Renewing an old interest: Pituitary folliculostellate cells

Anterior pituitary folliculostellate (FS) cells, first described almost 50 years ago, have a wide range of functions with respect to supporting and coordinating endocrine cell function, in particular through paracrine and gap junction‐mediated signalling. Our previous studies identified the morphological organisation of FS cells, which mediates coordinated calcium activity throughout the homotypic FS network and allows signalling across the whole pituitary gland. It is also clear that FS cells can modify endocrine output and feedback on pituitary axes over a range of timescales. Recently, several studies have defined FS cells as a source of anterior pituitary endocrine cell renewal, which has resulted in a renaming of FS cells as “Sox2+ve stem cells”. Here, we highlight the broader potential of the FS cell population in fine‐tuning and coordinating pituitary axes function. In addition, we identify a need for: the definition of the possible subtypes of FS cell and their relationship with the stem cell population; the potential role of FS cells in pulsatile hormone secretion and coordination of heterotypic cell networks; and the roles that FS cells may play in both early‐life programming of pituitary axes and in memory, or anticipation, of demand. Further studies of FS cells may demonstrate the fundamental importance of this cell type and its potential as a therapeutic target to correct pituitary gland dysfunction, one of which is stem cell therapy. Clearly, a thorough understanding of all of these interactions and relationships of FS and endocrine cells is required whatever therapeutic use is suggested by their various roles.


| INTRODUC TI ON
Increasingly, neuroendocrine research focuses on its potential for an impact of novel findings on highly-prevalent human health problems, in particular those where the underlying physiology and its dysfunction is poorly understood. Research on the pituitary, the master gland relaying hormonal information between the brain and many peripheral tissues, cannot escape this trend, as exemplified by the growing hope of regenerative medicine with the recent discovery of adult stem cells in the pituitary parenchyma. One exciting aspect of stem cell therapy is its potential to correct a number of pituitary disorders with aetiologies that are poorly understood. This focus on stem cells and their potential may explain how a subpopulation of pituitary cells, devoid of secretory granules, has changed from being described as folliculostellate (FS) cells (coined by Evelyne Vila-Porcile in 1972 1 ) to adult Sox2+ve stem cells. Recent studies have highlighted an important aspect of these stem cells, in that they are the source of secreted factors regulating not only their own function, but also those of other pituitary cells. 2,3 This aspect of the biology of pituitary stem cells resonates with roles described in previous studies of FS cells 4 and has prompted us to reassess the biology of this enigmatic cell type and its importance for pituitary gland physiology.
Because the stem cell potential of FS cells has been extensively reviewed recently, [5][6][7] we focus here on other important aspects of FS cell biology in the regulation of anterior pituitary gland function.

| FS CELL S AND THEIR CELL IDENTIT Y
FS cells were first described in the pioneering electron microscopy studies of the pituitary gland by Farquhar and Rinehart 8 as small agranular cells with long, slender processes ( Figure 1A). Although early work identified two types of cell, 9,10 these were unified as a single cell type by Vila-Porcile, who also described their organisation into a network in the rat. 1 Subsequently, they have been described in the anterior pituitaries in a range of mammalian and non-mammalian species, 11 as well as in anterior pituitaries with morphologically distinct organisation, such as teleosts. 12 Classically, they have been studied in animals such as the rat, where their expression of S100, a family of Ca 2+ -binding proteins transducing Ca 2+ signals, 13 enables identification based on gene expression as well as morphology. 14 This expression of S100 has subsequently been exploited in the generation of green flourescent protein (GFP) transgenic rats, allowing the isolation and ready identification of these cells. 15 FS cells can also be identified in ex vivo culture by their uptake of an alanine-lysine dipeptide conjugated to aminomethylcoumarin acetate (AMCA) fluorophore, which is dependent on FS cell expression of a protein peptide symporter. 16 Intriguingly, this uptake of dipeptide-conjugated AMCA was shown to be a feature of posterior pituitary pituicyctes, 17 which also express some of the key proteins related to FS cell function, suggesting that the distinct cell types of the two pituitary lobes may share common features and roles. More recently, specific pituitary expression of aldolase C in FS cells has been described in mice, which may allow increased genetic manipulation in this model species. 18 Dispersed FS cells in culture readily self-organise into aggregates of cells that resemble those of the intact pituitary and, in mixed cultures with other pituitary cell types, form clusters with hormone producing cells. 19 A role for the chemoattractant molecule CXCL12 and its receptor CXCR4, both expressed by FS cells, has been described in this in vitro recapitulation of cell organisation. 20 Homotypic FS cell interaction are then likely maintained by expression of adherence proteins such as E-cadherin and a differential expression of other adherence molecules likely mediates specific FS cell-heterotypic cell morphological relationships. 21 This suggests a role for FS cells in the organisation of the homotypic networks of all pituitary hormonal cell types studied to date, with important functional consequences. 22 An important interaction of FS cells with the extracellular matrix (ECM) has also been described, with matrix metalloprotease 9 mediating cell organisation, 23 integrin ß1 signalling regulating FS cell proliferation 24 and FS cell production of tissue inhibitors of metalloproteinases in turn regulating the ECM. 25

