REVIEW published: 07 October 2021 doi: 10.3389/fgwh.2021.742775 Characterizing the Brain Structural Adaptations Across the Motherhood Transition Magdalena Martínez-García 1,2*, María Paternina-Die 1,2 , Manuel Desco 1,2,3,4 , Oscar Vilarroya 5,6 and Susanna Carmona 1,2* 1 Instituto de Investigación Sanitaria Gregorio Marañón, Madrid, Spain, 2 Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM), Madrid, Spain, 3 Departamento de Bioingeniería e Ingeniería Aeroespacial, Universidad Carlos III de Madrid, Madrid, Spain, 4 Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain, 5 Department of Psychiatry and Legal Medicine, Autonomous University of Barcelona, Barcelona, Spain, 6 Hospital del Mar Medical Research Institute, Barcelona, Spain Women that become mothers face notable physiological adaptations during this life-period. Neuroimaging studies of the last decade have provided grounded evidence that women’s brains structurally change across the transition into motherhood. The characterization of this brain remodeling is currently in its early years of research. The current article reviews this scientific field by focusing on our longitudinal (pre-to-post Edited by: pregnancy) Magnetic Resonance Imaging (MRI) studies in first-time parents and other Lauren M. Osborne, longitudinal and cross-sectional studies of parents. We present the questions that are Johns Hopkins University, currently being answered by the parental brain literature and point out those that have United States not yet been explored. We also highlight potential confounding variables that need to be Reviewed by: Anne T. M. Konkle, considered when analyzing and interpreting brain changes observed during motherhood. University of Ottawa, Canada Keywords: mother, neuroplasticity, MRI, maternal behavior, pregnancy Orli Dahan, Tel Hai College, Israel *Correspondence: Magdalena Martínez-García INTRODUCTION mmartinez@hggm.es Susanna Carmona Motherhood is a transformative experience for women. It confers substantial anatomical and scarmona@hggm.es physiological changes in the endocrine, cardiovascular, respiratory, renal, and musculoskeletal systems of the mother (1). Neuroimaging studies of the last decade have confirmed that the women’s Specialty section: brain also undergoes visible structural changes during the motherhood transition. However, the This article was submitted to nature of this brain remodeling is still in its early years of research. Women’s Mental Health, In 2017, our group led one of the most notable longitudinal projects on this behalf: a study a section of the journal that tracked the neuroanatomical MRI data of primiparous women across three sessions: a few Frontiers in Global Women’s Health months before their first pregnancy, during the early postpartum, and at 2 years after parturition, Received: 16 July 2021 using fathers and non-parents as comparison groups (2). This study found prominent pre-to- Accepted: 10 September 2021 post pregnancy volumetric Gray Matter (GM) reductions in theory-of-mind regions in first-time Published: 07 October 2021 mothers. Since then, we have conducted several pre-to-post pregnancy longitudinal MRI studies Citation: to characterize the initially observed GM reductions and understand the potential underlying Martínez-García M, Paternina-Die M, mechanisms. Here, we cover the brain structural findings that we have found so far and discuss Desco M, Vilarroya O and Carmona S variables that need to be considered to correctly interpret brain changes during the motherhood (2021) Characterizing the Brain Structural Adaptations Across the transition. The manuscript is divided into the main questions that our research group and other Motherhood Transition. groups have aimed to directly or indirectly respond to through structural MRI. First, we summarize Front. Glob. Womens Health the neuroanatomic changes that have been detected through the motherhood transition. Then, 2:742775. we discuss the following topics: 1) which are the mediating factors behind these changes? 2) doi: 10.3389/fgwh.2021.742775 which are the underlying neural mechanisms? 3) what do we know about the changes’ temporal Frontiers in Global Women’s Health | www.frontiersin.org 1 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers course? and, 4) are deviations of these adaptations related to NEUROANATOMIC CHANGES DURING the emergence of postpartum mental health disorders? We also THE TRANSITION TO MOTHERHOOD contemplate other factors not directly related to motherhood as potential confounders of the observed changes. This review Existing human neuroimaging studies have applied both article aims to provide cohesion to the structural parental brain longitudinal and cross-sectional designs to track the short-term literature and set future directions for those critical questions that effects of pregnancy on the brain. Figure 1A and Table 1 sum remain unresolved. up and outline the most relevant findings from structural studies For the sake of efficiency, during the manuscript, we will that track brain changes during the transition to motherhood. refer to mothers as persons who identify as women that undergo The study of the women’s brain changes associated with pregnancy through natural or assisted processes. Future studies pregnancy dates to the first half of the twentieth century. In need to address the unique brain processes of mothers that do 1931, enlargements of the pituitary gland, one of the primary not undergo pregnancy. sources of prolactin during pregnancy, were first observed in FIGURE 1 | Schematic representation of the maternal brain GM volume findings and suggested trajectories. (A) Findings of the longitudinal MRI structural studies tracking the maternal brain. Represented studies are (2–7). (B) Hypothetical trajectories in GM volume across pregnancy and postpartum periods. Frontiers in Global Women’s Health | www.frontiersin.org 2 October 2021 | Volume 2 | Article 742775 Frontiers in Global Women’s Health | www.frontiersin.org Martínez-García et al. TABLE 1 | Summary of the characteristics and findings of the longitudinal MRI structural studies tracking the maternal brain. Oatridge et al. (7) Kim et al. (6) Hoekzema et al. (2) Lisofsky et al. (5) Luders et at. (4) Zhang et al. (3) Period * Pre-pregnancy (n = 2) 3 weeks pp ➤3.5 months Pre-pregnancy ➤2 months 2 months pp ➤5 months pp 1.5 days pp ➤5 weeks pp 1 year pp ➤2 years pp ➤22.5 weeks gestation (n = pp pp 4) ➤before delivery (n = 9) ➤6 weeks pp (n = 9) ➤24 weeks pp (n = 7) ➤40 weeks pp (n = 3) ➤1 year pp (n = 3) Participants 9 mothers 19 mothers 25 mothers and 25 24 mothers and 24 14 mothers 21 mothers nulliparous controls nulliparous women Primiparous Not reported 58% 100% 100% 50% 100% Age at baseline Mean: 31 years. Range: 33.27 ± 6.07 years Mothers: 33.36 ± 3.97 Mothers: 28.38 ± 3.41 32.8 ± 4.0 years 30.03 ± 2.75 years (mean ± sd) 20–38 years. years. Nulliparous: 31.10 ± years. Nulliparous: 25.42 ± 5.63 years 2.95 years Processing method Contour and thresholding Voxel-based morphometry Voxel-based morphometry Voxel-based morphometry Voxel-based morphometry Volume in Computational semiautomatic technique toolbox (SPM2) toolbox (SPM12) toolbox (SPM8) toolbox (SPM8) Anatomy toolbox (SPM12) Contrast and - Post > Pre (p < 0.05 Mothers (Post-Pre) < Group by time interaction (p Post > Pre (peak level p < Post > Pre (peak level p < threshold FDR-corr, extent > 100 Nulliparous women < 0.05 FWE-corr, exten t 0.001, FWE-corr) 0.05, FDR-corr) voxels) (Post-Pre) (peak level p < >200 voxels) 0.05, FWE-corr) Direction Brain size decreases until GM volume increases GM volume decreases GM volume increases GM volume increases GM volume decreases delivery, then increases 3 MODIFIED REGIONS Frontal regions R superior frontal Superior medial frontal Superior frontal Medial frontal Medial frontal Middle frontal L Middle frontal Middle frontal Middle frontal Medial orbitofrontal L inferior orbitofrontal L inferior frontal Inferior frontal L inferior frontal Inferior frontal L Central and frontal operculum L Superior Frontal Ventro medial prefrontal Precentral R medial precentral Precentral October 2021 | Volume 2 | Article 742775 R Paracentral Brain Structural Adaptations in Mothers Temporal regions L superior temporal Superior temporal Middle temporal Inferior temporal Fusiform Parietal regions Precuneus Precuneus R precuneus R precuneus Superior parietal L Superior parietal Inferior parietal (Continued) Frontiers in Global Women’s Health | www.frontiersin.org Martínez-García et al. TABLE 1 | Continued Oatridge et al. (7) Kim et al. (6) Hoekzema et al. (2) Lisofsky et al. (5) Luders et at. (4) Zhang et al. (3) R middle occipital Postcentral R postcentral Postcentral Insular cortex Insula L Insula L insula Cingulate cortex R anterior cingulate Anterior/posterior cingulates Anterior cingulate Cingulate Subcortical regions L hippocampus R Parahippocampal Parahippocampal R Parahippocampal L Nucleus accumbens R Thalamus Thalamus Thalamus Cerebellum Cerebellum R Cerebellum Caudate L caudate Caudate Substantia nigra R amygdala R Putamen Mammilary body R globus pallidum Hypothalamus L brainstem (pons, medulla) 4 Additional structural Ventricular size increases Cortical thickness and area Decreased BRAIN age WM increases and cortical findings ** until delivery, then decreases (Freesurfer v5.3) index (8) thickness decreases. decreases GM decreases maintained R hippocampus Cross-sectional analysis up to 2 years pp (subiculum,CA2 and mothers vs nulliparous CA3) (9) and amygdala women: lower GM and increases (10) cortical thickness, higher WM and gyrification index GM decreases maintained up to 6 years pp (11) Flattening of the cortex (12) L nucleus accumbens decreases (13) October 2021 | Volume 2 | Article 742775 Brain Structural Adaptations in Mothers *Periods were reported as the mean time-points when the scans were acquired. **The additional structural findings come from other studies or analyses that have used the same participants sample. pp, postpartum; FDR, false discovery rate; FWE, family wise error; GM, gray matter; WM, white matter; L, left hemisphere; R, right hemisphere; p, p-value. Martínez-García et al. Brain Structural Adaptations in Mothers deceased pregnant women (14) and were later corroborated by in number, likely due to the inclusion of a nulliparous control in-vivo MRI of this structure (15–17). The gradual increase of group. Moreover, mothers had smaller GM volumes at 1–2 the pituitary gland during pregnancy is thought to reflect an months postpartum compared to non-mothers, and this volume estrogen-stimulated proliferation and hypertrophy of prolactin difference diminished at 3–4 months postpartum (5). This cells (18). Besides the pituitary gland volume assessments, the finding suggests that the postpartum GM increases are preceded first longitudinal study that tracked the whole brain found by initial reductions, in agreement with the results of Hoekzema increased ventricles and reduced outer border of the brain et al. (2). Notably, this study also found that the younger the peaking at parturition in a small sample of pregnant women (7). mothers were, the stronger the GM increases from 1–2 to 3–4 In 2017, our research group longitudinally compared the months postpartum, suggesting that samples with older mothers brains of first-time mothers from preconception to 2 months might display different GM postpartum trajectories. Finally, postpartum (pre-to-post hereafter) and found pronounced GM Zhang et al. (3) scanned primiparous mothers at 8 months volume reductions within cortical regions that highly overlapped and 2 years after birth and found GM reductions and white with the theory-of-mind or Default Mode (DM) network matter (WM) increases in frontal areas, superior parietal gyrus, (2). The modified structures included frontal and temporal precuneus, and insula, WM increases in temporal regions and areas, precuneus, left insula, anterior cingulate cortex, and left GM reductions in subcortical regions such as parahippocampal hippocampus and parahippocampal regions. Besides, the pre-to- gyrus, thalamus, caudate, and cerebellum, which were positively post pregnancy cortical changes predicted the quality of mother- associated with the mother’s empathic abilities. Similarly, the to-infant attachment and the absence of hostility toward their authors found lower GM and higher WM volumes when babies. Such reductions were maintained at 2 years after giving comparing the mothers’ initial postpartum scans with non- birth, except for a partial left hippocampal volume recovery mothers’ scans (3). Most of the changing regions coincided with (2). We then extended this research by reporting that the pre- the longitudinal analysis with mothers, while middle