| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratory of Cell Pharmacology, Department of Molecular Cell Biology (J.C., A.C., H.V.), and Laboratory of Experimental Hematology, Department of Pathophysiology (V.V.D.), University of Leuven (K.U.Leuven), B-3000 Leuven, Belgium
Address all correspondence and requests for reprints to: Hugo Vankelecom, Laboratory of Cell Pharmacology, Department of Molecular Cell Biology, University of Leuven (K.U.Leuven), Campus Gasthuisberg O&N1, Herestraat 49, B-3000 Leuven, Belgium. E-mail: Hugo.Vankelecom{at}med.kuleuven.be.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-secretase inhibitor DAPT down-regulated Notch activity and reduced the proportion of SP cells. Activation of Notch signaling with the conserved DSL motif of Notch ligands, or with a soluble ligand, caused a rise in SP cell number, at least in part due to a proliferative effect. The SP also expanded in proportion when aggregates were treated with leukemia-inhibitory factor, basic fibroblast growth factor, and epidermal growth factor, again at least partly accounted for by a mitogenic action. These intrapituitary growth factors all activated Notch signaling, and DAPT abrogated the expansion of the SP by basic fibroblast growth factor and leukemia-inhibitory factor, thus exposing a possible cross talk. In conclusion, we show that the Notch pathway, typically situated in embryogenesis, is also present and active in the postnatal pituitary, that it is particularly expressed within the SP independent of age, and that it plays a role in the regulation of SP abundance. Whether our data indicate that Notch regulates renewal and fate decisions of putative stem/progenitor cells within the pituitary SP as found in other tissues, remains open for further exploration. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The Notch pathway embodies a cell-cell signaling system that primarily governs cell-fate choices, and more in particular, controls the balance between commitment to differentiate and repression of differentiation while permitting, or even promoting, proliferation of the undifferentiated cells (reviewed in Refs. 16, 17, 18). Notch signaling occurs when the transmembrane Notch receptor—of which four members have been identified in vertebrates—interacts with one of its ligands (Delta-like and Jagged in mammals), presented from the membrane of adjacent cells. Ligand binding triggers a series of proteolytic cleavages, finally resulting in the liberation of the Notch intracellular domain (NICD) by a membrane-associated -secretase-dependent protease (19, 20). NICD translocates into the nucleus where it forms a complex with RBP-J
, thereby converting this transcriptional repressor into a transcriptional activator that triggers expression of a range of Notch-responsive genes. The Hes (hairy and enhancer of split) genes form a principal class of targets and effectors downstream of Notch activation and represent basic helix-loop-helix transcriptional repressors (21). Some of these Hes genes (in particular Hes1 and Hes5) repress the expression or antagonize the function of differentiation-inducing basic helix-loop-helix factors during, for instance, neurogenesis, myogenesis, and hematopoiesis (5, 7, 12, 15, 22). In contrast, Hes6 is typically associated with the promotion of differentiation (21, 23, 24, 25). Notch signaling finally routes the fate of the sending (ligand dominant) cell to a particular phenotype which is dissimilar from that of the receiving (Notch dominant) cell, by a process generally known as lateral inhibition (16, 17, 18).
In recent work, we discovered in the adult mouse anterior pituitary a novel population of cells on the basis of their competence to extrude the DNA-binding dye Hoechst 33342 (26). These Hoechstlow cells are depicted as a small side branch in dual-wavelength fluorescence-activated cell sorting (FACS) analysis, therefore referred to as side population (SP). The SP phenotype was originally described in bone marrow a decade ago (27). Since then, SP cells have been identified in many other tissues, including skeletal muscle, brain, pancreas, heart, liver, testis, prostate, mammary gland, intestine, lung, kidney, retina, and epidermis (Refs. 7, 28, 29, 30, 31 , and reviewed in Ref. 32). In many of these tissues, the SP represents a population enriched for authentic or prospective resident stem/progenitor cells (7, 27, 28, 29, 30, 31, 32), suggesting that the SP phenotype marks a universal property of stem cells. Along the same line, we have found that the SP of the adult anterior pituitary (
1.5% of the total cell population) concentrates cells that express several markers linked with stem/progenitor cells in other tissues, including stem cell antigen 1 (Sca1), whereas it is depleted from cells expressing phenotypic markers of differentiated pituitary cells (hormones) (26). Furthermore, the Sca1high SP accumulates cells that express some components of the Notch (Notch1 and Hes1), Wnt (Frizzled8) and Shh (Patched1) pathways (26), signaling systems included in pituitary embryonic development (1, 2, 3).
