Search

This study takes the first step toward testing a Y chromosomal effect on both aggression and thermoregulatory nest-building behavior in mouse lines either bidirectionally selected for short (SAL) and long (LAL) attack latency or high (HIGH) and low (LOW) nest-building behavior. Using reciprocal crosses between SAL and LAL, and between HIGH and LOW, we found no indications for Y chromosomal effects on thermoregulatory nest-building behavior. As for aggression, we confirmed earlier studies on SAL and LAL, i.e., the origin of the Y chromosome influences attack latency, i.e., aggression. However, we did not find indications for a Y chromosomal effect on aggression in the HIGH and LOW lines. Since aggression and nest-building behavior have been shown to be characteristic parameters of two fundamentally different behavioral strategies, the present data underline the improbability of Y chromosomal genes underlying the genetic architecture of alternative behavioral strategies.

Genetically selected aggressive and nonaggressive male wild house mice were tested in the shock-probe/defensive burying test. Five distinct behaviors (burying, immobility, rearing, grooming, and exploration) were recorded in two environmental situations: fresh and home cage sawdust. Nonaggressive animals, characterized by a Long Attack Latency (LAL), showed more immobility in both test situations than animals having Short Attack Latencies (SAL), whereas SAL males displayed more defensive burying than LAL ones when tested with fresh sawdust. Testing with home cage sawdust, however, resulted in the same duration of defensive burying in SAL and LAL. These results support earlier findings about the existence of two heritable, fundamentally different strategies to cope with aversive situations. Aggressive (SAL) animals react actively to environmental challenges, whereas nonaggressive animals react actively or passively, depending on the characteristics of the stressful environment. These mouse lines, selected for attack latency, i.e., aggression, may, therefore, be important tools to unravel the genetic architecture underlying the physiological and neuronal mechanisms of behavioral strategies towards stressful events.

Genetically selected short attack latency (SAL) and long attack latency (LAL) male wild house mice behave differently in the defensive burying test. When challenged, SAL males respond actively with more time spent on defensive burying, whereas LAL males are more passive with more time remaining immobile. The first aim of this study was to find out whether the nonpairing part of the Y chromosome (Y(NPAR)) affects the behavioral stress response in this paradigm. Second, to determine if the differential behavioral profile found in males is also present in females, SAL and LAL females were tested. Third, nonattacking and attacking LAL males were compared. Five behavioral elements were recorded: defensive burying, immobility, rearing, grooming, and exploration. Males were first tested for attack latency. The results show that the Y(NPAR) influences defensive burying. However, the size of this effect is overshadowed by the background of the mice. Furthermore, although females differed from males, they tended to demonstrate the same behavioral profile as males. Nongenetic factors may also play a role, as attacking LAL males showed more defensive burying than nonattacking LAL males. Copyright (C) 1999 Elsevier Science Inc.

This paper reviews our present state of knowledge of genetic variation in (offensive) aggression in wild house mice. The basic tools in this research were lines bidirectionally selected for attack latency (fast attacking SAL and slow attacking LAL males), descended from a feral population. Using congenic lines for the nonpseudoautosomal region of the Y chromosome (Y-NPAR), reciprocal crosses between (parental) SAL and LAL, and crosses between parentals and congenics, an autosomally dependent Y chromosomal effect on aggression has been found. Both the pseudoautosomal (Y-PAR) region and the Y-NPAR play a role. As for environmental sources of variation, prenatal and postnatal maternal effects are of minor importance for the development of aggression differences. One of the physiological factors by which genetic effects may be mediated is testosterone (T). Besides quantitative aspects, the timing of T release seems crucial. Two important time frames are discussed: the perinatal and pubertal time periods. Finally, neurochemical and neuroanatomical correlates are considered. Differences in neostriatal dopaminergic activity, and sizes of the intra- and infrapyramidal messy fiber terminal fields, as well as Y chromosomal effects on the latter two, are discussed.