temporal, to-post reductions reflected an overall flattening of the cortex anterior cingulate, putamen, and globus pallidum were uniquely characterized by decreases in cortical thickness, surface area, modified. This suggests that such regions might change at an local gyrification index, sulcal depth, and sulcal length, as well as earlier postpartum state [in agreement with the results of Kim increases in sulcal width (12). In addition, a recent study using the et al. (6)] but are no longer modified during late postpartum. same prospective dataset of first-time mothers reported volume Of note, many of the modified regions found by these reductions within the left ventral striatum –a subcortical region longitudinal studies coincide (Table 1). Cortical regions that containing the nucleus accumbens that acts as the core of the consistently change in mothers include frontal areas, superior maternal reward system (13). temporal and parietal gyri, precuneus, postcentral, and precentral Other longitudinal studies have approached this question by gyri, anterior cingulate cortex and insula, most of which analyzing how the brain changes during the postpartum period are DM nodes, and common subcortical regions are the (3–6, 8–10). Luders et al. (8) applied a brain age algorithm hippocampus, amygdala, thalamus, caudate, nucleus accumbens, to a prospective dataset of mothers and found that mothers’ and cerebellum. brains were considerably younger at 4–6 weeks postpartum Cross-sectional designs have also found GM structural compared to at 1–2 days after childbirth and suggested that such variations in the maternal brain. Lisofsky et al. (19) found that, as “rejuvenation” reflected GM increases, rather than decreases. The compared to nulliparous women (N = 30), early postpartal first- same sample of mothers also presented GM increases within time mothers (N = 30) had lower volume in the left putamen, a the inferior frontal, precuneus, precentral, and postcentral gyri striatal region involved in navigation strategies. Also, the higher occipital and operculum regions, and subcortical regions such the prepartal estrogen levels and allocentric navigation strategy as the thalamus and caudate (4), the right CA1 and CA3 (flexible learning) vs. egocentric navigation strategy (stimulus- hippocampal regions (9), and the superficial and centromedian response learning) in mothers, the greater postpartal reductions regions of the amygdala (10). Kim et al. (6) scanned mothers the mothers displayed within the left putamen (19). Kim et al. at a later postpartum period (at 2–4 weeks and at 3–4 months (20) detected a positive association between postpartum months after birth) and found increasing GM trends in similar frontal, and prefrontal cortical thickness in a group of 39 first-time parietal and subcortical areas, but the GM increases further mothers with diverse socioeconomic and racial backgrounds, extended to the superior temporal gyrus, cingulate cortex, and these results were independent of the mother’s age, race, insula, hypothalamus, putamen, globus pallidum, cerebellum income levels or depressive symptoms. Moreover, higher cortical and brainstem. In addition, the higher the volume within the thickness within those prefrontal regions was associated with subcortical regions, the higher the maternal positive perception higher self-reported parental efficacy (20). of her baby (6). Lisofsky et al. (5) applied a similar postpartum Brain regions whose structure changes in mothers (Table 1) design to Kim et al. (6) and extended it by including a control are consistent with functional modifications reported during group of nulliparous women and by excluding multiparous postpartum. A group of event-related functional MR studies participants. Compared to non-mothers, mothers displayed GM scanned mothers at 48 h of delivery and again at 4–7 weeks volume increases in anterior cingulate cortex, middle frontal postpartum and found that the later postpartum session was and medial prefrontal regions, cerebellum, and left nucleus characterized by lower activity during response inhibition within accumbens, from 1–2 to 4–5 months postpartum (5). Changing inferior frontal, anterior cingulate and precentral regions (21), regions were similar to those reported by Kim et al. (6), but less higher emotion reactivity within the insula and frontal regions Frontiers in Global Women’s Health | www.frontiersin.org 5 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers (22), and decreased reactivity during emotional anticipation in Pregnancy Hormones the anterior cingulate cortex (23). Also, resting-state functional The transition to motherhood is characterized by unparalleled MR (rsfMRI) acquisitions of first-time mothers at 1 and 2 hormonal fluctuations that orchestrate the physiological changes year postpartum revealed changes in the local neural activity that mark pregnancy, parturition, and lactation (26, 27, 29, 30). strength and coherence of DM regions such as precuneus, Sex-steroid hormones such as progesterone and estradiol (E2 ) cingulate, insula, and frontal and temporal regions between are mainly produced by the placenta during pregnancy (31). these two time points (24). Finally, Dufford et al. (25) found Accordingly, they increase steadily across the three trimesters, fall an association between postpartum months and resting-state off rapidly at parturition after placental separation, and remain functional connectivity between the left amygdala and the low during postpartum (29, 30). Oxytocin and prolactin are anterior cingulate gyrus, left nucleus accumbens, caudate, right neurohormones synthesized by the hypothalamus and anterior putamen, left pallidum, and left cerebellum, and between the pituitary and are released both centrally and peripherally right amygdala and the caudate. These findings indicate that from the posterior and anterior pituitary, respectively (32, 33). structural and functional changes co-occur in overlapping DM Oxytocin levels remain inhibited during pregnancy and rise at and subcortical regions during the transition into motherhood. parturition, following a pulsatile secretion pattern to regulate Taken together, literature suggests that motherhood is uterine contractions and stimulate milk ejection (30). Prolactin associated with dynamic brain adaptations in cortical regions that levels increase during pregnancy, to prepare the breasts for overlap with the DM/ theory-of-mind network and subcortical lactation (29, 30). Prolactin also stimulates lactogenesis in the reward regions. These adaptations seem to differ in magnitude breast epithelial cells [(26), see section Lactation], but the and direction depending on the time frame studied. Specifically, high progesterone levels during pregnancy inhibit this process, longitudinal studies indicate a GM volume decrease (2, 7) and ensuring that lactation does not initiate beforehand [(34), a flattening of the cortex (12) in mothers scanned before and see section The prolactin system. Preparation for lactation]. after pregnancy, followed by volume increases during the early Progesterone drops after delivery releasing prolactin from postpartum (4–7, 9, 10), and volume decreases again during the inhibition and therefore enabling milk production [(34), see late postpartum (3). Finally, changes in GM volumes do not reach section The prolactin system. Preparation for lactation]. Once pre-pregnancy levels at 2 years (2) or 6 years (11) postpartum. established, lactation is sustained by the synchronized action If one were to unify these longitudinal results, several GM of oxytocin and prolactin, whose secretion is triggered by the structural trajectories emerge as plausible, which are outlined baby suckling (30). Prolactin maintains milk production while in Figure 1B. First, GM might decrease in volume during oxytocin stimulates the milk ejection (30). Aside from their pregnancy, increase right after birth until reaching baseline peripheral functions, these hormones are also synthesized and levels and fall again late postpartum (Figure 1B, Option 1). released in the brain, where they act as neuromodulators or Alternatively, the prenatal GM decrease might be more abrupt neurotransmitters of the maternal brain circuitry (35). during the peripartum period, and postnatal trajectories might In animals, maternal behavior is triggered by the hormonal not reach pre-pregnancy baseline levels (Figure 1B, Option 2). events of late pregnancy and parturition (35). Oxytocin Another possibility would be a combination of the prenatal and and estradiol (E2 ) prime the medial preoptic area (mPOA) postnatal trajectories reflected in Options 1 and 2 (Figure 1B, region so that maternal behavior emerges. The activation Option 3). of maternal behavior involves functional plasticity through synaptic strengthening and inhibition, and structural plasticity through modifications of the dendrites, soma size, glial cells, or WHICH ARE THE MEDIATING FACTORS? neurogenesis. These modifications affect critical regions of the maternal circuit (mPOA, hippocampus, olfactory bulb, and PFC) Researchers have not disentangled yet which factors trigger and translate into visible macroscopic structural changes with and coordinate the brain morphometric changes observed in neuroimaging techniques (36) as well as into behavioral changes. mothers. Theoretically, these changes could be mediated either In humans, the primary approach to investigate how by intrinsic pregnancy-related hormonal factors, by extrinsic hormones impact the brain is to analyze the women’s brain environmental factors that translate to internal signals, or a during natural steroid hormonal transitional periods. The combination of both. As discussed in the previous section, common thread of every woman’s hormonal transition, either the brain adaptations to motherhood are not linear but natural (menstrual cycle, puberty, pregnancy and menopause) or instead seem to differ during pregnancy and postpartum, two assisted (oral contraceptives and hormone replacement therapy) periods characterized by unique physiological events (gestation are fluctuations of the E2 steroid hormone. Although mainly vs. lactation), endocrine fluctuations (26, 27), and immune synthesized by the gonads, E2 can also be secreted within many responses (28). This difference is also observed from a brain regions, including the hippocampus, where it exerts rapid psychological point of view. Whereas an expectant mother often cell-to-cell paracrine effects and regulates memory and learning engages in imagination and simulation, the brain of a postpartum processes (37). E2 fluctuations have been associated with altered mother often engages in executive functions for planning the neuroplasticity in female rodents across their lifespan (38). In tasks necessary to take care of the newborn. It is therefore humans, transitional periods characterized by rapid sex-steroid reasonable to think that the factors that mediate the neural fluctuations such as puberty, the menstrual cycle, contraceptive changes in mothers might be distinct before and after the birth. use, menopause, and hormone therapy are characterized by Frontiers in Global Women’s Health | www.frontiersin.org 6 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers visible structural and functional modifications of the brain pregnancy and adolescence exert similar morphometric effects (39, 40). A recent dense-sampling rsfMRI study across one on the cortical mantle. entire menstrual cycle showed that peaks of estradiol confer increased coherence across the brain and increased within- network integration in the DM network (41). The same dense- Parenting Experience sampling protocol was repeated under contraceptive use (42). Although the above literature leads us to consider pregnancy- The authors found that high progesterone levels rendered induced hormones as the primary mediators of maternal brain altered GM volumes within medial temporal areas -including changes, there are also reasons to think that experience- the hippocampus- during the natural menstrual cycle, and dependent changes associated with approaching parenthood— progesterone suppression induced by contraceptives altered which include, but are not restricted to, the interaction these cycle-dependent effects. Cross-sectional studies have also with the baby—might also account for the pattern of neural found interactions between hormonal contraceptive use and GM changes observed in mothers. Brain plasticity can also be structure within the right putamen (43), left amygdala, and induced by environmental signals (51). In many rodent species, left parahippocampal regions (44). Notably, all these regions once the maternal circuit is primed by the endocrine