Expression of the Notch1 receptor and one of its key downstream effector molecules, Hes1, is suggestive of activated Notch signaling in the postnatal pituitary, and more in particular, in the SP compartment of the gland. To our knowledge, the Notch pathway has not yet been studied in the pituitary after birth. In the present study, we engaged in a more detailed evaluation of the Notch signaling system in the postnatal mouse anterior pituitary and addressed the questions of expression differences in SP vs. main population (MP), presence at different postnatal ages, and functional implication in the SP. In short, we found that the Notch pathway is significantly expressed in the postnatal pituitary, that it is predominantly present in the SP independent of age, and that it is involved in regulation of SP cell numbers.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
).
|
). Higher expression levels of Notch1 and Hes1 in the SP than in the MP (100- to 1000-fold; Fig. 1B
) confirmed our previous findings (26). In addition, we measured higher mRNA amounts of Notch2, Notch3 and Notch4 in the SP (10-, 5-, and 100- to 1000-fold above levels in the MP, respectively; Fig. 1, A and B). Of the direct transcriptional targets of the Notch signaling cascade, not only Hes1 but also Hes5 was more abundantly detected in the SP than the MP (between 10- and 50-fold higher mRNA levels; Fig. 1, A and B). In contrast, Hes6 mRNA was measured in a 10-fold higher amount in the MP. Interestingly, Hes6 is typically associated with inhibition of Hes1 activity (21, 23, 24, 25).
|
The Developmental Notch Pathway Remains Expressed in the Anterior Pituitary SP at Older Age
We previously identified a SP in the pituitary of adult (8 wk old) and immature (3 wk old) mice (26). Here, we extended this analysis to both younger (1 wk old) and older (16 wk old and 1 yr old) mice, and inspected Sca1 expression and the Notch pathway at different postnatal ages.
Also at the older ages tested, cells with SP phenotype were detected in the anterior pituitary (Fig. 2, A and B). SP proportion in the older mice did not differ from the one in adult mice (about 1.5% of the total cell population), whereas the SP fraction in immature animals was larger (2.4% at 3 wk, 5.4% at 1 wk of age; Fig. 2, A and B). We assessed Sca1 expression in the SP at three distant key time points of postnatal life, i.e. at immature (3 wk), adult (8 wk), and older (1 yr) age. At all three ages tested, the majority of the SP cells expressed Sca1 at a high level (between 52 and 70%; Fig. 2, C and D).
We then analyzed the Notch pathway by virtue of Notch1 and Hes1 expression at the same developmental stages (3, 8, and 52 wk of age). Sca1high SP as well as MP cells were collected by FACS and probed by real-time quantitative RT-PCR (qRT-PCR) for expression levels. Not unexpectedly, these members of the developmental Notch signaling pathway were also detected in immature mice. More remarkably, they remain expressed in the pituitary when mice become much older. Furthermore, levels of Notch1 and Hes1 mRNA were higher in the SP than in the MP at all three ages tested (Fig. 2E). When compared with adult mice, the SP/MP expression ratio was lower in immature mice (P < 0.001) and higher in 1-yr-old mice (P < 0.05) (Fig. 2E), mainly accounted for by higher and lower expression levels, respectively, in the MP (data not shown). Within the SP compartment, expression of Hes1 was higher in immature than in adult mice (Fig. 2F) but did not decline any further from adult to older age. It is noteworthy that Notch1 expression levels in the SP were not significantly different when immature, adult and older age groups were compared (Fig. 2F).
Together, these data indicate that the developmental Notch signaling pathway is expressed in the anterior pituitary throughout a very broad window of postnatal life, and that the segregation of expressing cells to the SP is independent of age.
Notch Signaling Is Active in the Anterior Pituitary SP and Affects SP Cell Number
Expression of receptors as well as target genes of the Notch pathway in the adult pituitary, and in particular in the SP, is indicative of active Notch signaling. To explore the functional implication of Notch in the SP compartment, cultures of three-dimensional reaggregates from adult mouse anterior pituitary cells were used and exposed to tools known to interfere with Notch signaling.
First, we characterized the SP proportion and Notch pathway gene expression in reaggregates after 10–14 d of culture. Although the SP size decreased in culture when compared with in vivo, a population with SP phenotype could clearly be discerned (0.25–0.35% of total cells; see Figs. 3–5). As for the cells ex vivo (26), the SP phenotype disappeared when verapamil, an inhibitor of the efflux mechanism, was added during incubation with Hoechst (data not shown). Also, Sca1 expression in the SP declined during culture, both with respect to proportion (25.2 ± 3.6% of the SP cells vs. 69.9 ± 1.6% in vivo; n = 3 and 6, respectively; P < 0.001) and intensity (10- to 50-fold difference in mean fluorescence as a parameter of average expression level in the SP). Notwithstanding these differences, all Notch receptors and target Hes genes detected in vivo remained expressed in the aggregates (Fig. 3, C and D, and data not shown). Moreover, differences in mRNA levels between SP and MP from the cultured cells were comparable to the ones observed in vivo as measured by semi-qRT-PCR. Notch1 and Hes1 were present at 100-fold higher mRNA levels in the SP than in the MP. Notch2, 3, and 4 and Hes5 were detected at 10-fold higher amounts in the SP, whereas Hes6 mRNA was 100-fold more abundant in the MP (all targets analyzed in two independent experiments, meaning RT-PCR performed one to two times on sortings from two independent cultures).
