For Manuscript Submission, Check or Review Login please go to Submission Websites List.
For the academic login, please select your country in the dropdown list. You will be redirected to verify your credentials.
Looking for Trouble: A Search for Developmental Defects of the HypothalamusCaqueret A. · Yang C. · Duplan S. · Boucher F. · Michaud J.L.
Research Center, Hôpital Sainte-Justine, Montréal, Canada Corresponding Author
Dr. Jacques L. Michaud
Research Center, Hôpital Sainte-Justine
3175 Côte Sainte-Catherine, Montréal, Qué., H3T 1C5 (Canada)
Tel. +1 514 345 4727, Fax +1 514 345 4766
The hypothalamus is a critical integrator of several homeostatic processes that are required for the survival of vertebrates. Disruption of the development of the hypothalamus thus has the potential of perturbing important physiological processes with lifelong consequences. We review current knowledge about how cell types are specified and circuits are formed within the developing hypothalamus. We emphasize the potential clinical impact of the perturbations of these pathways using the regulation of energy balance as a model. We predict that disruption of hypothalamic development is a common, previously unsuspected cause of disorders of homeostatic processes such as obesity and high blood pressure.
© 2005 S. Karger AG, Basel
The hypothalamus is an evolutionary ancient integrator of homeostasis that regulates basic processes such as food and water intake, energy expenditure, the response to stress, blood pressure, and reproduction. Structurally, the hypothalamus is composed of a dozen small nuclei interspaced between less defined regions. A great body of work involving physiological and genetic studies has assigned specific functions to each of these nuclei and regions (see summary in fig. 1). In contrast, relatively little attention has been paid to how the different components of the hypothalamus are specified and connected. Understanding the development of the hypothalamus has the potential of shedding light on frequent pathological processes that can manifest even late in life. An immediate illustration of this principle is provided by classical work done on the development of gonadotropin-releasing hormone (GnRH) neurons and its disruption in some conditions affecting puberty . Here, we review basic knowledge about other aspects of hypothalamic development focusing on how cell types are specified and axonal connections are established. The potential clinical impact of the perturbations of these pathways is illustrated using the regulation of energy balance as a model.
Axial signals provided by the mesendoderm underlying the rostral neural plate are required to divide the primordial single eye field into two eyes . Failure of this process causes holoprosencephaly, the most frequent brain malformation . Because these axial signals are also required to induce development of the hypothalamus between the two eyes, it is not surprising that it is either missing or lacking in structures in holoprosencephalic brains. In fact, endocrine deficiencies, caused by hypothalamic and/or pituitary dysfunction, can be the only clinical problem in milder forms of holoprosencephaly. Currently, mutations in eight genes have been shown to cause holoprosencephaly in humans . Three of these genes (SHH, PTC1,GLI2) encode members of the Sonic Hedgehog signaling pathway, whereas two others have the potential of interacting with this pathway (SIX3, ZIC2). The other genes (TDGF1, FAST1, TGIF) appear to interact with the Nodal/TGF-β signaling pathway. Consistent with the pathophysiology of holoprosencephaly, SHH and Nodal are axial signals that are required for hypothalamus induction (reviewed in ). Interestingly, these signals also appear to function for subsequent regionalization of the hypothalamus. Although the severe midline defect observed in Shh- and Nodal-mutant mice precluded investigation of such a role, recent studies in fish suggest that hedgehog and Nodal signaling pattern the prospective hypothalamus into distinct domains  (reviewed in ). Experiments performed in fish also indicate that inhibition of Wnt signaling in the neural plate is required, likewise, for the specification of the hypothalamus anlage and also for its regionalization . The identification of signals that regionalize the hypothalamus represents a critical step towards the elucidation of mechanisms underlying its organization into a set of small nuclei.