events coincide with those where women display structural changes associated with pregnancy and parturition, maternal behavior when becoming mothers for the first time (Table 1), suggesting is independent of hormonal control and is further maintained that sex-steroid hormones might modulate DM and subcortical by environmental cues derived from the interaction with the networks activity and structure across a woman’s lifespan. offspring [(52) Chapter 3]. When the hormonal events of Beyond the effects of estradiol and progesterone, pregnancy gestation wane, olfactive and tactile sensory inputs coming from outcomes -including the emergence of maternal behavior- are interacting with and nursing the pups stimulate the maternal indeed the result of a complex interplay between sex-steroid and circuit and maintain maternal behavior. This effect is not limited non-sex-steroid hormones (such as oxytocin and prolactin). For to dams that go through pregnancy and parturition. In rats, instance, at the end of a rat’s pregnancy, estrogen activates mPOA the mere exposure to foster pups through a “sensitization” neurons provided that progesterone levels are already low (45). process converts a female’s approach behavior from “avoidant” to Once primed, this region becomes responsive to oxytocin (46) “maternal” (53), and co-housed virgin females with dams further and prolactin (47), which are essential for the display of maternal boosts the “sensitization” in the virgins to behave maternally (54). behavior and breastfeeding (26, 47). Also, the bursts of oxytocin Moreover, when retrieving pups, “sensitized” females activate the and prolactin during peripartum are subject to the release same hypothalamic hubs that trigger maternal behavior in dams of magnocellular neurons and anterior pituitary, respectively, (54) and display signals of neural plasticity (55). from a tonic progesterone and opioid-mediated inhibition that In the human literature, researchers have tried to disentangle operates during pregnancy (48). Similar to these pregnancy the reproductive-induced and experience-induced effects by outcomes, neuroplasticity events occurring during motherhood studying the associations between the degree of maternal brain transition are likely to be mediated by the interaction of different changes and variables that account for the amount of time the fluctuating hormones. mother spends with her infant. Several studies, including our Adolescence is the transitional period closest to pregnancy in investigations, have addressed this approach by including in terms of hormonal milieu, since both periods are characterized by their analyses the time between the birth and the neuroimaging steady increases in sex-steroid hormones. Moreover, widespread assessment (2, 20). In Hoekzema et al. (2), we did not find GM brain changes (12) and increased risk for mental health associations between the longitudinal structural changes in disorders (49, 50) have been described in both transitions. mothers and the baby age. In contrast, the cross-sectional design Given these similarities, we hypothesized that a similar profile of Kim et al. (20) found a positive correlation between the of brain morphometric changes was operating in pregnancy mothers’ cortical thickness and the baby age during the first 6 and adolescence. To test this hypothesis, we compared the months postpartum. Although it works as an indirect estimate, longitudinal MR changes that take place in first-time mothers the baby’s age does not accurately indicate how much a mother with those occurring in a sample of female adolescents (12). has interacted with her infant after birth. The amount of mother- By doing so, we showed that the morphometric changes infant interaction might be better captured with devices that track occurring during the transition to motherhood are comparable the real-time physical proximity between mothers and infants, in magnitude, shape and direction to those occurring during the which are currently under development (56). Complementing adolescence period. Specifically, we quantified the change over such quantitative measures with qualitative assessments of time in a detailed set of metrics that fully characterize the brain’s mother-infant interactions such as the “still face” and “strange anatomy using MRI prospective data of three groups: 25 female situation” behavioral tasks would further characterize the quality adolescents who had never been pregnant, a group of 25 first- of that maternal investment. time adult mothers, and a group of 20 adult females who had One other approach to disentangle the reproductive-induced never been pregnant. Both adolescence and pregnancy groups from the experience-induced influence over the parental brain is displayed total brain volume reductions together with decreases to analyze the brain adaptations in individuals that experience in cortical thickness, surface area, local gyrification index, sulcal the parental transition without the reproductive experience. depth, and sulcal length, as well as increases in sulcal width. These Following that research line, we conducted a longitudinal study findings suggest that the sex-steroid hormonal fluctuations of in first-time fathers and a group of childless men as a control Frontiers in Global Women’s Health | www.frontiersin.org 7 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers group (57). We found that fatherhood entails “preconception- motherhood primarily relies on non-human animal experiments to-postpartum” cortical volume and thickness reductions that and a few human naturalistic experiments. were positively associated with the father’s neural response to In non-human animals, there is extensive evidence that pictures of his baby. These reductions were less pronounced and maternal behavior activation involves neural plasticity within affected fewer parts of the brain compared with those observed several maternal brain regions such as the hippocampus, in first-time mothers (57). Neural changes in fathers have also prefrontal cortex (PFC), basolateral amygdala, nucleus been reported during the postpartum period (58) and were accumbens, and hypothalamus [(52) Chapter 5]. Neural different in direction and affected regions compared to those modifications include changes in the number and morphology reported in mothers in a similar study by the same group (6). of neurons and glial cells (astrocytes, oligodendrocytes, and One plausible interpretation of these findings is that the maternal microglia) and synaptic plasticity events that vary depending brain changes are mediated by the cumulative effects of the on the studied species. This warns us that, beyond the similar drastic gestational factors and the continued interaction with maternal brain circuits found across mammals, the neural the baby during postpartum. In contrast, in fathers, the baby’s mechanisms behind the maternal brain might not be entirely postpartum interaction might be the most relevant operator of translatable among species. In humans, inferring neural the observed neural changes. Another plausible hypothesis is that mechanisms from structural MR imaging is challenging. the parental brain becomes sensitive to postpartum experience Although the ex vivo human brain has been recently scanned factors only in the absence of gestation, similar to what has with an ultra-high resolution of 100 µm isotropic (67), the been described in allomaternal rodent models [(52) Chapter resolution of conventional in vivo neuroimaging sequences 7]. Finally, researchers are beginning to shift the focus to the is still 1 mm isotropic. That resolution can capture major expectant parental brain (59). Besides the gestational factors and neuroanatomic trends but is insufficient to capture cellular- the parenting experience, the in-pregnancy simulation of certain level processes. These findings imply two things: 1) same aspects of parenting is also a candidate factor to trigger the neuroanatomic trends could mask different neural processes, parental brain circuits’ remodeling. Indeed, neural responses to and 2) neural processes behind the visible changes are more infant’s stimuli are known to start changing before birth both likely to involve processes that dramatically change the number in mothers (60) and fathers (60, 61), and to predict parenting of brain cells (such as apoptosis or altered proliferation) or attunement in the postpartum (62). the WM myelination, rather than morphological changes in those cells (which are likely to also be happening but would Other Mediating Factors not translate in macroscopic changes). Recalling the human Beyond parent-infant interaction, there are additional maternal brain literature, two major trends have consistently experiences often associated with parenting that might as been found in mothers: pre-to-post pregnancy GM reductions well contribute to the observed maternal brain changes. 44.5% and GM increases during postpartum (Figure 1A). Below we and 55% of mothers have poor sleep quality during pregnancy discuss several plausible neural mechanisms through which such and postpartum, respectively (63). A recent study in men GM changes could arise. Reduced proliferation of microglial and women in their twenties revealed that one night of sleep cells, synaptic pruning, and a myelination process are among the deprivation was sufficient to induce GM density increases discussed mechanisms. These neural processes are not necessarily and GM cortical thickness decreases (64). We thus cannot mutually exclusive but instead can be operating simultaneously. exclude with certainty a partial contribution of perinatal sleep During pregnancy, sex-steroid hormones coordinate changes deprivation to the observed GM changes. Stress, either arising in the mother’s immunological system to adapt it to the specific from a challenging perinatal environment or from complications needs of every gestational stage. The three main immune stages during gestation or childbirth, can also influence how women are a first pro-inflammatory burst during implantation, followed adjust to motherhood (65). Indeed, functional MRI studies find by an anti-inflammatory state to tolerate the fetus and a final pro- associations between high parenting stress and mothers’ brain inflammatory peak at parturition (28). The pregnancy-related responses to their infants (66). Future studies should test whether neuroimmune adaptations extend to the central nervous system, stress disrupts the neuroadaptations to parenting. populated by immunocompetent cells known as microglia (68). Taken together, current evidence suggests that both When activated by pro-inflammatory cues, these cells act as pregnancy-induced and experience-induced factors can macrophages and remove neuronal debris. A study by Haim trigger structural changes in the mother’s brain. Future studies et al. (69) showed that, compared to virgin females, dam rats collecting more specific reproductive-induced (such as hormonal displayed reduced microglial proliferation in maternal brain levels) as well as experience-induced factors (such as maternal regions such as the basolateral amygdala, the medial prefrontal care, or stress and sleep quality) and comparing the maternal cortex, the nucleus accumbens, and the hippocampus prior to brain adaptations between gestational and non-gestational parturition and at early postpartum. Notably, the location of mothers will help better discern the impact of gestational and such changes is highly resemblant to where GM reductions have environmental factors on the maternal brain. been found in mothers (2, 13), suggesting that neuroimmune- mediated microglial reductions might also operate in the human WHAT ARE THE NEURAL MECHANISMS? maternal brain. Given their immunosuppressive activity during gestation (70–72), perinatal rising estrogens, progesterone, and Our current understanding of the potential neural mechanisms glucocorticoids levels are candidates to suppress the proliferation that evoke the macroscopic changes observed during of microglia. In such a case, the pregnancy-induced microglial Frontiers in Global Women’s Health | www.frontiersin.org 8 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers reduction could buffer the pro-inflammatory activity triggered inhibition that down-regulates the stress axis during pregnancy by the sex-steroid rapid fall after delivery. A reduced microglial promotes myelin repair by enhancing the proliferation of coverage would lead to concurrent increases in brain water oligodendrocytes. The authors also suggested that microglia cells diffusivity, which has been reported in rats scanned before mating contributed to such re-myelination process by phagocytizing and during pregnancy (73). Diffusion and spectroscopy MRI myelin debris. Pregnancy seems to also induce myelination studies are required to corroborate this microglial phenomenon in healthy, free of WM lesions dams. For instance, Gregg in human mothers. et al. (87) found increased oligodendrocytes proliferation during Besides their role as immune macrophages, microglia also pregnancy, followed by increased axon myelination during sculpt neural circuits during brain maturational periods. During postpartum in dams compared to virgin females. Similarly to adolescence, microglia mediate synaptic pruning by promoting Kalakh and Mouihate (86), these authors found a pregnancy- synaptogenesis and engulfing faulty synapses (74–76). Synaptic induced increased myelin repair capacity after using an acute pruning selectively eliminates redundant synapses during this demyelinating injury. However, Gregg et al. (87) found that transition, which is thought to enhance the brain circuitry prolactin rises -instead of progesterone rises—triggered the efficiency in adolescents (77, 78). It has been suggested that observed oligodendrocyte proliferation. Chan et al. (73) found the reductions in neuropil and glial elements that surround the increased whole-brain diffusivity during rat pregnancies, which pruned synapses, rather than the pruned synapses themselves, are is a sign of a more diffusion-friendly environment for the water more likely to contribute to the cortical volume and thickness through brain tissues. Based on Gregg et al. (87) and Kalakh and reduction seen in adolescence (79). As previously mentioned, Mouihate (86) findings, it is plausible that pregnancy enhances we compared the cortical and sulcal morphometric changes that myelin repair processes, in turn increasing WM integrity, occur during pregnancy with those occurring during female myelination and water diffusivity. However, other processes such adolescence and observed the same pattern of morphometric as increased extracellular fluid or cell shrinkage might also changes in both female samples, that is, a loss of surface area contribute to these enhanced water diffusion properties. All in and cortical thickness, and a sulcal widening, causing an overall all, these rodent studies suggest that pregnancy promotes myelin flattening of the cerebral cortex (12). Based on the similar profile repair and that both prolactin and progesterone might play an of brain changes and sex-steroid burst between adolescence and essential role in this process. pregnancy, we hypothesized that similar neural mechanisms Increased myelination can also lead to voxels at the WM might be behind both life periods. and GM interface being misclassified as GM, thus inducing an Rat studies have revealed another pregnancy-induced apparent decrease in cortical volume. It is also plausible that neuroplasticity event that involves reductions. Physiological increases in WM are concomitant to the observed cortical GM events that stimulate oxytocin pulsatile firing (i.e., parturition, changes. Together with reporting GM volume and thickness lactation, or dehydration) reduce the astrocytic coverage of changes, Zhang et al. (3) also detected WM increases within oxytocin secreting magnocellular neurons in the hypothalamic the insula, postcentral gyrus, inferior parietal, and superior supraoptic nucleus of rats (80, 81). This increased juxtaposition and middle temporal gyri during late postpartum. Some of the of nearby neurons facilitates the formation of new synapses, modified regions overlapped with the GM volume reductions stimulating the contractile (82) and anti-diuretic (83) properties found by the same study (insula and postcentral gyri), while of the oxytocin during birth and lactation. In humans, the small other impacted areas were unique (inferior parietal and superior size of the hypothalamus and the lack of clear contrast with its and middle temporal gyri), suggesting that some GM and surrounding tissue makes the detection of changes within this WM changes in mothers are simultaneous while some others region very challenging. Combining more precise automatic are independent. Conversely, in Hoekzema et al. (2) we did segmentation tools of the hypothalamus (84) with higher not detect significant pre-to-post pregnancy changes in WM resolution T1 and T2-weighted MR acquisitions is promising for volume compared to the non-pregnant group. Other than investigating pregnancy-induced hypothalamic changes. anatomical T1-weighted estimations of WM volume, diffusion A third mechanism that might be operating in the observed MRI provides more precise estimates of WM integrity, fiber GM reductions is WM myelination. Pregnant women with orientation, myelin density, and axon diameters (88), which demyelinating illnesses such as Multiple Sclerosis (MS) have can improve the comprehension of the WM modifications of significantly fewer relapses during the last gestation trimester the maternal brain. Despite being a promising tool, no study (85) -when sex-steroid and prolactin hormonal levels are has applied this technique to study the human maternal brain highest-, suggesting that hormonal factors at the end of during gestation or postpartum so far. The only diffusion-based pregnancy ameliorate the demyelination of this disease. Rodent neuroimaging study in the maternal brain literature is a recent studies are revealing the mechanism through which such large-scale cross-sectional study with middle-aged mothers (54– MS improvement might occur. After inflecting WM lesions 81 years) of the UK Biobank neuroimaging database (89). to pregnant, virgin, and postpartum female rats, Kalakh This study revealed that the more parity the less estimated and Mouihate (86) found less demyelination in pregnant brain age based on their WM characteristics, which suggests a rats compared to the other two groups, an effect that protective effect of parity on WM later in life. However, the diminished after blocking the dam’s GABA receptors or when cross-sectional nature of the samples does not allow us to discern allopregnanolone (a progesterone metabolite) was antagonized. between the effect of reproductive experience and that of other This study evidenced that the progesterone-mediated GABAergic individual variables. Frontiers in Global Women’s Health | www.frontiersin.org 9 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers A last possibility would be that the GM reductions in neurons (97), and higher dendritic spine concentration within mothers reflect neurodegenerative processes. E2 drop after birth the anterodorsal medial amygdala (98) and within CA1 and provokes a microglial pro-inflammatory state that can lead CA3 hippocampal regions (55). At a cellular scale, neurogenesis to neural injury. It is thus possible that a disbalance of the changes have been consistently found in the hippocampus and neuroimmune environment during the motherhood transition olfactory bulb of dams. However, there is no consensual pattern could lead to neural injury and predispose for mental health of changes in neurogenesis among animal studies -the results disorders. Indeed, inflammatory levels were ranked as the are time-dependent and vary among species- [(99), Important second-best, after depression history, predictor for postpartum considerations], making the interpretation and translation to depression levels (90). However, we believe it is unlikely that the human mother brain very challenging. This comes as no the changes in GM that occur during a healthy motherhood surprise given the differences among species in terms of degree transition reflect a neurodegenerative process. Pregnant women of maturity and mobility (i.e., precociality) of the newborns, often report memory deficits while pregnant (91). However, and mating and parenting strategies, all of which can result the few cognitive assessments during postpartum have mixed in unique maternal brain adaptations. The widespread GM findings: some authors report poorer cognitive performance in increases observed in human mothers during postpartum involve postpartum mothers compared to non-mothers (92–94), while multiple cortical and subcortical regions, and therefore are likely other studies have not found any cognitive decline in mothers to reflect more complex processes. Some authors believe that (95) nor cognitive differences between mothers and non-mothers these postpartum changes might reflect a partial recovery from (95). The reported cognitive impairments could be biased by the reductions during pregnancy (5, 8). Following the above- the physical (i.e., sleep levels) or psychological stressors that exposed microglial hypothesis, it might be the case that the come with the motherhood transition, rather than reflecting a observed GM increase reflects a microglial proliferation after the neural-mediated cognitive incapacity. In support of this, scholars postpartum hormonal environment returns to its “normal” non- have recently proposed the “cognitive costs of reproduction inflammatory state. This hypothesis needs to be further tested model,” which hypothesizes that perinatal cognitive declines with neuroimaging modalities that can estimate cellular types result from the mother’s temporary reallocation of metabolic such as H1 spectroscopy and diffusion MRI. and attentional resources to boost the fetus’ development and The human maternal brain literature has mainly relied on the mother-infant relationship (96). Neuroimaging studies that T1-weighted images, a modality with limited ability to capture have compared the mother’s brain with the brain of nulliparous tissue contrasts other than the GM and WM ones. Multimodal women have not found cognitive differences among these two MR imaging (that is, combining information from different MR groups (2, 5), and rather suggest that the motherhood-related contrasts or modalities) is a promising approach to overcome the brain changes are associated with behavioral outcomes that limitations of each MR modality and provide a comprehensive improve the mother’s ability to deal with the challenges ahead. non-invasive histological picture of the brain. Complementing Specifically, the degree of structural brain adaptations in mothers T1-weighted MR sessions with other MR contrasts or modalities predict higher levels of mother’s positive perception (6) and will inform on different tissue macrostructure aspects such attachment (2) to the baby, and stronger functional activation as hippocampal and hypothalamic subfields (through high- to their babies’ signals (13). Also, brain reductions in DM resolution T2-weighted sequence), and microstructure aspects regions observed before and after pregnancy (2) have been such as the types of cells (neuronal or glial) and processes argued to reflect an in-pregnancy down-regulation of higher- (apoptosis, neurogenesis, differentiation) involved (through 1 H order cognitive functions in favor of low-order brain functions multivoxel spectroscopy). that provide mothers greater resilience to stress and pain during In conclusion, motherhood-induced anatomical brain childbirth (59). Altogether, current results suggest that maternal changes are likely to involve simultaneous neural processes brain changes reflect a neural adjustment or specialization rather affecting both GM and WM tissues, and microglia seems to than a neurodegenerative process. Neuroimaging studies with have a crucial role regulating these processes. Also, distinct larger and more heterogeneous samples and cognitive measures neural mechanisms seem to be operating under pregnancy that directly compare the brain changes in mothers and the and postpartum periods, resulting in opposite neuroanatomic brain changes in individuals with neurodegenerative disorders directional trends. MR advances that increase the anatomical will further clarify these preliminary indications. resolution of the images and better determine the microstructural In contrast to the GM reductions observed from before to after tissue properties are required to elucidate the neural mechanisms pregnancy, the postpartum period is characterized by increases behind the macroscopic brain changes observed in mothers. in GM volume (4–6, 9, 10). Such postpartum GM increases affect frontal cortical areas and extend to subcortical regions such as the hypothalamus, amygdala, hippocampus, and nucleus WHICH IS THE TEMPORAL COURSE OF accumbens, which together form the subcortical maternal core THE CHANGES? circuit. The hypothalamus, amygdala and hippocampus, among other subcortical regions, have also been recently detected to Another critical research line in the parental brain focuses on increase in volume in mice mothers (36). At a neuronal scale, determining the temporal course of the motherhood-induced