Abstract Male wild house mice selected for Short Attack Latency (SAL) show an active coping style to environmental challenges, whereas males selected for Long Attack Latency (LAL) generally exhibit a passive coping style. Previous studies showed that the aggressive SAL males develop routines more quickly and show more resistance to environmental changes than the less aggressive LAL ones; the latter behave more flexibly in non-social situations. Furthermore, using reciprocal F1s and backcross lines for SAL and LAL males, a Y chromosomal effect on aggression has been demonstrated. The aim of this study was to examine possible effects of the Y chromosome on behavioural flexibility. For this purpose six genotypes (SAL, LAL, their reciprocal F1s and their congenics for the non-pseudoautosomal part of the Y chromosome) were tested for their behavioural flexibility in a Y-maze. Differences between SAL and LAL were reproduced. Congenic lines showed identical scores to their parental ones. Regarding the reciprocal F1s, differences in aggression were not associated with behavioural flexibility. Therefore, it may be concluded that the Y chromosome does not influence aggression-related behavioural flexibility.

Congenic lines were made for the Y chromosome between aggressive and nonaggressive lines of house mice, which were previously established by artificial selection in wild mice for short attack latencies (SAL line) and long attack latencies (LAL line). The aggressiveness of the males in successive backcross generations of the congenic lines is reported. Results fit the hypothesis that the Y-chromosomal effect that is often found for aggression in house mice may be located on the pseudo-autosomal region of this chromosome.

Testosterone secretory capacity of testicular Leydig cells was determined in fetal males of an aggressive and a nonaggressive genetic selection line of wild house mice. They were studied at Days 15-18 of gestation and on the first day after birth. A previously described morphometric method was used to quantify 3 beta-hydroxy steroid dehydrogenase (3 beta-HSD)-stained Leydig cells in testicular sections to determine testosterone secretory capacity, which may be considered to reflect circulating plasma testosterone in the fetus. The results of this study show that the testosterone secretory capacity of Leydig cells in the testis changes differentially during intrauterine development in males of the aggressive and nonaggressive selection lines. The peak secretory capacity is reached at Day 17 of gestation for the males of the aggressive selection line, while the peak for the nonaggressive males is reached on the first neonatal day. The larger anogenital distance observed in aggressive males suggests a higher prenatal testosterone level in these males. The importance of the difference in timing of the perinatal 3 beta-HSD peak top individual variation in adult aggressive behavior is discussed.

The effect of exposure to inescapable long-duration shocks of moderate intensity on intershock activity and on subsequent escape or avoidance performance was studied in aggressive and non-aggressive male mice. The activity of the non-aggressive mice was severely suppressed during the inescapable shock session, while that of the aggressive males was hardly influenced. The decremental effect of prior shock exposure on subsequent response latency and activity in an active two-way escape or avoidance task was greater in the non-aggressive than in the aggressive mice. There was no evidence that learned inactivity or learned helplessness (an associative deficit) could explain the results. Instead, individual differences in behavioural strategy in response to threatening situations appeared to account for the effects of inescapable shock. Aggressive male mice predominantly adopted an active behavioural strategy in challenging situations, which resulted in persistent attempts to exercise control over the external situation and hence in a sustained tendency to initiate responses. Non-aggressive mice primarily assumed a passive strategy; their tendency to exercise control was low, which readily resulted in a reduced tendency to initiate responses.

Abstract To investigate the relationship between aggression and routine-like behaviour the response of male mice of bidirectionally selected lines for attack latency to a change in the social and non-social environment has been measured. In a non-social situation the extent of routine-like behaviour was measured in a Y-maze in which only one of the two arms gave access to the food compartments. The number of errors made in response to reversal of the arm that was blocked was taken as indicator for the degree of routine formation. Males of the short attack latency (SAL) line made significantly more errors, and hence were more routine-like in their performance, than mice of the long attack latency (LAL) line. Males of the LAL line that nevertheless had short attack latencies (i. e. aggressive LAL mice) turned out to be flexible in their behaviour; their response was similar to that of the non-aggressive LAL males. In a social situation SAL and aggressive LAL mice were used to investigate routine formation in attacking behaviour. The males were given different amounts of experience with male opponents after which their own female was introduced as opponent. The more extended the experience with male intruders was, the more SAL males subsequently attacked their female. In contrast, LAL mice appropriately changed their behaviour towards the female opponent. Thus, the attacking behaviour of SAL mice gets routine-like, whereas that of LAL males remains flexible. It is concluded that selection for attack latency generally coincides with selection for routine-like behaviour, suggesting that these two factors are influenced by many of the same genes. Regarding the fact that aggressive males of the LAL line show flexible behaviour, it may be proposed that with the phenotypic selection for attack latency there has in fact been selected for a mechanism that determines the organization (routine-like vs flexible) of behaviour.

