The perinatal period is critical for health and behavior later on life. The idea that the cause of a disease could be related to this critical period has been developed 30 years ago in the concept called “Primal Health research” (for review, Odent, 2001) that is now abandoned despite its relevance. The perinatal period is critical for pathologies related to feeding disorders or to sociability. An important factor to consider during this “primal period” is the neuroendocrine system (Lesage et al., 1996). The central oxytocin (OT) system, which projects to limbic and other central brain areas, appears to be a key regulator of the evolution and expression of different types of social systems in many species (for reviews: Choleris et al., 2009; Heinrichs et al., 2009; Insel, 2010). The current view is that OT modulates the coupling of central neuronal networks that process sensory cues activated during social interactions (Johnson and Young, 2017). More generally OT has been proposed as an allostatic hormone, acting in the brain and in peripheral organs, where it modulates both social and non-social vital behaviors by maintaining stability through changing environments (Quintana and Guastella, 2020). However, these assignments have been applied to the role of OT at adulthood.

During the last years, the scientific community studied the dynamic development of the neural OT circuitry. They have highlighted that how and when the neural circuit of the OT-system is organized in early development might have important consequences on behavior, a hypothesis indicates the “organizational effects of OT” (Eaton et al., 2012; Miller and Caldwell, 2015). More recently, studies revealed that an impact on OT-system ontogenesis during early-life period affects long-term behavioral expression (Onaka and Takayanagi, 2021). Furthermore, it is striking that the distribution of OT-receptor (OTR) in the brain undergoes dynamic changes in expression through the postnatal development with a peak of expression in the early infancy (in rodents and humans) and that this distribution correlates with brain regions controlling the sensory modalities primarily used to initiate social interactions (Hammock and Levitt, 2013; Tamborski et al., 2016).

Here, we discuss the early postnatal life role of OT shaping the sensory inputs and stimulating suckling, both contributing to the first social behaviors via the infant-mother bond. We also report the positive effects, in particular the long-term effects, of an early-life OT-treatment in neurodevelopmental diseases (NDD) such as Schaaf-Yang and Prader-Willi syndromes.

OT-system is defined by OT-producing cells and OT-target cells that together determine the qualitative and quantitative characteristics of OT responses. The action of OT is then controlled by the release of OT, the quantity and spatio-temporal expression of OT-binding sites on the target cells (OT-receptors and vasopressin receptors), and the intra-cellular signaling pathways triggered by OT. Many reviews have described the source of OT production (Althammer and Grinevich, 2017), the mechanisms of release (Johnson and Young, 2017; Brown et al., 2020), and the cellular signaling pathways via the OT-binding sites (Busnelli and Chini, 2018; Jurek and Neumann, 2018). Here, we will focus on developmental setup of the OT-system.

OT is produced by hypothalamic neurons of several regions: the Supraoptic Nucleus (SON), the Periventricular preoptic nucleus (PvPO) the Paraventricular Nucleus (PVN), the Accessory Nuclei (AN), the Bed Nucleus of the Stria Terminalis (BNST) and in a recently identified region named Antero-Lateral Preopticarea (ALPO; Soumier et al., 2022). Recently, a comprehensive map of OT/AVP neurons and projections has been generated in the mouse brain during development (Madrigal and Jurado, 2021) and at adulthood.¹ OT and AVP expression show distinct developmental dynamics in the mouse brain, with developmental maturation of AVP preceding that of OT (Madrigal and Jurado, 2021). The number of OT cells selectively increases during a critical window of postnatal development in several hypothalamic regions: in the PvPO and PVN, as well as in ALPO (Soumier et al., 2022). Noticeably, AVP-expressing cells in the different hypothalamic nuclei (PVN, SON, Medial Preoptic Nucleus, Tuberal Nucleus), as well as in the extra-hypothalamic regions, such as the Medial Amygdala (MEA) and BNST, remain stable over time, from birth to adulthood (Soumier et al., 2022). Interestingly, in most of hypothalamic nuclei, there is a high percentage of neurons coexpressing OT and AVP that decreases at adulthood (Madrigal and Jurado, 2021).

