical Implications. Front. Behav. Neurosci. 12.

Goel, N., Workman, J.L., Lee, T.T., Innala, L., and Viau, V. (2014). Sex Differences in the HPA Axis. In Comprehensive Physiology, (John Wiley & Sons, Inc.), pp. 1121–1155.

Goepfrich, A.A., Gluch, C., Friemel, C.M., and Schneider, M. (2013). Behavioral dif-ferences in three Wistar Han rat lines for emotional reactivity, cognitive pro-cessing and ethanol intake. Physiol. Behav. 110–111, 102–108.

Granholm, L. (2018). Stress, Drugs and Neuroscience : Neurobiological Effects of Social Stressors and Drug Exposure in Young and Adolescent Rats. PhD disser-tation. Acta Universitatis Upsaliensis , Uppsala University.

Granholm, L., Roman, E., and Nylander, I. (2015). Single housing during early ado-lescence causes time-, area- and peptide-specific alterations in endogenous opi-oids of rat brain. Br. J. Pharmacol. 172, 606–614.

Granholm, L., Todkar, A., Bergman, S., Nilsson, K., Comasco, E., and Nylander, I. (2017). The expression of opioid genes in non-classical reward areas depends on early life conditions and ethanol intake. Brain Res. 1668, 36–45.

Granholm, L., Segerström, L., and Nylander, I. (2018). Episodic Ethanol Exposure in Adolescent Rats Causes Residual Alterations in Endogenous Opioid Peptides. Front. Psychiatry 9.

Gray, K.M., and Squeglia, L.M. (2018). Research Review: What have we learned about adolescent substance use? J. Child Psychol. Psychiatry 59, 618–627.

Gustafsson, L., and Nylander, I. (2006). Time-Dependent Alterations in Ethanol In-take in Male Wistar Rats Exposed to Short and Prolonged Daily Maternal Sepa-ration in a 4-Bottle Free-Choice Paradigm. Alcohol. Clin. Exp. Res. 30, 2008–2016.

Gustafsson, L., Zhou, Q., and Nylander, I. (2007). Ethanol-induced effects on opioid peptides in adult male Wistar rats are dependent on early environmental factors. Neuroscience 146, 1137–1149.

Gustafsson, L., Oreland, S., Hoffmann, P., and Nylander, I. (2008). The impact of postnatal environment on opioid peptides in young and adult male Wistar rats. Neuropeptides 42, 177–191.

Hall, C.S. (1934). Emotional behavior in the rat. I. Defecation and urination as measures of individual differences in emotionality. J. Comp. Psychol. 18, 385–403.

Hallberg, M. (2015). Neuropeptides: Metabolism to Bioactive Fragments and the Pharmacology of Their Receptors. Med. Res. Rev. 35, 464–519.

Hallberg, M., and Nyberg, F. (2003). Neuropeptide conversion to bioactive frag-ments--an important pathway in neuromodulation. Curr. Protein Pept. Sci. 4, 31–44.

Haller, J., Harold, G., Sandi, C., and Neumann, I.D. (2014). Effects of Adverse Early-Life Events on Aggression and Anti-Social Behaviours in Animals and Humans. J. Neuroendocrinol. 26, 724–738.

Hånell, A., and Marklund, N. (2014). Structured evaluation of rodent behavioral tests used in drug discovery research. Front. Behav. Neurosci. 8, 252.

Hansen, C., and Spuhler, K. (1984). Development of the National Institutes of Health Genetically Heterogeneous Rat Stock. Alcohol. Clin. Exp. Res. 8, 477–479.

86

van Hasselt, F.N., Tieskens, J.M., Trezza, V., Krugers, H.J., Vanderschuren, L.J.M.J., and Joëls, M. (2012). Within-litter variation in maternal care received by individ-ual pups correlates with adolescent social play behavior in male rats. Physiol. Be-hav. 106, 701–706.

Heilig, M., Thorsell, A., Sommer, W.H., Hansson, A.C., Ramchandani, V.A., George, D.T., Hommer, D., and Barr, C.S. (2010). Translating the neuroscience of alco-holism into clinical treatments: From blocking the buzz to curing the blues. Neu-rosci. Biobehav. Rev. 35, 334–344.

Heim, C., Shugart, M., Craighead, W.E., and Nemeroff, C.B. (2010). Neurobiological and psychiatric consequences of child abuse and neglect. Dev. Psychobiol. 52, 671–690.

Heinrichs, S.C., and Koob, G.F. (2004). Corticotropin-Releasing Factor in Brain: A Role in Activation, Arousal, and Affect Regulation. J. Pharmacol. Exp. Ther. 311, 427–440.

Herman, J.P. (2009). Stress Response: Neural and Feedback Regulation of the HPA Axis. In Encyclopedia of Neuroscience, L.R. Squire, ed. (Academic Press), pp. 505–510.

Himmler, S.M., Modlinska, K., Stryjek, R., Himmler, B.T., Pisula, W., and Pellis, S.M. (2014). Domestication and diversification: a comparative analysis of the play fighting of the Brown Norway, Sprague-Dawley, and Wistar laboratory strains of (Rattus norvegicus). J. Comp. Psychol. Wash. DC 1983 128, 318–327.

Hwang, B.H., Stewart, R., Zhang, J.-K., Lumeng, L., and Li, T.-K. (2004). Cortico-tropin-releasing factor gene expression is down-regulated in the central nucleus of the amygdala of alcohol-preferring rats which exhibit high anxiety: a compar-ison between rat lines selectively bred for high and low alcohol preference. Brain Res. 1026, 143–150.

Jans, J.E., and Woodside, B.C. (1990). Nest temperature: effects on maternal behav-ior, pup development, and interactions with handling. Dev. Psychobiol. 23, 519–534.

Jarjour, S., Bai, L., and Gianoulakis, C. (2009). Effect of Acute Ethanol Administra-tion on the Release of Opioid Peptides from the Midbrain Including the Ventral Tegmental Area. Alcohol. Clin. Exp. Res. 33, 1033–1043.

