Maternal behavior definition: any behavior that contributes directly to the survival of offspring that have left the body of the female.
(PUBMED Neural basis of maternal behavior, 2013)
Oxytocin (OT) has been implicated in several aspects of reproduction, including sexual behavior, induction of labor, milk ejection, and maternal behavior. Oxytocin appears also to play a role in social behaviors, such as separation distress and affiliation as well as species-specific social behaviors, such as pair bonding in monogamous species.
Oxytocin is a nonapeptide produced primarily in the paraventricular (PVN) and supraoptic (SON) nuclei of the hypothalamus.
Although OT was originally characterized as a neurohypophyseal hormone, it is now clear that OT fibers project throughout the brain and that OT functions as a neurotransmitter. The central effects of OT are mediated by a seven transmembrain domain, G-protein-coupled receptor that is localized in discrete brain nuclei.
One of the most interesting features of brain oxytocin pathways is the species diversity in the neuroanatomical distribution of the receptor.
This plasticity in receptor distribution suggests that OT function may vary among species and may be related to species differences in social behavior. (PUBMED Species differences in central oxytocin receptor gene expression: comparative analysis of promoter sequences,1996.)
Transgenic technology now provides powerful tools for investigating the role of neuropeptides in controlling behaviors. Two strategies are used to investigate the function of OT in reproductive and social behaviors in rodents. First, characterizing the behavioral phenotype of a strain of mice that has been genetically engineered to lack the DNA sequence encoding the OT peptide. Second, beginning experiments which should ultimately allow the manipulation of the species-typical pattern of expression of oxytocin receptors in a targeted manner.
OT KNOCKOUT MICE
The oxytocin gene is composed of three exons which encode the oxytocin-neurophysin precursor peptide.
Using a targeted deletion vector, an OT knockout mouse was created by deleting the first exon of the OT precursor gene by homologous recombination in embryonic stem cells. This resulted in a mutant allele which completely lacks the nucleotide sequences encoding the OT peptide. As expected, no OT mRNA can be detected in the brain of homozygous knockout mice, while arginine vasopressin (AVP) mRNA appears to be unaffected
fig. A: OT mRNA in the paraventricular nucleus of an heterozygous mouse
fig. B: OT mRNA in the paraventicular nucleus of an homozygous mouse
fig. C: arginine vasopressin mRNA in the paraventricular nucleus of an heterozygous knock out mouse
fig. D: arginine vasopressin mRNA in the paraventricular nucleus of an heterozygous knock out mouse
The effect of the knockout allele on OT immunoreactivity in the PVN
and SON is illustrated in the picture below.
The density of OT-ir staining is reduced in heterozygotes (HET) compared to wild types (WT) and is absent in homozygote knockouts (HOM).
This graded, genotype-dependent decrement in OT peptide provides a useful opportunity to study the behavioral and physiological effects of modest as well as complete deficiencies in the OT system.
(PUBMED Gene targeting approaches to neuroendocrinology: oxytocin, maternal behavior, and affiliation, 1997.)
One of the most striking features of the OT and AVP system is the variability in the neuroanatomical distribution of the receptor system, both during development and among species.
It was predicted that if OT projections during development contribute to the final distribution of OT receptors in the adult brain, the OT knockout mouse should show an altered distribution of OT receptors.
Surprisingly, both the distribution and the concentration of OT receptors are unaffected in the OT knockout mouse (Fig. below), suggesting that presynaptic OT does not contribute to the neuroanatomical distribution of receptors during development and thus is unlikely to account for the species differences in receptor distribution.
figure: OT receptor autoradiography in heterozygote (A, C) and homozygote (B, D) OT knockout mice brains. No differences in receptor distribution or concentration of OT receptors were detected. ACo, anterior cortical amygdala; BL, basolateral amygdala; LS, lateral septum; Me, medial amygdala; Th, thalamus
(PUBMED Structure and expression of a human oxytocin receptor, 1992)
What about the behavior of the OT knockout mice?
As OT is thought to be crucial for several functions necessary for reproduction, we would predict that animals genetically deficient in OT would not mate readily, be unable to deliver, fail to lactate, and be deficient in maternal behavior. However, OT knockout mice mate, have normal gestation periods, build nests, and deliver pups normally. Pups from OT knockout mothers die within 24 hr due to the mother’s inability to eject milk.
Exogenous OT rescues milk ejection, allowing the pups to develop normally.
(PUBMED Oxytocin is required for nursing but is not essential for parturition or reproductive behavior, 1996.)