late gestation and parturient rats display increased cell body size brain changes. Existing longitudinal studies have not yet resolved (97) and higher dendritic complexity of hypothalamic mPOA whether this plasticity emerges during pregnancy, parturition, Frontiers in Global Women’s Health | www.frontiersin.org 10 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers or postpartum. Also, there are mixed results on whether the species, human maternal behavior extends beyond the “weaning” mother’s brain returns to baseline levels after the first 2 years period and is often everlasting. Therefore, more long-term neural postpartum or remains altered beyond this critical period of modifications are expected in human mothers compared to any maternal investment. Some studies are even considering that the other animal species. motherhood-induced changes might be permanent. Trying to fill this gap of knowledge, in a recent study we There is no solid evidence on when exactly human maternal followed the primiparous mothers of the study of Hoekzema brain changes start. Rodent studies give us some preliminary et al. (2) and scanned seven of them again at 6 years postpartum insights. Haim et al. (69) analyzed the brain’s immune cells (11). We compared the brains of the primiparous and nulliparous of dam rats at the beginning, middle, and end of gestation women before conception, during early postpartum and at 6 and found reduced microglial density from late pregnancy years postpartum, restricting the analysis to those regions that extending into the postpartum period. Besides, Barrière et al. underwent GM volume reductions between the pre- and post- (36) scanned longitudinally pregnant mice and found GM pregnancy sessions (2). We found that the brain of a mother increases in subcortical regions key for maternal behavior, some is still different from that of a nulliparous woman even at of which started during late gestation (mPOA, bed nucleus 6 years after delivery. Specifically, most of the GM volume of stria terminalis, and paraventricular nucleus), and some reductions were found to persist at 6 years postpartum, being of which started after birth (hippocampus and amygdala). able to classify women as having been pregnant or not with a Therefore, maternal neural plasticity in rodents is evident already 91.67% of total accuracy. These preliminary findings suggest that in pregnancy. In humans, the only study that has scanned the pregnancy-induced brain changes are long-lasting and open mothers before, during, and after pregnancy used a small sample the possibility that they are indeed permanent. There are two size (nine participants, of whose only two underwent an MR plausible explanations for these enduring effects of motherhood before conception), and did not include a control group of on the brain. On the one hand, these enduring brain changes nulliparous women (7). They found steady reductions in brain could be due to the organizational effects of hormones during size and increases in ventricular size starting around the second pregnancy and peripartum periods (meaning that the effects last trimester of pregnancy and peaking at parturition. To further beyond the period of hormonal exposure). On the other hand, validate these findings, researchers need to conduct similar other plausible mediators of the motherhood’s enduring GM studies with higher sample sizes and to include nulliparous effects might be the long-term alertness and sleep disruption that women to control for other potential variables and minimize the parents often experience. the noise induced by image acquisition, image processing and Other brain regions have been found to change during statistical analyses. Identifying the onset of the neural remodeling pregnancy but fully or partially reverse to baseline levels in mothers is necessary to determine when a mother’s brain is after the immediate postpartum. The gradual increase of the more vulnerable to environmental stressors (a plastic brain is a pituitary gland observed during pregnancy returns to normal size vulnerable brain) and would hint at the mediating factors behind coinciding with the end of breastfeeding (16, 17), a finding that the changes. was further replicated by Hoekzema et al. [(2); Supplementary The other question regarding the temporal course of the Figure 3]. Hoekzema et al. (2) also found that, while all regions changes is whether they are confined to the period of maximal that displayed volume changes remained reduced up to 2 maternal investment or instead persist beyond that period. years postpartum, the GM reductions within the hippocampus During the first 2 years of life, infants are highly dependent and partially recovered at the 2 year postpartum visit. require substantial parental investment. This period is equivalent Recent investigations have approached the long-term effects to the “pre-weaning” period of rodents. Several studies using of parenthood by analyzing how elderly subjects who became diverse timeframes have detected GM changes in mothers from parents decades ago differ neuroanatomically from those who did birth until 2 years postpartum, but most of the studies have not have children. In elder women, a higher number of previous not explored the maternal brain beyond that “weaning” period. children has been found to be associated with less apparent Literature in rodents indicates that reproductive experience brain aging in WM (89), cortical (89, 107) and subcortical confers both long-term behavioral and neural changes that are regions (108), larger global GM cortical volume (89) and evident beyond weaning, and short-term neural changes that thickness (109) and distinct patterns of resting state functional restore at late postpartum. On the one hand, post-weaned dam connectivity (110). In Ning et al. (111), middle-aged females rats are less anxious and fearful than virgin rats and have and males with a higher number of children displayed better better foraging strategies (100, 101), and spatial learning and visual memory, faster response time, and lower predicted brain memory skills (100–102). At the neural level, aged parous rats age, suggesting that lifestyles associated with parenting, rather present modifications in hippocampal plasticity (103, 104) and than the physiology of pregnancy and lactation themselves, might estrogen sensitivity (105) and less markers of brain aging (102). be beneficial for brain aging processes. Altogether, these studies On the other hand, microglia density changes detected in rats suggest that motherhood exerts neural changes that persist into during late gestation and early postpartum return to baseline older age. levels after weaning (69). In humans, similar processes are Given these findings, we believe it is necessary that human likely to happen: some neural changes might be lasting while neuroimaging databases and studies that track brain trajectories others not [reviewed in Duarte-Guterman et al. (106)]. With a across lifespan start including parity-related variables as part of notable difference: unlike the rest of the non-human altricial the participants’ relevant demographic information. Frontiers in Global Women’s Health | www.frontiersin.org 11 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers DO THESE STRUCTURAL CHANGES HAVE experienced mothers, stressing the importance of studying AN IMPLICATION IN POSTPARTUM primiparity and multiparity separately. Dudin et al. (124) were the first to report differences in amygdala responsivity between MENTAL HEALTH? mothers with PPD and women with major depression, using The physiological and endocrine signals around the peripartum as controls non-depressed mothers and non-depressed women. ensure labor and help establish lactation, while putting the This study was pioneer in characterizing the unique neural mother at risk for developing mental health disorders. Placental profiles of depression during the perinatal period compared to separation at birth causes rapid progesterone and estradiol depression arousals at other times in a woman’s life, which has withdrawal, whose levels remain low during the early postpartum been reviewed by Pawluski et al. (128). (112). This rapid drop triggers neurochemical deficiency in the Maternal brain researchers have also applied rsfMRI to serotonin and dopamine systems and affects the GABAergic capture the basal patterns of disrupted functional connectivity inhibitory system, a neurochemical pattern that resembles a in mothers with PPD. Given that this MRI modality does not depression-like scenario [discussed in Sacher et al. (113)]. require the cognitive engagement of the mothers while scanning, Postpartum mood changes resulting from such hormonal it offers unique insights into the neuroscience behind PPD. changes and increased stress and loss of sleep are expected in Studies investigating PPD through rsfMRI are diverse in analyses mothers after birth, but one in every five postpartum mothers but have consistently explored connectivity patterns from the experience a much more severe disorder: postpartum depression amygdala and cortical DM nodes. Besides, unlike task-based (PPD) (50). In the Diagnostic and Statistical Manual of Mental functional MRI studies, all resting-state studies have excluded Disorders (DSM 5th edition) (114), PPD is still vaguely defined medicated mothers to remove potential pharmacological as a “major depression event with peripartum onset.” However, interactions with the results. Five studies explored resting-state beyond the unique timing of PPD, this disorder involves a functional connectivity patterns in depressed and non-depressed different symptomatology, often characterized by avoidance, mothers at the whole-brain level (129) or using seed analyses intrusive behaviors or even psychosis toward the baby. If from the posterior cingulate cortex (130, 131), the anterior left unrecognized or untreated, the disorder can have serious cingulate cortex, amygdala, hippocampus, and dorsolateral consequences for both the mother and the infant’s well-being prefrontal cortex (dlPFC) (132), and the dorsomedial prefrontal (115). Unfortunately, in spite of the importance of early PPD cortex (dmPFC) (133). These studies found information flow identification and treatment, PPD is still often unrecognized direction changes (129), reduced interhemispheric connectivity in mothers. (131), and altered (increased, decreased or opposing) functional Neuroscience researchers are bringing efforts to identify connectivity (132, 133) among the seed regions (132), between the unique neurobiological and pharmacokinetic profiles of the seed regions and DM cortical nodes (133), thalamus, caudate PPD, and numerous authors have reviewed the neuroimaging (132) and amygdala (130), and among DM cortical nodes literature on the topic (113, 116–118). In humans, functional (129–131) and the amygdala (129). Two other studies analyzed MR Imaging is the current preferred choice of study of PPD. A changes in the regional or local resting-state homogeneity in PPD wide range of task-based functional studies have compared the and non-PPD mothers through Regional Homogeneity (ReHo) neural activations of mothers with PPD diagnosis (or depression or fractional Amplitude of Low Frequency Fluctuation (fALFF) predisposition) and non-depressed mothers upon infant stimuli. methods (134, 135), rendering mixed results of increased and Given its vast implication in depression symptomatology, most decreased DM regional homogeneity in PPD mothers compared studies chose the amygdala as the seed region of interest (119) to healthy mothers. The depression scores of mothers with PPD followed by the PFC (119, 120) anterior cingulate cortex (120) correlated with the regional (135), interhemispheric (131), and and ventral striatum (119). In contrast, two studies explored DM connectivity of the dmPFC (131, 133) or the dlPFC (135), activations at the whole-brain level (121, 122). Studies usually suggesting that functional connectivity in this region is especially compare the pattern of brain activation while exposing mothers relevant in the severity of PPD. to cues of their baby vs. of unrelated babies (121–125), although As observed, task-based and resting-state studies in some studies have also used as stimuli unrelated infants crying mothers with PPD find functional disruptions in regions (120), images of adult faces (119), and emotionally valenced that match with the frontal, temporal, and cingulate cortical words (126, 127). These studies also vary in the proportion regions and subcortical regions (amygdala, hippocampus, of primiparous and unmedicated mothers, diagnosis criteria, thalamus and striatum) that are structurally modified in and postpartum