|
|
|
-secretase that blocks cleavage of the NICD from all Notch receptors (19, 20, 33), clearly reduced Notch target gene expression. Hes1 transcript level in the SP dropped to 10% or less of the level seen in nontreated cultures as analyzed by semi-qRT-PCR (Fig. 3C). In addition, we measured a 10-fold reduction of Hes5 mRNA within the SP (Fig. 3C). After treatment of the aggregates with DAPT for 10–14 d, the SP proportion was reduced by 40% (from 0.34 ± 0.08% to 0.21 ± 0.06%; mean ± SD, n = 6; P < 0.05; Fig. 3, A and E). DAPT was not toxic to the pituitary cells as revealed by propidium iodide analysis (data not shown). It is well known that the membrane-bound -secretase does not only cleave Notch receptor but also other substrates like amyloid precursor protein. Addition of imatinib (Gleevec, Novartis, Basel, Switzerland) to the cultures at a final concentration of 7.5 µM, shown to have a maximal inhibitory effect on cleavage of amyloid precursor protein by -secretase without affecting NICD release (34), did not influence Hes1 mRNA level, neither did it change SP proportion (0.38% vs. 0.35% in control cultures; n = 2).
To stimulate Notch signaling, a Delta/Serrate/Lag-2 (DSL) peptide, corresponding to the conserved domain of the Notch ligands and shown to be the minimal unit for binding and activation of Notch receptors (9), was added to the aggregate cultures. Treatment with the DSL peptide (at a final concentration of 0.1 µM; see Ref. 9) caused a 1.7-fold up-regulation of Hes1 mRNA levels in the SP as measured by qRT-PCR (Fig. 3D). In addition, the SP proportion rose in response to treatment with the DSL peptide by 62% from 0.26 ± 0.07% to 0.42 ± 0.11% (mean ± SD, n = 7; P < 0.05; Fig. 3E). Remarkably, this effect declined at a 10-fold higher concentration of the DSL peptide, causing only a 1.3-fold expansion (from 0.30 ± 0.08% to 0.39 ± 0.07%; mean ± SD, n = 3; P < 0.05 from control; P < 0.05 from 0.1 µM DSL), and even disappeared at a dose of 10 µM (1 experiment). Also, other authors have described declining effects when Notch activation levels increased (14). We further tested another Notch-activating tool, i.e. a Notch ligand in soluble form that can act as an agonist of the pathway (9, 35, 36, 37). Treatment of the aggregates with the soluble Jagged1 (Jg1) ligand (at 4 µg/ml) caused elevated Hes1 mRNA levels (83% above control; Fig. 3D) and also resulted in enlargement of the SP fraction (by 77% from 0.26 ± 0.07% to 0.46 ± 0.13%; mean ± SD, n = 7; P < 0.05; Fig. 3, B and E). The number of Sca1-expressing cells within the SP tended to increase after treatment with Jg1 (34.2 ± 8.1% vs. 25.2 ± 6.2% in control cultures; n = 3). However, this increase was statistically not significant (P = 0.205).
As a possible mechanism, we explored whether the SP-expansive effect of Notch activation was due to a mitogenic activity. First, we discovered that analysis by bromodeoxyuridine (BrdU) incorporation was not feasible because BrdU+ cells artifactually segregated into the SP, possibly due to disruption of Hoechst binding to the DNA in those cells because the dye preferentially binds adenine-thymine base pairs and BrdU is incorporated in the place of thymidine. Therefore, we analyzed mitogenesis in the SP using the nuclear cell proliferation marker Ki67. After culture, SP cells were sorted by FACS, spotted by cytospin, and immunostained for Ki67. The Jg1 ligand increased the number of Ki67+ cells within the SP compartment by 45% above control (Fig. 3F), indicating that cell proliferation within the SP contributes, at least in part, to the expansion of the SP fraction after activation of the Notch pathway.
Collectively, these data indicate that the Notch signaling pathway is active in the adult pituitary as assessed in aggregate culture, and more in particular, that it regulates SP cell number by affecting cell proliferation.