Another source of organizing signals for the hypothalamus is the anlage of the anterior pituitary. Coculture experiments have shown that the development of ACTH/MSH cells of the prospective arcuate nucleus (ARC), located in the ventromedial hypothalamus, is enhanced by a signal provided by the Rathke’s pouch . Also, Hermesz et al.  have characterized a Hesx1 regulatory element that directs the expression of a reporter transgene in the ventral hypothalamus of mice, in a region that lies in direct contact with Rathke’s pouch . In tissue recombination experiments, the pouch can induce expression of the transgene in the neurectoderm isolated from other sites of the developing forebrain, suggesting that the Hesx1 element responds to a signal emanating from the pouch. It is not yet clear whether the signal produced by the pouch that enhance the development of ACTH/MSH cells is the same as that required for the transgene expression in the ventral hypothalamus. Based on these observations, a prediction can be made that disruption of the development of the pouch might affect the development of the ventral hypothalamus.
Differentiation of neurons typically involves cascades of transcription factors that progressively define their phenotypes. Such cascades have been characterized in cells that populate hypothalamic nuclei.
The homeobox-containing transcription factor OTP and the heterodimeric complex formed by the bHLH-PAS transcription factors SIM1 and ARNT2 are required for the development of (1) virtually all neurons of the paraventricular (PVN) and supraoptic (SON) nuclei, including those identifiable by the production of thyrotropin-releasing hormone (TRH), corticotropin-releasing hormone (CRH), oxytocin (OT), and arginine vasopressin (AVP); (2) somatostatin (SS) parvocellular neurons of the anterior periventricular nucleus, which is contiguous to the PVN; and (3) other populations of hypothalamic TRH neurons found in the lateral areas and preoptic region [9,10,11,12,13,14]. SIM1:ARNT2 and OTP function along parallel pathways to control multiple aspects of the development of these cell types, including the proliferation of their progenitors, differentiation, and migration [9,10,11,12,13,14]; Caqueret et al., unpubl. data].
SIM1:ARNT2 and OTP act upstream to maintain the expression of Brn2, a POU domain transcription factor, and Sim2, a close homolog of Sim1, which, in turn, control the differentiation of distinct populations of hypothalamic neurons (fig. 2). Brn2 is required for the differentiation of CRH, OT, and AVP neurons of the PVN/SON [15, 16]. The fact that Brn2 regulates the activity of the Crh promoter in vitro suggests that it acts at the final stage of differentiation . In contrast, these PVN populations of OT-, AVP-, and CRH-expressing cells are not affected in Sim2–/– embryos but TRH cells as well as the SS cells of the anterior periventricular nucleus are decreased by 43 and 83%, respectively, in these embryos . The interplay between Sim1 and Sim2 is complex. Sim1 acts upstream to maintain Sim2 expression, but it can also compensate, albeit ineffectively, for loss of Sim2. It is unclear whether TRH and SS cells that require Sim2 for their development and those that do not require Sim2 have different properties or function.
Brn2 and Sim2 are thus required for the differentiation of distinct cell types. The domain of Brn2 expression is located anteriorly compared to that of Sim2, correlating with the spatial distribution of the cell types in which they function . These observations provide novel insight into the emergence of the neuronal topography of the anterior hypothalamus.
The ARC is composed of distinct cell types whose identity and spatial distribution have not yet been fully resolved . Early on during development, signals produced by the axial mesendoderm induce the expression of the homeobox-containing transcription factor NKX2.1 in the overlying presumptive hypothalamus [19,20,21]. In turn, Nkx2.1 functions to specify the anlage of the ventral hypothalamus that will give rise to arcuate and the ventromedial nuclei [19, 22]. A set of homeobox-containing transcription factors has been shown to function downstream to specify some of the cell types populating the ARC (fig. 3). For instance, Gsh1 is required for Ghrh (growth-hormone-releasing hormone) expression in the ARC, whereas in vitro studies indicate that it can regulate Ghrh expression directly by interacting with its promoter . Hmx2 and Hmx3 together are also required for Ghrh expression, at least in part, by maintaining Gsh1 expression . Otp affects the development of another cell type in the ARC identifiable by the production of SS [13, 14]. As mentioned above, Otp also affects Ss expression in other regions of the diencephalon, including the anterior periventricular nucleus and zona incerta. It is likely that Gsh1, Hmx2/Hmx3, and Otp not only function to regulate the expression of Ghrh and Ss, but also have a more global impact on the differentiation of the cells in which they are expressed.