Jin, S., Zhao, Y., Jiang, Y., Wang, Y., Li, C., Zhang, D., Lian, B., Du, Z., Sun, H., and Sun, L. (2018). Anxiety-like behaviour assessments of adolescent rats after re-peated maternal separation during early life. Neuroreport 29, 643–649.

Kalinichev, M., Easterling, K.W., Plotsky, P.M., and Holtzman, S.G. (2002). Long-lasting changes in stress-induced corticosterone response and anxiety-like behav-iors as a consequence of neonatal maternal separation in Long–Evans rats. Phar-macol. Biochem. Behav. 73, 131–140.

Karlsson, O., and Roman, E. (2016). Dose-dependent effects of alcohol administration on behavioral profiles in the MCSF test. Alcohol 50, 51–56.

Karlsson, O., Roman, E., and Brittebo, E.B. (2009). Long-term Cognitive Impair-ments in Adult Rats Treated Neonatally with β-N-Methylamino-L-Alanine. Tox-icol. Sci. 112, 185–195.

van Kerkhof, L.W.M., Damsteegt, R., Trezza, V., Voorn, P., and Vanderschuren, L.J.M.J. (2013). Functional integrity of the habenula is necessary for social play behaviour in rats. Eur. J. Neurosci. 38, 3465–3475.

Khan, L. (2019). Puberty: Onset and Progression. Pediatr. Ann. Thorofare 48, 141–145.

Kim, J.J. (2001). Neural Basis of Classical Conditioning. In International Encyclope-dia of the Social & Behavioral Sciences, N.J. Smelser, and P.B. Baltes, eds. (Per-gamon), pp. 1946–1951.

87

de Kloet, E.R., Joëls, M., and Holsboer, F. (2005). Stress and the brain: from adapta-tion to disease. Nat. Rev. Neurosci. 6, 463–475.

Knop, J., Joëls, M., and van der Veen, R. (2017). The added value of rodent models in studying parental influence on offspring development: opportunities, limita-tions and future perspectives. Curr. Opin. Psychol. 15, 174–181.

Koolhaas, J.M., Boer, S.F. de, Buwalda, B., and Reenen, K. van (2007). Individual Variation in Coping with Stress: A Multidimensional Approach of Ultimate and Proximate Mechanisms. Brain. Behav. Evol. 70, 218–226.

Koolhaas, J.M., Bartolomucci, A., Buwalda, B., de Boer, S.F., Flügge, G., Korte, S.M., Meerlo, P., Murison, R., Olivier, B., Palanza, P., et al. (2011). Stress revis-ited: A critical evaluation of the stress concept. Neurosci. Biobehav. Rev. 35, 1291–1301.

Labots, M., Cousijn, J., Jolink, L.A., Kenemans, J.L., Vanderschuren, L.J.M.J., and Lesscher, H.M.B. (2018). Age-Related Differences in Alcohol Intake and Control Over Alcohol Seeking in Rats. Front. Psychiatry 9.

Ladd, C.O., Huot, R.L., Thrivikraman, K.V., Nemeroff, C.B., Meaney, M.J., and Plot-sky, P.M. (2000). Long-term behavioral and neuroendocrine adaptations to ad-verse early experience. In Progress in Brain Research, E.A.M. and C.B. Saper, ed. (Elsevier), pp. 81–103.

Lai, M.-C., and Huang, L.-T. (2011). Effects of early life stress on neuroendocrine and neurobehavior: mechanisms and implications. Pediatr. Neonatol. 52, 122–129.

Lamberts, R., and Goldsmith, P.C. (1986). Fixation, fine structure, and immunostain-ing for neuropeptides: perfusion versus immersion of the neuroendocrine hypo-thalamus. J. Histochem. Cytochem. 34, 389–398.

Le Merrer, J., Becker, J.A.J., Befort, K., and Kieffer, B.L. (2009). Reward Processing by the Opioid System in the Brain. Physiol. Rev. 89, 1379–1412.

Lebel, C., and Deoni, S. (2018). The development of brain white matter microstruc-ture. NeuroImage 182, 207–218.

Lehmann, J., and Feldon, J. (2000). Long-term biobehavioral effects of maternal sep-aration in the rat: consistent or confusing? Rev. Neurosci. 11, 383–408.

Levine, S. (2002). Regulation of the hypothalamic-pituitary-adrenal axis in the neo-natal rat: the role of maternal behavior. Neurotox. Res. 4, 557–564.

Levine, S., and Lewis, G.W. (1959). Critical period for effects of infantile experience on maturation of stress response. Science 129, 42–43.

Levine, S., Alpert, M., and Lewis, G.W. (1957). Infantile experience and the matura-tion of the pituitary adrenal axis. Science 126, 1347.

Li, T.K., and Lumeng, L. (1984). Alcohol preference and voluntary alcohol intakes of inbred rat strains and the National Institutes of Health heterogeneous stock of rats. Alcohol. Clin. Exp. Res. 8, 485–486.

Lightman, S.L. (2008). The neuroendocrinology of stress: a never ending story. J. Neuroendocrinol. 20, 880–884.

Lo Iacono, L., and Carola, V. (2018). The impact of adolescent stress experiences on neurobiological development. Semin. Cell Dev. Biol. 77, 93–103.

Lucion, A.B., and Bortolini, M.C. (2014). Mother-Pup Interactions: Rodents and Hu-mans. Front. Endocrinol. 5, 17.

Lundberg, S., Abelson, K.S.P., Nylander, I., and Roman, E. (2017). Few long-term consequences after prolonged maternal separation in female Wistar rats. PLOS ONE 12, e0190042.

Lundberg, S., Helene, L., and Roman, E. (2018). Left Alone – Effects of Acute and Prolonged Social Isolation on the Behavioral Profile in Adult Male and Female

88

Wistar Rats. In Measuring Behavior 2018, R. Grant, T. Allen, A. Spink, and M. Sullivan, eds. (Manchester, United Kingdom), pp. 288–289.

Lynn, D.A., and Brown, G.R. (2009). The ontogeny of exploratory behavior in male and female adolescent rats (Rattus norvegicus). Dev. Psychobiol. 51, 513–520.