A number of immunocytochemical studies show that the vasopressinergic (AVP) system in the rat brain is far from homogeneous. For example the AVP innervation of the lateral septum (LS), originating from the medial amygdala and bed nucleus of the stria terminalis, appears to be sexually dimorphic and testosterone (T) dependent. At an adult age, T appears to induce AVP synthesis in these neurons, whereas neonatal T enhances the degree of AVP innervation of the LS, as part of the sexual differentiation.' Adult as well as neonatal T is involved in behavior, not only in the behavioral differentiation between the sexes, but also in the individual differentiation of social behavior within the male sex. For example, at an adult age, male house mice selected for high levels of aggressive behavior (short attack latency, SAL) display a higher plasma-T level and androgen sensitivity compared to the nonaggressive males (long attack latency, LAL). Surprisingly the LAL males show the highest levels of neonatal T production, suggesting that high neonatal T leads to a downregulation of the pituitary gonadal axis together with a reduction in adult aggressive behavior.2 In view of this role of T in the individual differentiation in aggressive behavior within the male sex, it was expected to find an individual differentiation of LS-AVP as well. Recent studies both in SAL and LAL males indicate that also in the house mouse the LS-AVP innervation is T dependent. Moreover, quantitative immunocytochemical analyses revealed the highest density of AVP fibers in the LS, especially in the posterior part, of the adult LAL males. Not only a higher fiber density was found, but the fibers also contain more AVP. Present experiments are aimed at demonstrating the causal relationship between the variation in neonatal T and the individual differentiation in adult AVP in the LS within the male sex. The functional significance of the individual difference in the T-dependent AVP system with respect to the differences in aggressive behavior is not yet clarified. A number of behavioral studies on SAL and LAL males show that the individual differentiation in aggressive behavior reflects a more fundamental differentiation in active or passive coping behavior, based upon the degree in which behavior is guided by

Factors affecting health and disease are somewhat neglected in the study of the dynamics of mammalian populations. The present paper summarizes a number of studies on social stress in rats and mice. Social situations induce classical stress responses in terms of corticosterone and catecholamines. More important is the fact that there is a considerable individual differentiation in the magnitude of the responses in relation to the behavioural stategy in a social situation. These physiological differences render individuals differentially susceptible for stress pathologies such as cardiovascular diseases and immune deficiencies. Moreover, the behavioural strategy largely determines the position of an animal in the social structure and hence the frequency of behavioural and physiological activation. The data suggest that the individual behavioural and physiological reactivity determine in interaction with the social environtment the susceptibility for diseases. The relevance of these observations for the dynamics of a population is discussed.

Artificial selection for short and long attack latency levels in wild male Mus musculus over 11 generations was successful for short latencies. The realized heritability of 0.30 is comparable to those found in other selection studies on aggression. In part selection may have been for faster onto-genetic development of short attack latencies. Four attempts to select for longer attack latencies failed because the lines died out immediately or within two generations for unknown reasons. But neither the physical condition of the animals nor their behavior appeared to have been the cause. Female aggressiveness as measured in female-female encounters was not affected by the selection exerted on the males. This suggests that no genetic correlation exists between aggressiveness of males and females, confirming results of P. D. Ebert and J. S. Hyde [(1976). Behav. Genet. 6:291-304] obtained in a selection experiment on aggression using females.

This study tries to explain the discrepancy between the frequencies of t-alleles found in house mouse populations and those expected on the basis of a deterministic model, by hypothesizing a difference in trappability between + /+ and +/t genotypes. We found variation in t-genotypes to be of no influence on trappability of the individuals in ten experimental populations.

The hypothesis, partly based on findings in social interactions, that aggressive mice generally adopt an active behavioural strategy (cf. fight-flight) in threatening situations, while non-aggressive ones generally assume a passive strategy (cf. conservation-withdrawal) was tested using a two-way active shock avoidance paradigm. Overall, aggressive mice were found to be better active shock avoiders than non-aggressive animals; a finding that is consistent with our hypothesis. However, within the non-aggressive mice a clear dichotomy in high and low avoidance individuals was found. The high intertrial activity in the superior avoidance groups and the low activity in the poor avoidance group was interpreted as another indication of an active versus passive strategy respectively. Accordingly, it was concluded that not all non-aggressive mice assume a passive strategy, but that some mice adopt an active strategy, like all aggressive males.

