A 21st century perspective on hypothalamic–vascular–pituitary unit function
[Au: our journal style is to write 21st century in full as ‘twenty first’, which would take the title over the length limits for this type of article. We also use ‘perspective’ for a specific opinion-based article. To reduce the length of the title and avoid the use of perspective, I suggest the following title modification:
“An updated view of hypothalamic–vascular–pituitary unit function”. OK?]
[Au: you use a mixture of hypothalmo and hypothalamic in the main text. I suggested consistently using on hypothalamic, changes throughout main text also OK?]
Paul Le Tissier1, Pauline Campos2,3,4, Chrystel Lafont2,3,4, Nicola Romanò1, David J. Hodson5,6 and Patrice Mollard2,3,4
1Centre for Integrative Physiology, University of Edinburgh, Edinburgh, EH8 9XD, UK. 2CNRS, UMR-5203, Institut de Génomique Fonctionnelle, F-34000 Montpellier, France; 3INSERM, U661, F-34000 Montpellier, France; 4Université de Montpellier, UMR-5203, F-34000 Montpellier, France; 5Institute of Metabolism and Systems Research and Centre of Membrane Proteins and Receptors (COMPARE), University of Birmingham, Edgbaston, B15 2TT, UK. 6Centre for Endocrinology, Diabetes and Metabolism, Birmingham Health Partners, Birmingham, B15 2TH, UK. [Au: please include the street name for all addresses]
Abstract [Au: edits OK?]
The discovery of novel functional adaptations of the hypothalamus and anterior pituitary gland for physiological regulation has transformed our understanding their interaction. The activity of a small number of hypothalamic neurons can control complex hormonal signalling, which is disconnected from a simple stimulus and subsequent hormone secretion relationship but is dependent on physiological status. The interrelationship of hypothalamic neurons and pituitary cells with the vasculature has an important role in determining the pattern of neurohormone exposure. Cell in the pituitary gland form networks with distinct organizational motifs that are related to the duration and pattern of output. These networks are modified in different physiological states and can persist after cessation of demand, which results in enhanced function [Au: edit OK?]. Consequently, the hypothalamus and pituitary can no longer be considered as having a simple stratified relationship: they form a tripartite system with the vasculature, which must function in concert for appropriate hypothalamic regulation of physiological processes, in particular reproduction [Au: edit OK? I think it’s important to mention reproduction in the abstract here]. An improved understanding of the mechanisms underlying these regulatory features has implications for current and future therapies that correct defects in the hypothalamic–pituitary axes. In addition, recapitulating proper network organization will be an important challenge for regenerative stem cell treatment.
[Au:For your information, H1 and H3 refer to the level of heading and will be removed before proofs are made. H1 subheadings can have max 38 characters inc spaces. H3 subheads can be of any length. Subheads have been edited to fit these limits, where indicated]
[Au: I think the Review would benefit from a figures introducing the basics of the concepts you discuss. I think this will be important to increase the impact of the paper, and for our less informed readers who are endocrinologists, but not particularly familiar with the neuroendocrine aspects. I have included the details of the figures you might wish to consider adding at the end of this document. I have also indicated in the main text where I think new figures might be useful additions. When you submit your revisions, please submit any new figures and I will work with our art editor to generate the first versions. We can revise these before making the final proof. If you wish to discuss these in greater detail before submitting, please do not hesitate to contact me]
[H1] Introduction [Au: I have moved the second section (originally called ‘Hypothalamic–pituitary axes’ of your original draft to here to serve as an introduction (the opening section is always titled ‘Introduction’ in our style guide). I think this text is more appropriate here and sets up the rest of the article nicely. Please also include the appropriate references in this section]
To maximize reproductive success, via the appropriate timing of ovulation, lactation or body growth, the output of several hypothalamic–pituitary axes are dramatically altered [Au: edit OK?]. These adaptive changes occur over differing time scales, with varying frequencies and levels of predictability [Au: is there a general review you can cite here?]. For example, the increase in growth hormone (GH) output at puberty is largely predictable [Au: meaning the levels are predictable or simply that it occurs?]. On a relatively short time scale [Au: meaning measured in days?], the surge in luteinizing hormone (LH) secretion required for oestrus is an acute change that occurs regularly once every reproductive cycle and, in humans, continues for years in the absence of pregnancy. On a longer time scale [Au: meaning months/years?], the increase in prolactin required for lactation is maintained for a variable time (which depends on when offspring are weaned) and recurs at each pregnancy, but is unpredictable before gestation [Au: edit OK to avoid repetition of ‘pregnancy’?]. These large changes in output require modification of both hypothalamic and pituitary function, but whether this is reversed on cessation of physiological demand likely reflects the expectation that increased output will recur. [Au: edit OK is this what you?]. A mechanistic understanding of these alterations in hypothalamic–pituitary function is fundamental to interpret and treat defects that lead to endocrine diseases resulting from hormone deficiencies (eg dwarfism) or excess (eg polycystic ovarian syndrome) [Au: please be more specific here as to the diseases you mean in this context to set the scene for your review]. In this Review, we will focus on three pituitary axes that have roles in driving changes in physiology; the gonadatropin, prolactin and growth hormone axes [Au: addition OK?]. The level of our understanding varies for each of these axes and the features that might serve as general principles will be highlighted in the text [Au: edit OK?].