Maccari, S., Krugers, H.J., Morley-Fletcher, S., Szyf, M., and Brunton, P.J. (2014). The Consequences of Early-Life Adversity: Neurobiological, Behavioural and Epigenetic Adaptations. J. Neuroendocrinol. 26, 707–723.

Macrì, S., and Würbel, H. (2006). Developmental plasticity of HPA and fear responses in rats: a critical review of the maternal mediation hypothesis. Horm. Behav. 50, 667–680.

Magara, S., Holst, S., Lundberg, S., Roman, E., and Lindskog, M. (2015). Altered explorative strategies and reactive coping style in the FSL rat model of depres-sion. Front. Behav. Neurosci. 9.

Maniam, J., Antoniadis, C., and Morris, M.J. (2014). Early-Life Stress, HPA Axis Adaptation, and Mechanisms Contributing to Later Health Outcomes. Front. En-docrinol. 5.

Marais, L., van Rensburg, S.J., van Zyl, J.M., Stein, D.J., and Daniels, W.M.U. (2008). Maternal separation of rat pups increases the risk of developing depres-sive-like behavior after subsequent chronic stress by altering corticosterone and neurotrophin levels in the hippocampus. Neurosci. Res. 61, 106–112.

Marco, E.M., Peñasco, S., Hernández, M.-D., Gil, A., Borcel, E., Moya, M., Giné, E., López-Moreno, J.A., Guerri, C., López-Gallardo, M., et al. (2017). Long-Term Effects of Intermittent Adolescent Alcohol Exposure in Male and Female Rats. Front. Behav. Neurosci. 11.

Marmendal, M., Roman, E., Eriksson, C.J.P., Nylander, I., and Fahlke, C. (2004). Maternal separation alters maternal care, but has minor effects on behavior and brain opioid peptides in adult offspring. Dev. Psychobiol. 45, 140–152.

Mathè, A.A., Stenfors, C., Brodin, E., and Theodorsson, E. (1990). Neuropeptides in brain: Effects of microwave irradiation and decapitation. Life Sci. 46, 287–293.

McBride, W.J., Kimpel, M.W., McClintick, J.N., Ding, Z.-M., Hyytia, P., Colombo, G., Edenberg, H.J., Lumeng, L., and Bell, R.L. (2012). Gene expression in the ventral tegmental area of 5 pairs of rat lines selectively bred for high or low eth-anol consumption. Pharmacol. Biochem. Behav. 102, 275–285.

McBride, W.J., Rodd, Z.A., Bell, R.L., Lumeng, L., and Li, T.-K. (2014). The alcohol-preferring (P) and high-alcohol-drinking (HAD) rats--animal models of alcohol-ism. Alcohol Fayettev. N 48, 209–215.

McIlwain, K.L., Merriweather, M.Y., Yuva-Paylor, L.A., and Paylor, R. (2001). The use of behavioral test batteries: Effects of training history. Physiol. Behav. 73, 705–717.

McKinzie, D.L., McBride, W.J., Murphy, J.M., Lumeng, L., and Li, T.-K. (2002). Effects of amphetamine on locomotor activity in adult and juvenile alcohol-pre-ferring and -nonpreferring rats. Pharmacol. Biochem. Behav. 71, 29–36.

Meaney, M.J., Stewart, J., and Beatty, W.W. (1982). The influence of glucocorticoids during the neonatal period on the development of play-fighting in Norway rat pups. Horm. Behav. 16, 475–491.

Meisch, R.A. (2001). Oral drug self-administration: an overview of laboratory animal studies. Alcohol 24, 117–128.

Méndez, M., Barbosa-Luna, I.G., Pérez-Luna, J.M., Cupo, A., and Oikawa, J. (2010). Effects of acute ethanol administration on methionine–enkephalin expression and release in regions of the rat brain. Neuropeptides 44, 413–420.

89

Meyerson, B.J., Augustsson, H., Berg, M., and Roman, E. (2006). The Concentric Square Field: a multivariate test arena for analysis of explorative strategies. Be-hav. Brain Res. 168, 100–113.

Meyerson, B.J., Jurek, B., and Roman, E. (2013). A rank-order procedure applied to an ethoexperimental behavior model - the Multivariate Concentric Square Field (MCSF) test. J. Behav. Brain Sci. 2013, 350–361.

Miller, G.E., Chen, E., and Zhou, E.S. (2007). If it goes up, must it come down? Chronic stress and the hypothalamic-pituitary-adrenocortical axis in humans. Psychol. Bull. 133, 25–45.

Modlinska, K., and Pisula, W. (2020). The Norway rat, from an obnoxious pest to a laboratory pet. ELife 9, e50651.

Momeni, S., and Roman, E. (2014). Subgroup-dependent effects of voluntary alcohol intake on behavioral profiles in outbred Wistar rats. Behav. Brain Res. 275, 288–296.

Momeni, S., Sharif, M., Agren, G., and Roman, E. (2014). Individual differences in risk-related behaviors and voluntary alcohol intake in outbred Wistar rats. Behav. Pharmacol. 25, 206–215.

Momeni, S., Segerström, L., and Roman, E. (2015). Supplier-dependent differences in intermittent voluntary alcohol intake and response to naltrexone in Wistar rats. Front. Neurosci. 9, 424.

Montgomery, K.C. (1955). The relation between fear induced by novel stimulation and exploratory behavior. J. Comp. Physiol. Psychol. 48, 254–260.

Muhammad, A., and Kolb, B. (2011). Maternal separation altered behavior and neu-ronal spine density without influencing amphetamine sensitization. Behav. Brain Res. 223, 7–16.

Murphy, J.M., Stewart, R.B., Bell, R.L., Badia-Elder, N.E., Carr, L.G., McBride, W.J., Lumeng, L., and Ting-Kai, L. (2002). Phenotypic and Genotypic Charac-terization of the Indiana University Rat Lines Selectively Bred for High and Low Alcohol Preference. Behav. Genet. 32, 363–388.

Nemeroff, C.B. (2016). Paradise Lost: The Neurobiological and Clinical Conse-quences of Child Abuse and Neglect. Neuron 89, 892–909.

