The fallacy of behavioral phenotyping without standardisation

Animal models of behavioral dysfunctions: Basic concepts and classifications, and an evaluation strategy

In behavioral neurosciences, such as neurobiology and biopsychology, animal models make it possible to investigate brain-behavior relations, with the aim of gaining insight into normal and abnormal human behavior and its underlying neuronal and neuroendocrinological processes. Different types of animal models of behavioral dysfunctions are reviewed in this article. In order to determine the precise criteria that an animal model should fulfill, experts from different fields must define the desired characteristics of that model at the neuropathologic and behavioral level. The list of characteristics depends on the purpose of the model. The phenotype-abnormal behavior or behavioral dysfunctions-has to be translated into testable measures in animal experiments. It is essential to standardize rearing, housing, and testing conditions, and to evaluate the reliability, validity (primarily predictive and construct validity), and biological or clinical relevance of putative animal models of human behavioral dysfunctions. This evaluation, guided by a systematic strategy, is central to the development of a model. The necessity of animal models and the responsible use of animals in research are discussed briefly.

Behavioral phenotyping of transgenic and knockout mice: practical concerns and potential pitfalls

New technologies in molecular genetics have dramatically increased the number of targeted gene mutations available to the biomedical research community. Many mutant mouse lines have been generated to provide animal models for human genetic disorders, offering insights into anatomical, neurochemical, and behavioral effects of aberrant gene expression. A variety of assays have been developed to identify and characterize phenotypic changes. In the behavioral domain, our phenotyping strategy involves a comprehensive standardized methodological approach that assesses general health, reflexes, sensory abilities, and motor functions. This assessment is followed by a series of complementary tasks in the specific behavioral domain(s) hypothesized to reveal the function(s) of the gene. Our multitiered approach minimizes intersubject variability by standardizing the experimental history for all animals, improves interlaboratory reliability by providing a clearly defined experimental protocol, and minimizes artifactual interpretations of behavioral data by careful preliminary assessments of basic behaviors, followed by multiple tests within the behavioral domain of interest. Despite meticulous attention to experimental protocol, attention to environmental factors is essential. Differences in noise, light, home cage environment, handling, and diet can dramatically alter behavior. Baseline differences in the behaviors of inbred strains used to generate targeted mutant mouse lines can directly influence the behavioral phenotype of the mutant line. Strategies aimed at minimizing environmental variability and contributions of background genes will enhance the robustness of mouse behavioral phenotyping assays.

Low amplitude entrainment of mice and the impact of circadian phase on behavior tests

A tremendous increase in the use of genetically engineered mice as experimental animals has led to increased scrutiny of mouse models generally and mouse behavioral paradigms specifically. Although mice are nocturnal, for practical reasons, most experimental procedures, including behavioral studies, are conducted during their inactive, sleep phase. Accumulating evidence indicates that myriad behavioral, cellular and biochemical processes fluctuate with circadian rhythmicity; however, time of day at which experiments are conducted is rarely controlled. The impact of circadian phase on the reliability of experimental results has received little attention and the present data are conflicting. This study addressed two questions. First, will laboratory mice in a typical animal care facility entrain to a low amplitude light cycle using bright/dim rather than light/dark cycles? A positive answer will make reversing photocycle easy to implement in any facility as dim light suitable for animal husbandry and behavioral testing can substitute for darkness during work hours. By monitoring home cage wheel running, we examined the effectiveness of a dim/bright photocycle as a zeitgeiber. We found that mice subjected to dim/bright photocycles effectively entrained such that their subjective night and activity onset coincided with the beginning of the dim light period, suggesting a potential strategy for standardization and management of circadian phase in nocturnal animals. In a second experiment, we asked what effect circadian phase has on behavioral performance in commonly used mouse behavioral tests. We found no main effect of circadian phase on outcome in open field activity, elevated plus maze emotionality, water maze spatial memory, novel object exploration and hyperactivity in response to amphetamine; however, we observed occasional interactions between circadian phase and both strain and sex that were neither consistent nor systematic. These data suggest that the tests examined here are relatively impervious to circadian phase. In general, testing mice during their active phase is more suitable for behavioral studies; a reversed dim/bright photocycle potentially offers one practical strategy for managing rodents' circadian cycles.