Steroidogenic factor 1 (SF1), an orphan member of the nuclear receptor family, is required for the development of several endocrine systems. Sf1-homozygous mice lack gonads and adrenals and have impaired function of pituitary gonadotropes [25, 26]. The only site of Sf1 expression in the brain is the ventromedial nucleus (VMN). The VMN of Sf1–/–embryos, but not of heterozygotes, shows a complex phenotype [25,27,28,29,30,31]. VMN-mutant neurons are generated during development in appropriate number but their spatial distribution is altered with more neurons located medially and dorsally [30, 31]. The fact that the axonal projections originating from the VMN are absent in these mutants suggest that Sf1 not only affects migration of neurons but also their differentiation.
Loss of all known regulators of PVN/SON development – Sim1, Arnt2, Otp, Brn2 and Sim2 – in mice result in perinatal lethality. The severe PVN/SON defects found in mutants, such as Sim1 or Otp, might underlie their lethality. In contrast, Sim1+/– mice survive and develop early onset obesity, with increased linear growth, hyperinsulinemia, and hyperleptinemia. These mice are hyperphagic, even before they show increased weight gain [32, 33]. A balanced translocation interrupting SIM1 was found in a child with a similar phenotype . Like Sim1+/–mice, this child had early onset obesity, increased linear growth, and a voracious appetite, suggestive of hyperphagia. Mouse models suggest that SIM1 haploinsufficiency is responsible for the obesity in this child. The description of morbid obesity in children with chromosomal deletions in the 6q16 region, which contains SIM1, further strengthens this conclusion [35,36,37]. Interestingly, a genome-wide search for childhood obesity traits has shown significant linkage on chromosome 6q22–q23, near the SIM1 locus . The mechanisms underlying the hyperphagia associated with Sim1 haploinsufficiency are still under investigation. One possibility is that dysfunction of the PVN causes this hyperphagia (fig. 1). This dysfunction could be of developmental origin. Indeed, the PVN of Sim1+/– mice is hypocellular, containing 24% fewer cells . However, since Sim1 and its dimerizing partner Arnt2 are expressed continuously in the postnatal PVN, the possibility that Sim1 controls food intake physiologically is not excluded.
Brn2 is also expressed in the postnatal PVN/SON. The regulation of Crh expression by Brn2 in cultured cells and the fact that Avp and Ot mRNA levels are reduced in Brn2-heterozygous mice raise the possibility that Brn2 participates in physiological processes by controlling the expression of these effectors dynamically [15, 16]. However, overexpression of Brn2 in the PVN using adenoviral vectors does not affect expression levels of Avp and Crh . Also, the expression of Brn2 is not changed in a stress model associated with increased production of CRH . Nevertheless, the possibility that Brn2 regulates Crh expression in the postnatal PVN/SON in response to specific stimuli needs further exploration. For instance, it has been shown that Brn2 overexpression is not sufficient per se to induce CRH expression but is required for retinoic-acid-induced CRH expression in a neuroblastoma cell line .
Sf1–/–mice can be rescued from adrenal insufficiency at birth by corticosterone replacement or adrenal transplantation . These rescued mice show increased weight gain by 8 weeks of age and eventually weighed almost twice as much as wild-type controls. Sf1–/– mice are not hyperphagic but show decreased locomotor activity. This model will be useful for the dissection of the role of VMN in the regulation of energy balance.