NIAAA National Advisory Council (2004). Definition of binge drinking. Niesink, R.J., and Van Ree, J.M. (1989). Involvement of opioid and dopaminergic

systems in isolation-induced pinning and social grooming of young rats. Neuro-pharmacology 28, 411–418.

NobelPrize.org The Nobel Prize in Physiology or Medicine 1973. Noguchi, K., Gel, Y.R., Brunner, E., and Konietschke, F. (2012). nparLD: An R Soft-

ware Package for the Nonparametric Analysis of Longitudinal Data in Factorial Experiments. J. Stat. Softw. 50, 1–23.

Northcutt, K.V., and Nwankwo, V.C. (2018). Sex differences in juvenile play behav-ior differ among rat strains. Dev. Psychobiol. 60, 903–912.

Nowak, K.L., Ingraham, C.M., Mckinzie, D.L., Mcbride, W.J., Lumeng, L., Li, T.-K., and Murphy, J.M. (2000). An Assessment of Novelty-Seeking Behavior in Alcohol-Preferring and Nonpreferring Rats. Pharmacol. Biochem. Behav. 66, 113–121.

Nylander, I., and Roman, E. (2012). Neuropeptides as mediators of the early-life im-pact on the brain; implications for alcohol use disorders. Front. Mol. Neurosci. 5, 77.

Nylander, I., and Roman, E. (2013). Is the rodent maternal separation model a valid and effective model for studies on the early-life impact on ethanol consumption? Psychopharmacology (Berl.) 229, 555–569.

90

Nylander, I., Vlaskovska, M., and Terenius, L. (1995a). Brain dynorphin and enkeph-alin systems in Fischer and Lewis rats: effects of morphine tolerance and with-drawal. Brain Res. 683, 25–35.

Nylander, I., Vlaskovska, M., and Terenius, L. (1995b). The effects of morphine treat-ment and morphine withdrawal on the dynorphin and enkephalin systems in spra-gue-dawley rats. Psychopharmacology (Berl.) 118, 391–400.

Nylander, I., Stenfors, C., Tan-No, K., Mathé, A.A., and Terenius, L. (1997). A com-parison between microwave irradiation and decapitation: basal levels of dy-norphin and enkephalin and the effect of chronic morphine treatment on dy-norphin peptides. Neuropeptides 31, 357–365.

Odeon, M.M., Yamauchi, L., Grosman, M., and Acosta, G.B. (2017). Long-term ef-fects of repeated maternal separation and ethanol intake on HPA axis responsive-ness in adult rats. Brain Res. 1657, 193–201.

O’Leary, T.P., Gunn, R.K., and Brown, R.E. (2013). What are We Measuring When We Test Strain Differences in Anxiety in Mice? Behav. Genet. 43, 34–50.

Oreland, L., Lagravinese, G., Toffoletto, S., Nilsson, K.W., Harro, J., Cloninger, C.R., and Comasco, E. (2017). Personality as an intermediate phenotype for genetic dissection of alcohol use disorder. J. Neural Transm. 1–24.

Oswald, L.M., and Wand, G.S. (2004). Opioids and alcoholism. Physiol. Behav. 81, 339–358.

Overstreet, D.H. (2007). The Open Field Test for two. J. Psychopharmacol. (Oxf.) 21, 140–140.

Overstreet, D.H., Halikas, J.A., Seredenin, S.B., Kampov-Polevoy, A.B., Viglin-skaya, I.V., Kashevskaya, O., Badishtov, B.A., Knapp, D.J., Mormede, P., Ki-ianmaa, K., et al. (1997). Behavioral similarities and differences among alcohol-preferring and -nonpreferring rats: confirmation by factor analysis and extension to additional groups. Alcohol. Clin. Exp. Res. 21, 840–848.

Palanza, P., and Parmigiani, S. (2017). How does sex matter? Behavior, stress and animal models of neurobehavioral disorders. Neurosci. Biobehav. Rev. 76, Part A, 134–143.

Palm, S. (2014). Early Environment, Adolescent Alcohol Drinking and Neurobiolog-ical Responses to Drugs. PhD dissertation. Acta Universitatis Upsaliensis , Upp-sala University.

Palm, S., and Nylander, I. (2014a). Dopamine Release Dynamics Change during Ad-olescence and after Voluntary Alcohol Intake. PLOS ONE 9, e96337.

Palm, S., and Nylander, I. (2014b). Alcohol-induced changes in opioid peptide levels in adolescent rats are dependent on housing conditions. Alcohol. Clin. Exp. Res. 38, 2978–2987.

Palm, S., and Nylander, I. (2016). Alcohol and Endogenous Opioids. In Neuropathol-ogy of Drug Addictions and Substance Misuse, V.R. Preedy, ed. (Academic Press), pp. 401–410.

Palm, S., Roman, E., and Nylander, I. (2011a). Differences in voluntary ethanol con-sumption in Wistar rats from five different suppliers. Alcohol Fayettev. N 45, 607–614.

Palm, S., Hävermark, Å., Meyerson, B.J., Nylander, I., and Roman, E. (2011b). When is a Wistar a Wistar? Behavioral profiling of outbred Wistar rats from five differ-ent suppliers using the MCSF test. Appl. Anim. Behav. Sci. 135, 128–137.

Palm, S., Daoura, L., Roman, E., and Nylander, I. (2013). Effects of rearing conditions on behaviour and endogenous opioids in rats with alcohol access during adoles-cence. PloS One 8, e76591.

91

Panagiotakopoulos, L., and Neigh, G.N. (2014). Development of the HPA axis: Where and when do sex differences manifest? Front. Neuroendocrinol. 35, 285–302.

Pandey, S.C., Zhang, H., Roy, A., and Xu, T. (2005). Deficits in amygdaloid cAMP-responsive element–binding protein signaling play a role in genetic predisposition to anxiety and alcoholism. J. Clin. Invest. 115, 2762–2773.

Panksepp, J., Siviy, S., and Normansell, L. (1984). The psychobiology of play: theo-retical and methodological perspectives. Neurosci. Biobehav. Rev. 8, 465–492.