Multi-target strategies for the improved treatment of depressive states: Conceptual foundations and neuronal substrates, drug discovery and therapeutic application

Major depression is a debilitating and recurrent disorder with a substantial lifetime risk and a high social cost. Depressed patients generally display co-morbid symptoms, and depression frequently accompanies other serious disorders. Currently available drugs display limited efficacy and a pronounced delay to onset of action, and all provoke distressing side effects. Cloning of the human genome has fuelled expectations that symptomatic treatment may soon become more rapid and effective, and that depressive states may ultimately be "prevented" or "cured". In pursuing these objectives, in particular for genome-derived, non-monoaminergic targets, "specificity" of drug actions is often emphasized. That is, priority is afforded to agents that interact exclusively with a single site hypothesized as critically involved in the pathogenesis and/or control of depression. Certain highly selective drugs may prove effective, and they remain indispensable in the experimental (and clinical) evaluation of the significance of novel mechanisms. However, by analogy to other multifactorial disorders, "multi-target" agents may be better adapted to the improved treatment of depressive states. Support for this contention is garnered from a broad palette of observations, ranging from mechanisms of action of adjunctive drug combinations and electroconvulsive therapy to "network theory" analysis of the etiology and management of depressive states. The review also outlines opportunities to be exploited, and challenges to be addressed, in the discovery and characterization of drugs recognizing multiple targets. Finally, a diversity of multi-target strategies is proposed for the more efficacious and rapid control of core and co-morbid symptoms of depression, together with improved tolerance relative to currently available agents.

Two months makes a difference in spatial orientation learning in very old FBNF1 rats

Age-related changes in cognitive performance may be more pronounced in the period near or exceeding the median life span. Therefore, we compared the acquisition of a Morris water escape task by two groups of very old Fischer344 x Brown Norway hybrid rats. The mean age difference between the two groups of rats (a 33- to 34-month-old group versus a 35- to 36-month-old group) was about 2 months. Both groups of rats initially had the same level of performance, but then the younger group learned to escape onto the submerged platform faster, swimming a shorter distance, than the older group. By the fifth acquisition session, the younger rats needed only half the time and swam a shorter distance before they reached the platform than the older rats. These differences in learning were not due to different locomotor abilities as both groups had a similar swimming speed. These results suggest that age-related changes in cognitive performance are indeed more pronounced in the period around the median life span. We also discussed different set-ups to perform cross-sectional age-comparison studies. If there are not sufficient animals from one batch, it may be worthwhile to combine animals from different batches per age group, provided that breeding, rearing, housing, and testing conditions are highly standardized.

Common variations in the pretest environment influence genotypic comparisons in models of anxiety

The behavioral characterization of rodent strains in different studies and laboratories can provide unreplicable results even when genotypes are kept constant and environmental control is maximized. In the present study, the influence of common laboratory environmental variables and their interaction with genotype on the results of behavioral tests of anxiety/emotionality were investigated. To this end, the inbred rat strains Lewis (LEW) and spontaneously hypertensive rats (SHR), which are known to differ for numerous emotionality-related behaviors, were tested in the open field (OF), elevated plus maze (EPM) and black/white box (BWB), while three environmental factors were systematically controlled and analyzed: (1) the experimenter handling the animal (familiar or unfamiliar); (2) the position of the home cage (top or bottom shelf of the rack) and (3) the behavioral state of the animal immediately before the test (arousal or rest). Experimenter familiarity did not alter the behavior of rats in the OF. Cage position, on the other hand, influenced the behavior in the OF and BWB, with rats housed in top cages appearing less anxious than those housed in the bottom. In the BWB (but not in the OF), these effects were genotype dependent. Finally, the behavioral state of the animals prior to testing altered the results of the EPM in a strain-dependent manner, with some anxiety-related genotypic differences being found only among rats that were aroused in their home cages. This study showed that common variations in the laboratory environment interact with genotype in behavioral tests of anxiety/emotionality. Recognizing and understanding such variations can help in the design of more effective experiments.

Cage enrichment and mouse behaviour

Mice housed in standard cages show impaired brain development, abnormal repetitive behaviours (stereotypies) and an anxious behavioural profile, all of which can be lessened by making the cage environment more stimulating. But concerns have been raised that enriched housing might disrupt standardization and so affect the precision and reproducibility of behavioural-test results (for example, see ref. 4). Here we show that environmental enrichment increases neither individual variability in behavioural tests nor the risk of obtaining conflicting data in replicate studies. Our findings indicate that the housing conditions of laboratory mice can be markedly improved without affecting the standardization of results.

