imprinting

n.
A rapid learning process by which a newborn or very young animal establishes a behavior pattern of recognition and attraction to another animal of its own kind or to a substitute or an object identified as the parent.

Imprinting refers to the chemical modification of the DNA in some genes that affects how or whether those genes are expressed. One particular kind of DNA imprinting found in mammals is known as parental genomic imprinting, in which the sex of the parent from whom a gene is inherited determines how the gene is modified. While imprinting has been found in only about fifty human genes to date, some estimates suggest it may occur in several hundred more, in perhaps up to 1 percent of all genes. Imprinting defects are responsible for several human diseases, including some forms of cancer. Imprinting also occurs in other organisms, from yeast to plants to fruit flies.

Gene Expression in Imprinted and Nonimprinted Genes

Chromosomes, and the genes they contain, are inherited in pairs, with one copy of each supplied from each parent. For most genes, both members of the pair, called the maternal and paternal alleles, are used equally. Both are expressed (read by the transcription machinery to make protein) in roughly equal amounts.

In contrast, for most imprinted genes, only one allele is expressed, while the other copy is silenced by imprinting. For some genes it is the maternal copy, for others it is the paternal copy. This is an exception to the Mendelian assumption that the two parents contribute equally to the phenotype controlled by autosomal genes. For some genes, both alleles are expressed, but one copy is expressed much more than the other. For some genes, the silencing occurs in some tissues but not others.

Imprinted genes should not be confused with sex-linked genes, which are carried on the X or Y chromosome. Most imprinted alleles are located on autosomes, but are "stamped" with the sex of the parent that contributed them.

Imprinting should also not be confused with dominant and recessive alleles, in which one allele always controls the phenotype at the expense of the other, because of differences in the alleles themselves. The "dominance" seen in imprinting is determined by the sex of the parent contributing the allele, not any property of the allele itself. Thus, a particular allele will appear to be recessive when inherited from one parent, but dominant when inherited from the other. Such an effect, in which the expression difference is not due to the alleles but to forces acting on them from outside, is termed an "epigenetic effect."

Imprinting is thought to be responsible for many cases of incomplete penetrance, an inheritance pattern in which a dominant gene (as for a genetic disease) is not expressed in some individuals despite being present. Imprinting offers a mechanism by which a particular allele can be turned on or turned off as it is passed down through successive generations.

Timing and Mechanism of Imprinting

Although the details of imprinting are still unknown, it is clear that imprinting must occur either during the formation of the gametes or immediately after fertilization, while the two chromosome sets are still distinct. The imprint must be reliably passed on to each new daughter chromosome during DNA replication.

The exact molecular mechanism of imprinting is also unknown, but it is thought to involve the modification of a gene's promoter. The promoter is the upstream region to which RNA polymerase binds to begin transcription. Imprinting prevents or restricts binding of RNA polymerase, thus preventing gene transcription.

One method by which a gene becomes imprinted is believed to be by the addition of methyl groups (-CH₃) to cytosine nucleotides in the promoter region. The evidence for methylation is strong. Methylation is a common mechanism for gene silencing, because these bulky side groups interfere with the efficient binding of the various transcription factors required to attract the polymerase enzyme. Methylation patterns are known to be altered during gamete formation, and are reliably passed on during replication. Further evidence comes from the observation that altered methylation patterns in some imprinted genes are associated with the aberrant expression of the normally silent allele.

Example of Imprinting: the Igf2 Gene

One of the best-studied imprinted genes is the one that encodes an insulin-like growth factor called growth factor 2 (IGF2). In this gene, the paternal copy is active, whereas the maternal copy is inactive. Imagine that two parents have produced a female child. During egg formation in the mother (or shortly after fertilization), the mother's copy of the IGF2 gene is methylated, rendering it transcriptionally silent. The child uses only the paternal allele to make the growth factor. However, when this child makes her own eggs, neither copy of the gene will remain active, because the alleles will have been "restamped" as coming from a female. The active allele she used throughout life is passed on in an inactive form to her children.