Pape, H., Rossow, I., and Brunborg, G.S. (2018). Adolescents drink less: How, who and why? A review of the recent research literature. Drug Alcohol Rev. 37, S98–S114.

Parent, C.I., and Meaney, M.J. (2008). The influence of natural variations in maternal care on play fighting in the rat. Dev. Psychobiol. 50, 767–776.

Patchev, V.K., Hayashi, S., Orikasa, C., and Almeida, O.F. (1999). Ontogeny of gen-der-specific responsiveness to stress and glucocorticoids in the rat and its deter-mination by the neonatal gonadal steroid environment. Stress Amst. Neth. 3, 41–54.

Pavlov, P.I. (2010). Conditioned reflexes: An investigation of the physiological activ-ity of the cerebral cortex. First published 1927. Ann. Neurosci. 17, 136–141.

Paxinos, G., and Watson, C. (2007). The Rat Brain in Stereotaxic Coordinates (Aca-demic Press).

Paylor, R., Spencer, C.M., Yuva-Paylor, L.A., and Pieke-Dahl, S. (2006). The use of behavioral test batteries, II: Effect of test interval. Physiol. Behav. 87, 95–102.

Pearson, B.L., Crawley, J.N., Eilam, D., Pentkowski, N.S., and Summers, C.H. (2017). Curiosity as an approach to ethoexperimental analysis: Behavioral neuro-science as seen by students and colleagues of Bob Blanchard. Neurosci. Biobe-hav. Rev. 76, 415–422.

Pellis, S.M., Field, E.F., Smith, L.K., and Pellis, V.C. (1997). Multiple differences in the play fighting of male and female rats. Implications for the causes and func-tions of play. Neurosci. Biobehav. Rev. 21, 105–120.

Pellow, S., Chopin, P., File, S.E., and Briley, M. (1985). Validation of open : closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J. Neuro-sci. Methods 14, 149–167.

Piazza, P.V., Deminiere, J.-M., Le Moal, M., and Simon, H. (1989). Factors That Pre-dict Individual Vulnerability to Amphetamine Self-Administration. Sci. Wash. 245, 1511.

Ploj, K., Roman, E., and Nylander, I. (2002). Effects of maternal separation on brain nociceptin/orphanin FQ peptide levels in male Wistar rats. Pharmacol. Biochem. Behav. 73, 123–129.

Ploj, K., Roman, E., and Nylander, I. (2003). Long-term effects of maternal separation on ethanol intake and brain opioid and dopamine receptors in male Wistar rats. Neuroscience 121, 787–799.

Plotsky, P.M., and Meaney, M.J. (1993). Early, postnatal experience alters hypotha-lamic corticotropin-releasing factor (CRF) mRNA, median eminence CRF con-tent and stress-induced release in adult rats. Brain Res. Mol. Brain Res. 18, 195–200.

Poulin, J.-F., Arbour, D., Laforest, S., and Drolet, G. (2009). Neuroanatomical char-acterization of endogenous opioids in the bed nucleus of the stria terminalis. Prog. Neuropsychopharmacol. Biol. Psychiatry 33, 1356–1365.

Pryce, C.R., and Feldon, J. (2003). Long-term neurobehavioural impact of the post-natal environment in rats: manipulations, effects and mediating mechanisms. Neurosci. Biobehav. Rev. 27, 57–71.

92

Pryce, C.R., Bettschen, D., Bahr, N.I., and Feldon, J. (2001). Comparison of the ef-fects of infant handling, isolation, and nonhandling on acoustic startle, prepulse inhibition, locomotion, and HPA activity in the adult rat. Behav. Neurosci. 115, 71–83.

Quintanilla, M.E., Israel, Y., Sapag, A., and Tampier, L. (2006). The UChA and UChB rat lines: metabolic and genetic differences influencing ethanol intake. Ad-dict. Biol. 11, 310–323.

R Core Team (2020). R: A language and environment for statistical computing (Vi-enna, Austria: R Foundation for Statistical Computing).

Radley, J., Morilak, D., Viau, V., and Campeau, S. (2015). Chronic stress and brain plasticity: Mechanisms underlying adaptive and maladaptive changes and impli-cations for stress-related CNS disorders. Neurosci. Biobehav. Rev. 58, 79–91.

Raison, C.L., and Miller, A.H. (2003). When Not Enough Is Too Much: The Role of Insufficient Glucocorticoid Signaling in the Pathophysiology of Stress-Related Disorders. Am. J. Psychiatry 160, 1554–1565.

Rajalahti, T., and Kvalheim, O.M. (2011). Multivariate data analysis in pharmaceu-tics: A tutorial review. Int. J. Pharm. 417, 280–290.

Robinson, D.L., Brunner, L.J., and Gonzales, R.A. (2002). Effect of gender and es-trous cycle on the pharmacokinetics of ethanol in the rat brain. Alcohol. Clin. Exp. Res. 26, 165–172.

Rodd, Z.A., Bell, R.L., McKinzie, D.L., Webster, A.A., Murphy, J.M., Lumeng, L., Li, T.-K., and McBride, W.J. (2004). Low-Dose Stimulatory Effects of Ethanol During Adolescence in Rat Lines Selectively Bred for High Alcohol Intake. Al-cohol. Clin. Exp. Res. 28, 535–543.

Rodgers, R.J. (2007). More haste, considerably less speed. J. Psychopharmacol. (Oxf.) 21, 141–143.

Roman, E., and Colombo, G. (2009). Lower risk taking and exploratory behavior in alcohol-preferring sP rats than in alcohol non-preferring sNP rats in the multivar-iate concentric square field (MCSF) test. Behav. Brain Res. 205, 249–258.

Roman, E., and Nylander, I. (2005). The impact of emotional stress early in life on adult voluntary ethanol intake-results of maternal separation in rats. Stress Amst. Neth. 8, 157–174.

Roman, E., Gustafsson, L., Berg, M., and Nylander, I. (2006). Behavioral profiles and stress-induced corticosteroid secretion in male Wistar rats subjected to short and prolonged periods of maternal separation. Horm. Behav. 50, 736–747.