| ROLE A S A REL AYER OF PERIPHER AL FEEDBACK S IG NAL S?
Perhaps the most intriguing potential roles for FS cells, in terms of potential coordinators of endocrine function as well as stem cells, is their response to pituitary target organ feedback and mediators.
This would place them as key modifiers of pituitary function that fine-tune and adapt the various endocrine systems to ensure an optimal response to physiology, rather than as supportive cells as has been suggested previously. 35 For example, in the prolactin axis, secretion in response to corticotrophin-releasing hormone. 47,48 The regulation of ANXA1 activity may suggest one role for spatially localised calcium waves that we have described propagating through subsets of the FS cell network 26,27 : the structure of ANXA1 has been shown to be calcium sensitive, with increased biological activity in the presence of Ca 2+ . 49 The importance of ANXA1 in corticotroph regulation is also highlighted by a four-fold increase in corticotroph cell number in male ANXA1-knockout mice compared to wild-type controls, although, interestingly, this effect is much less pronounced or even absent in females. 50 Finally, FS cells may be programmed by early-life exposure to glucocorticoids because ANXA-1 expression is reduced in adult mice following prenatal dexamethasone exposure. 51

| PER S PEC TIVE S
If we combine the biology described above with their more recent We then speculate on the fundamental role that FS cells may have in fine-tuning physiology ranging from reproduction to metabolism and stress. Each of these may be defined through dual recording of FS and endocrine cell activity in ex vivo slices, as well as targeting specific molecules and pathways and determining in vivo consequences. This will require cell-type specific markers and targets, although the increasing availability of transcriptome data that includes FS cells suggests that these may be identified in the near future.

| Heterogeneity
FS cells were originally identified as two distinct cell types, each agranular and characterised by long cytoplasmic projections, but with distinct morphological arrangements based on whether they surround cavities filled with colloidal milieu. 9,10 The morphological heterogeneity is further emphasised by differential protein expression and this has led many to question whether FS cells are a single cell type, both in developmental origin and in function.

| Pulse generation
Perifusion studies of isolated pituitaries have shown that hormone output is spontaneously pulsatile in the absence of hypothalamic input. 61,62 Our descriptions of pituitary hormone networks 28 have provided, in part, a mechanism where these spontaneous pulses could occur. However, it was the FS cell network that pioneered the description of the endocrine networks: the finding of spontaneous pulsatile activity in these cells, and in particular the identification of a proportion that apparently acts as a pacemaker, 26 suggests that they may have a role. Indeed, the paracrine and gap junction-mediated communication between FS and endocrine cells would provide a mechanism allowing the coordination of pulsatile release over different timescales.
The interaction between endocrine and FS cells is, however, bidirectional, especially if peripheral feedback is considered, and the identification of which cell type is acting as a pulse generator is therefore complex

CO N FLI C T O F I NTE R E S T S
The author declares that they have no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Paul Le Tissier: Conceptualisation; Writing -original draft. Patrice Mollard: Conceptualisation; Writing -original draft.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/jne.13053.

DATA AVA I L A B I L I T Y S TAT E M E N T
Not required.