The protein encoded by the IGF2 gene is a growth factor, which stimulates the growth of target cells. Failure to properly imprint the maternal allele, or inheritance of two copies of the male allele, can have important consequences. For example, the expression of two copies of the IGF2 gene is associated with Beckwith-Wiedemann syndrome, a growth disorder, accompanied by an increase in a type of cancer called Wilms tumor. Other human cancers are also associated with improper imprinting (of other genes), causing either too much or too little gene expression.

Uniparental Disomy and Human Disease

Inheritance of two copies of one parent's chromosome (or part of it) is called uniparental disomy, a type of chromosome aberration. Detection of uni-parental disomy in individuals with genetic disorders was one of the first clues that imprinting had important developmental and medical consequences.

Prader-Willi syndrome and Angelman syndrome can both be caused by uniparental disomy of chromosome 15, which carries a maternally expressed, paternally imprinted gene. Two maternal copies of the gene causes Prader-Willi syndrome, which is marked by mild mental retardation, decreased growth of the gonads, and obesity. Two paternal copies of this same gene causes Angelman syndrome, marked by severe mental retardation, small head size, seizures, inappropriate laughter, and distinctive facial features. (The gene itself codes for a protein involved in degrading other proteins.) Imprinting defects can also cause these syndromes in the absence of uniparental disomy, since the result is the same: either zero or two copies of the gene are expressed.

Why Imprint?

The evolutionary reason for imprinting is not yet clear, although some scientists propose that, at least in mammals, it arose from an evolutionary tug of war between males and females. In this scheme, fathers (who contribute only sperm) benefit when the embryo grows as fast as possible. Thus, silencing genes that slow down embryonic growth is in their interest, even if it depletes resources from the mother. Mothers, on the other hand, need to conserve their resources. Silencing genes that promote rapid growth is therefore in their interest. Supporting this hypothesis is the fact that many of the known imprinted genes regulate growth. Paternally expressed (maternally imprinted) genes such as IGF2 tend to promote growth, whereas maternally expressed (paternally imprinted) genes tend to inhibit it.

Bibliography

Everman, David B., and Suzanne B. Cassidy. "Genomic Imprinting: Breaking the Rules." Journal of the American Academy of Child and Adolescent Psychiatry 39, no. 3 (March 2000): 386-389.

Greally, John M., and Matthew W. State. "Genomic Imprinting: The Indelible Mark of the Gamete." Journal of the American Academy of Child and Adolescent Psychiatry 39, no. 4 (April 2000): 532-535.

Paulsen, Martina, and Anne C. Ferguson-Smith. "DNA Methylation in Genomic Imprinting, Development, and Disease." Journal of Pathology 195, no. 1 (2001): 97-110.

Internet Resources

Geneimprint.com. http://www.geneimprint.com/.

Yale University School of Medicine and Yale-New Haven Hospital. http://info.med.yale.edu.

—Richard Robinson

For more information on imprinting, visit Britannica.com.

A form of learning that occurs in young animals, usually during a critical, sensitive period of their lives. During imprinting, a young animal learns to direct some of its social responses to a particular object, usually a parent.

Imprinting occurs when an animal learns to make a response to a particular animal or object. Usually the behavior is learned by a young animal through exposure to a stimulus early in its development. There are two types of imprinting: filial (social attachment) and sexual. Perhaps the most famous example of imprinting is that of Konrad Lorenz (1903-1989) and the goslings who imprinted him as their mother and followed him in a classic example of filial imprinting.

Previous question:	What does genetics have to do with behavior?
Next question:	Do all animals use imprinting?

---

When recently hatched birds such as ducklings are hand reared for a few days, they strongly prefer the company of their human keeper to that of their own species. This remarkable process, which can so dramatically influence the development of social relationships, is called 'filial imprinting'. A bird's experience later in life can also strikingly influence its sexual preferences. When exposed to another species at a certain stage in development, it may subsequently prefer to mate with that species. The process influencing mate choice is known as 'sexual imprinting'.