In order to measure maternal behavior quantitatively, behavioral tests were performed on the morning of parturition.
Two pups were removed from the nest and placed in opposite corners of the mother’s home cage and the behavior of the mother was videotaped for 30 min. The results reveal no differences in latency to retrieve pups into the nest, amount of time spent in the nest, and time spent grooming the pups, comparing mothers with and without OT.
Initially it was felt that it was possible that a related neuropeptide, such as vasopressin, might be interacting with the OT receptor to permit maternal behavior in the knockout mice.
To rule out this possibility, virgin OT knockout mice were fitted with micro-osmotic pumps which were used to centrally infuse either an oxytocin receptor antagonist [d(CH2)5,Tyr(Me)2,Thr4,Tyr-NH92]OVT at a constant rate of 75 ng/hr, or artificial CSF for 5 days. Maternal behavior was assessed as described above using foster pups on Days 1, 3, and 5 of treatment. On the last day, the brains were harvested and receptor autoradiography was used to determine the degree of receptor blockade.
This treatment effectively blocked 68% of the OT receptors in the lateral septum compared to CSF-infused controls; however, there was no difference in maternal behavior. Female mice lacking the gene for OT and with pharmacological blockade of the OT receptor appear to exhibit normal maternal behavior. These data further support the notion that neither OT nor the OT receptor is necessary for the normal expression of maternal behavior in mice. However, proof of this hypothesis awaits the development of an OT receptor knockout mouse.
For example, it is possible that the OT receptor could be activated by a ligand-independent mechanism as has been recently described for the progesterone receptor, although no evidence for such a mechanism in OT receptors has been reported.
(PUBMED Inhibition of post-partum maternal behaviour in the rat by injecting an oxytocin antagonist into the cerebral ventricles, 1987.)
Researchers also begun to characterize other social behaviors of the OT knockout mice (Table).
In a resident intruder paradigm, an ovariectomized, wildtype female was placed in the home cage of an adult male and the behavioral interactions were recorded.
Wildtype males typically spend much of the time in olfactory investigation of the intruder and show little aggression. Heterozygote and homozygote males show decreased olfactory investigation and increased aggressive behavior.
The levels of social investigation parallel, and the levels of aggression inversely parallel the OT immunoreactivity in each genotype.
Oxytocin has also been implicated in the ultrasonic isolation calls emitted by infant rodents when separated from the mother and littermates. Oxytocin injected into the brain reduces the frequency of these calls in a dose-dependent fashion. Separation distress was misured in OT knockout pups and found a significant decrease in distress calls in heterozygote and homozygote pups relative to wildtype pups. Again the change in behavior parallels the levels of OT immunoreactivity in each genotype. One interpretation of these paradoxical data is that animals deficient in OT fail to form social attachments early in life, and are therefore not distressed by the separation.
(PUBMED The social deficits of the oxytocin knockout mouse, 2002.)
An interesting parallel for these data comes from a comparison of social and asocial vole species. Prairie voles are highly social animals, spending over 50% of the time in side-by-side contact with conspecifics, while montane voles are virtually asocial. Isolation elicits a strong ultrasonic vocalization response in prairie vole pups, but little response in montane vole pups. It has been hypothesized that species differences in OT pathways may be associated with differences in social behavior in voles. The effects of the OT knockout appear to support this hypothesis and demonstrate the role of OT in the normal expression of social behavior.
IS MATERNAL BEHAVIOR INDEPENDENT OF OXYTOCIN?
The lack of a defect in maternal behavior in the OT knockout mouse suggests that the underlying neural circuits controlling maternal behavior are independent of oxytocin. However, this interpretation appears to contradict several studies in the rat (and in sheep) which report a role for oxytocin in the induction of maternal
Virgin rats do not show maternal behavior, but rather they ignore pups and sometimes exhibit infanticide.
Just prior to parturition, there is a rapid, dramatic shift in motivation from a lack of interest to a driven, relentless pursuit of nest-building, retrieval, licking, grouping of pups, and protection of pups. Central infusions of oxytocin in virgin rat facilitates this shift toward maternal behavior, whereas oxytocin antiserum and oxytocin receptor antagonists infused into the brain block the induction of maternal behavior.
However, once maternal behavior is established, oxytocin blockade has no effect.
In fact, established maternal behavior is unaffected by lesions of the PVN.
These data suggest that OT, released during parturition, acts as a switch to initiate the behavior and has little role in the maintenance of the behavior.