time-points assessed. Still, most of the analyses healthy mothers (Table 1), suggesting that PPD might disrupt find differential BOLD activation within the amygdala, anterior neuroanatomic trajectories during motherhood transition. cingulate cortex, prefrontal cortex, insula and ventral (nucleus Despite this significant overlap, no human study to date has accumbens), and dorsal (caudate and putamen) striatum, analyzed the brain structure of mothers with PPD. Instead, and disrupted connectivity between the amygdala and the human studies that track brain structural trajectories across insula (123), nucleus accumbens and anterior cingulate regions motherhood have focused on delineating the healthy pattern (120). Notably, these structures coincide with the changing of brain adaptations. The pre-to-post pregnancy sample of regions reported in Table 1. Wonch et al. (123) found that Hoekzema et al. (2) (N = 25) only included one mom that mothers’ amygdala response and functional connectivity between developed postpartum depression, and results were similar the amygdala and insula differed between primiparous and after excluding that woman from the analysis. Longitudinal Frontiers in Global Women’s Health | www.frontiersin.org 12 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers studies during the postpartum have excluded those mothers with the human maternal brain remodeling and its potential relation ongoing psychiatric conditions (3, 8) or have relied on mothers to postpartum mental health. As argued alongside the review, with minimal to moderate levels of depression [<13 Beck’s the answers to these questions are not entirely resolved, and Depression Inventory score in Kim et al. (6); <10 Edinburgh the function that the brain structural changes observed in Postnatal Depression Scale score in Lisofsky et al. (5)]. Moreover, mothers might serve is still under debate. Through longitudinal in those studies that did include depression questionnaires, neuroimaging designs with first-time parents, our research group mothers and non-mothers did not differ in depression levels, and has found preliminary evidence that maternal brain changes are thus the studies did not assess structural changes in relation to pronounced (2) and long-lasting (2, 11), that involve a flattening this variable (2, 5). The lack of PPD representation in structural of the cortex similar to the one occurring during adolescence MRI studies has a plausible reason why: the neuroadaptations (12), that predict mother-to-infant attachment (2), and that can across healthy motherhood are still unresolved, and to unravel be mediated by pregnancy-induced factors as well as by the them, studies prioritize minimizing the confounders in their parental experience with the infant (12, 57). Future investigations analyses. We believe that a grounded comprehension of the are needed to shed light on this nascent but exciting research area, typical trajectories of change in healthy mothers will frame especially studies with larger and more diverse samples in terms sound hypotheses on which patterns of disruptions are expected of culture and socioeconomic status, as well as those examining in mothers with postpartum depression. Nonetheless, maternal parents at multiple time points of their parental transition. brain neuroimaging studies must start including the depression and anxiety levels of the participants routinely and exploring AUTHOR CONTRIBUTIONS the neuroanatomic adaptations in relation to such levels. Postpartum mental health needs to be explored in relation MM-G, SC, and MP-D contributed to the conception and design to past or present complications during pregnancy. Also, of the review. MM-G wrote the first drafts of the manuscript. future studies should explore the implications of different SC designed the figure. All authors contributed to manuscript birth courses. Parturitions vary in terms of the degree of revision, read, and approved the submitted version. intervention, medicalization, and the mother’s psychological and physiological experience, and some of these events can impact FUNDING the mother’s mental health (136–139) and their neural response and bonding toward the infant (140, 141). Finally, uncovering This work was supported by Ministerio de Ciencia e Innovación the risk factors of this disease is paramount to developing project RTI2018-093952-B-100 and by Instituto de Salud Carlos prediction and early intervention strategies. Previous history III projects CP16/00096 and PI17/00064, and co-funded by of mental illness and stress, life-long vulnerability to hormonal European Regional Development Fund (ERDF), A way to changes (including premenstrual disorders and response to make Europe. MM-G was funded by Ministerio de Ciencia hormonal treatments or contraceptives), and perinatal poor e Innovación, Instituto de Salud Carlos III, Predoctorales sleep quality are considered risk factors for developing PPD de Formación en Investigación en Salud (PFIS) (contract (113). These variables should be tested as potential predictors for FI18/00255) and a predoctoral Fulbright grant. SC was funded by deviant brain morphometric trajectories during the transition a Miguel Servet Type I research contract (CP16/00096). MM-G to motherhood. and SC were co-funded by European Social Fund Investing in your future. The project leading to these results has received CONCLUSIONS funding from la Caixa Foundation under the project code LCF/PR/HR19/52160001, from the European Research Council Our understanding of the human maternal brain has grown (ERC) under the European Union’s Horizon 2020 research considerably during the last 20 years. Before the twentieth and innovation programme (grant agreement No 883069), and century, scholars established the neuroendocrinology of maternal from the Centro Nacional de Investigaciones Cardiovasculares behavior in rodents (52), and there were first indications that (CNIC). The CNIC is supported by Instituto de Salud Carlos pregnancy modified the pituitary gland of human mothers (15– III (ISCIII), Ministerio de Ciencia e Innovación and the Pro 18). Inspired by these initial works, new-age maternal brain CNIC Foundation, and is a Severo Ochoa Center of Excellence researchers have proved groundbreaking evidence that gestation (SEV-2015-050). impacts the mother’s brain at a structural and functional level. Besides, recent brain age prediction studies suggest that parity ACKNOWLEDGMENTS also affects the long-term trajectories of brain aging. Here, we have reviewed and discussed the current evidence about the The authors acknowledge all the participants that contributed to trajectories, mediating factors, and neural mechanisms behind our neuroimaging studies. REFERENCES 2. Hoekzema E, Barba-Müller E, Pozzobon C, Picado M, Lucco F, García-García D, et al. Pregnancy leads to long-lasting changes in 1. Thornburg KL, Bagby SP, Giraud GD. Maternal adaptations to pregnancy. human brain structure. Nat Neurosci. (2017) 20:287–96. doi: 10.1038/n In: Knobil and Neill’s Physiology of Reproduction. Elsevier (2015). p. 1927–55. n.4458 Frontiers in Global Women’s Health | www.frontiersin.org 13 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers 3. Zhang K, Wang M, Zhang J, Du X, Chen Z. Brain structural plasticity 21. Bannbers E, Gingnell M, Engman J, Morell A, Sylvén S, Skalkidou associated with maternal caregiving in mothers: a voxel- and surface- A, et al. Prefrontal activity during response inhibition decreases over based morphometry study. Neurodegenerative Dis. (2020) 361005:1–12. time in the postpartum period. Behav Brain Res. (2013) 241:132–8. doi: 10.1159/000506258 doi: 10.1016/j.bbr.2012.12.003 4. Luders E, Kurth F, Gingnell M, Engman J, Yong E-L, Poromaa IS, et al. From 22. Gingnell M, Bannbers E, Moes H, Engman J, Sylvén S, Skalkidou A, baby brain to mommy brain: widespread gray matter gain after giving birth. et al. Emotion reactivity is increased 4-6 weeks postpartum in healthy Cortex. (2020) 126:334–42. doi: 10.1016/j.cortex.2019.12.029 women: a longitudinal fMRI study. PLoS ONE. (2015) 10:e0128964. 5. Lisofsky N, Gallinat J, Lindenberger U, Kühn S. Postpartal neural plasticity doi: 10.1371/journal.pone.0128964 of the maternal brain: early renormalization of pregnancy-related decreases? 23. Gingnell M, Toffoletto S, Wikström J, Engman J, Bannbers E, Neurosignals. (2019) 27:12–24. doi: 10.33594/000000105 Comasco E, et al. Emotional anticipation after delivery–a longitudinal 6. Kim P, Leckman JF, Mayes LC, Feldman R, Wang X, Swain JE. The neuroimaging study of the postpartum period. Sci Rep. (2017) 7:114. plasticity of human maternal brain: longitudinal changes in brain anatomy doi: 10.1038/s41598-017-00146-3 during the early postpartum period. Behav Neurosci. (2010) 124:695–700. 24. Zhang K, Wang M, Zhang J, Xie J, Su X, Du X, et al. Dynamic doi: 10.1037/a0020884 alterations in spontaneous brain activity in mothers: a resting-state 7. Oatridge A, Holdcroft A, Saeed N, Hajnal JV, Puri BK, Fusi L, et al. Change in functional magnetic resonance imaging study. Neurosci Bull. (2019) 35:766– brain size during and after pregnancy: study in healthy women and women 70. doi: 10.1007/s12264-019-00392-7 with preeclampsia. AJNR Am J Neuroradiol. (2002) 23:19–26. 25. Dufford AJ, Erhart A, Kim P. Maternal brain resting-state connectivity 8. Luders E, Gingnell M, Poromaa IS, Engman J, Kurth F, Gaser C. Potential in the postpartum period. J Neuroendocrinol. (2019) 31:e12737. brain age reversal after pregnancy: younger brains at 4–6 weeks postpartum. doi: 10.1111/jne.12737 Neuroscience. (2018) 386:309–14. doi: 10.1016/j.neuroscience.2018.07.006 26. Anderson SM, MacLean PS, McManaman JL, Neville MC. Chapter 46- 9. Luders E, Gaser C, Gingnell M, Engman J, Sundström Poromaa I, lactation and its hormonal control. In Plant TM, Zeleznik AJ, editors. Knobil Kurth F. Gray matter increases within subregions of the hippocampal and Neill’s Physiology of Reproduction, 4th Edn. San Diego, CA: Academic complex after pregnancy. Brain Imaging Behav. (2021) 1–5. Press (2015). p. 2055–105. doi: 10.1016/B978-0-12-397175-3.00046-6 doi: 10.1007/s11682-021-00463-2 27. Kuo K, Hackney D, Mesiano S. The endocrine control of human pregnancy. 10. Luders E, Gaser C, Gingnell M, Engman J, Sundström Poromaa I, Kurth In: Bonora E, DeFronzo R editos. Diabetes. Epidemiology, Genetics, F. Significant increases of the amygdala between immediate and late Pathogenesis, Diagnosis, Prevention, and Treatment Endocrinology. Cham: postpartum: Pronounced effects within the superficial subregion. J Neurosci Springer International Publishing (2016). p. 1–33. Res. (2021) 99:2261–270. doi: 10.1002/jnr.24855 28. Mor G, Aldo P, Alvero AB. The unique immunological and microbial aspects 11. Martínez-García M, Paternina-Die M, Barba-Müller E, Martín de Blas D, of pregnancy. Nat Rev Immunol. (2017) 17:469–82. doi: 10.1038/nri.2017.64 Beumala L, Cortizo R, et al. Do pregnancy-induced brain changes reverse? 29. Albrecht ED, Pepe GJ. Placental endocrine function and hormone action. In: The brain of a mother six years after parturition. Brain Sci. (2021) 11:168. Knobil and Neill’s Physiology of Reproduction. Elsevier (2015). p. 1783–834. doi: 10.3390/brainsci11020168 30. Grattan DR, Ladyman SR. Neurophysiological and cognitive changes in 12. Carmona S, Martínez-García M, Paternina-Die M, Barba-Müller E, pregnancy. In: Handbook of Clinical Neurology. Elsevier (2020). p. 25–55. Wierenga LM, Alemán-Gómez Y, et al. Pregnancy and adolescence 31. Chatuphonprasert W, Jarukamjorn K, Ellinger I. Physiology entail similar neuroanatomical adaptations: a comparative analysis of and pathophysiology of steroid biosynthesis, transport and cerebral morphometric changes. Hum Brain Mapp. (2019) 40:2143–52. metabolism in the human placenta. Front Pharmacol. (2018) 9:1027. doi: 10.1002/hbm.24513 doi: 10.3389/fphar.2018.01027 13. Hoekzema E, Tamnes CK, Berns P, Barba-Müller E, Pozzobon C, 32. Grattan DR, Le Tissier P. Hypothalamic control of prolactin secretion, Picado M, et al. Becoming a mother entails anatomical changes in the and the multiple reproductive functions of prolactin. In: Knobil ventral striatum of the human brain that facilitate its responsiveness and Neill’s Physiology of Reproduction. Elsevier (2015). p. 469–526. to offspring cues. Psychoneuroendocrinology. (2020) 112:104507. doi: 10.1016/B978-0-12-397175-3.00012-0 doi: 10.1016/j.psyneuen.2019.104507 33. Furman BL. Posterior pituitary agents? In: Reference Module in Biomedical 14. Fischer F. V. Schwangerschaftsveränderung der hypophyse als ursache Sciences. Elsevier (2017). bitemporaler hemianopsie. Ophthalmologica. (1931) 75:343–54. 34. Brunton P, Russell J. Maternal brain adaptations in pregnancy. In: Knobil and doi: 10.1159/000297234 Neill’s Physiology of Reproduction: Two-Volume Set. (2015). 