Nestin-Immunoreactive Cells in Rat Pituitary Are neither Hormonal nor Typical Folliculo-Stellate Cells
http://www.100md.com
内分泌学杂志 2005年第5期
Laboratory of Cell Pharmacology, Department of Molecular Cell Biology, University of Leuven School of Medicine, Campus Gasthuisberg, B-3000 Leuven, Belgium
Address all correspondence and requests for reprints to: Dr. Hugo Vankelecom, Laboratory of Cell Pharmacology, Department of Molecular Cell Biology, University of Leuven, Campus Gasthuisberg (O&N), Herestraat 49, B-3000 Leuven, Belgium. E-mail: hugo.vankelecom@med.kuleuven.ac.be.
Abstract
Nestin is an intermediate filament protein that has originally been identified as a marker of neuroepithelial stem/progenitor cells. The present study explored whether nestin immunoreactivity (nestin-ir) is present in the rat pituitary and in which cell type(s). Nestin-ir was observed in scattered cells in the anterior, intermediate, and neural lobes. Nestin-ir cells were predominantly of stellate shape and were more numerous in immature than in adult animals. Nestin-ir did not colocalize with any pituitary hormone, and did not colocalize or only very sporadically with the folliculo-stellate cell marker S100. In the intermediate lobe, nestin-ir cells contained glial fibrillary acidic protein in an age-dependent manner. Nestin-ir cells were closely associated with endothelial and fibronectin-ir cells, but did mostly not coincide. Nestin-ir was not found in -smooth muscle actin-ir myofibroblasts or in microglial cells. Regardless of age, nestin-ir was detected in some unidentifiable cells that border the pituitary cleft. Nestin-ir remained present in pituitary cultured as three-dimensional aggregates. Treatment with basic fibroblast growth factor or leukemia inhibitory factor increased the number of nestin-ir cells. Starting from anterior lobe cell monolayer cultures, nestin-ir cells could be selected and propagated to a virtually pure population. These nestin-ir cells displayed remarkable motility and proliferative activity, and did not express hormones, glial fibrillary acidic protein, or S100, but contained vimentin-, fibronectin-, and -smooth muscle actin-ir. In conclusion, nestin-ir is present in the pituitary in cells that are neither hormonal nor typical folliculo-stellate. The expression pattern depends on age and lobe examined. Pericapillar localization suggests a pericyte phenotype for some of them. Whether the heterogeneous nestin-ir population also contains pituitary progenitor cells remains to be explored.
Introduction
THE MATURE PITUITARY is an endocrine organ that dynamically changes its hormone output in response to a variety of physiological and pathological triggers to meet the endocrine needs of the organism. Several of these conditions are associated with rapid changes in the number of certain cell types. Although cell proliferation is believed to occur within the various cell type populations, it remains unknown whether the expansion of a certain population may also result from cell development and differentiation from stem cells or progenitor cells. It also is not known whether steady cell turnover in the gland, estimated to be approximately 1.5%/d in the young adult male rat (1), is driven from a stem/progenitor cell pool.
Nestin is a class VI intermediate filament protein that has first been identified in the rat embryonic central nervous system, where it is abundantly expressed (2, 3). During postnatal development, nestin expression in the brain becomes more restricted and is particularly found in neuroepithelial stem and proliferative progenitor cells present in the (sub)ventricular zone (3, 4, 5, 6). Nestin has also been detected in putative stem/progenitor cells of a number of other tissues, such as dermis (7), pancreas (8, 9, 10), retina (11), dental tissue (12), and hair follicle (13).
Although nestin expression has attracted much attention as a marker in stem/progenitor cells, it is also present in other cell types, such as in newly developing vascular endothelial cells (14, 15, 16, 17, 18, 19, 20) and in activated stellate cells of the brain, pancreas, and liver (15, 21, 22, 23). In the adult human gastrointestinal tract, nestin expression has been found in CD34-positive cells, enteroglial cells, and interstitial cells of Cajal (24). Also Leydig and Sertoli cells have been reported to contain nestin (25). Furthermore, a number of findings do not support nestin as an in situ stem/progenitor cell marker in the pancreas (15, 16, 17, 18, 19, 26, 27). Rather, pancreatic nestin-positive cells are proposed to act as niche cells that induce or support endocrine cell neogenesis (17, 20). Thus, nestin-expressing cells may also behave as supporting cells in tissue homeostasis.
Nestin-positive stem/progenitor cells appear to behave peculiarly in culture. Neural stem/progenitor cells can be propagated in suspension cultures as floating colonies (neurospheres) when treated with basic fibroblast growth factor (bFGF) and/or epidermal growth factor (EGF) (28, 29, 30). Virtually all of the cells within a neurosphere express nestin. Nestin-positive putative precursor cells have also been obtained and multiplied in monolayer cultures. A monolayer of cells expressing nestin grows out from pancreatic islets of Langerhans [nestin-positive islet-derived progenitor cells (NIP)] (8). NIP cells are able to differentiate in vitro into multiple liver and pancreatic cell phenotypes.
To date, stem/progenitor cells in the postnatal pituitary gland have not been identified. Pinpointing these cells may be important to better unravel lineage relationships and cell type renewal and maintenance in the postnatal pituitary. Inspired by the expression profiles of nestin and its potential relationship to the stem/progenitor cell phenotype, the present investigation was intended to examine whether nestin is present in immature and adult pituitary glands, whether its presence is restricted to a certain cell type(s), and how nestin-ir cells behave in culture.
Materials and Methods
Immunofluorescence of pituitary Vibratome sections and analysis by confocal laser scanning microscopy
Female Wistar rats (5 and 14 d old or 8 wk old, random cycling) were purchased from Elevage Janvier (Schaijk, The Netherlands). Rats were killed by CO2 asphyxiation and decapitation. All animal experiments were approved by the university ethical committee and conducted in accordance with The Endocrine Society ethical guidelines. Pituitaries were carefully isolated under the stereomicroscope and immediately fixed for 3 h with 4% paraformaldehyde (Riedel-deHa?n, Seelze, Germany) in PBS (Invitrogen Life Technologies, Inc., Paisley, UK) or for 48 h with IHC Zinc Fixative (BD Biosciences, Erembodegem, Belgium) as needed for CD31/platelet endothelial cell adhesion molecule-1, also called CD31 (PECAM-1) immunostaining. Sometimes rats were perfused with PBS and 4% paraformaldehyde before separation of the gland. Pituitaries were embedded in agarose (ICN Biomedicals, Aurora, OH; 2% in PBS) before sectioning with the Vibratome (Leica VT 1000 S, Leica Microsystems, Wetzlar, Germany). Pituitaries were attentively positioned to cut 50-μm sections through all lobes and cleft. Sections were stored in PBS at 4 C until additional processing by immunofluorescence.
Each staining was performed on duplicate Vibratome sections at different, evenly distributed locations of the pituitary gland. Permeabilization of the sections was achieved with two 30-min incubations in saponin (Sigma-Aldrich Corp., St. Louis, MO; 0.5% in PBS; PBS-Sap). Aspecific binding sites were preadsorbed with 20% normal goat or normal donkey serum (DakoCytomation, Glostrup, Denmark). Sections were incubated overnight at room temperature with one or two of the following primary antibodies: mouse antirat nestin IgG1 (Rat-401, BD Biosciences, Mountain View, CA) (2) at a final dilution of 1:100 to 1:200 in PBS-Sap; rabbit anti-nestin (no. 130, provided by Dr. R. McKay, NIH, National Institute for Neurological Disorders and Stroke, Bethesda, MD; recognizes rat, mouse, and human nestin) at 1:2,000 dilution; mouse antirat PECAM-1/CD31 (BD Biosciences) at 1:20 dilution; mouse antirat Ox-42 (CD11b, Serotec, Oxford, UK) at 1:100 dilution; mouse antihuman -smooth muscle actin (SMA; DakoCytomation) at 1:100 dilution; rabbit antirat GH (UCB Bioproducts, Brussels, Belgium) at 1:2,500 dilution; rabbit anticow S100 at 1:100 dilution (from DakoCytomation or Zymed Laboratories, San Francisco, CA), anticow glial fibrillary acidic protein (GFAP) at 1:10,000 dilution (DakoCytomation); rabbit antirat vWF at 1:100 dilution (Chemicon Biognost, Heule, Belgium); mouse anti-pan cadherin at 1:200 dilution (Sigma-Aldrich Corp., Bornem, Belgium); goat antihuman vimentin at 1:20 dilution (Sigma-Aldrich Corp.); rabbit antihuman fibronectin at 1:200 dilution (Sigma-Aldrich Corp.); goat antihuman doublecortin (DCX; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at 1:400 dilution; and rabbit antihuman ACTH at 1:7,500 dilution, antirat prolactin (anti-rPRL-IC5) at 1:2,500 dilution, antirat LH at 1:10,000 dilution, antirat LH? at 1:50,000 dilution, antihuman FSH? at 1:1,250 dilution, antirat TSH? at 1:12,500 dilution (all provided by Dr. A. F. Parlow through the National Hormone and Pituitary Program; NIDDK, Torrance, CA). As negative controls, primary antibodies were omitted, or mouse nonsense IgG and normal rabbit or goat serum (both from DakoCytomation) were substituted for the primary antibodies at comparable dilutions. After thorough washing in PBS-Sap, sections were incubated for 1.5 h with the appropriate secondary antibodies: Alexa Fluor 488- or 555-conjugated goat antimouse or goat antirabbit (Molecular Probes, Eugene, OR; all at 1:1,000 dilution in PBS-Sap) or Cy3-conjugated donkey antigoat at 1:500 dilution (Jackson ImmunoResearch Laboratories, West Grove, PA). Nuclei were counterstained with the DNA-binding dye ToPro-3 (Molecular Probes). Sections were mounted on glass slides using Vectashield (Vector Laboratories, Inc., Burlingame, CA), gently covered with glass coverslips, and stored at –80 C until analysis by confocal laser microscopy.
Sections were scanned using a confocal laser scanning microscope (LSM 510, Zeiss, Zaventem, Belgium) in multitrack mode. Specifications were as follows: for Alexa Fluor 488, excitation at 488 nm and emission at 505–550 nm; for Alexa Fluor 555 and Cy3, excitation at 543 nm and emission at 560–615 nm through a NFT545; and for ToPro-3, excitation at 633 nm and emission above 650 nm. Bleedthrough of a fluorophore in a nonappropriate channel was not observed. For all stainings, at least one of the duplicate sections at different positions in the pituitary was thoroughly scanned, and stainings were repeated two or more times. Figures were prepared using Zeiss LSM Image Examiner or Browser and Microsoft PowerPoint.
RT-PCR analysis of nestin expression in the pituitary
Anterior (AL) and neurointermediate (NIL) lobes from adult rats were carefully dissected, and total RNA was prepared with TriPure reagent (Invitrogen Life Technologies, Inc.) according to the manufacturer’s recommendations. RT was carried out in the following reaction mixture: 2 μl RNA (2–20 ng), 4 μl MgCl2 (25 mM), 2 μl 10x PCR Gold buffer, 0.8 μl deoxy-NTPs (100 mM), 1 μl random hexamers (50 μM; all from Applied Biosystems, Lennik, Belgium), 1 μl Moloney murine leukemia virus (M-MuLV) reverse transcriptase (200 U/μl; Invitrogen Life Technologies, Inc.), 1 μl ribonuclease inhibitor (40 U/μl; RNasin, Promega Corp., Leiden, The Netherlands), and 8.2 μl H2O. RT was performed using the following temperature cycle: 10 min at 25 C, 50 min at 42 C, and 10 min at 95 C. For PCR, 1 μl of the RT reaction was added to 9 μl of a mixture containing 0.9 μl MgCl2 (25 mM), 1.1 μl 10x PCR Gold buffer, 0.1 μl deoxy-NTPs (100 mM), 0.15 μl AmpliTaq Gold DNA polymerase (5 U/μl; Applied Biosystems), and 0.1 μl of each oligonucleotide primer (100 μM; Invitrogen Life Technologies, Inc.). For amplification of nestin cDNA, the following primers were used: 5'-AAGCAGGGTCTACAGAGTCAGATCG-3' (sense) and 5'-GCTGTCACAGGAGTCTCAAGGGTAT-3' (antisense), yielding an amplified fragment of 299 bp. Samples were subjected to 7 min at 95 C, 40 cycles of 10 sec at 95 C, 20 sec at 58.4 C, and 25 sec at 72 C, and finally 7 min at 72 C in a GeneAmp PCR System 2400 (Applied Biosystems). To exclude genomic DNA amplification, primers were selected to span introns. In addition, RT was carried out without M-MuLV reverse transcriptase. As a negative control for PCR, H2O was used instead of RT reaction product. As a positive control, RT-PCR was performed for L19, a ribosomal protein constitutively expressed in all cells (data not shown). Amplified DNA products were electrophoresed in a 2% agarose gel containing 0.5 μg/ml ethidium bromide. Amplicons were cut from the gel, and their identities were confirmed by nucleotide sequence analysis (performed by Lark Technologies, Essex, UK).
Aggregate culture of pituitary cells and analysis by whole-aggregate immunofluorescence or FACS
Pituitary cells were cultured as aggregates as described in detail previously (31). In brief, pituitary glands (or the separately dissected AL or NIL) were dissociated into single cells using successive incubations with trypsin, deoxyribonuclease (DNase), trypsin inhibitor, and EDTA. Finally, tissue pieces were mechanically triturated, and cells were transferred to DNase-containing medium before harvesting by centrifugation. Cells were seeded at 2 x 106/2 ml in 35-mm petri dishes in chemically defined serum-free medium (SM; produced by Invitrogen Life Technologies, Inc.) supplemented, or not, with human recombinant bFGF (20–100 ng/ml), EGF (20 ng/ml; both from R&D Systems, Minneapolis, MN), rat leukemia inhibitory factor (LIF; 10 ng/ml; Chemicon Biognost), or a combination of these growth factors. Dishes were placed on a gyratory shaker in a 1.5% CO2 incubator (37 C), resulting in the formation of well characterized aggregates (31). On d 2, aggregates were transferred to 55-mm petri dishes in a final volume of 6 ml. Medium was renewed twice a week. Live aggregates were viewed and photographed using an inverted microscope (Nikon TMS-F, Analis, Namen, Belgium) and a digital camera (Nikon DXM 1200) connected to a computer with ACT-1 image capturing software (Nikon).
Immunofluorescent analysis of whole aggregates was performed as described above for Vibratome sections, except for the fixation time with paraformaldehyde (2 h). From all stainings, at least five aggregates were thoroughly scanned in the confocal setting, and stainings were repeated two or more times.
Quantification of nestin-immunopositive cells in the aggregates as well as in the pituitary ex vivo was performed by FACS. Pituitary glands and aggregates (after a 6-wk culture period) were dispersed into single cells as described above. Cells were fixed on ice for 1 h in 4% paraformaldehyde, permeabilized at room temperature during 10 min with PBS-Sap, and then consecutively incubated on ice with normal goat serum (20%; 5 min), mouse antirat nestin antibody (1:200; 1 h), and phycoerythrin-labeled goat antimouse Ig (1:20; 45 min; BD Pharmingen, San Diego, CA). Cells were analyzed on a FACSort (BD Biosciences), and nestin-positive cells were gated to calculate their proportion. Data were statistically analyzed using a one-way ANOVA with Tukey multiple comparison test (P < 0.05; statistical software from StatsDirect, Cheshire, UK).