Long-term individual housing in C57BL/6J and DBA/2 mice: assessment of behavioral consequences

The aim of the present study was to investigate the effects of individual housing on mouse behavior. The male mice of the C57BL/6J and DBA/2 strains were separated at the age of 4 weeks and kept in individual housing for 7 weeks until behavioral testing began. Their behavior was compared to the group-housed mice in a battery of tests during the following 7 weeks. The single-housed mice were hyperactive and displayed reduced habituation in the tests assessing activity and exploration. Reduced anxiety was established in the elevated plus-maze, but an opposite effect was observed in the dark-light (DL) and hyponeophagia tests. Immobility in the forced swimming test was reduced by social isolation. The DBA mice displayed higher anxiety-like behavior than the B6 mice in the plus-maze and DL exploration test, but hyponeophagia was reduced in the DBA mice. Moreover, all effects of individual housing on the exploratory and emotional behavior were more evident in the DBA than in the B6 mice. Novel object recognition and fear conditioning (FC) were significantly impaired in the single-housed mice, whereas water-maze (WM) learning was not affected. Marked strain differences were established in all three learning tests. The B6 mice performed better in the object recognition and FC tasks. Initial spatial learning in the WM was faster and memory retention slightly enhanced in the B6 mice. The DBA mice displayed lower preference to the new and enhanced preference to the old platform location than the B6 mice after reversal learning in the WM. We conclude that individual housing has strong strain- and test-specific effects on emotional behavior and impairs memory in certain tasks.

Flanking gene and genetic background problems in genetically manipulated mice

Mice carrying engineered genetic modifications have become an indispensable tool in the study of gene functioning. The interpretation of results obtained with targeted mutants is not completely straightforward, however, because of genetic complications due to linkage and epistasis. Effects of closely linked genes flanking the targeted locus might sometimes be responsible for phenotypic changes ascribed to the null mutation. The effects of the latter might also be modified by the general genetic background. This review presents some examples and discusses some simple strategies to deal with these complications.

The dark phase improves genetic discrimination for some high throughput mouse behavioral phenotyping

Dark-phase testing has previously been shown by others to improve the outcome of some 'classical' behavior test situations. However, the importance of such ethological correctness and the effect of the light/dark cycle on high throughput behavioral testing situations such as 'mutant vs. wild type' and 'screening', are less or unknown, respectively. These testing situations differ from the 'classical' in that they are designed primarily to discriminate between genetically different mice rather than provide a detailed assessment of ability or psychosocial state. Here we test the hypotheses that dark-phase testing affects the outcome of high throughput behavioral tests and that dark-phase testing improves discrimination between genetically distinct mice (C57BL/6J, 129S1/SvImJ and B6129F1) using high throughput behavioral tests. Our results demonstrate that, although all successful tests showed some effect of phase, only the SHIRPA primary screen, open-field test and motor learning on the rotarod showed improved strain discrimination in the dark phase. Surprisingly, the social interaction test did not show a clear benefit to either phase, and interestingly, the tail-flick test discriminated strains better in the light phase. However, since the preponderance of our data shows that dark-phase testing improves, or does not affect, strain discrimination, we conclude that for these strains and tests, dark-phase testing provided superior outcomes. If discrimination is not achieved in the dark phase, then light phase-testing would be undertaken.

Neurobehavioral assessment in the information age

The elucidation of the human and mouse genomes provides new opportunities for exploring the genetic underpinnings of complex mammalian behaviors. This information also provides new windows into the pathophysiology and treatment of neuropsychiatric diseases. Optimal use of the rapidly escalating numbers of mouse lines engineered for these purposes is hindered, however, by practical and theoretical limitations of common behavioral analyses. New strategies combining automated behavioral monitoring and information technologies are currently under development in both academic and industrial settings. These hold promise, both for improving the throughput of mouse behavioral assessment and for providing new insights into the neurobiology of mammalian behavioral regulation.

Enrichment of laboratory caging for rats: a review

Rats are a well-understood and widely used laboratory species that should be provided with environmentally enriched caging in line with modern animal welfare guidelines. This paper reviews which sources of enrichment are effective and should be prioritised, and how methods for providing enrichment might be selected using rats’ preferences as a guide. Rats demonstrate high demand for social contact and prefer larger cages, and cages with shelters, nesting material and foraging devices. Rats also discriminate between different methods of providing a given type of enrichment. It is clear that rats should be provided with enrichments such as social contact and shelter, and, in fact, that these should probably be considered basic husbandry requirements rather than optional improvements. It is still difficult, however, for animal caretakers to access proven, standardised methods for providing appropriately enriched caging, and the level of enrichment routinely provided to most rats in the laboratory appears to be low. Further research is required to assess the impact of enrichment upon research variables and to develop commercially viable enrichment products for rats in the laboratory.

Internal and External Factors Affecting the Development of Neuropathic Pain in Rodents. Is It All About Pain?