15. Gonzalez JG, Elizondo G, Saldivar D, Nanez H, Todd LE, Villarreal 35. Numan M, Woodside B. Maternity: neural mechanisms, motivational JZ. Pituitary gland growth during normal pregnancy: an in vivo study processes, and physiological adaptations. Behav Neurosci. (2010) 124:715–41. using magnetic resonance imaging. Am J Med. (1988) 85:217–20. doi: 10.1037/a0021548 doi: 10.1016/S0002-9343(88)80346-2 36. Barrière DA, Ella A, Szeremeta F, Adriaensen H, Même W, Chaillou 16. Elster AD, Sanders TG, Vines FS, Chen MY. Size and shape of the pituitary E, et al. Brain orchestration of pregnancy and maternal behavior in gland during pregnancy and post partum: measurement with MR imaging. mice: a longitudinal morphometric study. Neuroimage. (2021) 230:117776. Radiology. (1991) 181:531–5. doi: 10.1148/radiology.181.2.1924800 doi: 10.1016/j.neuroimage.2021.117776 17. Dinç H, Esen F, Demirci A, Sari A, Resit Gümele H. Pituitary dimensions and 37. Taxier LR, Gross KS, Frick KM. Oestradiol as a neuromodulator volume measurements in pregnancy and post partum. MR Assessment Acta of learning and memory. Nat Rev Neurosci. (2020) 21:535–50. Radiol. (1998) 39:64–9. doi: 10.1080/02841859809172152 doi: 10.1038/s41583-020-0362-7 18. Stefaneanu L, Kovacs K, Lloyd RV, Scheithauer BW, Young WF, Sano T, 38. Pawluski JL, Brummelte S, Barha CK, Crozier TM, Galea LAM. Effects et al. Pituitary lactotrophs and somatotrophs in pregnancy: a correlative in of steroid hormones on neurogenesis in the hippocampus of the adult situ hybridization and immunocytochemical study. Virchows Archiv B Cell female rodent during the estrous cycle, pregnancy, lactation and aging. Front Pathol. (1992) 62:291–6. doi: 10.1007/BF02899695 Neuroendocrinol. (2009) 30:343–57. doi: 10.1016/j.yfrne.2009.03.007 19. Lisofsky N, Wiener J, de Condappa O, Gallinat J, Lindenberger U, Kühn 39. Beltz AM, Moser JS. Ovarian hormones: a long overlooked but critical S. Differences in navigation performance and postpartal striatal volume contributor to cognitive brain structures and function. Ann N Y Acad Sci. associated with pregnancy in humans. Neurobiol Learn Mem. (2016) (2020) 1464:156–80. doi: 10.1111/nyas.14255 134:400–7. doi: 10.1016/j.nlm.2016.08.022 40. Rehbein E, Hornung J, Sundström Poromaa I, Derntl B. Shaping of the 20. Kim P, Dufford AJ, Tribble RC. Cortical thickness variation of the female human brain by sex hormones: a review. Neuroendocrinology. (2020) maternal brain in the first 6 months postpartum: associations with 111:183–206. doi: 10.1159/000507083 parental self-efficacy. Brain Structure and Function. (2018) 223:3267–77. 41. Pritschet L, Santander T, Taylor CM, Layher E, Yu S, Miller doi: 10.1007/s00429-018-1688-z MB, et al. Functional reorganization of brain networks across Frontiers in Global Women’s Health | www.frontiersin.org 14 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers the human menstrual cycle. Neuroimage. (2020) 220:117091. 63. Yang Y, Li W, Ma T-J, Zhang L, Hall BJ, Ungvari GS, et al. Prevalence doi: 10.1016/j.neuroimage.2020.117091 of poor sleep quality in perinatal and postnatal women: a comprehensive 42. Taylor CM, Pritschet L, Olsen RK, Layher E, Santander T, Grafton meta-analysis of observational studies. Front Psychiatry. (2020) 11:161. ST, et al. Progesterone shapes medial temporal lobe volume across doi: 10.3389/fpsyt.2020.00161 the human menstrual cycle. Neuroimage. (2020) 220:117125. 64. Sun J, Zhao R, Yang X, Deng H, Zhu Y, Chen Y, et al. Alteration of brain doi: 10.1016/j.neuroimage.2020.117125 gray matter density after 24 h of sleep deprivation in healthy adults. Front 43. Sharma R, Smith SA, Boukina N, Dordari A, Mistry A, Taylor BC, et al. Use of Neurosci. (2020) 14:754. doi: 10.3389/fnins.2020.00754 the birth control pill affects stress reactivity and brain structure and function. 65. Kim P. How stress can influence brain adaptations to motherhood. Front Horm Behav. (2020) 124:104783. doi: 10.1016/j.yhbeh.2020.104783 Neuroendocrinol. (2021) 60:100875. doi: 10.1016/j.yfrne.2020.100875 44. Lisofsky N, Riediger M, Gallinat J, Lindenberger U, Kühn S. 66. Noriuchi M, Kikuchi Y, Mori K, Kamio Y. The orbitofrontal cortex Hormonal contraceptive use is associated with neural and affective modulates parenting stress in the maternal brain. Sci Rep. (2019) 9:1658. changes in healthy young women. Neuroimage. (2016) 134:597–606. doi: 10.1038/s41598-018-38402-9 doi: 10.1016/j.neuroimage.2016.04.042 67. Edlow BL, Mareyam A, Horn A, Polimeni JR, Witzel T, Tisdall MD, et al. 7 45. Sheehan T, Numan M. Estrogen, progesterone, and pregnancy termination Tesla MRI of the ex vivo human brain at 100 micron resolution. Sci Data. alter neural activity in brain regions that control maternal behavior in rats. (2019) 6:244. doi: 10.1038/s41597-019-0254-8 Neuroendocrinology. (2002) 75:12–23. doi: 10.1159/000048217 68. Barth C, de Lange A-MG. Towards an understanding of women’s brain aging: 46. Meddle SL, Bishop VR, Gkoumassi E, van Leeuwen FW, Douglas the immunology of pregnancy and menopause. Front Neuroendocrinol. AJ. Dynamic changes in oxytocin receptor expression and activation (2020) 58:100850. doi: 10.1016/j.yfrne.2020.100850 at parturition in the rat brain. Endocrinology. (2007) 148:5095–104. 69. Haim A, Julian D, Albin-Brooks C, Brothers HM, Lenz KM, Leuner B. A doi: 10.1210/en.2007-0615 survey of neuroimmune changes in pregnant and postpartum female rats. 47. Bridges RS. Neuroendocrine regulation of maternal behavior. Front Brain Behav Immun. (2017) 59:67–78. doi: 10.1016/j.bbi.2016.09.026 Neuroendocrinol. (2015) 36:178–96. doi: 10.1016/j.yfrne.2014.11.007 70. Robinson DP, Klein SL. Pregnancy and pregnancy-associated hormones alter 48. Brunton PJ, Russell JA. The expectant brain: adapting for motherhood. immune responses and disease pathogenesis. Horm Behav. (2012) 62:263–71. Nature Reviews Neuroscience. (2008) 9:11–25. doi: 10.1038/nrn2280 doi: 10.1016/j.yhbeh.2012.02.023 49. Paus T, Keshavan M, Giedd JN. Why do many psychiatric disorders 71. Muzzio D, Zygmunt M, Jensen F. The role of pregnancy-associated emerge during adolescence? Nat Rev Neurosci. (2008) 9:947–57. hormones in the development and function of regulatory B cells. Front doi: 10.1038/nrn2513 Endocrinol. (2014) 5:39. doi: 10.3389/fendo.2014.00039 50. Howard LM, Khalifeh H. Perinatal mental health: a review of progress and 72. Deems NP, Leuner B. Pregnancy, postpartum and parity: Resilience and challenges. World Psychiatry. (2020) 19:313–27. doi: 10.1002/wps.20769 vulnerability in brain health and disease. Front Neuroendocrinol. (2020) 51. Mandolesi L, Gelfo F, Serra L, Montuori S, Polverino A, Curcio G, et al. 57:100820. doi: 10.1016/j.yfrne.2020.100820 Environmental factors promoting neural plasticity: insights from animal and 73. Chan RW, Ho LC, Zhou IY, Gao PP, Chan KC, Wu EX. Structural human studies. Neural Plast. (2017) 2017:1–10. doi: 10.1155/2017/7219461 and functional brain remodeling during pregnancy with diffusion tensor 52. Numan M. The Parental Brain: Mechanisms, Development, and Evolution. MRI and resting-state functional MRI. PLoS ONE. (2015) 10:e0144328. Oxford: Oxford University Press (2020). doi: 10.1371/journal.pone.0144328 53. Rosenblatt JS. Nonhormonal basis of maternal behavior in the rat. Science. 74. Schafer DP, Lehrman EK, Kautzman AG, Koyama R, Mardinly AR, (1967) 156:1512–3. Yamasaki R, et al. Microglia sculpt postnatal neural circuits in an 54. Carcea I, Caraballo NL, Marlin BJ, Ooyama R, Riceberg JS, Mendoza activity and complement-dependent manner. Neuron. (2012) 74:691–705. Navarro JM, et al. Oxytocin neurons enable social transmission of maternal doi: 10.1016/j.neuron.2012.03.026 behaviour. Nature. (2021) 596:553–7. doi: 10.1038/s41586-021-03814-7 75. Thion MS, Garel S. Microglia under the spotlight: activity and complement- 55. Pawluski JL, Galea LAM. Hippocampal morphology is differentially affected dependent engulfment of synapses. Trends Neurosci. (2018) 41:332–4. by reproductive experience in the mother. J Neurobiol. (2006) 66:71–81. doi: 10.1016/j.tins.2018.03.017 doi: 10.1002/neu.20194 76. Ikegami A, Haruwaka K, Wake H. Microglia: lifelong modulator of 56. Salo VC, Pannuto P, Hedgecock W, Biri A, Russo DA, Piersiak neural circuits: microglial machinery wires synapses. Neuropathology. (2019) HA, et al. Measuring naturalistic proximity as a window into 39:173–80. doi: 10.1111/neup.12560 caregiver-child interaction patterns. Behav Res Methods. (2021) 1–15. 77. Spear LP. Adolescent neurodevelopment. J Adolesc Health. (2013) 52:S7–13. doi: 10.3758/s13428-021-01681-8 doi: 10.1016/j.jadohealth.2012.05.006 57. Paternina-Die M, Martínez-García M, Pretus C, Hoekzema E, Barba-Müller 78. Riccomagno MM, Kolodkin AL. Sculpting neural circuits by axon E, Martín de Blas D, et al. The paternal transition entails neuroanatomic and dendrite pruning. Annu Rev Cell Dev Biol. (2015) 31:779–805. adaptations that are associated with the father’s brain response to his infant doi: 10.1146/annurev-cellbio-100913-013038 cues. Cereb Cortex Communi. (2020) 1:tgaa082. doi: 10.1093/texcom/tgaa082 79. Mills KL, Tamnes CK. Methods and considerations for longitudinal 58. Kim P, Rigo P, Mayes LC, Feldman R, Leckman JF, Swain JE. Neural structural brain imaging analysis across development. Dev Cogn Neurosci. plasticity in fathers of human infants. Soc Neurosci. (2014) 9:522–35. (2014) 9:172–90. doi: 10.1016/j.dcn.2014.04.004 doi: 10.1080/17470919.2014.933713 80. Theodosis DT, Poulain DA. Evidence for structural plasticity 59. Dahan O. The birthing brain: a lacuna in neuroscience. Brain Cogn. (2021) in the supraoptic nucleus of the rat hypothalamus in relation 150:105722. doi: 10.1016/j.bandc.2021.105722 to gestation and lactation. Neuroscience. (1984) 11:183–93. 60. Rutherford HJV, Bunderson M, Bartz C, Haitsuka H, Meins E, Groh doi: 10.1016/0306-4522(84)90222-7 AM, et al. Imagining the baby: neural reactivity to infant distress and 81. Hatton GI. Dynamic neuronal–glial interactions: an overview 20 years later. mind-mindedness in expectant parents. Biol Psychol. (2021) 161:108057. Peptides. (2004) 25:403–11. doi: 10.1016/j.peptides.2003.12.001 doi: 10.1016/j.biopsycho.2021.108057 82. Augustine RA, Seymour AJ, Campbell RE, Grattan DR, Brown CH. 61. Diaz-Rojas F, Matsunaga M, Tanaka Y, Kikusui T, Mogi K, Integrative neurohumoural regulation of oxytocin neurone activity Nagasawa M, et al. Development of the paternal brain in expectant in pregnancy and lactation. J Neuroendocrinol. (2018) 30:e12569. fathers during early pregnancy. Neuroimage. (2021) 225:117527. doi: 10.1111/jne.12569 doi: 10.1016/j.neuroimage.2020.117527 83. Sasaki S. Is oxytocin a player in antidiuresis? J Am Soc Nephrol. (2008) 62. Cardenas SI, Stoycos SA, Sellery P, Marshall N, Khoddam H, Kaplan J, et al. 19:189–90. doi: 10.1681/ASN.2007121319 Theory of mind processing in expectant fathers: associations with prenatal 84. Billot B, Bocchetta M, Todd E, Dalca AV, Rohrer JD, Iglesias JE. Automated oxytocin and parental attunement. Dev Psychobiol. (2021) 63: 1549–67. segmentation of the hypothalamus and associated subunits in brain MRI. doi: 10.1002/dev.22115 Neuroimage. (2020) 223:117287. doi: 10.1016/j.neuroimage.2020.117287 Frontiers in Global Women’s Health | www.frontiersin.org 15 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers 85. Alhomoud MA, Khan AS, Alhomoud I. The Potential Preventive Effect 105. Barha CK, Galea LAM. Motherhood alters the cellular response to estrogens of Pregnancy and Breastfeeding on Multiple Sclerosis. Eur Neurol. (2021) in the hippocampus later in life. Neurobiol Aging. (2011) 32:2091–5. 84:71–84. doi: 10.1159/000514432 doi: 10.1016/j.neurobiolaging.2009.12.004 86. Kalakh S, Mouihate A. Enhanced remyelination during late pregnancy: 106. Duarte-Guterman P, Leuner B, Galea LAM. The long and short term effects involvement of the GABAergic system. Sci Rep. (2019) 9:7728. of motherhood on the brain. Front Neuroendocrinol. (2019) 53:100740. doi: 10.1038/s41598-019-44050-4 doi: 10.1016/j.yfrne.2019.02.004 87. Gregg C, Shikar V, Larsen P, Mak G, Chojnacki A, Yong VW, et al. 107. de Lange A-MG, Kaufmann T, van der Meer D, Maglanoc LA, Alnæs White matter plasticity and enhanced remyelination in the maternal CNS. D, Moberget T, et al. Population-based neuroimaging reveals traces of J Neurosci. (2007) 27:1812–23. doi: 10.1523/JNEUROSCI.4441-06.2007 childbirth in the maternal brain. Proc Natl Acad Sci USA. (2019) 116:22341– 88. Alexander DC, Dyrby TB, Nilsson M, Zhang H. Imaging brain 6. doi: 10.1073/pnas.1910666116 microstructure with diffusion MRI: practicality and applications. NMR 108. Lange AG, Barth C, Kaufmann T, Anatürk M, Suri S, Ebmeier KP, Biomed. (2019) 32:e3841. doi: 10.1002/nbm.3841 et al. The maternal brain: region-specific patterns of brain aging are 89. Voldsbekk I, Barth C, Maximov II, Kaufmann T, Beck D, Richard G, traceable decades after childbirth. Hum Brain Mapp. (2020) 41:4718–29. et al. A history of previous childbirths is linked to women’s white matter doi: 10.1002/hbm.25152 brain age in midlife and older age. Hum Brain Mapp. (2021) 42:4372–86. 