(PUBMED Oxytocin activates the postpartum onset of rat maternal behavior in the ventral tegmental and medial preoptic area, 1994.)
In contrast to the rat (and wild house mice), the strain of laboratory mice used to create the OT knockout, as well as other strains of laboratory mice are spontaneously maternal. Virgin females exhibit full maternal behavior immediately upon their first exposure to a pup. Since there is no shift in maternal behavior which occurs at parturition in laboratory mice, it is not surprising that OT-deficient mice show normal maternal behavior.
A species comparison of the distribution and regulation of oxytocin receptors in the brain suggests a potential mechanism for the species difference in OT function (Fig. below).
The pattern of OT receptors in the brain varies among each of the species studied. For example,rats have high concentrations of OT receptors in the bed nucleus of the stria terminalis (BnST), with little binding in the lateral septum (LS). The concentration of OT recep-tors in the BnST increases significantly at parturition, leading to the hypothesis that this region is important for the OT induction of maternal behavior.
In contrast, mice have little OT binding in the BnST and high concentrations in the LS.
(PUBMED Oxytocin receptor distribution reflects social organization in monogamous and polygamous voles, 1992.)
The disparity in OT-sensitive brain structures among species suggests that we must be careful in making generalizations regarding the relationship between OT and specific behaviors. This, in addition to the behavioral results of the OT knockout mice, demonstrates that the role of oxytocin in the regulation of social behaviors must be considered on a species-by-species basis.
Therefore, the development of primate models will be necessary to elucidate the behavioral functions of OT in the primate and human brain.
A second transgenic approach for investigating the role of oxytocin in social behavior involves manipulating receptor expression. As discussed above, species differences in oxytocin mediated behavior may be related to species differences in oxytocin receptor distribution. Comparative studies in voles support this hypothesis.
Prairie voles (Microtus ochrogaster) are highly social and display high levels of affiliative behavior (prairie voles spend more than 50% of the time in side-by-side contact). In the field and in the laboratory, prairie voles nest in mated pairs, both mother and father contribute to the rearing of the offspring, and mated pairs form long-lasting social bonds with each other.
In contrast to prairie voles, montane voles (M. montanus) are asocial, are rarely in physical contact with other conspecifics, nest in isolation, and do not form social bonds between mates. These species differences in social behavior are striking because these species are closely related and exhibit relatively few differences in nonsocial behaviors, such as locomotor activity or responses to novelty.
What physiological differences between prairie voles and the nonmonogamous montane voles could account for the differences in social behavior? Pair bonding in females of the monogamous prairie voles appears to be mediated by OT. Although few differences in OT-producing cells or fibers have been found, the distribution of OT receptors in the brain is very different between the species.
This striking difference in OT receptor binding patterns led to the hypothesis that differences in OT pathways may be associated with species differences in social organization. Comparison of OT receptor patterns in other monogamous and nonmonogamous voles species revealed a correlation between receptor binding patterns and social organization. Species differences in binding patterns in voles could be due to species differences in posttranscriptional processing of the gene product (such as neuronal transport), or to differences in regional gene expression in the brain. Using in situ hybridization, it was determined that the pattern of OT receptor binding was similar to the pattern of OT mRNA expression in the brain in each species.
Therefore species differences in receptor distribution are due to species differences in region-specific gene expression.
(PUBMED Species differences in central oxytocin receptor gene expression: comparative analysis of promoter sequences, 1996.)_
Transgenic technology has added to the battery of techniques available for investigating the physiological function of oxytocin. The oxytocin knockout mice have provided surprising results regarding the neuroendocrine regulation of maternal behavior, emphasizing the importance of species diversity in OT function. The data demonstrate that the neural circuits underlying maternal behavior are independent of oxytocin; however, in some species, OT released at parturition may act as a switch to activate these circuits. This appears to be the case in rats as well as sheep. It is interesting to speculate whether the rat or mouse is a more appropriate model for maternal behavior in the human female. Nevertheless, the OT knockout mice demonstrate the significance of OT modulation of social and attachment behaviors.
The ability to manipulate OT receptor gene expression using target transgene expression may also provide a powerful tool for exploring the relationship between receptor distribution and OT mediated behaviors. The great diversity in receptor localization and regulation, even among closely related species, suggests that the plasticity in receptor gene expression may play a significant role in the control and the evolution of species-specific social behaviors. The ability to alter receptor expression of a species in a single generation using a transgenic approach offers an exciting possibility to manipulate social behavior in a species and to better understand the neural substrates controlling these behaviors.