Abstract
The female represents a study in contrasts. For example, the virgin or nulliparous (NULL) female, it can be argued, is a work in progress, possessing natural and inherent neuroplasticity that is expressed only after pregnancy and the initiation of lactation. These changes likely contribute to the mother's behavioral repertoire, economizing and increasing efficiencies directed at the care, protection, and nurturing of the vulnerable offspring. 1 The mother receives a boost to those behaviors required for both her and her offspring's survival; in exchange, the probability that her genes will survive is enhanced. We have argued that unfortunately, because the male/father may be compromised or not contributory because of the likelihood of competition-related injury, death, or unfaithfulness, the burden has fallen on the nervous systems of the female to care for the vulnerable genetic legacy. 2
What are the underlying neurobiologic alterations? During pregnancy and parturition, significant neurobiologic and neurochemical changes summate to create dramatic behavioral changes that may allow a mother to ensure the survival of her offspring. 3 Reproductive changes occur in the brain of a wide variety of species examined heretofore. Behavioral analyses coupled to neuroanatomical ones have highlighted the adaptations that occur during and after pregnancy, often times leading to better overall task performance in attention, spatial memory, and object recognition. 4
Using rodent models, researchers have been able to investigate the changing neuroanatomy and chemistry that accompany pregnancy. For example, the medial preoptic area (mPOA) has been implicated as one of the key brain regions associated with maternal behavior. In this region, the data suggest that pregnancy may enhance neuronal growth and activity. Lesions at this site in pregnant and parturient rats have reduced or abolished maternal behavior; pup-directed responses, such as nest building, grooming, and nursing, are affected. 3 In the human, Kim et al. 5 have shown architectural changes within the mPOA that are associated with maternal motivation and interest.
Primiparous (PRIM) and multiparous (MULT) rats showed better performance compared to NULL females in both the 8-arm radial maze and the dry land maze (which is analogous to the Morris water maze). 1,2,4 Compared to nonmothers, parous females are more efficient and faster to find the food reward in both types of memory tasks. Further, neuroanatomical analyses have shown relevant structural changes. 2,5,6 Further analyses of the neuroendocrinologic changes that occur during pregnancy have indicated an upregulation of estrogen (E2), prolactin, and oxytocin, 3 which are key modulators of hippocampal changes. Similar enhanced spatial memory was also shown in mice 7 and tied to oxytocin regulation of LTP. Last, memory enhancement is not isolated to pregnancy events alone: rodent models have shown that memory and spatial learning continue postpartum or in foster mothers through sensory stimulation provided by pups (e.g., sights, smells, sounds, suckling, and tactile stimulation). 1,8 Thus, maternal stimuli facilitation of long-term potentiation (LTP) in mothers may underlie some of their behavioral augmentation.
The hypothalamic-pituitary-adrenal (HPA) axis undergoes significant changes during pregnancy as well. 8 Studies in humans, rats, and mice have indicated that overall basal activity is reduced within the HPA axis; additionally, the threshold for response to physiologic, psychologic, and environmental stressors is raised, thereby enabling the mother to be better able to handle the intranest/extranest stressors that arise in the course of caring for young. Upregulation of prolactin has been shown to increase 15-fold in pregnant woman. Further studies of change to the HPA axis using prolactin receptor knockout mice have shown that defective prolactin binding is correlated with severe maternal behavior deficits. 3
Pregnancy significantly alters the rat's primary sensory system, olfaction, in that both rodent models and human studies 9 have shown that the sense of smell significantly regulates maternal behavioral responses. In rodent models, parity has been associated with hyperresponsiveness to olfactory cues, 10 with mothers preferring cages with pup-soiled bedding. Other studies examining the regulation of maternal behavior have shown that treatments that disrupt maternal behavior in lactating females (e.g., morphine) do so, in part, by increasing the aversive properties of the pups' olfactory signatures. 11
Exposure to the pregnancy hormones estradiol and progesterone (P) in rabbits increased stimulation in brain areas synonymous with maternal behavior (the mPOA) and learning/memory (the hippocampus 12,13 ). Zarrow et al. 14 reported that the ratio between E2 and P is crucial for maternal behavior, such as nest building, in rabbits. From building nests, protecting eggs and hatchlings, and feeding their offspring, examinations of bird behavioral models and brains suggest that reproduction has wide-ranging effects. In the bird, changing hormone levels and presence of sex steroids in reproductive females and egg yolk have suggested that the maternal sex steroids influence the release of steroids later in the offspring's life. Such changes may also, as in their mammalian counterparts, influence maternal behavior. 15,16
Working down the phylogenetic scale, observations of amphibians and certain species of insects have shown parental behaviors specific to the needs of their offspring (a kind of sliding scale), such as tadpole transport and protection seen in frogs. 17 Researchers investigating parental behavior in insects suggest that communicative chemical signaling and pheromones and their relevant molecular pathways may serve as mediators of reproductive change. This avenue of research, however, remains largely unexplored. 18 Our laboratory has examined reproduction-mediated effects in wolf spiders (which demonstrate care for spiderlings; S. Choi and C.H. Kinsley, unpublished observations) and tiger beetles, both of which show postreproduction changes in behavior. The effects of reproduction may, therefore, represent a common set of hormone-nervous system interactions that have evolved to facilitate reproduction regardless of species.