109. Orchard ER, Ward PGD, Sforazzini F, Storey E, Egan GF, Jamadar SD. doi: 10.1002/hbm.25553 Relationship between parenthood and cortical thickness in late adulthood. 90. Bränn E, Papadopoulos F, Fransson E, White R, Edvinsson Å, Hellgren C, PLoS ONE. (2020) 15:e0236031. doi: 10.1371/journal.pone.0236031 et al. Inflammatory markers in late pregnancy in association with postpartum 110. OrchardER, Ward PGD, Chopra S, Storey E, Egan GF, Jamadar SD. depression—A nested case-control study. Psychoneuroendocrinology. (2017) Neuroprotective effects of motherhood on brain function in late 79:146–59. doi: 10.1016/j.psyneuen.2017.02.029 life: a resting-state fMRI study. Cereb Cortex. (2020) 31:1270–83 91. Davies SJ, Lum JA, Skouteris H, Byrne LK, Hayden MJ. Cognitive doi: 10.1101/2020.06.09.143511 impairment during pregnancy: a meta-analysis. Med J Aust. (2018) 208:35– 111. Ning K, Zhao L, Franklin M, Matloff W, Batta I, Arzouni N, et al. Parity is 40. doi: 10.5694/mja17.00131 associated with cognitive function and brain age in both females and males. 92. Glynn LM. Giving birth to a new brain: Hormone exposures of pregnancy Sci Rep. (2020) 10:6100. doi: 10.1038/s41598-020-63014-7 influence human memory. Psychoneuroendocrinology. (2010) 35:1148–55. 112. Hendrick V, Altshuler LL, Suri R. Hormonal changes in the postpartum and doi: 10.1016/j.psyneuen.2010.01.015 implications for postpartum depression. Psychosomatics. (1998) 39:93–101. 93. Henry JF, Sherwin BB. Hormones and cognitive functioning during late doi: 10.1016/S0033-3182(98)71355-6 pregnancy and postpartum: a longitudinal study. Behav Neurosci. (2012) 113. Sacher J, Chechko N, Dannlowski U, Walter M, Derntl B. The peripartum 126:73–85. doi: 10.1037/a0025540 human brain: Current understanding and future perspectives. Front 94. Shin N-Y, Bak Y, Nah Y, Han S, Kim DJ, Kim SJ, et al. Disturbed retrieval Neuroendocrinol. (2020) 59:100859. doi: 10.1016/j.yfrne.2020.100859 network and prospective memory decline in postpartum women. Sci Rep. 114. American Psychiatric Association. dsm-v Manual 5th Edn. Washington, DC: (2018) 8:5476. doi: 10.1038/s41598-018-23875-5 American Psychiatric Association (2013). 95. Anderson MV, Rutherford MD. Cognitive reorganization during pregnancy 115. Slomian J, Honvo G, Emonts P, Reginster J-Y, Bruyère O. Consequences and the postpartum period: an evolutionary perspective. Evol Psychol. (2012) of maternal postpartum depression: a systematic review of maternal and 10:147470491201000420. doi: 10.1177/147470491201000402 infant outcomes. Womens Health (Lond). (2019) 15:174550651984404. 96. Ziomkiewicz A, Wichary S, Jasienska G. Cognitive costs of reproduction: life- doi: 10.1177/1745506519844044 history trade-offs explain cognitive decline during pregnancy in women. Biol 116. Barba-Müller E, Craddock S, Carmona S, Hoekzema E. Brain plasticity Rev. (2019) 94:1105–15. doi: 10.1111/brv.12494 in pregnancy and the postpartum period: links to maternal caregiving 97. Keyser-Marcus L, Stafisso-Sandoz G, Gerecke K, Jasnow A, Nightingale L, and mental health. Arch Womens Ment Health. (2019) 22:289–99. Lambert KG, et al. Alterations of medial preoptic area neurons following doi: 10.1007/s00737-018-0889-z pregnancy and pregnancy-like steroidal treatment in the rat. Brain Res Bull. 117. Bjertrup AJ, Friis NK, Miskowiak KW. The maternal brain: neural (2001) 55:737–45. doi: 10.1016/S0361-9230(01)00554-8 responses to infants in mothers with and without mood disorder. 98. Rasia-Filho AA, Fabian C, Rigoti KM, Achaval M. Influence of sex, Neurosci Biobehav Rev. (2019) 107:196–207. doi: 10.1016/j.neubiorev.2019. estrous cycle and motherhood on dendritic spine density in the rat medial 09.011 amygdala revealed by the Golgi method. Neuroscience. (2004) 126:839–47. 118. Pawluski JL, Swain JE, Lonstein JS. Chapter 5-Neurobiology of peripartum doi: 10.1016/j.neuroscience.2004.04.009 mental illness. In: Swaab DF, Buijs RM, Kreier F, Lucassen PJ, Salehi 99. Leuner B, Sabihi S. The birth of new neurons in the maternal brain: A, editors. Handbook of Clinical Neurology the Human Hypothalamus: hormonal regulation and functional implications. Front Neuroendocrinol. Neuropsychiatric Disorders. Amsterdam: Elsevier (2021). p. 63–82. (2016) 41:99–113. doi: 10.1016/j.yfrne.2016.02.004 doi: 10.1016/B978-0-12-819973-2.00005-8 100. Kinsley CH, Madonia L, Gifford GW, Tureski K, Griffin GR, Lowry C, 119. Moses-Kolko EL, Perlman SB, Wisner KL, James J, Saul AT, Phillips et al. Motherhood improves learning and memory. Nature. (1999) 402:137–8. ML. Abnormally Reduced dorsomedial prefrontal cortical activity and doi: 10.1038/45957 effective connectivity with amygdala in response to negative emotional 101. Love G, Torrey N, McNamara I, Morgan M, Banks M, Hester NW, et al. faces in postpartum depression. Am J Psychiatry. (2010) 167:1373–80. Maternal experience produces long-lasting behavioral modifications in the doi: 10.1176/appi.ajp.2010.09081235 rat. Behav Neurosci. (2005) 119:1084–96. doi: 10.1037/0735-7044.119.4.1084 120. Ho SS, Swain JE. Depression alters maternal extended amygdala 102. Gatewood JD, Morgan MD, Eaton M, McNamara IM, Stevens LF, Macbeth response and functional connectivity during distress signals in AH, et al. Motherhood mitigates aging-related decrements in learning and attachment relationship. Behav Brain Res. (2017) 325:290–6. memory and positively affects brain aging in the rat. Brain Res Bull. (2005) doi: 10.1016/j.bbr.2017.02.045 66:91–8. doi: 10.1016/j.brainresbull.2005.03.016 121. Laurent HK, Ablow JC. A cry in the dark: depressed mothers show reduced 103. Kinsley CH, Trainer R, Stafisso-Sandoz G, Quadros P, Marcus LK, Hearon neural activation to their own infant’s cry. Soc Cogn Affect Neurosci. (2012) C, et al. Motherhood and the hormones of pregnancy modify concentrations 7:125–34. doi: 10.1093/scan/nsq091 of hippocampal neuronal dendritic spines. Horm Behav. (2006) 49:131–42. 122. Lenzi D, Trentini C, Macaluso E, Graziano S, Speranza AM, Pantano P, doi: 10.1016/j.yhbeh.2005.05.017 et al. Mothers with depressive symptoms display differential brain activations 104. Galea LAM, Leuner B, Slattery DA. Hippocampal plasticity when empathizing with infant faces. Psychiatry Res Neuroimaging. (2016) during the peripartum period: influence of sex steroids, stress 249:1–11. doi: 10.1016/j.pscychresns.2016.01.019 and ageing. J Neuroendocrinol. (2014) 26:641–8. doi: 10.1111/jne. 123. Wonch KE, Medeiros CB, de Barrett JA, Dudin A, Cunningham WA, Hall 12177 GB, et al. Postpartum depression and brain response to infants: differential Frontiers in Global Women’s Health | www.frontiersin.org 16 October 2021 | Volume 2 | Article 742775 Martínez-García et al. Brain Structural Adaptations in Mothers amygdala response and connectivity. Soc Neurosci. (2016) 11:600–17. 135. Che K, Mao N, Li Y, Liu M, Ma H, Bai W, et al. Altered spontaneous doi: 10.1080/17470919.2015.1131193 neural activity in peripartum depression: a resting-state functional 124. Dudin A, Wonch KE, Davis AD, Steiner M, Fleming AS, Hall GB. Amygdala magnetic resonance imaging study. Front Psychol. (2020) 11:656. and affective responses to infant pictures: comparing depressed and non- doi: 10.3389/fpsyg.2020.00656 depressed mothers and non-mothers. J Neuroendocrinol. (2019) 31:e12790. 136. Chen H-H, Lai JC-Y, Hwang S-J, Huang N, Chou Y-J, Chien L-Y. doi: 10.1111/jne.12790 Understanding the relationship between cesarean birth and stress, anxiety, 125. Laurent HK, Ablow JC. A Face a mother could love: depression-related and depression after childbirth: a nationwide cohort study. Birth. (2017) maternal neural responses to infant emotion faces. Soc Neurosci. (2013) 44:369–76. doi: 10.1111/birt.12295 8:228–39. doi: 10.1080/17470919.2012.762039 137. Dekel S, Stuebe C, Dishy G. Childbirth induced posttraumatic stress 126. Silverman ME, Loudon H, Safier M, Protopopescu X, Leiter G, Liu X, et al. syndrome: a systematic review of prevalence and risk factors. Front Psychol. Neural dysfunction in postpartum depression: an fMRI pilot study. CNS (2017) 8:560. doi: 10.3389/fpsyg.2017.00560 Spectr. (2007) 12:853–62. doi: 10.1017/S1092852900015595 138. Dekel S, Ein-Dor T, Berman Z, Barsoumian IS, Agarwal S, Pitman 127. Silverman ME, Loudon H, Liu X, Mauro C, Leiter G, Goldstein MA. The RK. Delivery mode is associated with maternal mental health neural processing of negative emotion postpartum: a preliminary study of following childbirth. Arch Womens Ment Health. (2019) 22:817–24. amygdala function in postpartum depression. Arch Womens Ment Health. doi: 10.1007/s00737-019-00968-2 (2011) 14:355–9. doi: 10.1007/s00737-011-0226-2 139. Demarchi L, Pawluski JL, Bosch OJ. The brain oxytocin and 128. Pawluski JL, Lonstein JS, Fleming AS. The neurobiology of corticotropin-releasing factor systems in grieving mothers: what postpartum anxiety and depression. Trends Neurosci. (2017) 40:106–20. we know and what we need to learn. Peptides. (2021) 143:170593. doi: 10.1016/j.tins.2016.11.009 doi: 10.1016/j.peptides.2021.170593 129. Mao N, Che K, Xie H, Li Y, Wang Q, Liu M, et al. Abnormal 140. Swain JE, Tasgin E, Mayes LC, Feldman R, Todd Constable R, information flow in postpartum depression: a resting-state functional Leckman JF. Maternal brain response to own baby-cry is affected by magnetic resonance imaging study. J Affect Disord. (2020) 277:596–602. cesarean section delivery. J Child Psychol Psychiatry. (2008) 49:1042–52. doi: 10.1016/j.jad.2020.08.060 doi: 10.1111/j.1469-7610.2008.01963.x 130. Chase HW, Moses-Kolko EL, Zevallos C, Wisner KL, Phillips ML. Disrupted 141. Berman Z, Kaim A, Reed T, Felicione J, Hinojosa C, Oliver K, et al. Amygdala posterior cingulate–amygdala connectivity in postpartum depressed women response to traumatic childbirth recall is associated with impaired bonding as measured with resting BOLD fMRI. Soc Cogn Affect Neurosci. (2014) behaviors in mothers with childbirth-related ptsd symptoms: preliminary 9:1069–75. doi: 10.1093/scan/nst083 findings. Biol Psychiatry. (2020) 87:S142. doi: 10.1016/j.biopsych.2020.02.378 131. Zhang S, Wang W, Wang G, Li B, Chai L, Guo J, et al. Aberrant resting-state interhemispheric functional connectivity in patients Conflict of Interest: The authors declare that the research was conducted in the with postpartum depression. Behav Brain Res. (2020) 382:112483. absence of any commercial or financial relationships that could be construed as a doi: 10.1016/j.bbr.2020.112483 potential conflict of interest. 132. Deligiannidis KM, Sikoglu EM, Shaffer SA, Frederick B, Svenson AE, Kopoyan A, et al. GABAergic neuroactive steroids and resting-state Publisher’s Note: All claims expressed in this article are solely those of the authors functional connectivity in postpartum depression: a preliminary study. J and do not necessarily represent those of their affiliated organizations, or those of Psychiatr Res. (2013) 47:816–28. doi: 10.1016/j.jpsychires.2013.02.010 the publisher, the editors and the reviewers. Any product that may be evaluated in 133. Deligiannidis KM, Fales CL, Kroll-Desrosiers AR, Shaffer SA, Villamarin this article, or claim that may be made by its manufacturer, is not guaranteed or V, Tan Y, et al. Resting-state functional connectivity, cortical GABA, and neuroactive steroids in peripartum and peripartum depressed endorsed by the publisher. women: a functional magnetic resonance imaging and spectroscopy study. Neuropsychopharmacol. (2019) 44:546–54. doi: 10.1038/s41386-018- Copyright © 2021 Martínez-García, Paternina-Die, Desco, Vilarroya and Carmona. 0242-2 This is an open-access article distributed under the terms of the Creative Commons 134. Xiao-Juan W, Jian W, Zhi-Hong L, Yan M, Shi-Wei Z. Increased posterior Attribution License (CC BY). The use, distribution or reproduction in other forums cingulate, medial frontal and decreased temporal regional homogeneity is permitted, provided the original author(s) and the copyright owner(s) are credited in depressed mothers. a resting-state functional magnetic resonance and that the original publication in this journal is cited, in accordance with accepted study. Procedia Environ Sci. (2011) 8:737–43. doi: 10.1016/j.proenv.2011. academic practice. No use, distribution or reproduction is permitted which does not 10.112 comply with these terms. Frontiers in Global Women’s Health | www.frontiersin.org 17 October 2021 | Volume 2 | Article 742775