SP Proportion and Notch Signaling Increase in Response to Intrapituitary Growth Factors
To further define the SP-regulatory network within the adult anterior pituitary, we analyzed whether SP cells were responsive to growth factors that are produced within the pituitary, including basic FGF (bFGF) (38), epidermal growth factor (EGF) (39), leukemia-inhibitory factor (LIF) (40), and nerve growth factor (NGF) (41). It is noteworthy that, among other activities, LIF, bFGF, and EGF have in recent years been shown to stimulate growth of defined or putative stem/progenitor cells in a variety of tissues (11, 42, 43, 44, 45, 46).
Aggregate cultures of adult anterior pituitary cells were treated with LIF (10 ng/ml), bFGF (20 ng/ml), EGF (20 ng/ml) or NGF (20 ng/ml) for 10–14 d, and analyzed by FACS for SP proportion. As shown in Fig. 4, A and B, LIF and bFGF induced the SP to expand 2-fold (LIF, from 0.28 ± 0.05% of total cells to 0.56 ± 0.07%; mean ± SD, n = 6; P < 0.001; bFGF, from 0.34 ± 0.04% to 0.70 ± 0.10%, n = 9; P < 0.001), whereas EGF caused the SP proportion to triple (from 0.34 ± 0.02% to 0.98 ± 0.25%, n = 6; P < 0.001). In contrast, NGF did not affect the proportion of SP cells (0.37 ± 0.03% vs. 0.36 ± 0.04% in nontreated cultures, n = 3; P > 0.05; Fig. 4B). Identical outcomes were obtained in anterior pituitary cell aggregate cultures from rats (data not shown). LIF tended to enhance the number of Sca1-expressing SP cells when compared with control cultures (33.1 ± 9.3% vs. 25.2 ± 6.2%; n = 3). However, this increase was statistically not significant (P = 0.288).
We further explored a potential mitogenic basis for the expansive effect by scoring the Ki67+ cells in the SP fraction after culture. bFGF, LIF, and EGF raised the number of Ki67+ cells by 76, 91, and 183% above control, respectively (Fig. 4C). Therefore, expansion of the SP pool by these intrapituitary factors is, at least in part, due to a mitogenic activity.
Because both Notch receptor activation and growth factor signaling caused expansion of the SP, we examined a possible cross talk between these signaling systems. To study whether the growth factors affected Notch signaling in the SP, Hes1 mRNA levels were assessed by qRT-PCR as a parameter of activation of the Notch cascade. All three SP-expanding growth factors augmented Hes1 mRNA amounts in the SP (LIF: + 54%; bFGF: + 130%; EGF: + 300%; Fig. 5A). In contrast, NGF did not affect Hes1 mRNA levels in the SP (Fig. 5A). Concomitant addition of DAPT to the cultures abrogated the expansive effect on SP proportion as analyzed for LIF (data not shown) and bFGF (Fig. 5, B and C). Thus, Notch signaling seems to be involved in the growth-stimulatory effects that these intrapituitary factors exert on SP cells.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
During pituitary embryogenesis, several members of the Notch pathway have been detected in a spatially and temporally restricted manner (3). Hes1 was specifically found in the proliferative zone around the residual lumen (cleft) of the pituitary primordium, Rathke’s pouch, whereas Hes6 was only detected more ventrally in the developing anterior lobe. The peri-luminal zone is generally considered as the embryonic progenitor compartment from which cells originate and migrate to the developing anterior lobe where they differentiate (47, 48). Interestingly, our study similarly revealed a segregated expression pattern of Hes1 (more in the SP) and Hes6 (more in the MP of differentiated cells) in the postnatal gland. In other adult tissues, Hes1 expression has frequently been associated with a stem/progenitor cell phenotype (5, 7, 12, 15, 22), whereas Hes6 has more often been linked with the differentiation process by counteracting Hes1 activity (21, 23, 24, 25). It is tempting to interpret our findings within this context with Hes1 keeping (SP) cells in an undifferentiated state and Hes6 needed for differentiation to evolve (in the MP), roles also proposed by Raetzman et al. (3) in the embryonic gland. Moreover, our data suggest a similarity between progenitor cells in the embryonic pituitary and SP cells in the postnatal gland, further underscoring both the progenitor and embryonic character of the SP as proposed in our previous study (26). With respect to the Notch receptors, expression of Notch2 and Notch3 is also confined to the proliferative progenitor zone of Rathke’s pouch (3). Here, we also detected higher expression levels of Notch2 and Notch3 in the SP than in the MP. Expression of Notch1, however, was not observed by Raetzman and colleagues (3) in the developing pituitary at the early embryonic stages examined (3), whereas expression levels of Notch1 are high in the postnatal pituitary SP, revealing a potential difference between the embryonic (progenitor) cell and the SP cell phenotype.