OTR is a seven-transmembrane segment G protein-coupled receptor (GPCR). To date only one subtype of OTR has been described which is closely related to the three vasopressin (AVP) receptors, OT binds also AVPR-1a with a lower affinity. OTR mapping at the protein level is hampered by the lack of specific anti-OTR-antibody. Consequently, the distribution of OTR expression has been analyzed using different experimental approaches, such as receptor binding of radiolabeled OT analogs on tissue sections, in situ hybridization and transcriptomic analysis, and the use of transgenic mice expressing a fluorescent reporter under the control of the OTR promoter. In adult rodents, a comprehensive list of brain areas expressing OTR has been reported in the rat and mouse brain (Jurek and Neumann, 2018). The four main and best-studied OTR-expressing brain regions are the hypothalamus, prefrontal cortex, hippocampus, and amygdala (Jurek and Neumann, 2018; Carter et al., 2020). Son et al. (2022) have shown that there is no significant correlation comparing the quantity of OT projections and OTR expression across the whole brain, suggesting important indirect OT-signaling in OTR-expressing areas.

Comparative analysis of OTR distribution in brain rodents (i.e., prairie voles, rats, and mice) revealed species, sex, experiences, and developmental differences in OTR expression (Vaidyanathan and Hammock, 2017). In rats, using radiolabeled OT, it was shown that OTRs progressively appear in several brain regions throughout the development, being first detected in the presumptive embryonic region of the vagal motor nucleus from the embryonic day 14.5 (E14.5), and reaching a well-defined “infant” pattern of distribution after the Postnatal day 10 (PND10). OTR expression then follows a differential time course depending on the brain structure considered. Interestingly, during the early postnatal period some transient expressions are detected in several brain areas (Tribollet et al., 1989, 1991; Yoshimura et al., 1996). After PND13, the density of OTR drops sharply in several brain regions, and expression in new brain areas appears; this is referred to as the first transition to the adult pattern, which is almost reached at PND18. Around weaning and beyond, a second transition takes place that is characterized by a new remodeling of OTR expression, disappearing from some areas and increasing in others. Finally, the adult pattern of OTR expression is established at PND60-90 (Vaidyanathan and Hammock, 2017; Muscatelli et al., 2018).

A similar radioligand binding approach was performed in the mouse brain and reported an early detection of OT-binding sites atE16 (Tamborski et al., 2016), with a peak around 2 weeks after birth followed by a decrease in OT-binding sites in all brain regions thereafter (Hammock and Levitt, 2013); such a strong a transient expression of OTR around PND14 is particularly detected in different cortical regions (Hammock and Levitt, 2013). Yongsoo Kim’s lab used a fluorescent reporter mouse model (i.e., OTR Venus mice) and established a publicly available brain-wide map of the OTR in mice during postnatal development² (Newmaster et al., 2020). The drawback of this map is that it does not allow to visualize transitory expression. Studies in prairie voles also showed a dynamical expression profile of OTR over time (Prounis et al., 2018).

In conclusion, in all rodents examined (i.e., mice, rats, and prairie voles), Otr mRNA and OT-binding sites are present in embryos, even though the mature form of OT is not detected before birth, and the peak of OTR expression is observed around 2 weeks after birth. The distribution of OTRs during brain development is different from that of adult brains and can be classified into three expression profiles: 1) clusters of neurons with constant expression, where OTR expression begins to be detected during development and is maintained throughout life, 2) sites with transient expression where OTR is only observed during a developmental time window and its expression is no longer detectable after that, and 3) neurons where OTR expression begins to be detected during puberty and is maintained throughout life (Vaidyanathan and Hammock, 2017). OTR expression is sexually dimorphic, with Otr mRNA expression generally higher in female than in male brains (Tamborski et al., 2016).

In the human brain, similarly to rodents, OTR expression begins to accelerate just before birth, with a peak level expression occurring during infancy (Rokicki et al., 2022). There is, however, a sexual dimorphism of OTR expression. In girls, a peak of expression is observed around birth, with a decrease in expression during childhood, and a dip during adolescence. In boys, compared with girls, the peak of OTR expression is delayed in early childhood and the difference in expression between brain regions is more pronounced (Rokicki et al., 2022). This study used genome-wide exon-level transcriptome approach to define OXTR transcript density in 16 brain regions through different ages (pre-natal to 82 years old) and using males and females. The study is robust but there is a lack of quantification of the active transmembrane OT-receptors.