Roman, E., Meyerson, B.J., Hyytiä, P., and Nylander, I. (2007). The multivariate con-centric square field test reveals different behavioural profiles in male AA and ANA rats with regard to risk taking and environmental reactivity. Behav. Brain Res. 183, 195–205.

Roman, E., Stewart, R.B., Bertholomey, M.L., Jensen, M.L., Colombo, G., Hyytiä, P., Badia-Elder, N.E., Grahame, N.J., Li, T.-K., and Lumeng, L. (2012). Behav-ioral profiling of multiple pairs of rats selectively bred for high and low alcohol intake using the MCSF test. Addict. Biol. 17, 33–46.

Romeo, R.D., Patel, R., Pham, L., and So, V.M. (2016). Adolescence and the ontog-eny of the hormonal stress response in male and female rats and mice. Neurosci. Biobehav. Rev. 70, 206–216.

Rowland, N.E. (2007). Food or Fluid Restriction in Common Laboratory Animals: Balancing Welfare Considerations with Scientific Inquiry. Comp. Med. 57, 12.

Salvatore, J.E., Gottesman, I.I., and Dick, D.M. (2015). Endophenotypes for Alcohol Use Disorder: An Update on the Field. Curr. Addict. Rep. 2, 76–90.

Sanchis-Segura, C., and Becker, J.B. (2016). Why we should consider sex (and study sex differences) in addiction research. Addict. Biol. 21, 995–1006.

93

Sanchis-Segura, C., and Spanagel, R. (2006). Behavioural assessment of drug rein-forcement and addictive features in rodents: an overview. Addict. Biol. 11, 2–38.

Saunders, J.B., Degenhardt, L., Reed, G.M., and Poznyak, V. (2019). Alcohol Use Disorders in ICD-11: Past, Present, and Future. Alcohol. Clin. Exp. Res. 43, 1617–1631.

Sawyer, S.M., Azzopardi, P.S., Wickremarathne, D., and Patton, G.C. (2018). The age of adolescence. Lancet Child Adolesc. Health 2, 223–228.

Schneider, M. (2008). Puberty as a highly vulnerable developmental period for the consequences of cannabis exposure. Addict. Biol. 13, 253–263.

Schneider, M. (2013). Adolescence as a vulnerable period to alter rodent behavior. Cell Tissue Res. 354, 99–106.

Schramm-Sapyta, N.L., Francis, R., MacDonald, A., Keistler, C., O’Neill, L., and Kuhn, C.M. (2014). Effect of sex on ethanol consumption and conditioned taste aversion in adolescent and adult rats. Psychopharmacology (Berl.) 231, 1831–1839.

Schumann, G., Loth, E., Banaschewski, T., Barbot, A., Barker, G., Büchel, C., Conrod, P.J., Dalley, J.W., Flor, H., Gallinat, J., et al. (2010). The IMAGEN study: reinforcement-related behaviour in normal brain function and psycho-pathology. Mol. Psychiatry 15, 1128–1139.

Scott, H., Tjernström, N., and Roman, E. (2020). Effects of pair-housing on voluntary alcohol intake in male and female Wistar rats. Alcohol.

Segerström, L., Gustavsson, J., and Nylander, I. (2016). Minimizing Postsampling Degradation of Peptides by a Thermal Benchtop Tissue Stabilization Method. Bi-opreservation Biobanking 14, 172–179.

Selye, H. (1950). Stress and the General Adaptation Syndrome. Br. Med. J. 1, 1383–1392.

Shah, M., Chitforoushzadeh, Z., and Janes, K.A. (2016). Statistical data analysis and modeling. In Uncertainty in Biology: A Computational Modeling Approach, L. Geris, and D. Gomez-Cabrero, eds. (Springer), pp. 155–176.

Shalev, U., and Kafkafi, N. (2002). Repeated maternal separation does not alter su-crose-reinforced and open-field behaviors. Pharmacol. Biochem. Behav. 73, 115–122.

Shepherd, J.K., Grewal, S.S., Fletcher, A., Bill, D.J., and Dourish, C.T. (1994). Be-havioural and pharmacological characterisation of the elevated “zero-maze” as an animal model of anxiety. Psychopharmacology (Berl.) 116, 56–64.

Simms, J.A., Steensland, P., Medina, B., Abernathy, K.E., Chandler, L.J., Wise, R., and Bartlett, S.E. (2008). Intermittent access to 20% ethanol induces high ethanol consumption in Long-Evans and Wistar rats. Alcohol. Clin. Exp. Res. 32, 1816–1823.

Simpson, J., and Kelly, J.P. (2012). An investigation of whether there are sex differ-ences in certain behavioural and neurochemical parameters in the rat. Behav. Brain Res. 229, 289–300.

Söderpalm, B., and Ericson, M. (2011). Neurocircuitry Involved in the Development of Alcohol Addiction: The Dopamine System and its Access Points. In Behavioral Neurobiology of Alcohol Addiction, W.H. Sommer, and R. Spanagel, eds. (Springer Berlin Heidelberg), pp. 127–161.

Sommer, W., Hyytiä, P., and Kiianmaa, K. (2006). The alcohol-preferring AA and alcohol-avoiding ANA rats: neurobiology of the regulation of alcohol drinking. Addict. Biol. 11, 289–309.

Spanagel, R. (2009). Alcoholism: a systems approach from molecular physiology to addictive behavior. Physiol. Rev. 89, 649–705.

94

Spanagel, R. (2017). Animal models of addiction. Dialogues Clin. Neurosci. 19, 247–258.

Spear, L.P. (2000). The adolescent brain and age-related behavioral manifestations. Neurosci. Biobehav. Rev. 24, 417–463.

Spear, L.P. (2013). Adolescent Neurodevelopment. J. Adolesc. Health 52, S7–S13. Spear, L.P. (2015). Adolescent alcohol exposure: Are there separable vulnerable pe-

riods within adolescence? Physiol. Behav. 148, 122–130. Spear, L.P. (2018). Effects of adolescent alcohol consumption on the brain and be-

haviour. Nat. Rev. Neurosci. 19, 197–214. Spear, L.P., and Swartzwelder, H.S. (2014). Adolescent alcohol exposure and persis-

tence of adolescent-typical phenotypes into adulthood: a mini-review. Neurosci. Biobehav. Rev. 45, 1–8.