Monolayer culture of pituitary cells and immuno-fluorescent analysis
ALs from adult (8-wk-old) random cycling female Wistar rats were dissociated into single cells as described above. Cells were plated in SM in six- or 24-well plates on coverslips coated with collagen (5 μg/cm2; mouse collagen type IV, BD Biosciences) at a density of 0.3 x 106/cm2. Cultures were kept at 37 C in a 1.5% CO2 incubator. Other coatings, such as fibronectin and laminin, were also tested, but did not allow proper selection of nestin-ir cells. After 2 d in culture, strongly attached cells were selected by removing the less adhering cells with trypsin. Different protocols were tried and resulted in the following strict selection method. Coverslips were placed for 80 sec in trypsin solution (1% in DMEM/0.3% BSA; prewarmed at 37 C), then intensively rinsed with a stream of sterile prewarmed (37 C) PBS and placed for 5–10 min in a dish containing trypsin inhibitor solution (0.1% in DMEM/0.3% BSA). This solution was replaced with SM supplemented with 10% fetal calf serum (FCS; HyClone, Logan, UT). Selected cells (which mainly represented nestin-ir cells) were cultured in 1.5% CO2 at 37 C for 4–5 d until they reached 80–90% confluence. Then coverslips were processed for immunofluorescent analysis or cells were replated for additional passages. Passaging was performed as follows. Cells were trypsinized (0.5% trypsin in DMEM/0.3% BSA, containing 1 mM EDTA) for 10 min, spun down through a 3% BSA layer containing trypsin inhibitor, resuspended in fresh SM supplemented with 10% FCS, and plated at a dilution of 1:4 on collagen-coated coverslips if immunostaining was to be performed or on collagen-coated plastic dishes for propagation. Live cultures were viewed and photographed as described above for aggregates. Video recording of live cell cultures was performed using a Leica AS MDW workstation connected to the computer with Leica Deblur software. Cell separation by gravity sedimentation through a BSA gradient was performed as previously described (32). For immunofluorescent analysis, cells were fixed on coverslips with 4% paraformaldehyde in PBS for 15 min, permeabilized for 5 min in 4% paraformaldehyde with 0.4% Triton X-100 (Sigma-Aldrich Corp.), and incubated with primary antibodies (diluted in 1% BSA/PBS) for 1 h and with secondary antibodies for 50 min. The antibodies and dilutions used were as described above for Vibratome sections. For actin visualization, Alexa 488-conjugated phalloidin (Molecular Probes) was added to antibody mixtures at a dilution of 1:50. When necessary, positive control cells were simultaneously processed: rat heart vessel preparations for CD31 staining, rat brain microglial cell preparations for Ox-42 staining, and fetal rat brain for DCX staining. Finally, coverslips were mounted on glass slides using Vectashield containing 0.5 μg/ml 4',6-diamido-2-phenylindole hydrochloride (Roche, Mannheim, Germany) for nuclear counterstaining.
In some experiments the dipeptide ?-Ala-Lys-N-AMCA (7-amino-4-methylcoumarin-3-acetyl) was used as a marker for folliculo-stellate (FS) cells (33). ?-Ala-Lys-N-AMCA (provided by Dr. K. Bauer, Max Planck Institute, Hannover, Germany) was added to the culture medium at a concentration of 100 μM for 2 h at 37 C. Incubation was stopped on ice. Cells were rinsed three times with ice-cold Hanks’ balanced salt solution (Invitrogen Life Technologies, Inc.) and once with ice-cold PBS. Cells were then fixed for 20 min in 3% paraformaldehyde/1% glutaraldehyde (Merck, Darmstadt, Germany) at room temperature and treated three times for 5 min each time with 0.5 mg/ml NaBH4 (Sigma-Aldrich Corp.) on ice. Before immunostaining, performed as described above, cells were briefly permeabilized with 0,1% Triton X-100/PBS for 2 min.
To visualize cell proliferation, the nucleotide analog 5-bromo-2'-deoxyuridine (BrdU; BD Biosciences) was added to the cell cultures for 6–24 h at a final concentration of 10 μM. Cells were fixed, treated with DNase (100 U/ml DNase I in PBS; amplification grade, Invitrogen Life Technologies, Inc.) for 30 min at 37 C, incubated with mouse anti-BrdU (1:5; BD Biosciences), and processed for immunostaining as described above.
Slides with immunofluorescent samples were examined using a Leica DMRB epifluorescence microscope with appropriate filter sets. Pictures were taken with a digital camera (Nikon DXM 1200), connected to a computer with ACT-1 image-capturing software (Nikon). Figures and movies were prepared using ImageJ (1.30v, freely available at http://rsb.info.nih.gov/ij/), Adobe Photoshop (Adobe Systems, San Jose, CA), and PowerPoint (Microsoft, Redmond, CA).
Results
Nestin-ir in the pituitary gland of immature and adult rats
Pituitary Vibratome sections from immature (5 and 14 d old) and adult rats were immunofluorescently labeled using the monoclonal antirat nestin antibody Rat-401 (2) and analyzed by confocal laser scanning microscopy. Nestin-ir was present scattered over the three lobes and was associated with distinct cells as clear from nuclear counterstaining (Fig. 1). Simultaneous staining using the monoclonal Rat-401 and a polyclonal antinestin antiserum (gift from Dr. R. McKay) showed evident overlap in the vast majority of immunopositive cells, confirming the identity of the immunoreactive material (Fig. 1D). No staining was observed when the monoclonal Rat-401 was replaced with nonsense mouse IgG1 (data not shown).
FIG. 1. Nestin-ir is present in immature and adult pituitary glands and is mainly not localized in typical pituitary cell types. Nestin-ir is present in AL (A and C) and NL (B) and is more abundant in immature (A and B) than in adult (C) pituitary. The nestin-ir signals obtained using the monoclonal anti-nestin Rat-401 and the polyclonal anti-nestin antiserum overlapped (D, separate colors with merge). No colocalization of nestin-ir was found with ACTH (E), GH (F), PRL (G), LH? (H), TSH? (I; shown for 14-d-old AL), FSH?, or LH (not shown). Nestin-ir was only occasionally colocalized with S100 in adult AL [J, separate channels with merge; filled arrow, coexpressing cell; arrowhead, signals running over each other and yielding apparent, but not genuine, coexpression as is clear from split images and also from consecutive z-sections and three-dimensional reconstruction (data not shown); hollow arrow, nestin single cell]. In the 14-d-old IL, only some nestin-ir cells express GFAP (K, separate colors with merge; filled arrow, double-positive cell; hollow arrow, nestin single cell; arrowhead, GFAP single cell; immunoreactive spots or streaks represent optical sections through processes of the IL stellate glial cells). No colocalization is found in the NL. In the adult IL, most nestin- and GFAP-ir signals coincide (L, separate colors with merge). In all pictures, nestin-ir revealed by the monoclonal antibody is in red; nuclear stain (ToPro-3) is in blue. P14, Postnatal d 14; AD, adult. Original magnification of all pictures, x630.
Most nestin-ir cells had a stellate shape with long processes; some appeared more rounded or polygonal. Nestin-ir markedly declined between immature and adult ages (Fig. 1, A vs. C). About 0.5% (0.49 ± 0.19%; average ± SEM of three independent measurements) of freshly dispersed pituitary cells from 14-d-old rats were scored nestin-positive by flow cytometric analysis.
RT-PCR revealed the presence of nestin mRNA in the AL and NIL (Fig. 2). The amplified fragment was identified by nucleotide sequencing as part of rat nestin cDNA.
FIG. 2. The nestin gene is expressed in the pituitary gland. Nestin mRNA is detected by RT-PCR in adult AL and NIL, yielding an amplified fragment of 299 bp. M, 100-bp marker; + and –, RT with and without M-MuLV, respectively; –P, negative PCR control.
Identification of nestin-ir cells within pituitary lobes
First, double immunofluorescence was performed for nestin and pituitary hormones. At the ages analyzed, no colocalization of nestin-ir with ACTH, GH, PRL, LH?, FSH?, TSH?, or LH was detected (Fig. 1, E–I, for 14-d-old pituitary; adult pituitary data not shown). In the adult AL, nestin-ir did only sporadically colocalized with S100, a marker ofFS cells (34, 35) (Fig. 1J). Furthermore, colocalizationof nestin- and S100-ir was not observed in the neural lobe (NL) or in the 14-d-old pituitary (data not shown).
In the immature IL, some of the nestin-ir cells expressed GFAP (Fig. 1K). No colocalization was found in the immature NL, where GFAP is abundantly expressed in pituicytes (Fig. 1K). GFAP-ir was only very sporadically detected in the AL, and not together with nestin (data not shown). In the adult pituitary, colocalization of nestin- and GFAP-ir was detected in the great majority of nestin-positive IL cells (Fig. 1L), but not, or only very occasionally, in nestin-ir cells of the AL and NL (data not shown).
To examine whether nestin expression was associated with endothelial cells, sections were simultaneously stained for nestin and the endothelial cell markers von Willebrand factor (vWF) and CD31/PECAM-1. In the 14-d-old pituitary, nestin-ir often codistributed with vWF- and CD31-ir, but convincing colocalization in cells was only sporadically observed (Fig. 3, A and B). In the adult pituitary, nestin-ir cells also closely associated with CD31-ir cells, but did not overlap (Fig. 3C). vWF was not detected in adult pituitary, in contrast to CD31, which is considered to be a more general marker of endothelial cells (36).
FIG. 3. Additional characterization of nestin-ir cells in pituitary and in pituitary aggregate cell cultures. Nestin-ir often codistributes, but only sporadically overlaps, with the endothelial cell marker vWF in the immature pituitary (A and B, separate colors with merge). In the adult AL, no colocalization of nestin-ir and the pan-endothelial cell marker CD31 is found (C, separate colors with merge). In the (adult) AL, fibronectin-ir (FN) is also closely associated, but not wholly coinciding, with nestin-ir (D); single nestin-ir cells are observed (D; arrow). SMA-ir is occasionally detected, mostly around capillaries, but does not colocalize with nestin-ir (E). Nestin-ir is present in some cells of the layers bordering the cleft (arrows) at all ages analyzed (postnatal d 14, F and H; adult, G; not shown for postnatal d 5) and is not colocalized with hormones (shown for GH; H) or with S100, vWF, CD31, GFAP, fibronectin, vimentin, SMA, or Ox-42 (not shown). Nestin-ir persists in pituitary cell aggregate cultures for 4–7 wk and is present in similar cell types as in vivo. Nestin-ir is not localized in hormone- and GFAP-expressing cells [all hormones tested; shown are PRL (I) and GFAP (J) for aggregates from 14-d-old pituitaries after 5 wk in culture]. Treatment with bFGF induces the formation of hollow bulges in the aggregates (K; upper panel, control cultures after 5 wk; lower panel, after 5-wk treatment with bFGF). Some cells around these cavities contain nestin-ir (L), but not together with S100, GFAP, or hormones (not shown). In all pictures (except C and K), nestin-ir signal is in red; nuclear stain (ToPro-3) is in blue. C, Cleft; P14, postnatal d 14; AD, adult; AGG, aggregates. Original magnification, x630, except for K (x200).
Nestin-ir cells did not represent microglial cells, because nestin-ir was not colocalized with Ox-42 (CD11b), a marker for macrophage/microglial cells in the pituitary (37) (data not shown).
To examine whether nestin-ir cells displayed mesenchymal features, colocalization with fibronectin, vimentin, and SMA was tested. Again, a great part of nestin-ir closely associated, but not wholly overlapped, with fibronectin-ir (Fig. 3D) that was mainly found in or around endothelial cells in both immature and adult AL. In IL and NL, no colocalization of nestin- and fibronectin-ir was observed (data not shown). SMA-ir cells, present in small numbers around capillary structures, were also in close proximity with nestin-ir cells in all lobes at both ages, but signals did not overlap (Fig. 3E). SMA-ir cells, in contrast to nestin-ir cells, were lower in number in immature than adult pituitaries. Vimentin-ir was only sporadically detected in the AL (and not in the IL or NL), but no colocalization with nestin-ir was found (data not shown).
To address the question of whether some nestin-ir cells represented neuronal progenitor cells and whether the pituitary, in particular the NL, could function as a secondary germinal center for neurons, as has recently been demonstrated for the hypothalamus (38), sections were stained for DCX, a marker of migrating and differentiating neuronal cells during embryonic and postnatal development (39). No DCX-ir was found in the various lobes of the pituitary gland, at either early postnatal or adult ages (data not shown).
Nestin-ir cells in the layers bordering pituitary cleft
Nestin-ir was also detected in the cell layers lining the cleft, the remnant of Rathke’s pouch. Nestin-ir cells were observed at all ages analyzed (Fig. 3, F–H; postnatal d 5 not shown) in both the anterior wall (bordering the AL) and the posterior wall (flanking the IL), and either occurred as solitary cells or in clusters. No expression of hormones, S100, vWF, CD31, GFAP, fibronectin, vimentin, SMA, and Ox-42 (Fig. 3H and data not shown) was detected in these nestin-ir cells. Sometimes, hormone-, S100-, fibronectin-, vimentin-, and SMA-ir cells were also found in the cell layers bordering the cleft (Fig. 3H and data not shown).
Nestin-ir in pituitary cell aggregate cultures
As analyzed by whole-aggregate immunofluorescence and confocal laser scanning microscopy, nestin-ir was detected in aggregates formed from immature or adult AL cells and persisted in these aggregates up to at least 7 wk in serum-free culture (Fig. 3, I–J). Similar to that in vivo, nestin-ir was not detected in hormone-positive cells (Fig. 3I) and was not or only sporadically detected in GFAP- and S100-expressing cells, respectively (Fig. 3J and data not shown). vWF-, CD31-, Ox-42-, SMA-, and vimentin-ir were not found in pituitary cell aggregate cultures. In contrast, aggregates made from NIL cells alone showed colocalization of GFAP- and nestin-ir (data not shown).
bFGF (20 ng/ml) induced a 10-fold rise in the number of nestin-positive cells after a 6-wk treatment period (from 0.29% in untreated aggregates to 2.69% in bFGF-treated aggregates), and LIF (10 ng/ml) induced a 5-fold increase (from 0.29% to 1.44%; Fig. 4A). EGF (20 ng/ml) did not affect the number of nestin-ir cells (0.35% vs. 0.29% in controls), nor did EGF add to the effect of bFGF (2.45% for EGF plus bFGF vs. 2.69% for bFGF alone). bFGF and LIF together exerted a partially additive effect (3.72% for bFGF plus LIF vs. 2.69% for bFGF). The stimulatory effect of bFGF was dose dependent; a 10-fold increase in nestin-positive cells was obtained with 20 and 50 ng/ml bFGF, and a 20-fold rise was obtained with 100 ng/ml bFGF (Fig. 4B).
FIG. 4. bFGF and LIF increase the number of nestin-positive cells in pituitary cell aggregate cultures. Cell aggregate cultures from 14-d-old pituitaries were treated for 6 wk with bFGF, EGF, or LIF, alone or in combination. The percentage of nestin-ir cells was quantified by FACS analysis. LIF and bFGF enhance the number of nestin-positive cells, whereas EGF is not effective (A). LIF and bFGF exert a partially additive effect (A). The effect of bFGF is dose dependent (B; *, P < 0.0001 vs. control; **, P < 0.0001 vs. 20 and 50 ng/ml). The mean ± SEM (bars) of three independent experiments are shown in A and B.
A remarkable finding in the bFGF-treated, but not in the untreated, EGF-treated, or LIF-treated, AL cell aggregates was the development in time of hollow bulges (Fig. 3K). Some of the cells bordering the hollow bulges contained nestin-ir (Fig. 3L), but no hormones, S100, or GFAP (data not shown).
Nestin-ir cells in AL monolayer cultures
Dissociated cells from adult AL were cultured on collagen-coated coverslips in chemically defined SM without any supplements. In primary (1- to 4-d) cultures (Fig. 5A), nestin-ir cells were present and showed distinct morphological characteristics. Nestin-ir cells were large, fully spread cells with well developed actin structure and a large nucleus that was only weakly stained with 4',6-diamido-2-phenylindole hydrochloride (Fig. 5, D–H). They were mostly grouped in clusters (Fig. 5E) and often protruded few, but long, processes. None of the nestin-ir cells showed hormone-ir (Fig. 5G). Colocalization of nestin-ir with the FS cell marker S100 or labeling of nestin-ir cells with the FS cell-specific dipeptide ?-Ala-Lys-N-AMCA (33) (the uptake of which overlaps with S100) was found, but only very rarely (Fig. 5H). No expression of GFAP, Ox-42, or CD31 was detected in the nestin-ir cells (data not shown). The immunoreactivity of cadherin, a cell-cell adhesion protein, was not detectable in nestin-ir cells, but was present in the other pituitary cells (Fig. 5F), characterized as hormone- and S100-ir cells. Vimentin-ir colocalized with nestin-ir in the filament bundles (Fig. 5J). Vimentin-ir was also observed, although in a less characteristic filamentous format, in the other pituitary cell types (Fig. 5J) identified as FS cells and hormone-producing cells. In contrast, fibronectin was more exclusively found in nestin-ir cells, particularly in the clusters (Fig. 5L). Only rarely were nestin-ir cells in these primary cultures positive for SMA (Fig. 5N).