The relationship between testosterone level and attack latency was studied in genetically different wild house mice by means of castration and subsequent testosterone therapy. This was done to provide adequate physiological knowledge for further research on the genetic basis of individual differences in these mice. The findings show that individual variation in attack latency is related not only to variation in baseline plasma testosterone level (via a dose-response relation), but also to variation in responsiveness to testosterone that is induced before puberty. In addition it is shown that in fast-attacking mice the maintenance of the attack latency level reached by maturation is independent of testosterone, whereas this is not the case in mice that are reluctant to attack.

The biological significance, the origin and the genetical causation of differences in behaviour between inbred strains of mice have been studied. Three inbred strains with different origins were used, these were CPBs (Balb), CBA and C57Black. Exploratory behaviour, nestbuilding behaviour and agonistic behaviour have been observed in environments, which varied with respect to familiarity, presence or absence of nesting material and the possibility of social contact. A comparison of the behaviour of these strains showed the presence of differences in all aspects of the behaviour studied. None of these differences, however, was a qualitative one. The behaviour of each strain was strain specific as appeared from a comparison with earlier reports on related substrains. CPBs mice were territorial with a sensitive aggression-flight balance. They showed a relatively elaborate courtship. They were able to build spherical nests at the surface without support of walls because they treated the nesting material in a special way (fraying). Their exploratory behaviour was mainly directed to open space. C57Black mice were not territorial and did not show a sensitive aggression-flight balance. They hardly showed any courtship. They were not able to build a spherical nest at the surface without support of walls, because they hardly fraved the nesting material. They showed much digging, especially when confronted with nesting material. This digging was well directed and in a suitable substrate holes were made. Their exploratory behaviour was mainly directed to the surface. CBA mice were generally less aggressive than the CPBs mice, but some individuals showed clear territorial behaviour. Their courtship was intermediate to that of the other two strains. CBA mice as a rule did not build proper nests at the surface, either because they did not try to make a nest at all or because they treated the nesting material in such a way that they chopped it up to pieces which were too small to shape more than a simple platform at the surface. They showed much digging but this digging was randomly directed resulting in turning over the whole substrate; no holes resulted from such digging. The exploratory behaviour of the CBA mice was mainly directed to the surface. The nestbuilding behaviour element fraying was chosen for a genetical analysis to find out whether differences in the kind of behaviour studied here were genetically determined or not. Genetical determination appeared to play an important role. Heritabilities for F2 animals and their progeny varied between 17 and 47 per cent. A cross-fostering test between CPBs and C57 Black did not show any cultural influence on the fraying performance in these mice. CPBs mice, which showed a relatively high variability for fraying, appeared to be heterozygous for this character. The heritability found showed a similar size as was found for hybrids (43 per cent). Selection experiments on fraying in this strain were not successful because the lines died out. One selection line for high fraying survived for six generations. It showed no significant reaction to the selection exerted. Heterozygosity in CPBs mice for fraying appeared to be maintained by a kind of balanced system which worked partly by sterility of the high and low fraying types and partly by a negative influence of such types on the physical condition and thus on the fertility of their progeny. Intermediate fraying mice got selective advantage over high and low fraying types in this way, even under inbreeding conditions. Probably only one effective factor is involved in the determination of fraying. Hypotheses concerning the maintenance of genetic variability under inbreeding conditions have been discussed. An analysis of the function and the origin of the differences found between strains suggested that the differences between strains showed strain specific adaptations to habitats in which their ancestors lived. This holds for CPBs mice and for C57Black mice, the ancestors of the first being surface living mice, the ancestors of the second being hole living mice. The behaviour of the CBA mice appeared to be disrupted from a functional point of view, which is in agreement with the fact that CBA mice originate from a cross between two other inbred lines. A specially designed cage to test a differential habitat preference to be present in CPBs and C57Black mice confirmed the hypothesis about the habitat adaptations in these mice and clearly showed that preferences for these habitats are still present in the inbred mice. The results of CBA mice in this test confirmed the unpredictable character of these mice. A comparison for adaptations of this kind between laboratory mice and wild mice revealed that the differences have probably arisen in mice before they were taken into the laboratory. The scarce data available for wild mice suggested that similar adaptations as found here for inbred mice, can also be found in wild mice of different origin. It is discussed that surface living mice have had a better chance to adapt to living in buildings and become commensals, furthermore that the two behavioural types, showing differential adaptations, cannot have gone together as morphs in one polymorphic population. Breeding structure and homozygosity in wild mice and inbred mice have been compared and discussed. It is argued that some wild mice may show more mechanisms to protect them against loss of genetic variability than other organisms because their breeding structure in nature causes a high amount of inbreeding. The way in which differences in mouse behaviour may have evolved was discussed. A primary cause appears to be the differential habitat preference. The possibility of an ecological factor being a primary cause for speciation was suggested many times in studies comparing sympatric species. Differences in non-display behaviour seem to evolve in the same way as display behaviour, i.e. by small changes in threshold of behavioural mechanisms. This could be concluded from the fact that on a subspecies level, like here between different strains of mice, no qualitative differences in behaviour could be found, whereas this is always the case for some of the differences on the species level. It is argued that specific adaptations as found here for mice may have bearings on the outcome of experiments done with mice, furthermore that inbred mice, in addition to studies on wild mice, may be very useful, especially in studying the role of behaviour in regulating population density.