We further showed that the Notch pathway is functionally active in the adult pituitary using reaggregate cell cultures, and that signaling is implicated in the regulation of SP abundance. In search of a possible mechanism, we found that activation of Notch signaling increased cell proliferation within the SP compartment. In addition to this mitogenic activity, Notch signaling may also favor maintenance of SP cells by virtue of a pro-survival activity as shown in other tissues (15), or by preventing cells from choosing other paths than the SP phenotype, given its role in cell fate determination (16, 17, 18). It has now been demonstrated in many tissues including the brain (5, 10, 15, 22, 37), the hematopoietic system (6, 7, 51), the mammary gland (9), and the skeletal muscle (4, 11) that signaling through the Notch cascade results in expansion of the stem/progenitor cell pool by preventing differentiation and/or by promoting self-renewal. A similar role may be postulated for the Notch pathway in the SP of the postnatal pituitary, further supporting our previous idea that SP cells, at least part of them, represent stem-like cells within the postnatal gland (26). Stem/progenitor cells have been frequently proposed to exist in the pituitary, and to be implicated in growth of the early postnatal gland (52, 53) and in homeostatic cell turnover in the mature gland (52, 54, 55, 56), as well as in regenerative processes after partial hypophysectomy—a restorative capacity that has not been met with unanimity (57, 58)—and after transgenic cell ablation (59). Nonetheless, bona fide stem/progenitor cells have so far not been revealed in the pituitary. More and more attempts, however, are being made to pin-point these cells (26, 60, 61, 62). SP cells, or at least some of them, displaying molecular, regulatory and functional characteristics attributed to stem/progenitor cells in other tissues (Ref. 26 , and this paper), are important candidates.
Our finding that the developmental Notch pathway operates in the adult pituitary may further add to the emerging picture that certain pieces of the embryonic developmental puzzle are either retained or reinstated in mature tissues to nourish genesis of cells during homeostatic turnover throughout life as well as during regeneration/repair (8, 11, 63). Thus, the embryonic phenotype of SP cells may result from a recapitulation of the embryonic developmental program during cell neogenesis within the adult gland, or alternatively, SP cells may embody residual embryonic cells that persist in the mature gland to feed cell turnover. The finding of a higher proportion of SP cells shortly after birth may be in favor of the latter hypothesis, together with our previous observation of Lhx4 expression in the SP (26), a LIM homeobox transcription factor that has been shown to be important for maintenance of the embryonic progenitor cells (48). No further alterations in SP proportion (as well as in Sca1 and Notch1/Hes1 expression) from adult to much older age refutes the possibility of the SP as a simple leftover population from embryonic development that then would gradually disappear during further life. In contrast, the SP is maintained at a remarkably constant size for large part of life once adult, which clearly requires active processes, likely including Notch signaling. By definition, stem/progenitor cells are expected to be present and available for cell replacement during most part of life.
The higher SP abundance at early postnatal age may also reflect elevated activity during the process of vivid molding and maturation of the pituitary cell composition during this period. Although all hormonal cell types are formed during embryonic development (64), it is well established that the majority of the hormone-producing pituitary cells only develop the first weeks after birth (52, 53, 56). Knowledge about the mechanisms underlying these phases of development and growth are limited (56). The SP may be involved as a source population, explaining its higher abundance at this stage. Along the same line, a more active Notch signaling cascade at immature age, as epitomized by higher Hes1 expression levels, is in favor of elevated activity within the SP at this age.
A further appealing finding of our study is the responsiveness of SP cells, or at least part of them, to growth factors produced within the pituitary. SP proportion in adult anterior pituitary cell aggregates increased in response to LIF, bFGF, and EGF (but not NGF), at least partly due to a proliferative activity. These factors have in the pituitary mainly been characterized to affect differentiation, proliferation, and hormone production of the endocrine cells (39, 40, 65, 66). Here, we identify a novel action in the adult pituitary by showing that they affect the nonhormonal SP. With respect to a possible link between pituitary embryonic progenitor cells and postnatal SP cells, it is worth noting that in Rathke’s pouch explants, bFGF exerts a negative control on the generation of differentiating cells and maintains the cells in a proliferative progenitor state (65, 67). Along the same line, more and more evidence is presented that bFGF, LIF, and EGF act as potent stimulators of stem/progenitor cell growth in a variety of embryonic as well as postnatal tissues (11, 42, 43, 44, 45, 46). Responsiveness of the pituitary SP to these factors further supports our hypothesis that the SP contains stem/progenitor-like cells.
Remarkably, the SP-expansive effects of LIF, bFGF, and EGF seem to involve Notch signaling. Blocking the Notch cascade with a -secretase inhibitor abrogated the increase in SP proportion, suggesting that Notch is either an essential downstream signaler of growth factor activity in the SP, or a crucial interacting pathway. A growing number of studies demonstrate cooperative cross talk between the growth factors analyzed here and Notch, particularly with respect to stem/progenitor cell maintenance and proliferation (14, 15, 36, 63, 68, 69, 70). Taken together, our findings support the presence of a SP-regulatory network within the adult pituitary, comprising Notch and the paracrine growth factors bFGF, EGF, and LIF, and possible cross talk between both signaling systems.