FIG. 5. Nestin-ir cells propagate in AL monolayer cultures. Cells from adult AL were seeded in monolayer culture on collagen-coated surface. Primary cultures (A; 2 d in serum-free SM) were briefly treated with trypsin to select for the highly adherent cells (B; immediately after trypsinization). These selected cells were further cultured in SM with 10% FCS (C; after 4 d in SM with 10% FCS). Primary cultures contain nestin-ir cells with well developed actin structure (D), mainly grouped in clusters (E; actin in green, and nestin in red). These nestin-ir cells (in red) do not express cadherin (F) or hormones (G; stained with a mixture of all hormone antibodies) and only sporadically express S100 (H; no coexpressing cell shown). Nestin-ir cells as well as the other cells in primary culture contain vimentin (J; vimentin in red, and nestin in green). Fibronectin is more restricted to nestin-ir cells in primary culture (L; fibronectin in green, and nestin in red). After selection and replating, predominantly cells with nestin-ir are left (I; 1 d after first passage; nestin in red, and actin in green). These enriched nestin-ir cells also contain vimentin (K; vimentin in red, and nestin in green) and fibronectin (M; fibronectin in green, and nestin in red). SMA is only rarely expressed by nestin-ir cells in primary culture (N) and immediately after selection (O), but expression is up-regulated with time in culture (P; 1 d after selection) and after additional propagation and passaging (Q; N–Q, SMA in red, and nestin in green). For motility and proliferative activity of nestin-ir cells in primary and selected (enriched) cultures, see Video Files 1 and 3 provided as supplemental data, respectively. I corresponds to the last frame of Video File 3. A–C, Phase contrast; D–Q, epifluorescence. Bar, 100 μm.
Using live cell imaging and subsequent immunocytochemical identification, it was found that nestin-ir cells, in contrast to the other pituitary cells, were remarkably motile (see Video File 1, provided as supplemental data published on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org). They displayed multiple focal adhesion points, which could be visualized by vinculin and ?1 integrin staining (data not shown). Motile nestin-ir cells were found to be small cells of low density, because they were collected in the upper layers of a BSA gradient after gravity sedimentation of freshly isolated AL cells through that gradient (see Video File 2, provided as supplemental data). These cells became well spread shortly after plating and were different from S100-ir cells present in the same BSA layers. Moreover, in contrast to the other cell types, nestin-ir cells proliferated rapidly in the primary cultures, as revealed by live cell imaging (see Video Files 1 and 2) and BrdU incorporation (data not shown).
Brief (80-sec) trypsinization of primary cultures removed most of the hormone- and S100-positive cells, but preserved the firmly attached nestin-ir cells, especially those organized in clusters (Fig. 5B). Additional expansion of the nestin-ir cells after selection required the addition of FCS (10%), which could not be substituted for by EGF or bFGF or by conditioned medium from primary pituitary cell cultures. At present, nestin-ir cells (Fig. 5, C and I) could be passaged with an estimated doubling time of 48 h for 1.5 months (10 passages) and could be frozen and thawed again, without any significant changes in morphology or proliferative activity (see Video File 3, provided as supplemental data). S100- and hormone-ir cells became practically undetectable after two or three passages. In contrast, virtually all of the selected and propagated nestin-ir cells showed vimentin-ir, neatly colocalized with nestin-ir (Fig. 5K). Fibronectin-ir was also detected in these propagated nestin-ir cells, but clearly in a different cellular arrangement than nestin (Fig. 5M). SMA-ir was observed only sporadically in the nestin-ir cells immediately after selection, particularly in solitary, but not in clustered, cells (Fig. 5O). However, 1 d later, the clustered nestin-ir cells started to express SMA (Fig. 5P), and after additional propagation, virtually all nestin-ir cells expressed this myofibroblast marker (Fig. 5Q).
Discussion
The present investigation shows expression of nestin-ir in scattered cells in the three lobes of the postnatal rat pituitary gland. To our knowledge, this is the first report of nestin expression in normal pituitary. Pituitary nestin-ir is more abundant in immature than in adult animals. The most striking observation is that nestin-ir in the AL and IL is not localized in hormone-expressing cells, confining its expression, rather, to undifferentiated cells or supporting tissue cells. The cells are mostly stellate in shape with long processes, consistent with a nonhormonal phenotype. The best known stellate cells in the AL and IL are the FS cells. However, nestin-ir is only sporadically found in FS cells identified by S100 immunostaining. S100 is a widely recognized marker for FS cells in the AL (34, 35, 40). However, FS cells represent a heterogeneous cell population (34, 41, 42, 43, 44), and S100 may not stain all of the FS cells (43), suggesting that nestin-ir is localized in an S100-negative FS cell subpopulation. In this respect it should be noted that we have previously shown that the cytokine interferon- induces the expression of inducible nitric oxide synthase not only in some S100-ir cells, but also in a large number of unidentifiable cells characterized as small cells of low density (32). These cells in which inducible nitric oxide synthase is induced may overlap with the nestin-ir cells. In addition, we previously detected transferrin-like immunoreactivity in a considerable number of cells in the AL and IL different from hormone- and S100-ir cells (45). Unfortunately, transferrin immunostaining failed in vibratome sections and whole aggregates, and double staining with nestin was not possible. On the other hand, colocalization of nestin and S100 has recently been reported in human pituitary adenomas (46).
A remarkable characteristic of some of the nestin-ir staining is its occurrence in close proximity to endothelial cells. The pericapillar localization and stellate shape of these nestin-ir cells are reminiscent of a pericyte phenotype (15, 47, 48). Pericytes were also found to be nestin positive in the pancreas (15). As pericytes, nestin-ir cells may regulate capillary perfusion in the pituitary by their contractile movements (47). Pericytes have also recently been demonstrated to be important in angiogenesis (48). Alternatively, the close endothelial association of nestin-ir cells may be interpreted in the context of recent findings that nestin-expressing stem cells can give rise to endothelial cells (49), and that endothelial cells seem to be essential players in the stem cell niche (12, 50, 51). In addition to the close proximity of nestin-ir cells to endothelial cells, we found some endothelial cells that contained nestin-ir in the immature (but not adult) pituitary. The expression of nestin in endothelial cells has also been reported in other tissues during both embryonic and early postnatal ages (14, 15, 18, 21, 24) and is strongly reduced in adult organs (15).
In the IL, part of the nestin-ir overlaps with GFAP-ir in the immature pituitary, whereas virtually all nestin-ir coincides with GFAP in the adult IL. GFAP in the IL marks stellate glial cells (52, 53, 54). The function of these cells is not clear (52), but they undergo dynamic changes under certain experimental conditions (53, 54). Nestin may be expressed in these stellate glial cells depending on the activation state. Indeed, nestin expression has also been observed in stellate cells of other tissues when activated (15, 21, 22, 23). In contrast, expression of GFAP in the developing IL may indicate maturation of these glial-like (nestin-ir) cells. It has been reported that glial-like IL cells shift from vimentin to GFAP expression during postnatal development (55).
In the NL, the great majority of nestin-ir cells do not express GFAP, a marker for astrocyte-like pituicytes (56), and only sporadically is nestin-ir localized to endothelial cells. No DCX was detected within the NL, suggesting that the neurohypophysis does not contain neuronal progenitor cells, which, in contrast, have been found in the hypothalamus (38). Because the nucleated cells in the NL mainly represent pituicytes, we suggest that nestin is expressed in pituicyte precursors, and that during differentiation, nestin is replaced by filament proteins of mature pituicytes, such as GFAP. Stage-dependent expression of nestin and GFAP has also been reported in the developing brain (2, 3). The sporadic colocalization of nestin and GFAP in (adult) NL cells may reflect transient overlap of both proteins en route to differentiation, which is different from the apparently more permanent coexpression in (adult) IL cells. The considerably lower expression of nestin in adult vs. immature NL is reconcilable with full maturation of pituicytes of adult animals. Also in the AL, GFAP is only very sporadically found together with nestin. GFAP in the AL labels a subset of FS cells (34, 53).
The expression of nestin was also observed in a number of cells in the layers bordering the pituitary cleft, the lumen remaining from Rathke’s pouch (57). This finding was independent of postnatal age examined. These cells also did not belong to known cell types; they did not express any of the pituitary hormones and markers analyzed. Hormone-, S100-, vimentin-, fibronectin-, and SMA-positive cells were intermingled between these bordering cells. The marginal cell layer indeed is not a homogeneous structure. It contains cells with different morphologies and ultrastructural features (57, 58), and S100-ir in the layer has also been reported by others (35, 59). The marginal cell layers around the cleft have in the past repeatedly been proposed to embody a stem/progenitor cell compartment in the postnatal pituitary (57, 60, 61, 62, 63), and the presence of nestin may be consistent with such an assumption. Compelling evidence, however, is still lacking.
In our well-characterized aggregate culture system (31), nestin-ir was detected in similar cell types as in the ex vivo sections, and most nestin-ir cells also did not belong to known pituitary cell lineages. Growth factors such as bFGF and LIF, which are well documented to support maintenance and proliferation of stem/progenitor cells in culture (7, 8, 11, 28, 29, 30, 64), augmented the number of nestin-ir cells in the aggregates. The partially additive effect of bFGF and LIF may suggest a partly overlapping, partly different target population. A remarkable observation was that treatment with bFGF resulted in the formation of hollow bulges in the aggregates. These cavities were bordered by a layer of cells, some of which showed nestin-ir, reminiscent of the nestin-ir cells around the cleft.
When examined in monolayer culture, nestin-ir also did not colocalize with hormones or S100. A striking characteristic is that in these cultures, cells that express nestin-ir were found to move actively and proliferate rapidly. On the basis of their firm attachment to collagen-coated dishes (as apparent from the flattened morphology and the prominent focal adhesion points), nestin-ir cells could be cleared from the less adhering hormone- and S100-positive cells. Adhesive properties have previously been applied by others to isolate progenitor cells from certain tissues (65, 66). Nestin-ir cells could be additionally enriched because of their proliferative activity, which was much higher than the division rate of the few remaining hormone- and S100-ir cells. High proliferative activity has also been reported in monolayer cell cultures originating from pancreatic islets that principally consisted of NIP cells with multipotential capacity (8). The morphology of the enriched pituitary nestin-ir cells strikingly resembles that of NIP cells, with wave-like structures formed in the culture dish, and is also reminiscent of the morphological appearance of pluripotent bone marrow stromal cells (mesenchymal stem cells) in culture, also purified on the basis of tight adhesion (65). Conspicuous structural similarity was also found with the rat fetal brain stem cell monolayer cultures as used by U et al. (67). Of note, the nestin-ir cells obtained in culture did not represent activated microglial cells or some motile endothelial cells, because no Ox-42 or vWF/CD31 was detected in these cells, respectively.
Motility is an outstanding characteristic of pituitary nestin-ir cells in culture, which first may be related to their contractile properties as pericytes. On the other hand, motility and, accordingly, expression of specific adhesion molecules have in the last few years recurrently been linked to a stem/progenitor cell phenotype, allowing such cells to be anchored in a niche and to migrate when needed for formation of new cells during physiological cell turnover or tissue repair (68, 69). In epidermal stem cells, high levels of ?1 integrin and low levels of E-cadherin were found to be typical (70). We report here a similar expression pattern of ?1 integrin and cadherin in cultured nestin-ir cells from the AL, characteristics not present in the other AL cells. Whether nestin-ir cells also exhibit high motility in the pituitary in vivo is at present not clear. We found motility in cells shortly after isolation from the AL and further characterized these cells by BSA gradient separation as small cells of low density. To our knowledge, no data are available on motile cells within the pituitary gland in situ. It is tempting to speculate that motile cells in the pituitary gland, if present, play a role in formation of new cells under physiological conditions and during tissue repair as they do in the brain (68, 69).
The propagated nestin-ir cells in culture contain vimentin, fibronectin, and SMA and may thus represent mesenchymal cells of the pituitary or, alternatively, adopt mesenchymal features in culture. In support of the latter hypothesis, we observed no convincing colocalization of fibronectin, vimentin, and SMA in nestin-ir cells in pituitary sections. Moreover, SMA expression is not present in the nestin-ir cells after selection, but only gradually appears in the propagated nestin-ir cells. Mesenchymal features and markers have been found in nestin-positive cells in other tissues. For instance, a number of researchers described nestin-ir cells in the pancreas as mesenchymal-like cells with no stem/progenitor cell characteristics (15, 16, 26). In contrast, nestin-ir cells derived from skin displaying multipotential capacity also contained fibronectin (7). The class III intermediate filament protein vimentin, although generally considered a mesenchymal marker, has been reported to be an obligatory partner of nestin for proper assembly of intermediate filaments, at least in astrocytes (71). Nestin and vimentin coexpression has been found in putative stem/progenitor cells, such as neurosphere cells (6), neuroepithelial and radial glial cells (2), and pancreatic islet precursor cells (10), as well as in other cell types, such as pancreatic mesenchymal and endothelial cells (16, 18, 27). SMA, a recognized marker of tissue myofibroblasts, has been described as a partner for nestin in putative pancreatic progenitor cells (72) and has been found in cultured papillary cells, which were proposed to play a role as adult kidney stem cells (73). Last, but not least, coexpression of vimentin and nestin in motile cells may point toward a role in repair processes, as reported in regenerating skeletal muscle (74). Considering all of these data, it would be tempting to speculate that the enriched nestin-ir cells represent multipotential progenitor cells of the pituitary. However, attempts to date to drive differentiation of these cells into pituitary cell phenotypes failed (Krylyshkina, O., and H. Vankelecom, unpublished observations).
Finally, it is at present unclear how our enriched nestin-ir cell cultures relate to the pituitary follicular cell cultures reported by Ferrara et al. (75), seminal cell cultures of FS cells that have led to the identification and isolation of the growth factors bFGF and vascular endothelial growth factor. There are some striking differences in the culture protocols used, but their impact on the final outcome is unknown. Ferrara et al. (75) immediately seeded cells in high serum conditions. Follicular cell cultures were obtained by natural overgrowth of proliferating cells and did not require extracellular matrix components to be established. In contrast, collagen (not fibronectin or laminin) and cell culture during the first days (primary culture) without serum was in our hands essential to allow selective (dis)attachment for obtaining the almost homogeneous nestin-ir cell cultures. Moreover, we found that the FS cell markers S100 (not checked by Ferrara et al.) and ?-Ala-Lys-N-AMCA do not label nestin-ir cells. In contrast, FS cells have also been shown to express fibronectin (76) and vimentin (77). However, we did not find substantial colocalization of vimentin and fibronectin with S100 or ?-Ala-Lys-N-AMCA in pituitary sections (unpublished observations).
In conclusion, nestin is expressed in the pituitary gland as well as in pituitary cell cultures. Within the different lobes, the level of expression and the colocalization profile change with age. Nestin-ir is not present in hormone and typical FS cells, but is present in a variety of other not well defined pituitary cell types, including putative pericytes and some uncharacterized cells around the pituitary cleft. Nestin-ir cells rapidly propagate and are strikingly motile in monolayer cell cultures; at least these nestin-ir cells show some mesenchymal characteristics. The biological significance of these enriched nestin-positive cells as well as of the nestin-expressing cells in the pituitary in vivo, particularly in terms of a stem/progenitor cell function or a supportive role, remains to be explored.
Acknowledgments
We thank Dr. A. F. Parlow (Harbor-University of California-Los Angeles Medical Center, Torrance, CA) to supply us with the hormone antisera. We are also very grateful to Dr. R. McKay (National Institutes of Health, Bethesda, MD) for generously providing the polyclonal antinestin antiserum, and to Dr. K. Bauer (Max Planck Institute, Hannover, Germany) for supplying the ?-Ala-Lys-N-AMCA dipeptide. We appreciate the kind sharing of the Leica live imaging workstation by Drs. P. Zimmermann and G. David (Department of Human Genetics, University of Leuven, Leuven, Belgium). We acknowledge the help of Nicole Hersmus (own laboratory) with RT-PCR analysis, of Nathalie De Vos (student in Pharmacy, University of Leuven) for confirming the reliability of the nestin staining, and of Dr. J. van den Oord (Department of Morphology and Medical Imaging, University of Leuven) for help interpreting the immunocytochemical data.