OTR are also strongly detected in the peripheral tissues of neonatal mice and prairie voles. In particular, OTR is transiently and highly expressed in the oro-facial region of mouse with marked sex and species differences (Greenwood and Hammock, 2017).

Noticeably, the distribution of OT and vasopressin (VP) receptors shows overlapping expression in many regions and specific non-overlapping expression in some brain structures as shown in the lemurs (Grebe et al., 2021). These two neural systems play a role in the same signaling pathway, and have either complementary or opposing functions depending on the species, sex, age, and social experiences.

The dynamic of OT release in postnatal development is poorly studied and, paradoxically, although there is a huge number of OTR in the infant brain, the synaptogenesis of OT neurons is not fully completed. Interestingly, Hoffiz et al. (2021) showed that birth triggers specifically c-Fos activation of VP and OT neurons from the suprachiasmatic (VP only), supraoptic, and paraventricular nuclei of the hypothalamus. These activations are high at 3 h postnatal and returned to baseline levels at P1.

In conclusion, the pattern of OT-system is highly dynamic throughout development, particularly from birth with transition phases at weaning time, puberty, and young adulthood (Figure 1).

At adulthood, the role of OT as a modulator of sensory processing in relation with the social behavior has been clearly shown for olfaction (Oettl et al., 2016; Oettl and Kelsch, 2018) and touch (Tang et al., 2020; Yu et al., 2022). OT has been shown to modulate pain perception (Poisbeau et al., 2018). In dams, OT modulates audition allowing them to recognize the vocalizations of their pups and to adapt maternal behavior accordingly (Schiavo et al., 2020).

During postnatal development, numerous studies show a role of OT-system as an integrator and modulator of sensory inputs, allowing to shape brain circuits and connectivity (Grinevich and Stoop, 2018). Interestingly, VP and OT neurons of the supraoptic and paraventricular nuclei of the hypothalamus were specifically activated in the 30 min following birth in mice and rats (Hoffiz et al., 2021).

Some of the pioneer studies showed that in rat (Nelson and Alberts, 1997; Lenz and Sengelaub, 2010), rabbit (Caba et al., 2003), or prairie voles pups (Barrett et al., 2015), mimicking parental licking stimulates the activity of hypothalamic OT neurons. In the rat, this stimulation also induced an increase in OT concentration in the spinal cord, suggesting that maternal licking may affect the maturation of sensory and autonomic centers (Lenz and Sengelaub, 2010). Affiliative stimuli, defined by gentle tactile stimuli of the pups during developmental period, activate also OT neurons (Walker et al., 2017). Maternal care (skin to skin contacts) stimulates the central OT in pups creating the conditions for inducing a preference for maternal odor (Kojima and Alberts, 2011a; Kojima et al., 2012) and for establishing a social affiliation in rat pup’s filial huddling preference (Kojima and Alberts, 2011a; Kojima et al., 2012). The increase in OT induced in pups by contact with the mother also facilitates the development of thermal-seeking huddling behavior (Kojima and Alberts, 2011b).

One of the most convincing demonstration of the sensory cues-dependent release of OT in the pups’ brain comes from experiments based on blocking early sensory stimulation (Zheng et al., 2014). Deprivation of sensory inputs, such as whisker deprivation, right after birth (ending at PND 14) results to reductions of the firing rates of OT PVN neurons and lower released of OT in the brain of juvenile mice (Zheng et al., 2014). This deprivation also affects other sensory cortices in the brain and caused a reduction in excitatory synaptic transmission across the other sensory cortices, demonstrating that OT promotes cross-modal, experience-dependent cortical development. Similar results were obtained in dark rearing. Injection of OT into the brains of sensory-deprived animals enabled them to recover this deficiency and improved the brain’s response to other sensory inputs. Conversely, increased sensory stimulation from birth through environmental enrichment increases at PND14 the level of OT, excitatory synaptic transmission in several sensory cortices, and corrects the effects of sensory deprivation. This study thus demonstrates that during the period of synaptogenesis, OT promotes excitatory synaptic transmission and, in sensory cortices, mediates early experience-dependent multimodal plasticity. Noticeably, this action of OT, which regulates excitatory synaptic transmission in pyramidal neurons of sensory cortices, is maximal around PND14 and ceases after PND18, revealing a sensitive/critical period (Zheng et al., 2014). Similarly, sensory experience regulates the expression of OTRs with a similar time course, an elevation at PND14, and essentially no change at PND18 (Maldonado et al., 2021). More generally, OTR expression in the pyramidal glutamatergic neurons of sensory cortex peaks around PND14 and drops significantly at PND21 (Hammock and Levitt, 2013; Newmaster et al., 2020; Maldonado et al., 2021). Then, at PND28, OTR expression is more important in GABAergic neurons, especially somatostatin interneurons (Zheng et al., 2014). Thus, a reduction in OTRs expression may promote the closure of the sensitive/critical period (Zheng et al., 2014).