We observed that SP cells persist in pituitary cell reaggregate cultures during at least 2 wk, but that their number as well as their Sca1 expression declines, which is suggestive of a deficiency in trophic influences in the cultures. Despite the demonstrated organotypic behavior and topographical organization of anterior pituitary cells in the aggregates (71, 72, 73), it is not unthinkable that not all structural and microenvironmental cues are reformed given the intricate cytoarchitecture of the gland (74). More in particular, with regard to the proposed stem/progenitor cell phenotype of the SP, it is not surprising that a niche with its highly regulated interactions to support stem cells (6, 8, 10, 13, 37, 75, 76), is not completely reestablished in the aggregates. In addition, pituitary-extrinsic trophic factors may be missing in the aggregate cultures. A final remark here is that we treat the SP in the present study as a cellular and functional unit. However, as shown in other tissues (11, 28, 30, 77, 78) and also in our previous characterization study (26), the SP represents a population that is not homogeneous. Thus, the pituitary SP may not only contain pituitary-phenotypic cells but also other (pituitary-nonspecific) cell types, some of which may not survive in the aggregate culture conditions used. Despite these apparent shortcomings of the in vitro aggregate culture system with respect to SP behavior, findings from Notch signal manipulation are very likely to be relevant because Notch expression profiles after culture (with respect to the members expressed and the differential expression in SP vs. MP) remain similar to the ones observed in vivo.
Certainly, it would be very valuable to know whether Notch signaling also affects SP proportion in the postnatal pituitary in vivo. However, transgenic studies so far were unable to answer this question (Vankelecom, H., A. Crabbe, and J. Chen, unpublished observations). Transgenic activation of Notch signaling by Cre-mediated NICD expression (using RosaNotch mice; Ref. 79) in nestin- (80) and 
-glycoprotein hormone subunit-expressing cells (81) resulted in perinatal and embryonic death, respectively, of the bitransgenic pups. Lethality is probably due to widespread and/or ectopic activation of the Notch system because of transcription from the nestin (80) or the -glycoprotein hormone subunit promoter (also activated in the heart, e.g. Ref. 81). To investigate the role of Notch in the postnatal pituitary, a more sophisticated system is required to induce Notch activation in a spatiotemporal manner, i.e. specifically in the pituitary and only after birth.
In conclusion, our data provide the first indications that Notch signaling is present and functional in the postnatal pituitary. In particular, the Notch pathway is prominently expressed in the SP subset and signaling plays a role in the regulation of SP abundance, partly as a result of a mitogenic drive. In several other adult tissues, the assembled qualities of SP phenotype, expression of Sca1 (as well as of other markers; see Ref. 26), and regulation by Notch (and by LIF, bFGF, and EGF) are recurrently associated with a stem/progenitor cell phenotype. Whether SP cells represent stem/progenitor cells in the postnatal pituitary, potentially involved in the gland’s well-known but not well-understood plasticity (56), is a question of great interest that poses a formidable challenge for further studies. Despite this pressing question, it becomes more and more clear that the newly discovered SP represents a biologically significant population in the postnatal pituitary, as substantiated by its presence at different key postnatal ages, its responsiveness to regulatory factors such as Notch and intrapituitary growth factors (and their interplay), and its occurrence in evolutionarily distant rodents and birds (26).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Culture of Anterior Pituitary Cells as Three-Dimensional Aggregates
Freshly dispersed anterior pituitary cells from adult (8 wk old) female mice were seeded in 35-mm nontreated polystyrene dishes at 2 x 106 cells per 2.5 ml of either serum-free chemically defined culture medium (49, 71) or advanced DMEM-F12 (Invitrogen), containing Albumax and supplemented with Glutamax and antibiotics. Culture dishes were incubated on a gyratory shaker (64 rpm) in a water-saturated, 1.9% or 6.0% CO2 incubator at 37 C. Under these conditions, anterior pituitary cells reassociate to form three-dimensional aggregates. This aggregate culture system has been thoroughly characterized in our previous work and found very suitable to study organotypic cell behavior with respect to function (hormone secretion and regulation) and development (proliferation and differentiation) (49, 71, 72, 73).