It is important to know the factors that will influence animal models of neuropathic pain. A good reproducibility and predictability in different strains of animals for a given test increases the clinical relevance and possible targeting. An obligatory requirement for enabling comparisons of results of different origin is a meticulous definition of the specific sensitivities of a model for neuropathic pain and a description of the test conditions. Factors influencing neuropathic pain behavior can be subdivided in external and internal factors. The most important external factors are; timing of the measurement of pain after induction of neuropathy, circadian rhythms, seasonal influences, air humidity, influence of order of testing, diet, social variables, housing and manipulation, cage density, sexual activity, external stress factors, and influences of the experimenter. The internal factors are related to the type of animal, its genetic background, gender, age, and the presence of homeostatic adaptation mechanisms to specific situations or stress. In practice, the behavioral presentations to pain depend on the combination of genetic and environmental factors such as accepted social behavior. It also depends on the use of genetic manipulation of the animals such as in transgenic animals. These make the interpretation of data even more difficult. Differences of pain behavior between in- and outbred animals will be better understood by using modern analysis techniques. Substrains of animals with a high likelihood for developing neuropathic pain make the unraveling of specific pathophysiological mechanisms possible. Concerning the effect of stress on pain, it is important to differentiate between external and internal stress such as social coping behavior. The individual dealing with this stress is species sensitive, and depends on the genotype and the social learning. In the future, histo-immunological and genetic analysis will highlight similarities of the different pathophysiological mechanisms of pain between different species and human subjects. The final objective for the study of pain is to describe the genetics of the eliciting pain mechanisms in humans and to look for correlations with the knowledge from basic research. Therefore, it is necessary to know the genetic evolution of the different mechanisms in chronic pain. In order to be able to control the clinical predictability of a putative treatment the evolutionary pharmacogenomic structure of specific transmitters and receptors must be clarified.

SEE locomotor behavior test discriminates C57BL/6J and DBA/2J mouse inbred strains across laboratories and protocol conditions

Conventional tests of behavioral phenotyping frequently have difficulties differentiating certain genotypes and replicating these differences across laboratories and protocol conditions. This study explores the hypothesis that automated tests can be designed to quantify ethologically relevant behavior patterns that more readily characterize heritable and replicable phenotypes. It used SEE (Strategy for the Exploration of Exploration) to phenotype the locomotor behavior of the C57BL/6 and DBA/2 mouse inbred strains across 3 laboratories. The 2 genotypes differed in 15 different measures of behavior, none of which had a significant genotype-laboratory interaction. Within the same laboratory, most of these differences were replicated in additional experiments despite the test photoperiod phase being changed and saline being injected. Results suggest that well-designed tests may considerably enhance replicability across laboratories.

Functional genomics of neural and behavioral plasticity

How does the environment, particularly the social environment, influence brain and behavior and what are the underlying physiologic, molecular, and genetic mechanisms? Adaptations of brain and behavior to changes in the social or physical environment are common in the animal world, either as short-term (i.e., modulatory) or as long-term modifications (e.g., via gene expression changes) in behavioral or physiologic properties. The study of the mechanisms and constraints underlying these dynamic changes requires model systems that offer plastic phenotypes as well as a sufficient level of quantifiable behavioral complexity while being accessible at the physiological and molecular level. In this article, I explore how the new field of functional genomics can contribute to an understanding of the complex relationship between genome and environment that results in highly plastic phenotypes. This approach will lead to the discovery of genes under environmental control and provide the basis for the study of the interrelationship between an individual's gene expression profile and its social phenotype in a given environmental context.

Different data from different labs: lessons from studies of gene-environment interaction

It is sometimes supposed that standardizing tests of mouse behavior will ensure similar results in different laboratories. We evaluated this supposition by conducting behavioral tests with identical apparatus and test protocols in independent laboratories. Eight genetic groups of mice, including equal numbers of males and females, were either bred locally or shipped from the supplier and then tested on six behaviors simultaneously in three laboratories (Albany, NY; Edmonton, AB; Portland, OR). The behaviors included locomotor activity in a small box, the elevated plus maze, accelerating rotarod, visible platform water escape, cocaine activation of locomotor activity, and ethanol preference in a two-bottle test. A preliminary report of this study presented a conventional analysis of conventional measures that revealed strong effects of both genotype and laboratory as well as noteworthy interactions between genotype and laboratory. We now report a more detailed analysis of additional measures and view the data for each test in different ways. Whether mice were shipped from a supplier or bred locally had negligible effects for almost every measure in the six tests, and sex differences were also absent or very small for most behaviors, whereas genetic effects were almost always large. For locomotor activity, cocaine activation, and elevated plus maze, the analysis demonstrated the strong dependence of genetic differences in behavior on the laboratory giving the tests. For ethanol preference and water escape learning, on the other hand, the three labs obtained essentially the same results for key indicators of behavior. Thus, it is clear that the strong dependence of results on the specific laboratory is itself dependent on the task in question. Our results suggest that there may be advantages of test standardization, but laboratory environments probably can never be made sufficiently similar to guarantee identical results on a wide range of tests in a wide range of labs. Interpretations of our results by colleagues in neuroscience as well as the mass media are reviewed. Pessimistic views, prevalent in the media but relatively uncommon among neuroscientists, of mouse behavioral tests as being highly unreliable are contradicted by our data. Despite the presence of noteworthy interactions between genotype and lab environment, most of the larger differences between inbred strains were replicated across the three labs. Strain differences of moderate effects size, on the other hand, often differed markedly among labs, especially those involving three 129-derived strains. Implications for behavioral screening of targeted and induced mutations in mice are discussed.