References
Nolan LA, Kavanagh E, Lightman SL, Levy A 1998 Anterior pituitary cell population control: basal cell turnover and the effects of adrenalectomy and dexamethasone treatment. J Neuroendocrinol 10:207–215
Hockfield S, Mckay RDG 1985 Identification of major cell classes in the developing mammalian nervous system. J Neurosci 5:3310–3328
Lendahl U, Zimmerman LB, Mckay RDG 1990 CNS stem cells express a new class of intermediate filament protein. Cell 60:585–595
Johansson CB, Momma S, Clarke DL, Risling M, Lendahl U, Frisen J 1999 Identification of a neural stem cell in the adult mammalian central nervous system. Cell 96:25–34
Morshead CM, van der Kooy D 2004 Disguising adult neural stem cells. Curr Opin Neurobiol 14:1–7
Lobo MVT, Alonso FJM, Redondo C, Lopez-Toledano MA, Caso E, Herranz AS, Paino CL, Reimers D, Bazan E 2003 Cellular characterization of epidermal growth factor-expanded free-floating neurospheres. J Histochem Cytochem 51:89–103
Toma JG, Akhavan M, Fernandes KJL, Barnabe-Heider F, Sadikot A, Kaplan DR, Miller FD 2001 Isolation of multipotent adult stem cells from the dermis of mammalian skin. Nat Cell Biol 3:778–784
Zulewski H, Abraham EJ, Gerlach MJ, Daniel PB, Moritz W, Muller B, Vallejo M, Thomas MK, Habener JF 2001 Multipotential nestin-positive stem cells isolated from adult pancreatic islets differentiate ex vivo into pancreatic endocrine, exocrine, and hepatic phenotypes. Diabetes 50:521–533
Kim SY, Lee SH, Kim BM, Kim EH, Min BH, Bendayan M, Park IS 2004 Activation of nestin-positive duct stem (NPDS) cells in pancreas upon neogenic motivation and possible cytodifferentiation into insulin-secreting cells from NPDS cells. Dev Dynam 230:1–11
Wu FD, Jagir MR, Powell JS 2003 Long-term correction of hyperglycemia in diabetic mice after implantation of cultured human cells derived from fetal pancreas. Blood 102:942A
Tropepe V, Coles BLK, Chiasson BJ, Horsford DJ, Elia AJ, McInnes RR, van der Kooy D 2000 Retinal stem cells in the adult mammalian eye. Science 287:2032–2036
Miura M, Gronthos S, Zhao MR, Lu B, Fisher LW, Robey PG, Shi ST 2003 SHED: stem cells from human exfoliated deciduous teeth. Proc Natl Acad Sci USA 100:5807–5812
Li LN, Mignone J, Yang M, Matic M, Penman S, Enikolopov G, Hoffman RM 2003 Nestin expression in hair follicle sheath progenitor cells. Proc Natl Acad Sci USA 100:9958–9961
Mokry J, Nemecek S 1999 Cerebral angiogenesis shows nestin expression in endothelial cells. Gen Physiol Biophys 18:25–29
Lardon J, Rooman I, Bouwens L 2002 Nestin expression in pancreatic stellate cells and angiogenic endothelial cells. Histochem Cell Biol 117:535–540
Humphrey RK, Bucay N, Beattie GM, Lopez A, Messam CA, Cirulli V, Hayek A 2003 Characterization and isolation of promoter-defined nestin-positive cells from the human fetal pancreas. Diabetes 52:2519–2525
Treutelaar MK, Skidmore JM, Dias-Leme CL, Hara M, Zhang LZ, Simeone D, Martin DM, Burant CF 2003 Nestin-lineage cells contribute to the microvasculature but not endocrine cells of the islet. Diabetes 52:2503–2512
Klein T, Ling ZD, Heimberg H, Madsen OD, Heller RS, Serup P 2003 Nestin is expressed in vascular endothelial cells in the adult human pancreas. J Histochem Cytochem 51:697–706
Delacour A, Nepote V, Trumpp A, Herrera PL 2004 Nestin expression in pancreatic exocrine cell lineages. Mech Dev 121:3–14
Joanette EA, Reusens B, Arany E, Thyssen S, Remacle RC, Hill DJ 2004 Low-protein diet during early life causes a reduction in the frequency of cells immunopositive for nestin and CD34 in both pancreatic ducts and islets in the rat. Endocrinology 145:3004–3013
Frisen J, Johansson CB, Torok C, Risling M, Lendahl U 1995 Rapid, widespread, and long-lasting induction of nestin contributes to the generation of glial scar tissue after CNS injury. J Cell Biol 131:453–464
Clarke SR, Shetty AK, Bradley JL, Turner DA 1994 Reactive astrocytes express the embryonic intermediate neurofilament nestin. Neuroreport 5:1885–1888
Niki T, Pekny M, Hellemans K, De Bleser P, Van den Berg K, Vaeyens F, Quartier E, Schuit F, Geerts A 1999 Class VI intermediate filament protein nestin is induced during activation of rat hepatic stellate cells. Hepatology 29:520–527
Vanderwinden JM, Gillard K, De Laet MH, Messam CA, Schiffmann SN 2002 Distribution of the intermediate filament nestin in the muscularis propria of the human gastrointestinal tract. Cell Tissue Res 309:261–268
Lobo MV, Arenas MI, Alonso FJ, Gomez G, Bazan E, Paino CL, Fernandez E, Fraile B, Paniagua R, Moyano A, Caso E 2004 Nestin, a neuroectodermal stem cell marker molecule, is expressed in Leydig cells of the human testis and in some specific cell types from human testicular tumours. Cell Tissue Res 316:369–376
Gao R, Ustinov J, Pulkkinen MA, Lundin K, Korsgren O, Otonkoski T 2003 Characterization of endocrine progenitor cells and critical factors for their differentiation in human adult pancreatic cell culture. Diabetes 52:2007–2015
Street CN, Lakey JRT, Seeberger K, Helms L, Rajotte RV, Shapiro AMJ, Korbutt GS 2004 Heterogenous expression of nestin in human pancreatic tissue precludes its use as an islet precursor marker. J Endocrinol 180:213–225
Reynolds BA, Weiss S 1992 Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 255:1707–1710
Morshead CM, Reynolds BA, Craig CG, Mcburney MW, Staines WA, Morassutti D, Weiss S, Vanderkooy D 1994 Neural stem cells in the adult mammalian forebrain: a relatively quiescent subpopulation of subependymal cells. Neuron 13:1071–1082
Gritti A, Parati EA, Cova L, Frolichsthal P, Galii R, Wanke E, Faravelli L, Morassutti DJ, Roisen F, Nickel DD, Vescovi AL 1996 Multipotential stem cells from the adult mouse brain proliferate and self-renew in response to basic fibroblast growth factor. J Neurosci 16:1091–1100
Denef C, Maertens P, Allaerts W, Mignon A, Robberecht W, Swennen L, Carmeliet P 1989 Cell-to-cell communication in peptide target cells of anterior pituitary. Methods Enzymol 168:47–71
Vankelecom H, Matthys P, Denef C 1997 Inducible nitric oxide synthase in the anterior pituitary gland: induction by interferon- in a subpopulation of folliculostellate cells and in an unidentifiable population of non-hormone-secreting cells. J Histochem Cytochem 45:847–857
Otto C, Dieck T, Bauer K 1996 Dipeptide uptake by adenohypophyseal folliculo-stellate cells. Am J Physiol 271:C210–C217
Allaerts W, Carmeliet P, Denef C 1990 New perspectives in the function of pituitary folliculo-stellate cells. Mol Cell Endocrinol 71:73–81
Cocchia D, Miani N 1980 Immunocytochemical localization of the brain-specific S-100 protein in the pituitary gland of adult rat. J Neurocytol 9:771–782
Muller AM, Hermanns MI, Skrzynski C, Nesslinger M, Muller KM, Kirkpatrick CJ 2002 Expression of the endothelial markers PECAM-1, vWf, and CD34 in vivo and in vitro. Exp Mol Pathol 72:221–229
Sato T, Inoue K 2000 Dendritic cells in the rat pituitary gland evaluated by the use of monoclonal antibodies and electron microscopy. Arch Histol Cytol 63:291–303
Markakis EA, Palmer TD, Randolph-Moore L, Rakic P, Gage FH 2004 Novel neuronal phenotypes from neural progenitor cells. J Neurosci 24:2886–2897
Gleeson JG, Lin PT, Flanagan LA, Walsh CA 1999 Doublecortin is a microtubule-associated protein and is expressed widely by migrating neurons. Neuron 23:257–271
Tsuchida T, Nagao S, Ohmoto T 1991 The fine-structure of the S-100 protein positive cells in the rat pituitary gland: an immunoelectron microscopic study. Brain Res 564:164–166
Vankelecom H, Denef C 1997 Paracrine communication in the anterior pituitary as studied in reaggregate cell cultures. Microsc Res Technol 39:150–156
Soji T, Mabuchi Y, Kurono C, Herbert DC 1997 Folliculo-stellate cells and intercellular communication within the rat anterior pituitary gland. Microsc Res Technol 39:138–149
Allaerts W, Jeucken PHM, Debets R, Hoefakker S, Claassen E, Drexhage HA 1997 Heterogeneity of pituitary folliculo-stellate cells: implications for interleukin-6 production and accessory function in vivo. J Neuroendocrinol 9:43–53
Inoue K, Couch EF, Takano K, Ogawa S 1999 The structure and function of folliculo-stellate cells in the anterior pituitary gland. Arch Histol Cytol 62:205–218
Tilemans D, Vande Vijver V, Verhoeven G, Denef C 1995 Production of transferrin-like immunoreactivity by rat anterior pituitary and intermediate lobe. J Histochem Cytochem 43:657–664
Osamura Y, Ishii Y, Messam CA, Umeoka K, Sanno N, Teramoto A Expression of neural stem cell marker nestin in human pituitary glands and pituitary adenomas. Program of the 84th Annual Meeting of The Endocrine Society, San Francisco, CA, 2002, Poster 2–597
Kawamura H, Kobayashi M, Li Q, Yamanishi S, Katsumura K, Minami M, Wu DM, Puro DG 2004 Effects of angiotensin II on the pericyte-containing microvasculature of the rat retina. J Physiol 561:671–683
Gerhardt H, Betsholtz C 2003 Endothelial-pericyte interactions in angiogenesis. Cell Tissue Res 314:15–23
Amoh Y, Li L, Yang M, Moossa AR, Katsuoka K, Penman S, Hoffman RM 2004 Nascent blood vessels in the skin arise from nestin-expressing hair-follicle cells. Proc Natl Acad Sci USA 101:13291–13295
Palmer TD, Willhoite AR, Gage FH 2000 Vascular niche for adult hippocampal neurogenesis. J Comp Neurol 425:479–494
Shen Q, Goderie SK, Jin L, Karanth N, Sun Y, Abramova N, Vincent P, Pumiglia K, Temple S 2004 Endothelial cells stimulate self-renewal and expand neurogenesis of neural stem cells. Science 304:1338–1340
Boyd WH 1987 Nonsecretory neuronal elements in the bovine pituitary intermediate lobe. Anatomischer Anzeiger 164:117–128
Gary KA, Chronwall BM 1995 Regulation of GFAP expression in glial-like cells of the rat pituitary intermediate lobe by lactation, salt-loading, and adrenalectomy. Glia 13:272–282
Redecker P, Morgenroth C 1989 Comparative immunohistochemical study of the presence of glial fibrillary acidic protein in the pituitary of several vertebrates. Anatomischer Anzeiger 168:37–47
Gary KA, Sands SA, Chronwall BM 1995 Glial-like cells of the rat pituitary intermediate lobe change morphology and shift from vimentin to GFAP expression during development. Int J Dev Neurosci 13:555–565
Suess U, Pliska V 1981 Identification of the pituicytes as astroglial cells by indirect immunofluorescence staining for the glial fibrillary acidic protein. Brain Res 221:27–33
Correr S, Motta PM 1981 The rat pituitary cleft: a correlated study by scanning and transmission electron microscopy. Cell Tissue Res 215:515–529
Soji T, Yashiro T, Herbert DC 1989 Granulated ‘marginal cell layer’ in the rat anterior pituitary gland. Tissue Cell 21:849–856
Shirasawa N, Kihara H, Yamaguchi S, Yoshimura F 1983 Pituitary folliculo-stellate cells immunostained with S-100 protein anti-serum in postnatal, castrated and thyroidectomized rats. Cell Tissue Res 231:235–249
Yoshimura F, Soji T, Sato S, Yokoyama M 1977 Development and differentiation of rat pituitary follicular cells under normal and some experimental conditions with special reference to an interpretation of renewal cell system. Endocrinol Jpn 24:435–449
Gon G, Nakamura F, Ishikawa H 1987 Cystlike structures derived from the marginal cells of Rathke’s cleft in rat pituitary grafts. Cell Tissue Res 250:29–33
Wilson DB 1986 Distribution of 3H-thymidine in the postnatal hypophysis of the C57BL mouse. Acta Anat 126:121–126
Carbajo S, Hernandez JL, Carbajoperez E 1992 Proliferative activity of cells of the intermediate lobe of the rat pituitary during the postnatal period. Tissue Cell 24:829–834
Williams RL, Hilton DJ, Pease S, Willson TA, Stewart CL, Gearing DP, Wagner EF, Metcalf D, Nicola NA, Gough NM 1988 Myeloid leukemia inhibitory factor maintains the developmental potential of embryonic stem cells. Nature 336:684–687
Hung SC, Chen NJ, Hsieh SL, Li H, Ma HL, Lo WH 2002 Isolation and characterization of size-sieved stem cells from human bone marrow. Stem Cells 20:249–258
Lim DA, Alvarez-Buylla A 1999 Interaction between astrocytes and adult subventricular zone precursors stimulates neurogenesis. Proc Natl Acad Sci USA 96:7526–7531
U HS, Alilain W, Saljooque F 2002 Fetal brain progenitor cells transdifferentiate to fates outside the nervous system. Mol Endocrinol 16:2645–2656
Imitola J, Raddassi K, Park KI, Mueller FJ, Nieto M, Teng YD, Frenkel D, Li J, Sidman RL, Walsh CA, Snyder EY, Khoury SJ Directed migration of neural stem cells to sites of CNS injury by the stromal cell-derived factor 1/CXC chemokine receptor 4 pathway. Proc Natl Acad Sci USA 101:18117–18122
Prestoz L, Relvas JB, Hopkins K, Patel S, Sowinski P, Price J, ffrench-Constant C 2001 Association between integrin-dependent migration capacity of neural stem cells in vitro and anatomical repair following transplantation. Mol Cell Neurosci 18:473–484
Moles JP, Watt FM 1997 The epidermal stem cell compartment: variation in expression levels of E-cadherin and catenins within the basal layer of human epidermis. J Histochem Cytochem 45:867–874
Eliasson C, Sahlgren C, Berthold CH, Stakeberg J, Celis JE, Betsholtz C, Eriksson JE, Pekny M 1999 Intermediate filament protein partnership in astrocytes. J Biol Chem 274:23996–24006
Taguchi M, Otsuki M 2004 Co-localization of nestin and PDX-1 in small evaginations of the main pancreatic duct in adult rats. J Mol Histol 35:785–789
Oliver JA, Maarouf O, Cheema FH, Martens TF, Al Awqati Q 2004 The renal papilla is a niche for adult kidney stem cells. J Clin Invest 114:795–804
Vaittinen S, Lukka R, Sahlgren C, Hurme T, Rantanen J, Lendahl U, Eriksson JE, Kalimo H 2001 The expression of intermediate filament protein nestin as related to vimentin and desmin in regenerating skeletal muscle. J Neuropathol Exp Neurol 60:588–597
Ferrara N, Schweigerer L, Neufeld G, Mitchell R, Gospodarowicz D 1987 Pituitary follicular cells produce basic fibroblast growth factor. Proc Natl Acad Sci USA 84:5773–5777
Liu YC, Tanaka S, Inoue K, Kurosumi K 1989 Localization of fibronectin in the folliculo-stellate cells of the rat anterior pituitary by the double bridge peroxidase-antiperoxidase method. Histochemistry 92:43–45
Marin F, Boya J, Lopez-Carbonell A 1989 Immunocytochemical localization of vimentin in stellate cells (folliculo-stellate cells) of the rat, cat and rabbit pituitary pars distalis. Anat Embryol 179:491–495(Olga Krylyshkina, Jiangha)
Address all correspondence and requests for reprints to: Dr. Hugo Vankelecom, Laboratory of Cell Pharmacology, Department of Molecular Cell Biology, University of Leuven, Campus Gasthuisberg (O&N), Herestraat 49, B-3000 Leuven, Belgium. E-mail: hugo.vankelecom@med.kuleuven.ac.be.