More specifically, a considerable body of data from rodents has shown that reproductive experience positively alters females' cognitive functions, markedly spatial, 1,7,19,20 object recognition, 21 and prospective memory. 22 Admittedly, improved memory brings a number of benefits to the mother. it allows her to save time and energy, economies that increase the likelihood of her descendants'/genes' survival. Hence, it is important to investigate the possibility that similar alterations may extend to human mothers. It becomes challenging, however, to determine appropriate or parallel experimental models, for example, rodents vs. humans, given the large number of potential individual covariates that could interfere with the results (e.g., diet, age, education level, personal and medical histories).
In this issue of the Journal of Women's Health, Glynn 23 herein presents a controlled, large-sample longitudinal study to assess verbal memory performances in PRIM and MULT women in different stages of pregnancy. Additionally, the author tested the subjects at 3 months, postpartum. The behavior examined was a paired-association learning task, which consists of recalling word-pair lists presented verbally. Such tasks involving associations of unrelated but familiar words are widely employed in neuropsychologic studies; they may require complex interactions between phonologic short-term and long-term memories, as well as semantic coding strategies. 24,25 The larger question is: Are they relevant to the human mother's care toward her offspring?
Glynn's results 23 clearly show that PRIMs outperformed MULTs across gestation and that a cumulative effect of multiple pregnancies was associated with poor performance in the last weeks of pregnancy. Finally, the data showed that the effects were persistent even after 3 months postpartum. de Groot et al. 26 also found that pregnant women performed significantly worse on a word-learning task compared to nonpregnant women, and the effect persisted for 32 weeks after delivery. Conversely, Christensen et al. 27 found no differences related to maternal experience in subjects exposed to a working memory task. The discrepancies in the aforementioned work may be related to the differences among the experimental groups analyzed or the memory task used. For instance, Christensen et al. 27 and de Groot et al. 26 compared memory performance of mothers and nonmothers, whereas Glynn 23 investigated the same functions in groups of mothers with different parity histories (PRIM vs. MULT). The memory tasks employed may require the participation of different neural systems; that is, a memory task that requires the subjects to repeat digits backwards emphasizes components of a working memory system, whereas a word-pair learning task using unrelated familiar words allows the interaction between short-term and long-term memory and even additional semantic strategies. Therefore, the results of Glynn 23 demonstrate a strong cumulative effect of multiple pregnancies on a specific memory function (verbal recall memory) and raises important questions. Additional studies comparing similar groups with NULL women would be necessary to establish a baseline performance, thereby examining if multiparity has the same deleterious effect compared to the NULL condition. In other words, without a NULL group for comparison, we are not certain if the effects reported relate to improvements or decrements.
To conclude, as suggested by the Glynn article, 23 the maternal brain and the accompanying behavioral effects represent adaptations to the increasing demands of motherhood. We have argued that the female nervous system has evolved a potentiality for many strong and long-lasting modifications during and after pregnancy, 1,2 including aging-related enhancements. 28 Such changes regulate successful reproduction and impinge on ancillary maternal behaviors, motor function, and the aging brain and nervous system. In fact, the reproduction-induced changes, as in the case of early developmental and sexual differentiation, together with puberty, likely mark the female for the rest of her life. Such events are life defining, and we would argue, too, that reproduction is akin to other neurobiologic epochs in the animal's life. For example, when one considers the constellation of maternal behaviors induced by pregnancy and lactation, infant recognition and motivation are primary effects. There are permanent modifications in the female's behavior toward her own or unfamiliar young, a general facilitation of maternal responsiveness and cognition. Such change suggests likely permanent alterations to underlying neural structures that are associated with reproduction (Moltz et al.) 29 reported that reproductively experienced females subsequently young retained the memory for pups and acted maternal toward them many months after their initial exposure, an effect similarly reported by Bridges 30 and referred to as maternal memory. Thus, a single early reproductive event and associated maternal experience is sufficient to mark the female for many months, 28 the equivalent of decades in the human, in the intensity of her maternal responsiveness.
It is important to bear in mind that the period of pregnancy is not some quiescent interlude in the female's transition from nulliparous to parous female, where both mother and child are adrift on some peaceful sailing bark, slowly propelled by a light zephyr across the silent seas of pending maternal bliss. Rather, this period is marked in the mother by dramatic increases in the concentration of powerful steroid (E2 and P) and other hormones (prolactin) washing over a sensitive and plastic brain and, in the fetus, by an explosion of neural development. For the new mother, the striking influence exerted by the levels of these natural steroid and protein hormones apparently alters many neural areas, among them regions that govern maternal and supporting behaviors.