Identification and ranking of genetic and laboratory environment factors influencing a behavioral trait, thermal nociception, via computational analysis of a large data archive

Laboratory conditions in biobehavioral experiments are commonly assumed to be 'controlled', having little impact on the outcome. However, recent studies have illustrated that the laboratory environment has a robust effect on behavioral traits. Given that environmental factors can interact with trait-relevant genes, some have questioned the reliability and generalizability of behavior genetic research designed to identify those genes. This problem might be alleviated by the identification of the most relevant environmental factors, but the task is hindered by the large number of factors that typically vary between and within laboratories. We used a computational approach to retrospectively identify and rank sources of variability in nociceptive responses as they occurred in a typical research laboratory over several years. A machine-learning algorithm was applied to an archival data set of 8034 independent observations of baseline thermal nociceptive sensitivity. This analysis revealed that a factor even more important than mouse genotype was the experimenter performing the test, and that nociception can be affected by many additional laboratory factors including season/humidity, cage density, time of day, sex and within-cage order of testing. The results were confirmed by linear modeling in a subset of the data, and in confirmatory experiments, in which we were able to partition the variance of this complex trait among genetic (27%), environmental (42%) and genetic x environmental (18%) sources.

Assessment of age-associated cognitive deficits in rats: a tricky business

Every living organism is affected by changes as a consequence of aging. Perhaps the most appropriate concept to describe age-related changes is that of 'functional age'. Laboratory rodents are especially suited as models of cognitive aging in humans, because they have a relatively short life-span and because many tests have been developed to investigate their cognitive performance. Examples from studies using the Morris water escape task were chosen to describe and discuss factors which must be considered before drawing conclusions about age-related cognitive deficits. In particular, the roles of rearing and housing conditions, of sensorimotor impairments, and of motivational differences between young and old rats are discussed. Conclusions are drawn about how aging studies should be conceived and performed.

Knockout mice: simple solutions to the problems of genetic background and flanking genes

Inducing null mutations by means of homologous recombination provides a powerful technique to investigate gene function and has found wide application in many different fields. However, it was realized some time ago that the specific way in which such knockout mutants are generated can be confounding, making it impossible to separate the effects of the induced null mutation from those of alleles originating from the embryonic stem cell donor. In addition, effects from null mutations can be altered on different genetic backgrounds. Here we present some simple breeding strategies to test for flanking gene effects that are compatible with the recommendations of the Banbury Conference on Genetic Background in Mice and with common practices of creating and maintaining mouse knockout lines.

Behavioral phenotyping enhanced--beyond (environmental) standardization

It is basic biology that the phenotype of an animal is the product of a complex and dynamic interplay between nature (genotype) and nurture (environment). It is far less clear, however, how this might translate into experimental design and the interpretation of animal experiments. Animal experiments are a compromise between modelling real world phenomena with maximal validity (complexity) and designing practicable research projects (abstraction). Textbooks on laboratory animal science generally favour abstraction over complexity. Depending on the area of research, however, abstraction can seriously compromise information gain, with respect to the real world phenomena an experiment is designed to model. Behavioral phenotyping of mouse mutants often deals with particularly complex manifestations of life, such as learning, memory or anxiety, that are strongly modulated by environmental factors. A growing body of evidence indicates that current approaches to behavioral phenotyping might often produce results that are idiosyncratic to the study in which they were obtained, because the interactive nature of genotype-environment relationships underlying behavioral phenotypes was not taken into account. This paper argues that systematic variation of genetic and environmental backgrounds, instead of excessive standardization, is needed to control the robustness of the results and to detect biologically relevant interactions between the mutation and the genetic and environmental background of the animals.