Aggregates were treated during the whole culture period of 10–14 d with one of the following growth factors: LIF (recombinant mouse LIF; Chemicon, Temecula, CA), bFGF (recombinant human bFGF; R&D Systems, Minneapolis, MN), epidermal growth factor (recombinant human EGF; R&D Systems) or NGF (mouse NGF-2.5S from submaxillary glands; Sigma-Aldrich, St. Louis, MO). In other experiments, cultures were treated from d 2 of culture (after re-aggregation of the cells) with the -secretase inhibitor DAPT (Calbiochem, Darmstadt, Germany) (33), the Delta/Serrate/Lag-2 Notch-binding domain (DSL peptide; synthesized by Pepscan Systems, AB Lelystad, The Netherlands) (9), a soluble Notch receptor ligand (Jagged1 as a Fc fusion chimera; R&D Systems; used at concentrations recommended by the manufacturer) (35, 37), or imatinib (Gleevec) (34). Culture medium was renewed every 2–3 d. After culture, aggregates were redispersed into single cells by trypsinization as previously described (73), and cells were subjected to further analysis.
Flow-Cytometric Analysis of the SP and of Sca1 Expression
Analysis of the SP phenotype was done essentially as described before (26). In short, cells from dispersed anterior lobes or aggregates were incubated with Hoechst 33342 dye (bisbenzimide at 2.5 µg/ml; Sigma-Aldrich) in chemically defined culture medium for 90 min at 37 C. Then, cells were immunostained for Sca1 expression using phycoerythrin-labeled rat antimouse Sca1 (BD Biosciences, Erembodegem, Belgium). Finally, cells were resuspended in ice-cold, Ca2+/Mg2+-free PBS (pH 7.4; Invitrogen) with fetal calf serum (2%; Invitrogen) and propidium iodide (2 µg/ml; Sigma-Aldrich) for examination on a FACSVantage (BD Biosciences). SP cells efficiently extrude the Hoechst dye and are identified as a streak of Hoechstlow cells in a dual-wavelength (blue and red fluorescence) emission plot after UV excitation. The bulk of Hoechsthigh cells represents the MP. Only cells within the living (PIneg) cell population were analyzed or sorted. To control for the active efflux of Hoechst and confirm the SP phenotype, verapamil (50–100 µM; Sigma-Aldrich) was added together with Hoechst, resulting in the disappearance of the SP (see Ref. 26).
Cell Proliferation Analysis by Ki67 Immunodetection
SP cells from dissociated anterior pituitary cell aggregates were sorted by FACS into chemically defined culture medium, and deposited by cytospin onto Polysine glass slides (Menzel-Gläser, Braunschweig, Germany). Cells were rinsed with PBS and fixed with 4% paraformaldehyde for 10 min at room temperature. After another rinse with PBS, cells were immersed into citrate buffer (pH 6.0; 10 mM) and antigens retrieved by boiling in the microwave (550 W; 20 min). Samples were slowly cooled down to room temperature, and cells permeabilized with Triton X-100 (0.4% in PBS; Sigma-Aldrich) for 10 min, blocked with normal goat serum (10% in PBS with 0.1% Triton X-100 or PBS-Tx) for 20 min, incubated overnight at room temperature with rabbit anti-Ki67 antiserum (1:50 in PBS-Tx; Abcam, Cambridge, UK), and for 1.5 h with AlexaFluor 488-labeled goat antirabbit antibody (1:1000 in PBS-Tx; Molecular Probes, Eugene, OR), and finally covered with Vectashield (Vector Laboratories, Burlingame, CA). As a negative control, the primary antibody was replaced by normal rabbit serum at a comparable dilution. No staining signals were observed in these controls (data not shown). Pictures of 10 random fields of each culture condition per independent experiment were taken at a 400-fold magnification using a Nikon Digital Camera DXM 1200 (Analis, Namen, Belgium), mounted on an epifluorescence microscope (DMRB; Leica, Wetzlar, Germany). Pictures were imported from Nikon ACT-1 image capturing software into Adobe Photoshop 6.0 (Adobe Systems, San Jose, CA) to produce overlays of immunofluorescence and light-microscopic images. The number of Ki67-immunoreactive (Ki67+) cells on total SP cells was counted with the help of ImageJ analysis tools (http://rsb.info.nih.gov/ij/; version 1.33 C) in all the pictures taken. Typically, a total of 150–300 SP cells were counted per culture condition per independent experiment.
Gene-Expression Analysis by RT-PCR
Reverse Transcription (RT).
RT was performed essentially as described before (26). In brief, (Sca1high) SP cells and MP cells within the living (PIneg) population were sorted by FACS. Total RNA was extracted using TriPure isolation reagent (Roche Diagnostics, Vilvoorde, Belgium) and quantified with RiboGreen Quantitation Reagent (Molecular Probes). RT was performed on specified amounts of RNA (2 µl of RNA dilution in a final volume of 20 µl RT mixture) in the presence of random hexamers (Applied Biosystems, Foster City, CA) and Moloney murine leukemia virus reverse transcriptase (Invitrogen), using a temperature program of 10 min at 23 C, 50 min at 42 C, and 10 min at 95 C. As a positive control for the RT reaction, 10 ng of mouse anterior pituitary RNA was used, and as a negative control, ribonuclease-free H2O replaced the RNA sample.