Thus, the hormones of pregnancy change behavior and physiologic regulation in the female, making maternal behavior and its immediate onset—characteristic of the postpartum maternal female—among the most striking. Such rapid and intense behavioral effects are preceded by significant actions of the hormones on the female's maternal neural substrate. A Golgi-Cox silver-staining examination revealed an example of neural plasticity related to pregnancy and its hormonal exposure. 31 Neurons in the mPOA, a region that strongly regulates maternal behavior, were examined in a group of females from different hormonal groups (ovariectomized [OVX], diestrus, sequential P and E2, late pregnant, and lactating [d5] rats). There was no difference between OVX and diestrus females, and both had smaller neuronal cell body areas compared to P-treated and E2-treated and late pregnant females. 31 The area of the soma returned to diestrus/OVX levels in lactating females, suggesting a return to baseline that follows the female's pregnancy hormonal state.
There were similar hormone-induced effects on a number of dendritic branches and cumulative dendritic length in pregnant and hormone-treated groups compared to the OVX, diestrus, and lactating females. The increase in somal area denotes increased cellular activity (for instance, protein synthesis 32 ), whereas the stimulatory effects on additional neuronal variables represent modifications in information processing capacity. Pregnancy and its consequent hormone stimulation may alter neurons in the mPOA and possibly other regions, which then contribute to the display of maternal behavior and its supporting activities. For example, whereas the mPOA regulates pup-directed maternal behavior, ancillary sites (e.g., hippocampus) are also liable to undergo changes in their own right. Data show that the effects may extend to other behaviors and additional brain regions on which the female relies for performing her maternal duties. Reproductive experience and exposure to offspring significantly modify the brain and behavior of the female, particularly those required for effective care and rearing of offspring.
Several studies have reported that E2 and P modify other substructures of the neuron in the adult female brain, for example, increasing the concentration of apical dendritic spines in hippocampal neurons. 33 –37 These effects occur with comparatively short exposure to the hormones, primarily E2, during the female's 4–5-day estrous cycle. Thus, should the level or pattern of E2 and P associated with the estrous cycle be prolonged or increased, as occurs during pregnancy, effects on the morphology of the neuron may be even greater or long lasting. Such effects have been reported in the CA1 region of the hippocampus, with a higher concentration of dendritic spines in late pregnant, lactating, and pregnancy hormone-exposed rats compared to NULL females. 38 Thus, regions that regulate learning and memory appear to undergo significant modification with alterations in reproductive status, but the effects in human mothers appear to be different or more subtle compared to those described in lower animals.
As we have argued, the transition from virginity to motherhood is a time of great change, marked by powerful hormonal fluctuations, their neuronal interactions, and consequent change in maternal interest, motivation, and behavior. Whether rat or human, the presence of young requires a shift in the mother's focus from self to other. In order to ensure her survival, and to secure her offspring's genetic legacy, the mother is fundamentally altered in ways that require a different kind of focus, one that involves infant-directed cognitive changes and, hence, their examination. Making connections between basic research and more clinical applications—the Venn diagram of animal and human relevance—is becoming more and more crucial. Glynn's current work 23 focuses on the overlap and parallels between human and the rodent experiences, which for the rodent (and for other species), are well-described behavioral brain process modifications. Glynn focuses on tasks (verbal memory) that are (as far as we know) not part of the rodent repertoire. How important they are to the human mother's care and protection of young is arguable, and Glynn's work begins to address this question in her thorough examination of the cognition of parous women.
Keep in mind that context is important. We and others have observed alterations to behavior and fundamental brain processes, and some relevant neurobiologic effects occur in the human maternal brain. As mentioned, Glynn focuses on tasks that obviously are not part of the rodent repertoire (verbal memory), nor, we would argue, are they requisite for the human mother's care of young, falling to lower rungs on the hierarchy of survival mechanisms. We would further argue that to understand the kinds of changes that would be relevant to the mother, one needs to step back in time to where humans were tens of thousands years ago, when work, as we define it today, was very different. Verbal memory may be important today, as are other cognitive features, but we would not necessarily define them as maternal behavior per se. That is, we would argue that only a behavior with ecologic adaptive relevance for the offspring is likely to be improved in mothers. We must, therefore, look at the 21st century woman through the eyes of our Pliocene Epoch ancestors.
Further examinations of diverse forms of maternal memory or of adult reproduction-related developmental offspring-relevant tasks (e.g., facial emotional processing, attentional mechanisms, potential sharpening of other sensory systems) are required to comprehensively address the wholesale renovation of the female brain into the maternal brain. Glynn's article 23 is a very good start, and it raises many questions worth following up in humans, other mammals, and additional postreproduction models.
Footnotes
Disclosure Statement
The authors have no conflicts of interest to report.