Abstract
Nestin is an intermediate filament protein that has originally been identified as a marker of neuroepithelial stem/progenitor cells. The present study explored whether nestin immunoreactivity (nestin-ir) is present in the rat pituitary and in which cell type(s). Nestin-ir was observed in scattered cells in the anterior, intermediate, and neural lobes. Nestin-ir cells were predominantly of stellate shape and were more numerous in immature than in adult animals. Nestin-ir did not colocalize with any pituitary hormone, and did not colocalize or only very sporadically with the folliculo-stellate cell marker S100. In the intermediate lobe, nestin-ir cells contained glial fibrillary acidic protein in an age-dependent manner. Nestin-ir cells were closely associated with endothelial and fibronectin-ir cells, but did mostly not coincide. Nestin-ir was not found in -smooth muscle actin-ir myofibroblasts or in microglial cells. Regardless of age, nestin-ir was detected in some unidentifiable cells that border the pituitary cleft. Nestin-ir remained present in pituitary cultured as three-dimensional aggregates. Treatment with basic fibroblast growth factor or leukemia inhibitory factor increased the number of nestin-ir cells. Starting from anterior lobe cell monolayer cultures, nestin-ir cells could be selected and propagated to a virtually pure population. These nestin-ir cells displayed remarkable motility and proliferative activity, and did not express hormones, glial fibrillary acidic protein, or S100, but contained vimentin-, fibronectin-, and -smooth muscle actin-ir. In conclusion, nestin-ir is present in the pituitary in cells that are neither hormonal nor typical folliculo-stellate. The expression pattern depends on age and lobe examined. Pericapillar localization suggests a pericyte phenotype for some of them. Whether the heterogeneous nestin-ir population also contains pituitary progenitor cells remains to be explored.
Introduction
THE MATURE PITUITARY is an endocrine organ that dynamically changes its hormone output in response to a variety of physiological and pathological triggers to meet the endocrine needs of the organism. Several of these conditions are associated with rapid changes in the number of certain cell types. Although cell proliferation is believed to occur within the various cell type populations, it remains unknown whether the expansion of a certain population may also result from cell development and differentiation from stem cells or progenitor cells. It also is not known whether steady cell turnover in the gland, estimated to be approximately 1.5%/d in the young adult male rat (1), is driven from a stem/progenitor cell pool.
Nestin is a class VI intermediate filament protein that has first been identified in the rat embryonic central nervous system, where it is abundantly expressed (2, 3). During postnatal development, nestin expression in the brain becomes more restricted and is particularly found in neuroepithelial stem and proliferative progenitor cells present in the (sub)ventricular zone (3, 4, 5, 6). Nestin has also been detected in putative stem/progenitor cells of a number of other tissues, such as dermis (7), pancreas (8, 9, 10), retina (11), dental tissue (12), and hair follicle (13).
Although nestin expression has attracted much attention as a marker in stem/progenitor cells, it is also present in other cell types, such as in newly developing vascular endothelial cells (14, 15, 16, 17, 18, 19, 20) and in activated stellate cells of the brain, pancreas, and liver (15, 21, 22, 23). In the adult human gastrointestinal tract, nestin expression has been found in CD34-positive cells, enteroglial cells, and interstitial cells of Cajal (24). Also Leydig and Sertoli cells have been reported to contain nestin (25). Furthermore, a number of findings do not support nestin as an in situ stem/progenitor cell marker in the pancreas (15, 16, 17, 18, 19, 26, 27). Rather, pancreatic nestin-positive cells are proposed to act as niche cells that induce or support endocrine cell neogenesis (17, 20). Thus, nestin-expressing cells may also behave as supporting cells in tissue homeostasis.
Nestin-positive stem/progenitor cells appear to behave peculiarly in culture. Neural stem/progenitor cells can be propagated in suspension cultures as floating colonies (neurospheres) when treated with basic fibroblast growth factor (bFGF) and/or epidermal growth factor (EGF) (28, 29, 30). Virtually all of the cells within a neurosphere express nestin. Nestin-positive putative precursor cells have also been obtained and multiplied in monolayer cultures. A monolayer of cells expressing nestin grows out from pancreatic islets of Langerhans [nestin-positive islet-derived progenitor cells (NIP)] (8). NIP cells are able to differentiate in vitro into multiple liver and pancreatic cell phenotypes.
To date, stem/progenitor cells in the postnatal pituitary gland have not been identified. Pinpointing these cells may be important to better unravel lineage relationships and cell type renewal and maintenance in the postnatal pituitary. Inspired by the expression profiles of nestin and its potential relationship to the stem/progenitor cell phenotype, the present investigation was intended to examine whether nestin is present in immature and adult pituitary glands, whether its presence is restricted to a certain cell type(s), and how nestin-ir cells behave in culture.
Materials and Methods
Immunofluorescence of pituitary Vibratome sections and analysis by confocal laser scanning microscopy
Female Wistar rats (5 and 14 d old or 8 wk old, random cycling) were purchased from Elevage Janvier (Schaijk, The Netherlands). Rats were killed by CO2 asphyxiation and decapitation. All animal experiments were approved by the university ethical committee and conducted in accordance with The Endocrine Society ethical guidelines. Pituitaries were carefully isolated under the stereomicroscope and immediately fixed for 3 h with 4% paraformaldehyde (Riedel-deHa?n, Seelze, Germany) in PBS (Invitrogen Life Technologies, Inc., Paisley, UK) or for 48 h with IHC Zinc Fixative (BD Biosciences, Erembodegem, Belgium) as needed for CD31/platelet endothelial cell adhesion molecule-1, also called CD31 (PECAM-1) immunostaining. Sometimes rats were perfused with PBS and 4% paraformaldehyde before separation of the gland. Pituitaries were embedded in agarose (ICN Biomedicals, Aurora, OH; 2% in PBS) before sectioning with the Vibratome (Leica VT 1000 S, Leica Microsystems, Wetzlar, Germany). Pituitaries were attentively positioned to cut 50-μm sections through all lobes and cleft. Sections were stored in PBS at 4 C until additional processing by immunofluorescence.
Each staining was performed on duplicate Vibratome sections at different, evenly distributed locations of the pituitary gland. Permeabilization of the sections was achieved with two 30-min incubations in saponin (Sigma-Aldrich Corp., St. Louis, MO; 0.5% in PBS; PBS-Sap). Aspecific binding sites were preadsorbed with 20% normal goat or normal donkey serum (DakoCytomation, Glostrup, Denmark). Sections were incubated overnight at room temperature with one or two of the following primary antibodies: mouse antirat nestin IgG1 (Rat-401, BD Biosciences, Mountain View, CA) (2) at a final dilution of 1:100 to 1:200 in PBS-Sap; rabbit anti-nestin (no. 130, provided by Dr. R. McKay, NIH, National Institute for Neurological Disorders and Stroke, Bethesda, MD; recognizes rat, mouse, and human nestin) at 1:2,000 dilution; mouse antirat PECAM-1/CD31 (BD Biosciences) at 1:20 dilution; mouse antirat Ox-42 (CD11b, Serotec, Oxford, UK) at 1:100 dilution; mouse antihuman -smooth muscle actin (SMA; DakoCytomation) at 1:100 dilution; rabbit antirat GH (UCB Bioproducts, Brussels, Belgium) at 1:2,500 dilution; rabbit anticow S100 at 1:100 dilution (from DakoCytomation or Zymed Laboratories, San Francisco, CA), anticow glial fibrillary acidic protein (GFAP) at 1:10,000 dilution (DakoCytomation); rabbit antirat vWF at 1:100 dilution (Chemicon Biognost, Heule, Belgium); mouse anti-pan cadherin at 1:200 dilution (Sigma-Aldrich Corp., Bornem, Belgium); goat antihuman vimentin at 1:20 dilution (Sigma-Aldrich Corp.); rabbit antihuman fibronectin at 1:200 dilution (Sigma-Aldrich Corp.); goat antihuman doublecortin (DCX; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at 1:400 dilution; and rabbit antihuman ACTH at 1:7,500 dilution, antirat prolactin (anti-rPRL-IC5) at 1:2,500 dilution, antirat LH at 1:10,000 dilution, antirat LH? at 1:50,000 dilution, antihuman FSH? at 1:1,250 dilution, antirat TSH? at 1:12,500 dilution (all provided by Dr. A. F. Parlow through the National Hormone and Pituitary Program; NIDDK, Torrance, CA). As negative controls, primary antibodies were omitted, or mouse nonsense IgG and normal rabbit or goat serum (both from DakoCytomation) were substituted for the primary antibodies at comparable dilutions. After thorough washing in PBS-Sap, sections were incubated for 1.5 h with the appropriate secondary antibodies: Alexa Fluor 488- or 555-conjugated goat antimouse or goat antirabbit (Molecular Probes, Eugene, OR; all at 1:1,000 dilution in PBS-Sap) or Cy3-conjugated donkey antigoat at 1:500 dilution (Jackson ImmunoResearch Laboratories, West Grove, PA). Nuclei were counterstained with the DNA-binding dye ToPro-3 (Molecular Probes). Sections were mounted on glass slides using Vectashield (Vector Laboratories, Inc., Burlingame, CA), gently covered with glass coverslips, and stored at –80 C until analysis by confocal laser microscopy.
Sections were scanned using a confocal laser scanning microscope (LSM 510, Zeiss, Zaventem, Belgium) in multitrack mode. Specifications were as follows: for Alexa Fluor 488, excitation at 488 nm and emission at 505–550 nm; for Alexa Fluor 555 and Cy3, excitation at 543 nm and emission at 560–615 nm through a NFT545; and for ToPro-3, excitation at 633 nm and emission above 650 nm. Bleedthrough of a fluorophore in a nonappropriate channel was not observed. For all stainings, at least one of the duplicate sections at different positions in the pituitary was thoroughly scanned, and stainings were repeated two or more times. Figures were prepared using Zeiss LSM Image Examiner or Browser and Microsoft PowerPoint.
RT-PCR analysis of nestin expression in the pituitary
Anterior (AL) and neurointermediate (NIL) lobes from adult rats were carefully dissected, and total RNA was prepared with TriPure reagent (Invitrogen Life Technologies, Inc.) according to the manufacturer’s recommendations. RT was carried out in the following reaction mixture: 2 μl RNA (2–20 ng), 4 μl MgCl2 (25 mM), 2 μl 10x PCR Gold buffer, 0.8 μl deoxy-NTPs (100 mM), 1 μl random hexamers (50 μM; all from Applied Biosystems, Lennik, Belgium), 1 μl Moloney murine leukemia virus (M-MuLV) reverse transcriptase (200 U/μl; Invitrogen Life Technologies, Inc.), 1 μl ribonuclease inhibitor (40 U/μl; RNasin, Promega Corp., Leiden, The Netherlands), and 8.2 μl H2O. RT was performed using the following temperature cycle: 10 min at 25 C, 50 min at 42 C, and 10 min at 95 C. For PCR, 1 μl of the RT reaction was added to 9 μl of a mixture containing 0.9 μl MgCl2 (25 mM), 1.1 μl 10x PCR Gold buffer, 0.1 μl deoxy-NTPs (100 mM), 0.15 μl AmpliTaq Gold DNA polymerase (5 U/μl; Applied Biosystems), and 0.1 μl of each oligonucleotide primer (100 μM; Invitrogen Life Technologies, Inc.). For amplification of nestin cDNA, the following primers were used: 5'-AAGCAGGGTCTACAGAGTCAGATCG-3' (sense) and 5'-GCTGTCACAGGAGTCTCAAGGGTAT-3' (antisense), yielding an amplified fragment of 299 bp. Samples were subjected to 7 min at 95 C, 40 cycles of 10 sec at 95 C, 20 sec at 58.4 C, and 25 sec at 72 C, and finally 7 min at 72 C in a GeneAmp PCR System 2400 (Applied Biosystems). To exclude genomic DNA amplification, primers were selected to span introns. In addition, RT was carried out without M-MuLV reverse transcriptase. As a negative control for PCR, H2O was used instead of RT reaction product. As a positive control, RT-PCR was performed for L19, a ribosomal protein constitutively expressed in all cells (data not shown). Amplified DNA products were electrophoresed in a 2% agarose gel containing 0.5 μg/ml ethidium bromide. Amplicons were cut from the gel, and their identities were confirmed by nucleotide sequence analysis (performed by Lark Technologies, Essex, UK).
Aggregate culture of pituitary cells and analysis by whole-aggregate immunofluorescence or FACS
Pituitary cells were cultured as aggregates as described in detail previously (31). In brief, pituitary glands (or the separately dissected AL or NIL) were dissociated into single cells using successive incubations with trypsin, deoxyribonuclease (DNase), trypsin inhibitor, and EDTA. Finally, tissue pieces were mechanically triturated, and cells were transferred to DNase-containing medium before harvesting by centrifugation. Cells were seeded at 2 x 106/2 ml in 35-mm petri dishes in chemically defined serum-free medium (SM; produced by Invitrogen Life Technologies, Inc.) supplemented, or not, with human recombinant bFGF (20–100 ng/ml), EGF (20 ng/ml; both from R&D Systems, Minneapolis, MN), rat leukemia inhibitory factor (LIF; 10 ng/ml; Chemicon Biognost), or a combination of these growth factors. Dishes were placed on a gyratory shaker in a 1.5% CO2 incubator (37 C), resulting in the formation of well characterized aggregates (31). On d 2, aggregates were transferred to 55-mm petri dishes in a final volume of 6 ml. Medium was renewed twice a week. Live aggregates were viewed and photographed using an inverted microscope (Nikon TMS-F, Analis, Namen, Belgium) and a digital camera (Nikon DXM 1200) connected to a computer with ACT-1 image capturing software (Nikon).
Immunofluorescent analysis of whole aggregates was performed as described above for Vibratome sections, except for the fixation time with paraformaldehyde (2 h). From all stainings, at least five aggregates were thoroughly scanned in the confocal setting, and stainings were repeated two or more times.
Quantification of nestin-immunopositive cells in the aggregates as well as in the pituitary ex vivo was performed by FACS. Pituitary glands and aggregates (after a 6-wk culture period) were dispersed into single cells as described above. Cells were fixed on ice for 1 h in 4% paraformaldehyde, permeabilized at room temperature during 10 min with PBS-Sap, and then consecutively incubated on ice with normal goat serum (20%; 5 min), mouse antirat nestin antibody (1:200; 1 h), and phycoerythrin-labeled goat antimouse Ig (1:20; 45 min; BD Pharmingen, San Diego, CA). Cells were analyzed on a FACSort (BD Biosciences), and nestin-positive cells were gated to calculate their proportion. Data were statistically analyzed using a one-way ANOVA with Tukey multiple comparison test (P < 0.05; statistical software from StatsDirect, Cheshire, UK).