During development, attractive and repulsive cues guide axons towards their targets. Very little, however, is known about the molecules that guide hypothalamic axons. Recent studies have shed some light on how one set of axons, those originating from GnRH neurons, reach their target. GnRH neurons originate from the olfactory epithelium, migrate following olfactory/vomeronasal axons and settle in a continuum from the olfactory bulbs to the hypothalamus. Specification and migration of GnRH neurons has been reviewed elsewhere and will not be further discussed here . The majority of GnRH neurons projects to the median eminence (ME). Explant and transplantation studies indicate that the ME produces a diffusible attractive signal for GnRH axons . A prototypical guidance molecule is netrin-1, a secreted protein that attracts axons by interacting with its receptor DCC. Netrin-1 is produced in medial areas flanking the third ventricle in which GnRH axons progress as well as in the prospective ME, whereas GnRH neurons express DCC . Loss of netrin-1 or DCC reduces dramatically the number of GnRH axons that reach the ME, suggesting that netrin-1 is a major cue for the guidance of these axons towards their target . In contrast, CRH and AVP axons grow normally into the ME and posterior pituitary, respectively, of these mutants. Previous studies have established that dopaminergic, GHRH, and SS neurons restrict their projections to certain subregions of the ME (reviewed in ). Altogether, these observations suggest the existence of a complex network of cues produced by distinct domains of the prospective ME that each attracts different sets of hypothalamic axons.
Whereas axon projections to the ME and infundibulum are established by the time of birth in rodents, recent studies indicate that this is not the case for all hypothalamic projections. Bouret et al.  have found that ARC projections to the PVN, lateral hypothalamic area, and the dorsal medial nucleus develop during the first 2 weeks of postnatal life in mice. These ARC projections are critical mediators of the action of leptin on food intake. The fact that these projections are not yet completely developed at this stage of life might explain the insensitivity of the neonatal hypothalamus to the massive surge of leptin levels observed during the first week of life in rodents . Interestingly, Bouret et al.  found that leptin actually acts as a trophic signal that stimulates ARC axon outgrowth. More specifically, the density of ARC projections to the PVN, lateral hypothalamic area, and dorsal medial nucleus is decreased in leptin-deficient mice during the first 2 weeks of life, before weaning, and remains so in adulthood. Administration of leptin during the first 2 weeks, but not later in life, increases the number of ARC projections in leptin-deficient mice, which results in decreased food intake in the long term. Obesity secondary to leptin deficiency thus appears to involve a developmental component.
Interspecies comparisons of various features of neural development indicate that the brain of newborn mice is immature, corresponding to that of human fetuses at the beginning of the third trimester of pregnancy [48,49,50]. From a clinical point of view, factors that modulate levels of leptin in human fetuses or premature babies would thus be predicted to affect the development of ARC projections, with the potential of having long-lasting effect on energy balance. Alternatively, the development of these ARC projections after birth could represent a conserved feature of mammalian development, also present in humans, that defines a phase during which food intake would not be subjected to the same regulatory circuitries than later on, possibly reflecting the need to maximize food intake in the immediate postnatal period.
Projections of parvocellular OT neurons of the PVN to the dorsal vagal complex, a group of brainstem nuclei that comprises the dorsal motor nucleus of the vagus and the nucleus of the solitary tract, also increases during the first 2 weeks of postnatal life in rodents . These axons have also been shown to mediate in part the effect of leptin on food intake . An interesting possibility will be that the development of these projections is also stimulated by leptin during the first weeks of life. A functional correlate of the immaturity of this circuit has been proposed by Rinaman . There is good evidence that OT-containing projections from the PVN to the dorsal vagal complex are required for the anorexic response that accompanies dehydration . Interestingly, this response does not emerge until after the first 2 weeks of postnatal development, possibly because of the lack of OT projections .
Analysis of chimeric embryos and mice containing normal and uniparental cells have shown that parthenogenetic (complete maternal disomy) and androgenetic (complete paternal disomy) cells contribute to complementary regions of the brain [54, 55]. Although at 10 days of embryonic development, androgenetic cells appear widely distributed in the brain of chimeras, they subsequently become restricted to the hypothalamus. Conversely, parthogenetic cells are progressively excluded from the hypothalamus during development of chimeric brains, while they are found abundantly in the cortex. The stage during which these complementary distributions of uniparental cells are established coincides with the proliferation and differentiation of hypothalamic neural cells. This observation suggests that paternal expression of some genes is required for development of the hypothalamus.