In many ways the term 'imprinting' is a misnomer since it implies an instantaneous stamping into the bird's memory of information that will determine its social preferences. In fact, the process is more gradual and is better likened to the painting of a portrait in which the broad outlines are first sketched and fine details are filled in by degrees. Furthermore, despite the remarkable distortions of social preferences that can occur under artificial conditions, some stimulus objects are much more effective than others. In general, objects that are most like the animal's own species are most effective in influencing its subsequent preferences.

The procedures used for studying imprinting are not associative. The experimenter merely exposes the bird at the right age to a single conspicuous object and subsequently shows that its preferences have become restricted to that object. Even so, the underlying process might be associative. If this were the case, initially significant and attractive features of the stimulus object would be linked to initially neutral features in the course of learning. Deciding whether or not this is really what happens is not easy. Even if the imprinting process that leads to the identification of parents (or future mates) is not associative, it operates simultaneously with both classical and operant conditioning. What is more, the objects that work best for purposes of imprinting can be used with the associative procedures as rewards. Thus, the imprinting stimulus is highly effective as a reinforcer, and a young duckling can easily be trained to press a pedal that switches on a motor and brings an object into motion. However, enough is now known about the neural mechanisms to dissociate by lesions in the brain the operant conditioning which involves learning to press a pedal from the imprinting which involves the narrowing of preferences to a familiar object.

As it happens, filial imprinting in domestic chicks has proved to be a useful model for studying the neural basis of learning. Using biochemical measures of neural activity and autoradiography, a site in the intermediate part of the medial hyperstriatum ventrale (IMHV) has been found to be intimately involved in filial imprinting. When the neurons in this area are destroyed before imprinting, a chick fails to identify the object with which it has been trained; furthermore, when they are destroyed immediately after imprinting, a chick no longer responds preferentially to the object with which it was trained. Nevertheless, lesioned birds are just as capable as intact animals of learning discriminations in which reward is involved, and will work just as readily for an imprinting object — even though they do not learn its characteristics.

Some other discoveries of the work on the neural mechanisms involved in imprinting are also striking. Storage only remains localized in the IMHV on the left side of the brain. On the right side, information is stored initially in the IMHV but within 24 hours it has dispersed to other, and as yet unknown, sites in the brain. In the left IMHV, where information about the imprinting object remains stored, the dendritic synapses increase in size. It would be an exciting as well as demanding task to discover whether the pattern of synaptic enlargement codes the information stored in the course of imprinting.

One of the striking features of filial imprinting is that it is easiest to get birds to learn about an object at a particular stage early in life. This stage is referred to as a 'sensitive period' in development. However, it is important not to confuse the descriptive evidence for such a period with explanations for how the evidence has been generated. The nature of the timing mechanisms has proved to be subtle, and the image of a clock, opening a window onto the external world and then shutting it again, is not good enough.

Generally, the onset of sensitivity is measured in terms of time after hatching. However, birds can differ by as much as 30 hours in the stage of development at which they hatch. In other words, when eggs have been incubated under identical conditions, the time from the beginning of incubation to hatching varies greatly. It is possible, therefore, to have birds of the same post-hatch age which are at different stages of development, and birds which are at the same stage of development but of different post-hatch ages. The influence on imprinting of the general stage of development can be separated from the influence of experience occurring at hatching and after it. It turns out that both the age from beginning of embryonic development and the age from hatching influence the results. In other words, it looks as though the general stage of development of the bird plays a part in the onset of sensitivity, but that the events associated with hatching, or experiences subsequent to it, also play their part. Part of the increase in sensitivity is attributable to the changes in the efficiency of the visual system. This being the case, the interaction between internal and external influences is particularly easy to understand. Visual experience with patterned light has a general facilitating effect on the development of visually guided behaviour. It also serves to strengthen connections in neural pathways. Thus, it is probable that the development of the visual pathways, on which filial imprinting must depend, can be accelerated by early hatching if this means that the bird receives more experience from patterned light than the bird that is still inside the egg.