Monolayer culture of pituitary cells and immuno-fluorescent analysis
ALs from adult (8-wk-old) random cycling female Wistar rats were dissociated into single cells as described above. Cells were plated in SM in six- or 24-well plates on coverslips coated with collagen (5 μg/cm2; mouse collagen type IV, BD Biosciences) at a density of 0.3 x 106/cm2. Cultures were kept at 37 C in a 1.5% CO2 incubator. Other coatings, such as fibronectin and laminin, were also tested, but did not allow proper selection of nestin-ir cells. After 2 d in culture, strongly attached cells were selected by removing the less adhering cells with trypsin. Different protocols were tried and resulted in the following strict selection method. Coverslips were placed for 80 sec in trypsin solution (1% in DMEM/0.3% BSA; prewarmed at 37 C), then intensively rinsed with a stream of sterile prewarmed (37 C) PBS and placed for 5–10 min in a dish containing trypsin inhibitor solution (0.1% in DMEM/0.3% BSA). This solution was replaced with SM supplemented with 10% fetal calf serum (FCS; HyClone, Logan, UT). Selected cells (which mainly represented nestin-ir cells) were cultured in 1.5% CO2 at 37 C for 4–5 d until they reached 80–90% confluence. Then coverslips were processed for immunofluorescent analysis or cells were replated for additional passages. Passaging was performed as follows. Cells were trypsinized (0.5% trypsin in DMEM/0.3% BSA, containing 1 mM EDTA) for 10 min, spun down through a 3% BSA layer containing trypsin inhibitor, resuspended in fresh SM supplemented with 10% FCS, and plated at a dilution of 1:4 on collagen-coated coverslips if immunostaining was to be performed or on collagen-coated plastic dishes for propagation. Live cultures were viewed and photographed as described above for aggregates. Video recording of live cell cultures was performed using a Leica AS MDW workstation connected to the computer with Leica Deblur software. Cell separation by gravity sedimentation through a BSA gradient was performed as previously described (32). For immunofluorescent analysis, cells were fixed on coverslips with 4% paraformaldehyde in PBS for 15 min, permeabilized for 5 min in 4% paraformaldehyde with 0.4% Triton X-100 (Sigma-Aldrich Corp.), and incubated with primary antibodies (diluted in 1% BSA/PBS) for 1 h and with secondary antibodies for 50 min. The antibodies and dilutions used were as described above for Vibratome sections. For actin visualization, Alexa 488-conjugated phalloidin (Molecular Probes) was added to antibody mixtures at a dilution of 1:50. When necessary, positive control cells were simultaneously processed: rat heart vessel preparations for CD31 staining, rat brain microglial cell preparations for Ox-42 staining, and fetal rat brain for DCX staining. Finally, coverslips were mounted on glass slides using Vectashield containing 0.5 μg/ml 4',6-diamido-2-phenylindole hydrochloride (Roche, Mannheim, Germany) for nuclear counterstaining.
In some experiments the dipeptide ?-Ala-Lys-N-AMCA (7-amino-4-methylcoumarin-3-acetyl) was used as a marker for folliculo-stellate (FS) cells (33). ?-Ala-Lys-N-AMCA (provided by Dr. K. Bauer, Max Planck Institute, Hannover, Germany) was added to the culture medium at a concentration of 100 μM for 2 h at 37 C. Incubation was stopped on ice. Cells were rinsed three times with ice-cold Hanks’ balanced salt solution (Invitrogen Life Technologies, Inc.) and once with ice-cold PBS. Cells were then fixed for 20 min in 3% paraformaldehyde/1% glutaraldehyde (Merck, Darmstadt, Germany) at room temperature and treated three times for 5 min each time with 0.5 mg/ml NaBH4 (Sigma-Aldrich Corp.) on ice. Before immunostaining, performed as described above, cells were briefly permeabilized with 0,1% Triton X-100/PBS for 2 min.
To visualize cell proliferation, the nucleotide analog 5-bromo-2'-deoxyuridine (BrdU; BD Biosciences) was added to the cell cultures for 6–24 h at a final concentration of 10 μM. Cells were fixed, treated with DNase (100 U/ml DNase I in PBS; amplification grade, Invitrogen Life Technologies, Inc.) for 30 min at 37 C, incubated with mouse anti-BrdU (1:5; BD Biosciences), and processed for immunostaining as described above.
Slides with immunofluorescent samples were examined using a Leica DMRB epifluorescence microscope with appropriate filter sets. Pictures were taken with a digital camera (Nikon DXM 1200), connected to a computer with ACT-1 image-capturing software (Nikon). Figures and movies were prepared using ImageJ (1.30v, freely available at http://rsb.info.nih.gov/ij/), Adobe Photoshop (Adobe Systems, San Jose, CA), and PowerPoint (Microsoft, Redmond, CA).
Results
Nestin-ir in the pituitary gland of immature and adult rats
Pituitary Vibratome sections from immature (5 and 14 d old) and adult rats were immunofluorescently labeled using the monoclonal antirat nestin antibody Rat-401 (2) and analyzed by confocal laser scanning microscopy. Nestin-ir was present scattered over the three lobes and was associated with distinct cells as clear from nuclear counterstaining (Fig. 1). Simultaneous staining using the monoclonal Rat-401 and a polyclonal antinestin antiserum (gift from Dr. R. McKay) showed evident overlap in the vast majority of immunopositive cells, confirming the identity of the immunoreactive material (Fig. 1D). No staining was observed when the monoclonal Rat-401 was replaced with nonsense mouse IgG1 (data not shown).
FIG. 1. Nestin-ir is present in immature and adult pituitary glands and is mainly not localized in typical pituitary cell types. Nestin-ir is present in AL (A and C) and NL (B) and is more abundant in immature (A and B) than in adult (C) pituitary. The nestin-ir signals obtained using the monoclonal anti-nestin Rat-401 and the polyclonal anti-nestin antiserum overlapped (D, separate colors with merge). No colocalization of nestin-ir was found with ACTH (E), GH (F), PRL (G), LH? (H), TSH? (I; shown for 14-d-old AL), FSH?, or LH (not shown). Nestin-ir was only occasionally colocalized with S100 in adult AL [J, separate channels with merge; filled arrow, coexpressing cell; arrowhead, signals running over each other and yielding apparent, but not genuine, coexpression as is clear from split images and also from consecutive z-sections and three-dimensional reconstruction (data not shown); hollow arrow, nestin single cell]. In the 14-d-old IL, only some nestin-ir cells express GFAP (K, separate colors with merge; filled arrow, double-positive cell; hollow arrow, nestin single cell; arrowhead, GFAP single cell; immunoreactive spots or streaks represent optical sections through processes of the IL stellate glial cells). No colocalization is found in the NL. In the adult IL, most nestin- and GFAP-ir signals coincide (L, separate colors with merge). In all pictures, nestin-ir revealed by the monoclonal antibody is in red; nuclear stain (ToPro-3) is in blue. P14, Postnatal d 14; AD, adult. Original magnification of all pictures, x630.
Most nestin-ir cells had a stellate shape with long processes; some appeared more rounded or polygonal. Nestin-ir markedly declined between immature and adult ages (Fig. 1, A vs. C). About 0.5% (0.49 ± 0.19%; average ± SEM of three independent measurements) of freshly dispersed pituitary cells from 14-d-old rats were scored nestin-positive by flow cytometric analysis.
RT-PCR revealed the presence of nestin mRNA in the AL and NIL (Fig. 2). The amplified fragment was identified by nucleotide sequencing as part of rat nestin cDNA.
FIG. 2. The nestin gene is expressed in the pituitary gland. Nestin mRNA is detected by RT-PCR in adult AL and NIL, yielding an amplified fragment of 299 bp. M, 100-bp marker; + and –, RT with and without M-MuLV, respectively; –P, negative PCR control.
Identification of nestin-ir cells within pituitary lobes
First, double immunofluorescence was performed for nestin and pituitary hormones. At the ages analyzed, no colocalization of nestin-ir with ACTH, GH, PRL, LH?, FSH?, TSH?, or LH was detected (Fig. 1, E–I, for 14-d-old pituitary; adult pituitary data not shown). In the adult AL, nestin-ir did only sporadically colocalized with S100, a marker ofFS cells (34, 35) (Fig. 1J). Furthermore, colocalizationof nestin- and S100-ir was not observed in the neural lobe (NL) or in the 14-d-old pituitary (data not shown).
In the immature IL, some of the nestin-ir cells expressed GFAP (Fig. 1K). No colocalization was found in the immature NL, where GFAP is abundantly expressed in pituicytes (Fig. 1K). GFAP-ir was only very sporadically detected in the AL, and not together with nestin (data not shown). In the adult pituitary, colocalization of nestin- and GFAP-ir was detected in the great majority of nestin-positive IL cells (Fig. 1L), but not, or only very occasionally, in nestin-ir cells of the AL and NL (data not shown).
To examine whether nestin expression was associated with endothelial cells, sections were simultaneously stained for nestin and the endothelial cell markers von Willebrand factor (vWF) and CD31/PECAM-1. In the 14-d-old pituitary, nestin-ir often codistributed with vWF- and CD31-ir, but convincing colocalization in cells was only sporadically observed (Fig. 3, A and B). In the adult pituitary, nestin-ir cells also closely associated with CD31-ir cells, but did not overlap (Fig. 3C). vWF was not detected in adult pituitary, in contrast to CD31, which is considered to be a more general marker of endothelial cells (36).
FIG. 3. Additional characterization of nestin-ir cells in pituitary and in pituitary aggregate cell cultures. Nestin-ir often codistributes, but only sporadically overlaps, with the endothelial cell marker vWF in the immature pituitary (A and B, separate colors with merge). In the adult AL, no colocalization of nestin-ir and the pan-endothelial cell marker CD31 is found (C, separate colors with merge). In the (adult) AL, fibronectin-ir (FN) is also closely associated, but not wholly coinciding, with nestin-ir (D); single nestin-ir cells are observed (D; arrow). SMA-ir is occasionally detected, mostly around capillaries, but does not colocalize with nestin-ir (E). Nestin-ir is present in some cells of the layers bordering the cleft (arrows) at all ages analyzed (postnatal d 14, F and H; adult, G; not shown for postnatal d 5) and is not colocalized with hormones (shown for GH; H) or with S100, vWF, CD31, GFAP, fibronectin, vimentin, SMA, or Ox-42 (not shown). Nestin-ir persists in pituitary cell aggregate cultures for 4–7 wk and is present in similar cell types as in vivo. Nestin-ir is not localized in hormone- and GFAP-expressing cells [all hormones tested; shown are PRL (I) and GFAP (J) for aggregates from 14-d-old pituitaries after 5 wk in culture]. Treatment with bFGF induces the formation of hollow bulges in the aggregates (K; upper panel, control cultures after 5 wk; lower panel, after 5-wk treatment with bFGF). Some cells around these cavities contain nestin-ir (L), but not together with S100, GFAP, or hormones (not shown). In all pictures (except C and K), nestin-ir signal is in red; nuclear stain (ToPro-3) is in blue. C, Cleft; P14, postnatal d 14; AD, adult; AGG, aggregates. Original magnification, x630, except for K (x200).
Nestin-ir cells did not represent microglial cells, because nestin-ir was not colocalized with Ox-42 (CD11b), a marker for macrophage/microglial cells in the pituitary (37) (data not shown).
To examine whether nestin-ir cells displayed mesenchymal features, colocalization with fibronectin, vimentin, and SMA was tested. Again, a great part of nestin-ir closely associated, but not wholly overlapped, with fibronectin-ir (Fig. 3D) that was mainly found in or around endothelial cells in both immature and adult AL. In IL and NL, no colocalization of nestin- and fibronectin-ir was observed (data not shown). SMA-ir cells, present in small numbers around capillary structures, were also in close proximity with nestin-ir cells in all lobes at both ages, but signals did not overlap (Fig. 3E). SMA-ir cells, in contrast to nestin-ir cells, were lower in number in immature than adult pituitaries. Vimentin-ir was only sporadically detected in the AL (and not in the IL or NL), but no colocalization with nestin-ir was found (data not shown).
To address the question of whether some nestin-ir cells represented neuronal progenitor cells and whether the pituitary, in particular the NL, could function as a secondary germinal center for neurons, as has recently been demonstrated for the hypothalamus (38), sections were stained for DCX, a marker of migrating and differentiating neuronal cells during embryonic and postnatal development (39). No DCX-ir was found in the various lobes of the pituitary gland, at either early postnatal or adult ages (data not shown).
Nestin-ir cells in the layers bordering pituitary cleft
Nestin-ir was also detected in the cell layers lining the cleft, the remnant of Rathke’s pouch. Nestin-ir cells were observed at all ages analyzed (Fig. 3, F–H; postnatal d 5 not shown) in both the anterior wall (bordering the AL) and the posterior wall (flanking the IL), and either occurred as solitary cells or in clusters. No expression of hormones, S100, vWF, CD31, GFAP, fibronectin, vimentin, SMA, and Ox-42 (Fig. 3H and data not shown) was detected in these nestin-ir cells. Sometimes, hormone-, S100-, fibronectin-, vimentin-, and SMA-ir cells were also found in the cell layers bordering the cleft (Fig. 3H and data not shown).
Nestin-ir in pituitary cell aggregate cultures
As analyzed by whole-aggregate immunofluorescence and confocal laser scanning microscopy, nestin-ir was detected in aggregates formed from immature or adult AL cells and persisted in these aggregates up to at least 7 wk in serum-free culture (Fig. 3, I–J). Similar to that in vivo, nestin-ir was not detected in hormone-positive cells (Fig. 3I) and was not or only sporadically detected in GFAP- and S100-expressing cells, respectively (Fig. 3J and data not shown). vWF-, CD31-, Ox-42-, SMA-, and vimentin-ir were not found in pituitary cell aggregate cultures. In contrast, aggregates made from NIL cells alone showed colocalization of GFAP- and nestin-ir (data not shown).
bFGF (20 ng/ml) induced a 10-fold rise in the number of nestin-positive cells after a 6-wk treatment period (from 0.29% in untreated aggregates to 2.69% in bFGF-treated aggregates), and LIF (10 ng/ml) induced a 5-fold increase (from 0.29% to 1.44%; Fig. 4A). EGF (20 ng/ml) did not affect the number of nestin-ir cells (0.35% vs. 0.29% in controls), nor did EGF add to the effect of bFGF (2.45% for EGF plus bFGF vs. 2.69% for bFGF alone). bFGF and LIF together exerted a partially additive effect (3.72% for bFGF plus LIF vs. 2.69% for bFGF). The stimulatory effect of bFGF was dose dependent; a 10-fold increase in nestin-positive cells was obtained with 20 and 50 ng/ml bFGF, and a 20-fold rise was obtained with 100 ng/ml bFGF (Fig. 4B).
FIG. 4. bFGF and LIF increase the number of nestin-positive cells in pituitary cell aggregate cultures. Cell aggregate cultures from 14-d-old pituitaries were treated for 6 wk with bFGF, EGF, or LIF, alone or in combination. The percentage of nestin-ir cells was quantified by FACS analysis. LIF and bFGF enhance the number of nestin-positive cells, whereas EGF is not effective (A). LIF and bFGF exert a partially additive effect (A). The effect of bFGF is dose dependent (B; *, P < 0.0001 vs. control; **, P < 0.0001 vs. 20 and 50 ng/ml). The mean ± SEM (bars) of three independent experiments are shown in A and B.
A remarkable finding in the bFGF-treated, but not in the untreated, EGF-treated, or LIF-treated, AL cell aggregates was the development in time of hollow bulges (Fig. 3K). Some of the cells bordering the hollow bulges contained nestin-ir (Fig. 3L), but no hormones, S100, or GFAP (data not shown).
Nestin-ir cells in AL monolayer cultures
Dissociated cells from adult AL were cultured on collagen-coated coverslips in chemically defined SM without any supplements. In primary (1- to 4-d) cultures (Fig. 5A), nestin-ir cells were present and showed distinct morphological characteristics. Nestin-ir cells were large, fully spread cells with well developed actin structure and a large nucleus that was only weakly stained with 4',6-diamido-2-phenylindole hydrochloride (Fig. 5, D–H). They were mostly grouped in clusters (Fig. 5E) and often protruded few, but long, processes. None of the nestin-ir cells showed hormone-ir (Fig. 5G). Colocalization of nestin-ir with the FS cell marker S100 or labeling of nestin-ir cells with the FS cell-specific dipeptide ?-Ala-Lys-N-AMCA (33) (the uptake of which overlaps with S100) was found, but only very rarely (Fig. 5H). No expression of GFAP, Ox-42, or CD31 was detected in the nestin-ir cells (data not shown). The immunoreactivity of cadherin, a cell-cell adhesion protein, was not detectable in nestin-ir cells, but was present in the other pituitary cells (Fig. 5F), characterized as hormone- and S100-ir cells. Vimentin-ir colocalized with nestin-ir in the filament bundles (Fig. 5J). Vimentin-ir was also observed, although in a less characteristic filamentous format, in the other pituitary cell types (Fig. 5J) identified as FS cells and hormone-producing cells. In contrast, fibronectin was more exclusively found in nestin-ir cells, particularly in the clusters (Fig. 5L). Only rarely were nestin-ir cells in these primary cultures positive for SMA (Fig. 5N).