Aggressive and non-aggressive male mice differ in their reaction to a changing social environment. In order to investigate if this differentiation holds also for non-social situations male mice are trained in a standard maze task, whereafter a change (extramaze and intramaze, respectively) is introduced. The results indicate that aggressive males fulfil their task fairly routine-like and do not react to a change which is in contrast to the non-aggressive individuals. In a second experiment a more continuously changing situation is created by testing the animals every 3 trials in a different maze configuration. In this situation in which a routine cannot be developed c.q. used, the aggressive males performed worse than the non-aggressive animals. It is suggested that the behaviour of aggressive males is mainly controlled by intrinsic factors whereas the behaviour of non-aggressive males is more dependent on external factors. Similar results are obtained when repeating the experiments with rats. This indicates that the relation between aggression and behavioural reaction to a changing environment has more general validity. The possibly underlying mechanism is discussed as well as the consequences for the functioning of the animals in a social setting.

Publisher Summary This chapter presents a novel, behavioral physiological stress concept that originates from the classical view that stress is a response. This new concept is extended to environment, behavior, and physiology, and it incorporates the novel neuroendocrine views including the neuropeptide concept. Stress is viewed as a general biological and usually functional response to environmental and bodily demands. A stress depends on interactions among environment, individual characteristics and the properties of stressors, stress, and the physiological systems, and also among the nervous system, peripheral organ systems, and the neuroendocrine system. To adapt to the altering social and physical environmental demands, man and other animals require a chain of behavioral, neuroendocrine, and autonomic physiological and metabolic responses to maintain bodily and mental homeostasis. The neuroendocrine state of the brain is given a central position in determining the state of health or disease of mind and body.

One of the major concerns in research on the aetiology of coronary heart disease (CHD) and hypertension has been the question why some individuals suffer from these diseases while others are healthy, although exposed to seemingly identical external conditions. Since the early studies by Friedman and Roseman (1959), it has become increasingly clear that apart from environmental circumstances, personality factors are also important. In order to study the complex interplay between environmental variables and personality factors in the pathophysiology of the cardiovascular system, animal research is indispensable.

In wild house mice, the hypothesis that in the organization of behavior the relative contribution of intrinsic factors is more important in aggressive males, while that of extrinsic factors is more important in nonaggressive individuals was confirmed using the circadian rhythmicity of activity. The faster rate of reentrainment and the suppression of activity during a subjective night and during adaptation to the new LD cycle in the nonaggressive males indicate that their circadian rhythmicity of activity is to a large extent determined by the Zeitgeber, an extrinsic factor. The slower reentrainment rate, the lack of response to a subjective night and the normal activity level seen during reentrainment in the aggressive mice suggest strong control by the pacemaker, an intrinsic factor. Tau differences between the aggressive and nonaggressive mice provide some evidence that the pacemaker of nonaggressive males is fairly labile and is easily influenced by external factors, whereas the pacemaker of aggressive animals is rather stable.

Eines der Nestbauverhaltenselemente bei Mausen (Beknappern von Papierstreifen) wurde bei 2 Inzuchtstammen und deren F1- und F2 Hybriden verglichen. F1-Tiere zeigten im Vergleich zu den Eltern eine grossere Variationsbreite des Verhaltenselementes. Ein grosser und besonders signifikanter Teil dieser Variabilitat war einer altersabhangigen Verschiebung zuzuschreiben. Diese kommt wahrscheinlich nicht infolge eines Lernprozesses zustande, sondern durch Anderungen der genetisch-regulierenden Mechanismen (wie. z.B. Dominanz und Epistasie).