In a similar manner, it has been shown that OT affects spontaneous activity patterns in the mouse developing visual cortex, recruiting Somatostatin interneurons and allowing to refine the synaptic connections in the visual cortex prior to eye opening (Maldonado et al., 2021). OT is also required in early infancy to impose the positive quality of olfactory imprinted memory in the nursed neonates, in which imprinted memory is associated with a pleasant mental state. Importantly, there is a critical period (first week of life) during which OT, by imposing positive quality on imprinted odor memory, contributes to smooth social interactions in the adult life (Inoue et al., 2021). Noticeably, during early infancy, the transient strong oro-facial expression of OTR (see above) suggests also an effect of OT-system in the peripheral control of sensory inputs. In our team, we have shown a role of OT in thermo-sensory reactivity of neonates. Indeed, in a cool environment, neonate mice emit ultrasonic vocalizations that trigger a “pup retrieval” behavior from their mothers. Inactivation of OT neurons prevents this thermo-sensory reactivity of neonates (Da Prato et al., 2022).

Altogether, these data reveal a stimulation of OT-system in response to all types of sensory inputs (tactile, visual, olfactive, thermal), in particular those linked to the oro-facial stimulation, triggered at birth and during a critical period of time of infancy (Figure 2). In this period, sensory experience influences the production of OT and OT shapes neuronal circuits by modulating spontaneous and evoked activity.

In all mammals, the mother-infant bond is the first social experience. Many studies report the role of OT in mother-infant bonding induced by lactation and maternal behavior in the mother (Nagasawa et al., 2012; Okabe et al., 2012). In addition, an important effect of maternally produced OT at the time of birth could protect the neonatal brain of rodents from birth stress by decreasing perinatal neuronal activity (Tyzio et al., 2014). On the other side, the role of OT produced by the newborn in stimulating bonding with its mother has been much less studied. The newborn’s OT-system is activated by parental care which is responsible for all of the sensory inputs described above (Paragraph 2); deficiencies in parental care, as it happens in maternal separation, impact the OT-system and alter social and emotional behaviors later on in adulthood (Onaka and Takayanagi, 2021).

The first sensorimotor and social behavior of a newborn mammal is to identify its mother, find the nipple, and suckle the milk. It is therefore not surprising that OT, by integrating sensory inputs in newborns, plays an important role in these processes. The OT-system is involved, from birth, in the first social interactions associated with feeding behavior on the mother’s side (lactation, suckling) and on the baby’s side (suckling) allowing the creation of a strong mother-infant bond. Thus, in addition to promoting the development of the sensory-motor system, OT has a concerted action in the mother and the baby, promoting the initiation of feeding behavior (Figure 5). How these two roles are related and dependent on OT circuits is illuminated by the pathologies in which both behaviors are affected.

Presumably, eating disorders in early childhood may lead to alterations in social behavior later in life. Early eating disorders could be an early marker of neurodevelopmental disorders such as ASD and could be an indication for early OT-treatment of these patients.

Considering the topic of this article, the main general questions to address in rodent models now are:

These questions can be specifically applied to the role of OT in the initiation of sucking in rodents. To answer these questions, it would be necessary to develop efficient tools to study, around birth and in the first days of life, the connectivity of OT neurons and to manipulate OT neurons by optogenetics and chemogenetics.

In addition, studies to further analyze the alterations in early feeding and sucking behavior in ASD patients would be needed.

FM designed and wrote the manuscript. VM and BC participated in the writing. All authors contributed to the article and approved the submitted version.

This work was supported by the Foundation for Prader-Willi Research, Prader-Willi France, and Tonix Pharmaceuticals.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.