PCR.
PCR was performed on 1 µl of the RT end mixture in a final volume of 10 µl, using AmpliTaq Gold DNA polymerase (Applied Biosystems) and gene-specific oligonucleotide primers. Primers were designed with Vector NTI (InforMax; Invitrogen) to span at least one intron to distinguish between amplification from genomic DNA and from cDNA, and synthesized by Invitrogen. Primer sequences are available upon request. PCR temperature cycling was optimized using total RNA from mouse anterior pituitary and consisted of the following steps: 7 min at 95 C, followed by 40–45 cycles of denaturation (10–15 sec) at 95 C, annealing (15–20 sec) at 59–64 C, and extension (25–45 sec) at 72 C, and a final incubation for 7 min at 72 C before cooling down to 15 C. The amplicons obtained from anterior pituitary-derived RNA were purified, cloned into the pCR 2.1-TOPO vector using the TOPO-TA Cloning System (Invitrogen), and sequenced by Lark Technologies (Essex, UK) to verify that the primers amplified the intended gene’s cDNA. This was achieved without exception.
The PCR products were size-fractionated by 2% agarose gel electrophoresis and detected by staining with ethidium bromide (0.5 µg/ml). Pictures were taken using a charge-coupled device camera and edited with Microsoft (Redmond, WA) Paint 5.1.
Semi-qRT-PCR.
To compare expression levels in SP and MP, a semi-qRT-PCR method was applied as described before in detail (26). The reliability of this method has been confirmed by real-time qRT-PCR (26). Briefly, serially diluted samples (10- or 5-fold, typically from 10 ng/µl if available) of SP and MP RNA were subjected to RT-PCR. Expression of L19, a constitutive ribosomal protein in all cells, was used as internal standard to normalize the RNA dilution series of the SP and MP. SP/MP expression ratios were calculated from the detection limits, defined as the lowest concentration of RNA showing an amplification signal (26). To minimize influences of extrinsic factors, RT of the SP and MP dilution series to be compared (i.e. from the same cell sorting) were simultaneously subjected to RT using the same RT mixture, and per marker to PCR using the same PCR mixture (and in the same GeneAmp PCR System, model 2400 or 9700; Applied Biosystems). The positive PCR control consisted of cDNA that was reverse transcribed from 10 ng RNA isolated from mouse anterior pituitary. As a negative control, 1 µl ribonuclease-free H2O was used instead of cDNA sample. Analysis of each target was performed one or two times on RNA dilution series from each of three or four independent cell sortings. It should be noted that expression levels cannot be compared between different targets because PCR amplification efficiency is not necessarily equal for the different targets.
qRT-PCR.
qRT-PCR was performed for Hes1 and Notch1 expression, essentially as described before (26, 50). Briefly, an appropriate amount of RNA was subjected to RT and further analyzed by qPCR on an ABI PRISM 7700 Sequence Detector (Applied Biosystems) using TaqMan Universal PCR Mix (Applied Biosystems) and the gene-specific primers and probes included in the TaqMan Gene Expression Assays (purchased from Applied Biosystems) for mouse Notch1 (Assay ID Mm00435245_m1) and mouse Hes1 (Assay ID Mm00468601_m1). Each sample was measured in triplicate, and data were analyzed according to the 
Ct (threshold cycle)/standard curve method (26, 50). 18S rRNA was used as internal (normalization) standard.
Statistical Analysis
Statistical significance was assessed by one-way ANOVA followed by the Tukey-Kramer multiple-comparison test (Number Cruncher Statistical System 2000 statistical software; Statistical Solutions, Saugus, MA). A probability of less than 5% (P < 0.05) was considered statistically significant.
|
FOOTNOTES |
|---|
Disclosure Summary: The authors have nothing to disclose.
First Published Online September 7, 2006
Abbreviations: bFGF, Basic FGF; BrdU, bromodeoxyuridine; DAPT, N-[N-(3,5-difluorophenacetyl)-L-alanyl]-(S)-phenylglycine-t-butyl ester; DSL, Delta/Serrate/Lag-2; EGF, epidermal growth factor; FACS, fluorescence-activated cell sorting; FGF, fibroblast growth factor; Jg1, Jagged1; Hes, hairy and enhancer of split; LIF, leukemia-inhibitory factor; MP, main population; NICD, Notch intracellular domain; NGF, nerve growth factor; RT, reverse transcription; qRT-PCR; quantitative (real-time) RT-PCR; Sca1, stem cell antigen 1; semi-q, semiquantitative; Shh, sonic hedgehog; SP, side population.
Received for publication July 18, 2006. Accepted for publication August 31, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