FIG. 5. Nestin-ir cells propagate in AL monolayer cultures. Cells from adult AL were seeded in monolayer culture on collagen-coated surface. Primary cultures (A; 2 d in serum-free SM) were briefly treated with trypsin to select for the highly adherent cells (B; immediately after trypsinization). These selected cells were further cultured in SM with 10% FCS (C; after 4 d in SM with 10% FCS). Primary cultures contain nestin-ir cells with well developed actin structure (D), mainly grouped in clusters (E; actin in green, and nestin in red). These nestin-ir cells (in red) do not express cadherin (F) or hormones (G; stained with a mixture of all hormone antibodies) and only sporadically express S100 (H; no coexpressing cell shown). Nestin-ir cells as well as the other cells in primary culture contain vimentin (J; vimentin in red, and nestin in green). Fibronectin is more restricted to nestin-ir cells in primary culture (L; fibronectin in green, and nestin in red). After selection and replating, predominantly cells with nestin-ir are left (I; 1 d after first passage; nestin in red, and actin in green). These enriched nestin-ir cells also contain vimentin (K; vimentin in red, and nestin in green) and fibronectin (M; fibronectin in green, and nestin in red). SMA is only rarely expressed by nestin-ir cells in primary culture (N) and immediately after selection (O), but expression is up-regulated with time in culture (P; 1 d after selection) and after additional propagation and passaging (Q; N–Q, SMA in red, and nestin in green). For motility and proliferative activity of nestin-ir cells in primary and selected (enriched) cultures, see Video Files 1 and 3 provided as supplemental data, respectively. I corresponds to the last frame of Video File 3. A–C, Phase contrast; D–Q, epifluorescence. Bar, 100 μm.
Using live cell imaging and subsequent immunocytochemical identification, it was found that nestin-ir cells, in contrast to the other pituitary cells, were remarkably motile (see Video File 1, provided as supplemental data published on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org). They displayed multiple focal adhesion points, which could be visualized by vinculin and ?1 integrin staining (data not shown). Motile nestin-ir cells were found to be small cells of low density, because they were collected in the upper layers of a BSA gradient after gravity sedimentation of freshly isolated AL cells through that gradient (see Video File 2, provided as supplemental data). These cells became well spread shortly after plating and were different from S100-ir cells present in the same BSA layers. Moreover, in contrast to the other cell types, nestin-ir cells proliferated rapidly in the primary cultures, as revealed by live cell imaging (see Video Files 1 and 2) and BrdU incorporation (data not shown).
Brief (80-sec) trypsinization of primary cultures removed most of the hormone- and S100-positive cells, but preserved the firmly attached nestin-ir cells, especially those organized in clusters (Fig. 5B). Additional expansion of the nestin-ir cells after selection required the addition of FCS (10%), which could not be substituted for by EGF or bFGF or by conditioned medium from primary pituitary cell cultures. At present, nestin-ir cells (Fig. 5, C and I) could be passaged with an estimated doubling time of 48 h for 1.5 months (10 passages) and could be frozen and thawed again, without any significant changes in morphology or proliferative activity (see Video File 3, provided as supplemental data). S100- and hormone-ir cells became practically undetectable after two or three passages. In contrast, virtually all of the selected and propagated nestin-ir cells showed vimentin-ir, neatly colocalized with nestin-ir (Fig. 5K). Fibronectin-ir was also detected in these propagated nestin-ir cells, but clearly in a different cellular arrangement than nestin (Fig. 5M). SMA-ir was observed only sporadically in the nestin-ir cells immediately after selection, particularly in solitary, but not in clustered, cells (Fig. 5O). However, 1 d later, the clustered nestin-ir cells started to express SMA (Fig. 5P), and after additional propagation, virtually all nestin-ir cells expressed this myofibroblast marker (Fig. 5Q).
Discussion
The present investigation shows expression of nestin-ir in scattered cells in the three lobes of the postnatal rat pituitary gland. To our knowledge, this is the first report of nestin expression in normal pituitary. Pituitary nestin-ir is more abundant in immature than in adult animals. The most striking observation is that nestin-ir in the AL and IL is not localized in hormone-expressing cells, confining its expression, rather, to undifferentiated cells or supporting tissue cells. The cells are mostly stellate in shape with long processes, consistent with a nonhormonal phenotype. The best known stellate cells in the AL and IL are the FS cells. However, nestin-ir is only sporadically found in FS cells identified by S100 immunostaining. S100 is a widely recognized marker for FS cells in the AL (34, 35, 40). However, FS cells represent a heterogeneous cell population (34, 41, 42, 43, 44), and S100 may not stain all of the FS cells (43), suggesting that nestin-ir is localized in an S100-negative FS cell subpopulation. In this respect it should be noted that we have previously shown that the cytokine interferon- induces the expression of inducible nitric oxide synthase not only in some S100-ir cells, but also in a large number of unidentifiable cells characterized as small cells of low density (32). These cells in which inducible nitric oxide synthase is induced may overlap with the nestin-ir cells. In addition, we previously detected transferrin-like immunoreactivity in a considerable number of cells in the AL and IL different from hormone- and S100-ir cells (45). Unfortunately, transferrin immunostaining failed in vibratome sections and whole aggregates, and double staining with nestin was not possible. On the other hand, colocalization of nestin and S100 has recently been reported in human pituitary adenomas (46).
A remarkable characteristic of some of the nestin-ir staining is its occurrence in close proximity to endothelial cells. The pericapillar localization and stellate shape of these nestin-ir cells are reminiscent of a pericyte phenotype (15, 47, 48). Pericytes were also found to be nestin positive in the pancreas (15). As pericytes, nestin-ir cells may regulate capillary perfusion in the pituitary by their contractile movements (47). Pericytes have also recently been demonstrated to be important in angiogenesis (48). Alternatively, the close endothelial association of nestin-ir cells may be interpreted in the context of recent findings that nestin-expressing stem cells can give rise to endothelial cells (49), and that endothelial cells seem to be essential players in the stem cell niche (12, 50, 51). In addition to the close proximity of nestin-ir cells to endothelial cells, we found some endothelial cells that contained nestin-ir in the immature (but not adult) pituitary. The expression of nestin in endothelial cells has also been reported in other tissues during both embryonic and early postnatal ages (14, 15, 18, 21, 24) and is strongly reduced in adult organs (15).
In the IL, part of the nestin-ir overlaps with GFAP-ir in the immature pituitary, whereas virtually all nestin-ir coincides with GFAP in the adult IL. GFAP in the IL marks stellate glial cells (52, 53, 54). The function of these cells is not clear (52), but they undergo dynamic changes under certain experimental conditions (53, 54). Nestin may be expressed in these stellate glial cells depending on the activation state. Indeed, nestin expression has also been observed in stellate cells of other tissues when activated (15, 21, 22, 23). In contrast, expression of GFAP in the developing IL may indicate maturation of these glial-like (nestin-ir) cells. It has been reported that glial-like IL cells shift from vimentin to GFAP expression during postnatal development (55).
In the NL, the great majority of nestin-ir cells do not express GFAP, a marker for astrocyte-like pituicytes (56), and only sporadically is nestin-ir localized to endothelial cells. No DCX was detected within the NL, suggesting that the neurohypophysis does not contain neuronal progenitor cells, which, in contrast, have been found in the hypothalamus (38). Because the nucleated cells in the NL mainly represent pituicytes, we suggest that nestin is expressed in pituicyte precursors, and that during differentiation, nestin is replaced by filament proteins of mature pituicytes, such as GFAP. Stage-dependent expression of nestin and GFAP has also been reported in the developing brain (2, 3). The sporadic colocalization of nestin and GFAP in (adult) NL cells may reflect transient overlap of both proteins en route to differentiation, which is different from the apparently more permanent coexpression in (adult) IL cells. The considerably lower expression of nestin in adult vs. immature NL is reconcilable with full maturation of pituicytes of adult animals. Also in the AL, GFAP is only very sporadically found together with nestin. GFAP in the AL labels a subset of FS cells (34, 53).
The expression of nestin was also observed in a number of cells in the layers bordering the pituitary cleft, the lumen remaining from Rathke’s pouch (57). This finding was independent of postnatal age examined. These cells also did not belong to known cell types; they did not express any of the pituitary hormones and markers analyzed. Hormone-, S100-, vimentin-, fibronectin-, and SMA-positive cells were intermingled between these bordering cells. The marginal cell layer indeed is not a homogeneous structure. It contains cells with different morphologies and ultrastructural features (57, 58), and S100-ir in the layer has also been reported by others (35, 59). The marginal cell layers around the cleft have in the past repeatedly been proposed to embody a stem/progenitor cell compartment in the postnatal pituitary (57, 60, 61, 62, 63), and the presence of nestin may be consistent with such an assumption. Compelling evidence, however, is still lacking.
In our well-characterized aggregate culture system (31), nestin-ir was detected in similar cell types as in the ex vivo sections, and most nestin-ir cells also did not belong to known pituitary cell lineages. Growth factors such as bFGF and LIF, which are well documented to support maintenance and proliferation of stem/progenitor cells in culture (7, 8, 11, 28, 29, 30, 64), augmented the number of nestin-ir cells in the aggregates. The partially additive effect of bFGF and LIF may suggest a partly overlapping, partly different target population. A remarkable observation was that treatment with bFGF resulted in the formation of hollow bulges in the aggregates. These cavities were bordered by a layer of cells, some of which showed nestin-ir, reminiscent of the nestin-ir cells around the cleft.
When examined in monolayer culture, nestin-ir also did not colocalize with hormones or S100. A striking characteristic is that in these cultures, cells that express nestin-ir were found to move actively and proliferate rapidly. On the basis of their firm attachment to collagen-coated dishes (as apparent from the flattened morphology and the prominent focal adhesion points), nestin-ir cells could be cleared from the less adhering hormone- and S100-positive cells. Adhesive properties have previously been applied by others to isolate progenitor cells from certain tissues (65, 66). Nestin-ir cells could be additionally enriched because of their proliferative activity, which was much higher than the division rate of the few remaining hormone- and S100-ir cells. High proliferative activity has also been reported in monolayer cell cultures originating from pancreatic islets that principally consisted of NIP cells with multipotential capacity (8). The morphology of the enriched pituitary nestin-ir cells strikingly resembles that of NIP cells, with wave-like structures formed in the culture dish, and is also reminiscent of the morphological appearance of pluripotent bone marrow stromal cells (mesenchymal stem cells) in culture, also purified on the basis of tight adhesion (65). Conspicuous structural similarity was also found with the rat fetal brain stem cell monolayer cultures as used by U et al. (67). Of note, the nestin-ir cells obtained in culture did not represent activated microglial cells or some motile endothelial cells, because no Ox-42 or vWF/CD31 was detected in these cells, respectively.
Motility is an outstanding characteristic of pituitary nestin-ir cells in culture, which first may be related to their contractile properties as pericytes. On the other hand, motility and, accordingly, expression of specific adhesion molecules have in the last few years recurrently been linked to a stem/progenitor cell phenotype, allowing such cells to be anchored in a niche and to migrate when needed for formation of new cells during physiological cell turnover or tissue repair (68, 69). In epidermal stem cells, high levels of ?1 integrin and low levels of E-cadherin were found to be typical (70). We report here a similar expression pattern of ?1 integrin and cadherin in cultured nestin-ir cells from the AL, characteristics not present in the other AL cells. Whether nestin-ir cells also exhibit high motility in the pituitary in vivo is at present not clear. We found motility in cells shortly after isolation from the AL and further characterized these cells by BSA gradient separation as small cells of low density. To our knowledge, no data are available on motile cells within the pituitary gland in situ. It is tempting to speculate that motile cells in the pituitary gland, if present, play a role in formation of new cells under physiological conditions and during tissue repair as they do in the brain (68, 69).
The propagated nestin-ir cells in culture contain vimentin, fibronectin, and SMA and may thus represent mesenchymal cells of the pituitary or, alternatively, adopt mesenchymal features in culture. In support of the latter hypothesis, we observed no convincing colocalization of fibronectin, vimentin, and SMA in nestin-ir cells in pituitary sections. Moreover, SMA expression is not present in the nestin-ir cells after selection, but only gradually appears in the propagated nestin-ir cells. Mesenchymal features and markers have been found in nestin-positive cells in other tissues. For instance, a number of researchers described nestin-ir cells in the pancreas as mesenchymal-like cells with no stem/progenitor cell characteristics (15, 16, 26). In contrast, nestin-ir cells derived from skin displaying multipotential capacity also contained fibronectin (7). The class III intermediate filament protein vimentin, although generally considered a mesenchymal marker, has been reported to be an obligatory partner of nestin for proper assembly of intermediate filaments, at least in astrocytes (71). Nestin and vimentin coexpression has been found in putative stem/progenitor cells, such as neurosphere cells (6), neuroepithelial and radial glial cells (2), and pancreatic islet precursor cells (10), as well as in other cell types, such as pancreatic mesenchymal and endothelial cells (16, 18, 27). SMA, a recognized marker of tissue myofibroblasts, has been described as a partner for nestin in putative pancreatic progenitor cells (72) and has been found in cultured papillary cells, which were proposed to play a role as adult kidney stem cells (73). Last, but not least, coexpression of vimentin and nestin in motile cells may point toward a role in repair processes, as reported in regenerating skeletal muscle (74). Considering all of these data, it would be tempting to speculate that the enriched nestin-ir cells represent multipotential progenitor cells of the pituitary. However, attempts to date to drive differentiation of these cells into pituitary cell phenotypes failed (Krylyshkina, O., and H. Vankelecom, unpublished observations).
Finally, it is at present unclear how our enriched nestin-ir cell cultures relate to the pituitary follicular cell cultures reported by Ferrara et al. (75), seminal cell cultures of FS cells that have led to the identification and isolation of the growth factors bFGF and vascular endothelial growth factor. There are some striking differences in the culture protocols used, but their impact on the final outcome is unknown. Ferrara et al. (75) immediately seeded cells in high serum conditions. Follicular cell cultures were obtained by natural overgrowth of proliferating cells and did not require extracellular matrix components to be established. In contrast, collagen (not fibronectin or laminin) and cell culture during the first days (primary culture) without serum was in our hands essential to allow selective (dis)attachment for obtaining the almost homogeneous nestin-ir cell cultures. Moreover, we found that the FS cell markers S100 (not checked by Ferrara et al.) and ?-Ala-Lys-N-AMCA do not label nestin-ir cells. In contrast, FS cells have also been shown to express fibronectin (76) and vimentin (77). However, we did not find substantial colocalization of vimentin and fibronectin with S100 or ?-Ala-Lys-N-AMCA in pituitary sections (unpublished observations).
In conclusion, nestin is expressed in the pituitary gland as well as in pituitary cell cultures. Within the different lobes, the level of expression and the colocalization profile change with age. Nestin-ir is not present in hormone and typical FS cells, but is present in a variety of other not well defined pituitary cell types, including putative pericytes and some uncharacterized cells around the pituitary cleft. Nestin-ir cells rapidly propagate and are strikingly motile in monolayer cell cultures; at least these nestin-ir cells show some mesenchymal characteristics. The biological significance of these enriched nestin-positive cells as well as of the nestin-expressing cells in the pituitary in vivo, particularly in terms of a stem/progenitor cell function or a supportive role, remains to be explored.
Acknowledgments
We thank Dr. A. F. Parlow (Harbor-University of California-Los Angeles Medical Center, Torrance, CA) to supply us with the hormone antisera. We are also very grateful to Dr. R. McKay (National Institutes of Health, Bethesda, MD) for generously providing the polyclonal antinestin antiserum, and to Dr. K. Bauer (Max Planck Institute, Hannover, Germany) for supplying the ?-Ala-Lys-N-AMCA dipeptide. We appreciate the kind sharing of the Leica live imaging workstation by Drs. P. Zimmermann and G. David (Department of Human Genetics, University of Leuven, Leuven, Belgium). We acknowledge the help of Nicole Hersmus (own laboratory) with RT-PCR analysis, of Nathalie De Vos (student in Pharmacy, University of Leuven) for confirming the reliability of the nestin staining, and of Dr. J. van den Oord (Department of Morphology and Medical Imaging, University of Leuven) for help interpreting the immunocytochemical data.