The end of sensitivity to novel objects arises from a property of the imprinting process. When a bird has formed a preference for an object, as a consequence it ignores or even escapes from other objects. Therefore, imprinting with one object prevents further imprinting with other objects from taking place. While some objects are much more effective than others in eliciting social behaviour from naive birds, domestic chicks can form social preferences for suboptimal stimuli such as the static cages in which they were isolated, although this takes several days. It follows, that if some birds are reared with near-optimal stimuli, such as their siblings, and some are reared in isolation, it should be possible to imprint the isolated birds with a novel object at an age when the socially reared birds escape from, or are indifferent to, new things. This is clearly the case with both domestic chicks and mallard ducklings. Investigators find it difficult to get older birds to respond socially to an object, because the birds have already developed a preference for something else.

It seems, then, that the end of sensitivity of a bird to a wide range of objects is generated by a runaway process in which the consequences of an initial change in state generate the conditions for further and more rapid changes. When the bird is poised to learn, all that is required is the presence of a suitable external stimulus to trigger the process. As a result of learning about a particular object, the bird escapes from those it can detect as being different. This escape from strange objects intensifies the young bird's contact with a familiar object, and the range of acceptable companions rapidly contracts so that in the natural world the bird ends up with a distinct preference for its own mother. Enforced contact with something other than the preferred familiar object may wear down the behavioural constraints to the point where the bird does form a new preference. Escape from novel objects can be habituated, and domestic chicks that have developed a preference for object A can be induced to prefer object B to object A by sufficiently long exposure to the second object. It turns out that the early preference is not forgotten and may resurface after isolation from both types of object. So, it would seem, the protection of preferences from subsequent disruption is accomplished not only by behavioural means, but also by internal processes as well.

For many years it was thought that imprinting enabled an animal to learn about its species. However, it is becoming apparent that imprinting serves a more subtle but equally important purpose, enabling the young to recognize one or both of their parents as individuals. The species that are feathered and active at hatching must learn the parental characteristics quickly because, on leaving the place where it was hatched, the young bird must stay near a parent. If it approaches another adult of its own species, it may be attacked or even killed. For this reason, as soon as the young animal is able to recognize its parent (or a substitute parent in an experiment), it escapes from anything that is noticeably different. The rapid attachment to the first conspicuous objects encountered after hatching is characteristic of the precocious species. In birds like the swallow, which are hatched naked and helpless, learning occurs later in development, and the young respond selectively to their parents only when they have left the nest about two weeks after hatching. Sexual imprinting probably occurs later in development because birds need to know the characteristics of their siblings' adult appearance when they come to choose a mate. The ideal mate must be a bit different but not too different from close relatives in order to avoid the dangers of inbreeding and, at the other extreme, the dangers of mating with a genetically incompatible partner such as a member of another species. Some animals are so careful in their choice of sexual partner that, having been reared with their siblings, they then choose first cousins as mates. This has been discovered in Japanese quail and also in mice. Indeed, it is becoming apparent that processes similar to those first described in birds are equally important in mammals. The biological need to recognize close kin, both early in life and when adult, is a general one.

(Published 1987)

— P. P. G. Bateson

Bibliography
• Bateson, P. (2000). 'What must be known in order to understand imprinting?' In Heyes, C., and Huber, L. (eds.), The Evolution of Cognition.
• Horn, G. (1986). Memory, Imprinting and the Brain: An Inquiry into Mechanisms.
• — —  (1991). 'Cerebral function and behaviour investigated through a study of filial imprinting'. In Bateson, P. (ed.), The Development and Integration of Behaviour: Essays in Honour of Robert Hinde.

---

A kind of learning in the very young based mainly on maternal attachment and acquisition of basic behavior patterns.

This entry is from Wikipedia, the leading user-contributed encyclopedia. It may not have been reviewed by professional editors (see full disclaimer)