[H1] Beyond stimulus-secretion coupling [Au: I think you could include a simple figure in this section on stimulus-secretion as an addition to BOX 1. And based on the summary figure you have already submitted. I think these concepts will work well in graphical form for our less informed readers]
The speed of communication between the brain and peripheral tissues is highlighted by muscle contraction, which requires the transfer of electrical signals from axons via the neuromuscular junction [Au: edit OK? Is this what you mean here?]. This sequence of events, known as excitation-contraction coupling1, takes <1 s in mammals and is highly plastic [Au: please elaborate on what you mean by plastic in this instance]. Similarly, in the hypothalamus just a few thousand neurons can also send signals to the periphery, in this case toward the median eminence (ME) via a specialized neurohaemal junction [Au: edits OK?]. Here, nerve terminal depolarization, either originating from the perikarya2 or the terminal itself3, allows the sufficiently rapid entry of calcium ions to trigger exocytosis of neurohormones towards the first loop of the portal fenestrated capillaries [Au: please reference this statement here]. This rapid (< 1 s) sequence of events was termed ‘stimulus-secretion coupling’ due to the clear similarities with excitation-contraction coupling4,5. Soon after release, neurohormones pervade the second loop of fenestrated capillaries within the downstream pituitary gland, before binding to cognate receptors on endocrine cells to induce pituitary hormone exocytosis through a second ‘stimulus-secretion coupling’ event6-9. [Au: I think here might be a nice place to include a figure detailing these events for our less informed readers. A simple outline/schematic would be sufficient and that will set the scene for understanding the different axes later in the text]
In the second half of the twentieth century (and building on Harris’ work on the hypothalamus–pituitary axis [Au: please reference this statement with a reference to Harris’ work for our less informed readers]) the analogy between excitation-contraction and stimulus-secretion coupling was developed further [Au: edit OK?]10. However, important and fundamental differences exist between the two processes. Specifically, in the hypothalamus, endogenous [Au: circadian] rhythms exist and the time scale for pituitary hormone release is much longer (measured in minutes to several hours) [Au: edit OK? Please reference this statement]. Nevertheless, the analogy with neural control of locomotor activity led to a generally accepted model of hypothalamic regulation of pituitary function. Specifically, the excitation of specific hypothalamic neuron populations, determined by higher brain centres and peripheral feedback, is relayed as an unmodified series of signals to drive balanced pituitary hormone output [Au: please reference this statement]. The release of neurohormones and subsequent transportation and the effects on target cells were previously considered to be passive events in the regulation of pituitary hormone secretion, with only variation in the number of endocrine cells seeming to affect response levels11,12. Similarly, the alterations in gene expression and cell proliferation, which support maintenance of hormone output, were simply considered a correlated response to hypothalamic regulation of secretion [Au: please reference this statement].
In the early twentyfirst century, a series of paradigm shifts in our understanding of the hypothalamic–pituitary system was established as a consequence of newly developed tools and techniques, including calcium imaging, fluorescent protein identification of specific cell types and 2-photon microscopy, [Au: such as?] for use in genetically modified mice (e.g. xxx) [Au: please reference this statement with some examples]. The use of these methods have shown that both the pituitary gland and portal system can no longer be considered as static structures simply responding to neurohormonal regulation (BOX 1) [Au: edits OK to shorten sentence? I think the text box should be cited later in the text as more of a summary section]. In addition, hypothalamic neuron function has been found to be more dynamic than initially thought, which might, rather than changes in excitation, contribute to modifications in its regulation of the pituitary under different physiological states [Au: edit OK? Please reference this statement].