Interestingly, at least two genes with monoallelic expression from the paternal allele, Peg3 and Ndn, might be required for the development of the hypothalamus. Peg3 is a critical regulator of several apoptotic pathways. Loss of the paternal copy of Peg3 in mice causes a decrease of activation of the sympathetic nervous system, presumably via the hypothalamus, that is associated with lower core body temperature and metabolic activity as well as with an increase of white and brown adipose tissue . Peg3 is expressed at high levels during the development of several regions of the hypothalamus that regulate energy balance. It is thus tempting to speculate that the loss of Peg3 disrupts normal hypothalamic development through its role in apoptotic pathways, leading to a cascade of energy balance dysfunctions.
NDN (encoding necdin) and MAGEL2 are two of four protein-coding genes lost in individuals with Prader-Willi syndrome (PWS), a sporadic chromosome deletion disorder characterized by hypothalamic dysfunction with hyperphagia and severe obesity . Necdin and MAGEL2, which are part of a multiprotein family related by a MAGE homology domain, are expressed in hypothalamic areas, such as the ARC and PVN, that are involved in the regulation of energy balance [58,]. Lee et al.  have recently found that necdin physically interacts with components of the centrosomes, which play a role in organizing microtubules during axonal outgrowth. As it might be expected from this observation, axons from necdin-null neurons elongate more slowly and are less bundled when grown in culture and several axonal tracts are reduced or misrouted in the brain of necdin-null embryos . These interesting observations raise the possibility that the hypothalamic dysfunction of PWS patients is caused by abnormal axonal connections. Hypothalamic axonal defects were not described in these mice, possibly because these projections do not form easily detectable bundles. Their visualization requires specific labeling, the use of which has not yet been reported. Another possibility is that MAGEL2 plays a similar role during development and that disruption of hypothalamus connections are observed only when both necdin and MAGEL2 expression is lost, explaining the lack of single mutations in PWS. Of note, necdin physically interacts with BBS4, one of the several genes mutated in Bardet-Biedl syndrome, raising the possibility that similar axonal defects exist in this condition also characterized by early onset obesity . Finally, a decrease of OT and GnRH cells has been observed in hypothalamus of Necdin-mutant mice, also possibly contributing to PWS phenotype .
Although loss of these genes individually does not have a dramatic effect on hypothalamic development, it is possible that the combined loss of several genes that are only expressed from the paternal allele, as found in parthenogenetic cells, could have more severe consequences, explaining the absence of these cells in chimeric hypothalamus late during development. Why then are imprinted genes required for hypothalamus development? One theory interprets imprinting as a tug-of-war between the interests of each parent through their genome . Paternally derived alleles would enhance the offspring’s own fitness, favoring greater maternal investment whether or not this investment occurs before or after weaning. In the context of this theory, building a functional hypothalamus would allow optimal growth and survival of offspring, which is to the advantage of the father.
The study of hypothalamus development can foster insight into critical physiological processes and their disorders, as it has been shown for other regulatory centers of homeostasis such as the pituitary and pancreas [63, 64].First, pathways that control differentiation, especially at the later stages, can be maintained in mature cells for the regulation of physiological effectors, like hormones or neurotransmitters, representing critical targets for therapies. Second, subtle developmental defects affecting the number of hypothalamic cells, their differentiation, or their axonal projections have the potential of causing metabolic and physiological disorders which can be isolated and expressed relatively late in life. As discussed in this review, instances of developmental defects of the hypothalamus affecting energy balance have been reported. We predict that disruption of hypothalamic development can also cause other phenotypes, such as high blood pressure or abnormal responses to stress, that are not as readily detectable as dysregulation of energy balance in animal models. Given that small groups of hypothalamic neurons like the PNV or ARC can participate in the regulation of multiple homeostatic processes, we would also predict that disruption of a single developmental pathway could generate several phenotypes that may coexist in an animal model or in a patient.
Dr. Jacques L. Michaud
Research Center, Hôpital Sainte-Justine
3175 Côte Sainte-Catherine, Montréal, Qué., H3T 1C5 (Canada)
Tel. +1 514 345 4727, Fax +1 514 345 4766
Copyright: All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher or, in the case of photocopying, direct payment of a specified fee to the Copyright Clearance Center.
Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.