Spencer, R.L., and Deak, T. (2016). A users guide to HPA axis research. Physiol. Behav.

Spoelder, M., Hesseling, P., Baars, A.M., Lozeman-van ‘t Klooster, J.G., Rotte, M.D., Vanderschuren, L.J.M.J., and Lesscher, H.M.B. (2015). Individual Variation in Alcohol Intake Predicts Reinforcement, Motivation, and Compulsive Alcohol Use in Rats. Alcohol. Clin. Exp. Res. 39, 2427–2437.

van der Staay, F.J. (2006). Animal models of behavioral dysfunctions: Basic concepts and classifications, and an evaluation strategy. Brain Res. Rev. 52, 131–159.

Stamps, J., and Groothuis, T.G.G. (2010a). The development of animal personality: relevance, concepts and perspectives. Biol. Rev. Camb. Philos. Soc. 85, 301–325.

Stamps, J.A., and Groothuis, T.G.G. (2010b). Developmental perspectives on person-ality: implications for ecological and evolutionary studies of individual differ-ences. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 365, 4029–4041.

Stanford, S. (2017). Confusing preclinical (predictive) drug screens with animal ‘models’ of psychiatric disorders, or ‘disorder-like’ behaviour, is undermining confidence in behavioural neuroscience. J. Psychopharmacol. (Oxf.) 31, 641–643.

Stanford, S.C. (2007a). The Open Field Test: reinventing the wheel. J. Psychophar-macol. (Oxf.) 21, 134–135.

Stanford, S.C. (2007b). Open fields (unlike wheels) can be any shape but still miss the target. J. Psychopharmacol. (Oxf.) 21, 144–144.

Stead, J.D.H., Clinton, S., Neal, C., Schneider, J., Jama, A., Miller, S., Vazquez, D.M., Watson, S.J., and Akil, H. (2006). Selective Breeding for Divergence in Novelty-seeking Traits: Heritability and Enrichment in Spontaneous Anxiety-related Be-haviors. Behav. Genet. 36, 697.

Stewart, A.M., and Kalueff, A.V. (2015). Developing better and more valid animal models of brain disorders. Behav. Brain Res. 276, 28–31.

Stewart, R.B., Gatto, G.J., Lumeng, L., Li, T.-K., and Murphy, J.M. (1993). Compar-ison of alcohol-preferring (P) and nonpreferring (NP) rats on tests of anxiety and for the anxiolytic effects of ethanol. Alcohol 10, 1–10.

Sturman, D.A., and Moghaddam, B. (2011). The neurobiology of adolescence: changes in brain architecture, functional dynamics, and behavioral tendencies. Neurosci. Biobehav. Rev. 35, 1704–1712.

Switonski, M. (2014). Dog as a model in studies on human hereditary diseases and their gene therapy. Reprod. Biol. 14, 44–50.

Sze, Y., and Brunton, P.J. (2019). Sex, stress and steroids. Eur. J. Neurosci. epub ahead of print, 1–29.

Tabakoff, B., and Hoffman, P.L. (2013). The neurobiology of alcohol consumption and alcoholism: an integrative history. Pharmacol. Biochem. Behav. 113, 20–37.

95

Tawa, E.A., Hall, S.D., and Lohoff, F.W. (2016). Overview of the Genetics of Alcohol Use Disorder. Alcohol Alcohol 51, 507–514.

Teicher, M.H., Samson, J.A., Anderson, C.M., and Ohashi, K. (2016). The effects of childhood maltreatment on brain structure, function and connectivity. Nat. Rev. Neurosci. 17, 652–666.

Thompson, R.F. (2001). Behavioral Neuroscience. In International Encyclopedia of the Social & Behavioral Sciences, N.J. Smelser, and P.B. Baltes, eds. (Pergamon), pp. 1118–1125.

Tinbergen, N. (1963). On aims and methods of Ethology. Z. Für Tierpsychol. 20, 410–433.

Tirelli, E., Laviola, G., and Adriani, W. (2003). Ontogenesis of behavioral sensitiza-tion and conditioned place preference induced by psychostimulants in laboratory rodents. Neurosci. Biobehav. Rev. 27, 163–178.

Tjernström, N., and Roman, E. (2018). Characterization of behavior and voluntary alcohol intake in Wistar and Lister Hooded rats. In Measuring Behavior 2018, R. Grant, T. Allen, A. Spink, and M. Sullivan, eds. (Manchester, United Kingdom), pp. 437–442.

Tottenham, N., and Galván, A. (2016). Stress and the adolescent brain: Amygdala-prefrontal cortex circuitry and ventral striatum as developmental targets. Neu-rosci. Biobehav. Rev. 70, 217–227.

Trent, N.L., and Menard, J.L. (2013). Lateral septal infusions of the neuropeptide Y Y2 receptor agonist, NPY13–36 differentially affect different defensive behaviors in male, Long Evans rats. Physiol. Behav. 110–111, 20–29.

Trezza, V., Baarendse, P.J.J., and Vanderschuren, L.J.M.J. (2010). The pleasures of play: pharmacological insights into social reward mechanisms. Trends Pharma-col. Sci. 31, 463–469.

Trezza, V., Baarendse, P.J.J., and Vanderschuren, L.J.M.J. (2014). On the interaction between drugs of abuse and adolescent social behavior. Psychopharmacology (Berl.) 231, 1715–1729.

Trigo, J.M., Martin-García, E., Berrendero, F., Robledo, P., and Maldonado, R. (2010). The endogenous opioid system: a common substrate in drug addiction. Drug Alcohol Depend. 108, 183–194.

Tsigos, C., and Chrousos, G.P. (2002). Hypothalamic-pituitary-adrenal axis, neuroen-docrine factors and stress. J. Psychosom. Res. 53, 865–871.

Vanderschuren, L.J., Niesink, R.J., Spruijt, B.M., and Van Ree, J.M. (1995). Influence of environmental factors on social play behavior of juvenile rats. Physiol. Behav. 58, 119–123.