[H1] Gonadotroph axis
The reproductive system is critically dependent upon pulsatile secretion of gonadotrophin-releasing hormone (GnRH) and LH; however, the understanding of pulse generation has been hampered by the complexity of the regulatory mechanisms, many of which are lost in in vitro preparations [Au: please include the references here also as examples]. Investigators working in the late 1980s using pituitary portal bleeding and microdialysis documented the pulsatile nature of GnRH release into the portal vasculature of the sheep, monkey and rat25-31, and showed a strong correlation between GnRH and LH pulses28,32,33. However, the scattered distribution and relative paucity of GnRH cell bodies limited the investigation of the cellular events that lead to pulsatile secretion of LH in vivo. In the past few years, the development of optogenetic techniques in rats and mice and an ultra-sensitive ELISA capable of measuring LH levels in small whole blood samples34 has enabled investigators to dissect the GnRH neuron excitation parameters that generate LH pulses35. In these studies, the stimulation of just 60 GnRH neurons can trigger short-lived increases in LH secretion that resemble endogenous pulses [Au: Edits OK? please reference this statement]. Given the critical importance of GnRH neurons to the survival of all mammalian species, a degree of functional redundancy within this cell population is expected. Indeed, activating ~5% of the GnRH hypophysiotropic neurons seems to be sufficient to generate an LH pulse [Au: please reference this statement]. This finding is consistent with studies in which just 10% of the GnRH neuron population is sufficient to maintain pulsatile LH secretion36,37. When the timing and frequency of stimulation is varied a brief (2 min) optogenetic stimulation at high frequency (10 Hz) evokes an LH pulse, whereas shorter periods and lower frequencies cannot elicit LH output that resembles endogenous pulses35. [Au: edit OK?] This finding was also the case for a bursting pattern of stimulation [Au: meaning of LH secretion? How is a bursting pattern defined in this context?], which had been assumed to be effective for pulse generation and the focus of many previous studies38 [Au: meaning that this mechanism was the only one thought to be necessary of LH output?]. Whether such a stimulatory signal exists in situ [Au: meaning within the organism?] and where its origin might be is unknown, although a ‘GnRH pulse driver’ might be located in the mediobasal–hypothalamus, specifically at the level of neurons co-expressing kisspeptin, neurokinin B, as well as dynorphin A (so-called ‘KNDy’ neurons)39,40. [Au: Edits OK? perhaps you can expand on that specific evidence here for clarify?]
Pulsatile secretion of GnRH requires synchronization [Au: of the secretion and/or pulsatilty itself? What do you mean in this context?] within the GnRH neuron population. While the cell bodies of GnRH neurons are scattered throughout the basal forebrain, their projections have dendrodendritic bundling, that is they share synapses [Au: edit OK, is this what you mean?]41, and become highly concentrated around the ME [Au: please reference this statement]. Fascinatingly, these projections simultaneously receive and integrate synaptic inputs — they possess both axonal and dendritic characteristics, leading to their description as ‘dendrons’, before finally acquiring an axonal morphology within the ME and ramifying into numerous terminals that appose blood vessels42. [Au: I think this could be represented in the Figure also. It will then also give context to the vasculature aspect of the article] Dendrons might be an ideal location for putative afferent axons to modulate the excitability of multiple GnRH neuron dendrites, and for multiple GnRH neurons to align their firing pattern, which thereby provides a potential mechanism for their synchronized activity directly in the mediobasal hypothalamus [Au: is this your opinion?]. An additional source of pulse synchronization is in the ME, where hypophysiotropic GnRH neurons terminate within the external zone close to endothelial cells of the portal vasculature43. Endothelial cells in the ME might modulate GnRH release through nitric oxide secretion (which has been reviewed elsewhere44) [Au: edit OK? I would include some of the original reference here also to reduce the nu,ber of other reviews cited]. At the ME, nitric oxide is spontaneously released from an endothelial source and follows a pulsatile and cyclic pattern of secretion45, and inhibition of nitric oxide synthesis [Au: by local do you mean specifically in the ME or a wider area?] can disrupt reproductive cyclicity46. Conversely, in the GnRH neuron perikarya, basal nitric oxide synthase activity might provide the tonic inhibition of the GnRH neural system required to maintain nadir levels of LH [Au: edit OK?] 47. [Au: this mechanism to be represented in the figure also?]
Once released into the ME, the transport of GnRH to the pituitary, and the pattern of gonadotroph exposure to the neurohormone, have been largely assumed to represent a simple linear process [Au: please reference this statement]. However, the use of fluorescent tracking using 4 kDa dextran, which mimics the size of most hypothalamic neurohormones, has shown that the diffusion processes, both at the level of the ME and the pituitary capillaries, are complex and non-linear 7. [Au: edit OK?] Consequently, the portal vessel network might function as a ‘physical integrator’, enabling neurohormones to be transferred from the ME to the gonadotroph within a few seconds [Au: can you cite a reference here?]. Once in the blood stream, the moderately rapid clearance rate [Au: can you define this rate here?] of LH underlies [Au: meaning it generates it?] the specific asymmetric pulse shape of this hormone, which is characterized by a fast increase immediately followed by a slower decrease34. Importantly, a faithful delivery of the pulsatile pattern of GnRH secretion to the pituitary is crucial for gonadotroph function48 [Au: please include some of the original key references for this finding]. For example, high GnRH pulse frequencies (>1 pulse per h) activate LH production, whereas low frequencies (<1 pulse per 2–3 h) preferentially induce follicle-stimulating hormone (FSH) synthesis and release49. Overall, the intricate relationships between pulsatile GnRH release, secretory competency of the pituitary gonadotrophs and regulatory mechanisms within the vasculature, generate the rhythmic fluctuations in LH secretion.