References
Nolan LA, Kavanagh E, Lightman SL, Levy A 1998 Anterior pituitary cell population control: basal cell turnover and the effects of adrenalectomy and dexamethasone treatment. J Neuroendocrinol 10:207–215
Hockfield S, Mckay RDG 1985 Identification of major cell classes in the developing mammalian nervous system. J Neurosci 5:3310–3328
Lendahl U, Zimmerman LB, Mckay RDG 1990 CNS stem cells express a new class of intermediate filament protein. Cell 60:585–595
Johansson CB, Momma S, Clarke DL, Risling M, Lendahl U, Frisen J 1999 Identification of a neural stem cell in the adult mammalian central nervous system. Cell 96:25–34
Morshead CM, van der Kooy D 2004 Disguising adult neural stem cells. Curr Opin Neurobiol 14:1–7
Lobo MVT, Alonso FJM, Redondo C, Lopez-Toledano MA, Caso E, Herranz AS, Paino CL, Reimers D, Bazan E 2003 Cellular characterization of epidermal growth factor-expanded free-floating neurospheres. J Histochem Cytochem 51:89–103
Toma JG, Akhavan M, Fernandes KJL, Barnabe-Heider F, Sadikot A, Kaplan DR, Miller FD 2001 Isolation of multipotent adult stem cells from the dermis of mammalian skin. Nat Cell Biol 3:778–784
Zulewski H, Abraham EJ, Gerlach MJ, Daniel PB, Moritz W, Muller B, Vallejo M, Thomas MK, Habener JF 2001 Multipotential nestin-positive stem cells isolated from adult pancreatic islets differentiate ex vivo into pancreatic endocrine, exocrine, and hepatic phenotypes. Diabetes 50:521–533
Kim SY, Lee SH, Kim BM, Kim EH, Min BH, Bendayan M, Park IS 2004 Activation of nestin-positive duct stem (NPDS) cells in pancreas upon neogenic motivation and possible cytodifferentiation into insulin-secreting cells from NPDS cells. Dev Dynam 230:1–11
Wu FD, Jagir MR, Powell JS 2003 Long-term correction of hyperglycemia in diabetic mice after implantation of cultured human cells derived from fetal pancreas. Blood 102:942A
Tropepe V, Coles BLK, Chiasson BJ, Horsford DJ, Elia AJ, McInnes RR, van der Kooy D 2000 Retinal stem cells in the adult mammalian eye. Science 287:2032–2036
Miura M, Gronthos S, Zhao MR, Lu B, Fisher LW, Robey PG, Shi ST 2003 SHED: stem cells from human exfoliated deciduous teeth. Proc Natl Acad Sci USA 100:5807–5812
Li LN, Mignone J, Yang M, Matic M, Penman S, Enikolopov G, Hoffman RM 2003 Nestin expression in hair follicle sheath progenitor cells. Proc Natl Acad Sci USA 100:9958–9961
Mokry J, Nemecek S 1999 Cerebral angiogenesis shows nestin expression in endothelial cells. Gen Physiol Biophys 18:25–29
Lardon J, Rooman I, Bouwens L 2002 Nestin expression in pancreatic stellate cells and angiogenic endothelial cells. Histochem Cell Biol 117:535–540
Humphrey RK, Bucay N, Beattie GM, Lopez A, Messam CA, Cirulli V, Hayek A 2003 Characterization and isolation of promoter-defined nestin-positive cells from the human fetal pancreas. Diabetes 52:2519–2525
Treutelaar MK, Skidmore JM, Dias-Leme CL, Hara M, Zhang LZ, Simeone D, Martin DM, Burant CF 2003 Nestin-lineage cells contribute to the microvasculature but not endocrine cells of the islet. Diabetes 52:2503–2512
Klein T, Ling ZD, Heimberg H, Madsen OD, Heller RS, Serup P 2003 Nestin is expressed in vascular endothelial cells in the adult human pancreas. J Histochem Cytochem 51:697–706
Delacour A, Nepote V, Trumpp A, Herrera PL 2004 Nestin expression in pancreatic exocrine cell lineages. Mech Dev 121:3–14
Joanette EA, Reusens B, Arany E, Thyssen S, Remacle RC, Hill DJ 2004 Low-protein diet during early life causes a reduction in the frequency of cells immunopositive for nestin and CD34 in both pancreatic ducts and islets in the rat. Endocrinology 145:3004–3013
Frisen J, Johansson CB, Torok C, Risling M, Lendahl U 1995 Rapid, widespread, and long-lasting induction of nestin contributes to the generation of glial scar tissue after CNS injury. J Cell Biol 131:453–464
Clarke SR, Shetty AK, Bradley JL, Turner DA 1994 Reactive astrocytes express the embryonic intermediate neurofilament nestin. Neuroreport 5:1885–1888
Niki T, Pekny M, Hellemans K, De Bleser P, Van den Berg K, Vaeyens F, Quartier E, Schuit F, Geerts A 1999 Class VI intermediate filament protein nestin is induced during activation of rat hepatic stellate cells. Hepatology 29:520–527
Vanderwinden JM, Gillard K, De Laet MH, Messam CA, Schiffmann SN 2002 Distribution of the intermediate filament nestin in the muscularis propria of the human gastrointestinal tract. Cell Tissue Res 309:261–268
Lobo MV, Arenas MI, Alonso FJ, Gomez G, Bazan E, Paino CL, Fernandez E, Fraile B, Paniagua R, Moyano A, Caso E 2004 Nestin, a neuroectodermal stem cell marker molecule, is expressed in Leydig cells of the human testis and in some specific cell types from human testicular tumours. Cell Tissue Res 316:369–376
Gao R, Ustinov J, Pulkkinen MA, Lundin K, Korsgren O, Otonkoski T 2003 Characterization of endocrine progenitor cells and critical factors for their differentiation in human adult pancreatic cell culture. Diabetes 52:2007–2015
Street CN, Lakey JRT, Seeberger K, Helms L, Rajotte RV, Shapiro AMJ, Korbutt GS 2004 Heterogenous expression of nestin in human pancreatic tissue precludes its use as an islet precursor marker. J Endocrinol 180:213–225
Reynolds BA, Weiss S 1992 Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 255:1707–1710
Morshead CM, Reynolds BA, Craig CG, Mcburney MW, Staines WA, Morassutti D, Weiss S, Vanderkooy D 1994 Neural stem cells in the adult mammalian forebrain: a relatively quiescent subpopulation of subependymal cells. Neuron 13:1071–1082
Gritti A, Parati EA, Cova L, Frolichsthal P, Galii R, Wanke E, Faravelli L, Morassutti DJ, Roisen F, Nickel DD, Vescovi AL 1996 Multipotential stem cells from the adult mouse brain proliferate and self-renew in response to basic fibroblast growth factor. J Neurosci 16:1091–1100
Denef C, Maertens P, Allaerts W, Mignon A, Robberecht W, Swennen L, Carmeliet P 1989 Cell-to-cell communication in peptide target cells of anterior pituitary. Methods Enzymol 168:47–71
Vankelecom H, Matthys P, Denef C 1997 Inducible nitric oxide synthase in the anterior pituitary gland: induction by interferon- in a subpopulation of folliculostellate cells and in an unidentifiable population of non-hormone-secreting cells. J Histochem Cytochem 45:847–857
Otto C, Dieck T, Bauer K 1996 Dipeptide uptake by adenohypophyseal folliculo-stellate cells. Am J Physiol 271:C210–C217
Allaerts W, Carmeliet P, Denef C 1990 New perspectives in the function of pituitary folliculo-stellate cells. Mol Cell Endocrinol 71:73–81
Cocchia D, Miani N 1980 Immunocytochemical localization of the brain-specific S-100 protein in the pituitary gland of adult rat. J Neurocytol 9:771–782
Muller AM, Hermanns MI, Skrzynski C, Nesslinger M, Muller KM, Kirkpatrick CJ 2002 Expression of the endothelial markers PECAM-1, vWf, and CD34 in vivo and in vitro. Exp Mol Pathol 72:221–229
Sato T, Inoue K 2000 Dendritic cells in the rat pituitary gland evaluated by the use of monoclonal antibodies and electron microscopy. Arch Histol Cytol 63:291–303
Markakis EA, Palmer TD, Randolph-Moore L, Rakic P, Gage FH 2004 Novel neuronal phenotypes from neural progenitor cells. J Neurosci 24:2886–2897
Gleeson JG, Lin PT, Flanagan LA, Walsh CA 1999 Doublecortin is a microtubule-associated protein and is expressed widely by migrating neurons. Neuron 23:257–271
Tsuchida T, Nagao S, Ohmoto T 1991 The fine-structure of the S-100 protein positive cells in the rat pituitary gland: an immunoelectron microscopic study. Brain Res 564:164–166
Vankelecom H, Denef C 1997 Paracrine communication in the anterior pituitary as studied in reaggregate cell cultures. Microsc Res Technol 39:150–156
Soji T, Mabuchi Y, Kurono C, Herbert DC 1997 Folliculo-stellate cells and intercellular communication within the rat anterior pituitary gland. Microsc Res Technol 39:138–149
Allaerts W, Jeucken PHM, Debets R, Hoefakker S, Claassen E, Drexhage HA 1997 Heterogeneity of pituitary folliculo-stellate cells: implications for interleukin-6 production and accessory function in vivo. J Neuroendocrinol 9:43–53
Inoue K, Couch EF, Takano K, Ogawa S 1999 The structure and function of folliculo-stellate cells in the anterior pituitary gland. Arch Histol Cytol 62:205–218
Tilemans D, Vande Vijver V, Verhoeven G, Denef C 1995 Production of transferrin-like immunoreactivity by rat anterior pituitary and intermediate lobe. J Histochem Cytochem 43:657–664
Osamura Y, Ishii Y, Messam CA, Umeoka K, Sanno N, Teramoto A Expression of neural stem cell marker nestin in human pituitary glands and pituitary adenomas. Program of the 84th Annual Meeting of The Endocrine Society, San Francisco, CA, 2002, Poster 2–597
Kawamura H, Kobayashi M, Li Q, Yamanishi S, Katsumura K, Minami M, Wu DM, Puro DG 2004 Effects of angiotensin II on the pericyte-containing microvasculature of the rat retina. J Physiol 561:671–683
Gerhardt H, Betsholtz C 2003 Endothelial-pericyte interactions in angiogenesis. Cell Tissue Res 314:15–23
Amoh Y, Li L, Yang M, Moossa AR, Katsuoka K, Penman S, Hoffman RM 2004 Nascent blood vessels in the skin arise from nestin-expressing hair-follicle cells. Proc Natl Acad Sci USA 101:13291–13295
Palmer TD, Willhoite AR, Gage FH 2000 Vascular niche for adult hippocampal neurogenesis. J Comp Neurol 425:479–494
Shen Q, Goderie SK, Jin L, Karanth N, Sun Y, Abramova N, Vincent P, Pumiglia K, Temple S 2004 Endothelial cells stimulate self-renewal and expand neurogenesis of neural stem cells. Science 304:1338–1340
Boyd WH 1987 Nonsecretory neuronal elements in the bovine pituitary intermediate lobe. Anatomischer Anzeiger 164:117–128
Gary KA, Chronwall BM 1995 Regulation of GFAP expression in glial-like cells of the rat pituitary intermediate lobe by lactation, salt-loading, and adrenalectomy. Glia 13:272–282
Redecker P, Morgenroth C 1989 Comparative immunohistochemical study of the presence of glial fibrillary acidic protein in the pituitary of several vertebrates. Anatomischer Anzeiger 168:37–47
Gary KA, Sands SA, Chronwall BM 1995 Glial-like cells of the rat pituitary intermediate lobe change morphology and shift from vimentin to GFAP expression during development. Int J Dev Neurosci 13:555–565
Suess U, Pliska V 1981 Identification of the pituicytes as astroglial cells by indirect immunofluorescence staining for the glial fibrillary acidic protein. Brain Res 221:27–33
Correr S, Motta PM 1981 The rat pituitary cleft: a correlated study by scanning and transmission electron microscopy. Cell Tissue Res 215:515–529
Soji T, Yashiro T, Herbert DC 1989 Granulated ‘marginal cell layer’ in the rat anterior pituitary gland. Tissue Cell 21:849–856
Shirasawa N, Kihara H, Yamaguchi S, Yoshimura F 1983 Pituitary folliculo-stellate cells immunostained with S-100 protein anti-serum in postnatal, castrated and thyroidectomized rats. Cell Tissue Res 231:235–249
Yoshimura F, Soji T, Sato S, Yokoyama M 1977 Development and differentiation of rat pituitary follicular cells under normal and some experimental conditions with special reference to an interpretation of renewal cell system. Endocrinol Jpn 24:435–449
Gon G, Nakamura F, Ishikawa H 1987 Cystlike structures derived from the marginal cells of Rathke’s cleft in rat pituitary grafts. Cell Tissue Res 250:29–33
Wilson DB 1986 Distribution of 3H-thymidine in the postnatal hypophysis of the C57BL mouse. Acta Anat 126:121–126
Carbajo S, Hernandez JL, Carbajoperez E 1992 Proliferative activity of cells of the intermediate lobe of the rat pituitary during the postnatal period. Tissue Cell 24:829–834
Williams RL, Hilton DJ, Pease S, Willson TA, Stewart CL, Gearing DP, Wagner EF, Metcalf D, Nicola NA, Gough NM 1988 Myeloid leukemia inhibitory factor maintains the developmental potential of embryonic stem cells. Nature 336:684–687
Hung SC, Chen NJ, Hsieh SL, Li H, Ma HL, Lo WH 2002 Isolation and characterization of size-sieved stem cells from human bone marrow. Stem Cells 20:249–258
Lim DA, Alvarez-Buylla A 1999 Interaction between astrocytes and adult subventricular zone precursors stimulates neurogenesis. Proc Natl Acad Sci USA 96:7526–7531
U HS, Alilain W, Saljooque F 2002 Fetal brain progenitor cells transdifferentiate to fates outside the nervous system. Mol Endocrinol 16:2645–2656
Imitola J, Raddassi K, Park KI, Mueller FJ, Nieto M, Teng YD, Frenkel D, Li J, Sidman RL, Walsh CA, Snyder EY, Khoury SJ Directed migration of neural stem cells to sites of CNS injury by the stromal cell-derived factor 1/CXC chemokine receptor 4 pathway. Proc Natl Acad Sci USA 101:18117–18122
Prestoz L, Relvas JB, Hopkins K, Patel S, Sowinski P, Price J, ffrench-Constant C 2001 Association between integrin-dependent migration capacity of neural stem cells in vitro and anatomical repair following transplantation. Mol Cell Neurosci 18:473–484
Moles JP, Watt FM 1997 The epidermal stem cell compartment: variation in expression levels of E-cadherin and catenins within the basal layer of human epidermis. J Histochem Cytochem 45:867–874
Eliasson C, Sahlgren C, Berthold CH, Stakeberg J, Celis JE, Betsholtz C, Eriksson JE, Pekny M 1999 Intermediate filament protein partnership in astrocytes. J Biol Chem 274:23996–24006
Taguchi M, Otsuki M 2004 Co-localization of nestin and PDX-1 in small evaginations of the main pancreatic duct in adult rats. J Mol Histol 35:785–789
Oliver JA, Maarouf O, Cheema FH, Martens TF, Al Awqati Q 2004 The renal papilla is a niche for adult kidney stem cells. J Clin Invest 114:795–804
Vaittinen S, Lukka R, Sahlgren C, Hurme T, Rantanen J, Lendahl U, Eriksson JE, Kalimo H 2001 The expression of intermediate filament protein nestin as related to vimentin and desmin in regenerating skeletal muscle. J Neuropathol Exp Neurol 60:588–597
Ferrara N, Schweigerer L, Neufeld G, Mitchell R, Gospodarowicz D 1987 Pituitary follicular cells produce basic fibroblast growth factor. Proc Natl Acad Sci USA 84:5773–5777
Liu YC, Tanaka S, Inoue K, Kurosumi K 1989 Localization of fibronectin in the folliculo-stellate cells of the rat anterior pituitary by the double bridge peroxidase-antiperoxidase method. Histochemistry 92:43–45
Marin F, Boya J, Lopez-Carbonell A 1989 Immunocytochemical localization of vimentin in stellate cells (folliculo-stellate cells) of the rat, cat and rabbit pituitary pars distalis. Anat Embryol 179:491–495(Olga Krylyshkina, Jiangha)