Varlinskaya, E.I., and Spear, L.P. (2002). Acute Effects of Ethanol on Social Behavior of Adolescent and Adult Rats: Role of Familiarity of the Test Situation. Alcohol. Clin. Exp. Res. 26, 1502–1511.

Vashishtha, R., Livingston, M., Pennay, A., Dietze, P., MacLean, S., Holmes, J., Her-ring, R., Caluzzi, G., and Lubman, D.I. (2019). Why is adolescent drinking de-clining? A systematic review and narrative synthesis. Addict. Res. Theory 0, 1–14.

Veenema, A.H., and Neumann, I.D. (2009). Maternal separation enhances offensive play-fighting, basal corticosterone and hypothalamic vasopressin mRNA expres-sion in juvenile male rats. Psychoneuroendocrinology 34, 463–467.

Vendruscolo, L.F., and Roberts, A.J. (2014). Operant alcohol self-administration in dependent rats: Focus on the vapor model. Alcohol 48, 277–286.

Vengeliene, V., Bilbao, A., Molander, A., and Spanagel, R. (2008). Neuropharmacol-ogy of alcohol addiction. Br. J. Pharmacol. 154, 299–315.

96

Verhulst, B., Neale, M.C., and Kendler, K.S. (2015). The heritability of alcohol use disorders: a meta-analysis of twin and adoption studies. Psychol. Med. 45, 1061–1072.

Vetter-O’Hagen, C., Varlinskaya, E., and Spear, L. (2009). Sex Differences in Ethanol Intake and Sensitivity to Aversive Effects during Adolescence and Adulthood. Alcohol Alcohol. Oxf. Oxfs. 44, 547–554.

Vetulani, J. (2013). Early maternal separation: a rodent model of depression and a prevailing human condition. Pharmacol. Rep. 65, 1451–1461.

Viglinskaya, I.V., Overstreet, D.H., Kashevskaya, O.P., Badishtov, B.A., Kampov-Polevoy, A.B., Seredenin, S.B., and Halikas, J.A. (1995). To drink or not to drink: Tests of anxiety and immobility in alcohol-preferring and alcohol-nonpreferring rat strains. Physiol. Behav. 57, 937–941.

Volkow, N.D., Koob, G.F., Croyle, R.T., Bianchi, D.W., Gordon, J.A., Koroshetz, W.J., Pérez-Stable, E.J., Riley, W.T., Bloch, M.H., Conway, K., et al. (2018). The conception of the ABCD study: From substance use to a broad NIH collaboration. Dev. Cogn. Neurosci. 32, 4–7.

Walker, B.M., Roth, J.L., and Ehlers, C.L. (2008). Dissociable Effects of Ethanol Consumption during the Light and Dark Phase in Adolescent and Adult Wistar Rats. Alcohol Fayettev. N 42, 83–89.

WHO (2014). Health for the world’s adolescents: a second chance in the second dec-ade (World Health Organization).

WHO (2018). Global Status Report on Alcohol and Health 2018 (World Health Or-ganization).

WHO (2019). International Classification of Diseases 11th Revision. Widiger, T.A., Crego, C., Rojas, S.L., and Oltmanns, J.R. (2018). Basic personality

model. Curr. Opin. Psychol. 21, 18–22. Wille-Bille, A., Ferreyra, A., Sciangula, M., Chiner, F., Nizhnikov, M.E., and Pau-

tassi, R.M. (2017). Restraint stress enhances alcohol intake in adolescent female rats but reduces alcohol intake in adolescent male and adult female rats. Behav. Brain Res. 332, 269–279.

Wille-Bille, A., Miranda-Morales, R.S., Pucci, M., Bellia, F., D’Addario, C., and Pau-tassi, R.M. (2018). Prenatal ethanol induces an anxiety phenotype and alters ex-pression of dynorphin & nociceptin/orphanin FQ genes. Prog. Neuropsychophar-macol. Biol. Psychiatry 85, 77–88.

Windle, R.J., Wood, S.A., Shanks, N., Lightman, S.L., and Ingram, C.D. (1998). Ul-tradian Rhythm of Basal Corticosterone Release in the Female Rat: Dynamic In-teraction with the Response to Acute Stress. Endocrinology 139, 443–450.

Wise, R.A. (1973). Voluntary ethanol intake in rats following exposure to ethanol on various schedules. Psychopharmacologia 29, 203–210.

Wood, C.M., Nicolas, C.S., Choi, S.-L., Roman, E., Nylander, I., Fernandez-Teruel, A., Kiianmaa, K., Bienkowski, P., de Jong, T.R., Colombo, G., et al. (2016). Prev-alence and influence of cys407* Grm2 mutation in Hannover-derived Wistar rats: mGlu2 receptor loss links to alcohol intake, risk taking and emotional behaviour. Neuropharmacology.

Yurgelun-Todd, D. (2007). Emotional and cognitive changes during adolescence. Curr. Opin. Neurobiol. 17, 251–257.

Zha, X., and Xu, X. (2015). Dissecting the hypothalamic pathways that underlie innate behaviors. Neurosci. Bull. 31, 629–648.

Zhang, H., Sakharkar, A.J., Shi, G., Ugale, R., Prakash, A., and Pandey, S.C. (2010). Neuropeptide Y Signaling in the Central Nucleus of Amygdala Regulates Alco-hol-Drinking and Anxiety-Like Behaviors of Alcohol-Preferring Rats. Alcohol. Clin. Exp. Res. 34, 451–461.

97

Zimmerberg, B., and Sageser, K.A. (2011). Comparison of Two Rodent Models of Maternal Separation on Juvenile Social Behavior. Front. Psychiatry 2.

Zuckerman, M. (1996). The Psychobiological Model for Impulsive Unsocialized Sen-sation Seeking: A Comparative Approach. Neuropsychobiology 34, 125–129.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Pharmacy 284

Editor: The Dean of the Faculty of Pharmacy

A doctoral dissertation from the Faculty of Pharmacy, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofPharmacy. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Pharmacy”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-405313

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020