Pablo G. Hofmann and Gabriel Gutiérrez-Ospina, Departamento de Biología Celular y Fisiología, Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México, Mexico D.F., Ciudad de México 04510. T: + 52 (55) 56229195; F: + 52 (55) 56229198; E: firstname.lastname@example.org and email@example.com
Neuroendocrine cells are present in virtually all organs of the vertebrate body; however, it is yet uncertain whether they exist in the ovaries. Previous reports of ovarian neurons and neuron-like cells in mammals and birds might have resulted from misidentification. The aim of the present work was to determine the identity of neuron-like cells in immature ovaries of the domestic fowl. Cells immunoreactive to neurofilaments, synaptophysin, and chromogranin-A, with small, dense-core secretory granules, were consistently observed throughout the sub-cortical ovarian medulla and cortical interfollicular stroma. These cells also displayed immunoreactivity for tyrosine, tryptophan and dopamine β-hydroxylases, as well as to aromatic L-DOPA decarboxylase, implying their ability to synthesize both catecholamines and indolamines. Our results support the argument that the ovarian cells previously reported as neuron-like in birds, are neuroendocrine cells.
The diffuse neuroendocrine system comprises scattered cells or groups of cells distributed in virtually all organs of vertebrates. Neuroendocrine (NE) cells are commonly interspersed within epithelia, sub-epithelial linings, and/or stromal compartments. Cells of this lineage share the ability to synthesize biogenic amines, produce and release neuropeptidic messengers, and display dense-core secretory granules, and neuronal and endocrine molecular markers (Pearse, 1968, 1986; Flatmark, 2000; Montuenga et al. 2003; Toni, 2004).
Neuroendocrine cells have been extensively characterized along the gastrointestinal and respiratory tracts of vertebrates, particularly mammals. More recent studies have reported their presence along the urogenital tract of higher vertebrates. This cellular phenotype has been identified in epithelia, submucosae and/or interstitial spaces of the urethra, prostate, vas deferens, epididymis and testis (di Sant'Agnese & De Mesy Jensen, 1984; Hanyu et al. 1987; Davidoff et al. 1993, 2005; Jiménez-Trejo et al. 2007). Several reports have suggested the existence of neuron-like cells or intrinsic neurons in the ovarian interstitial space of some birds and mammals (Dees et al. 1995; D'Albora & Barcia, 1996; Mayerhofer et al. 1996; Anesetti et al. 2001; D'Albora et al. 2002; Kimaro & Madekurozwa, 2006; Madekurozwa, 2008). However, proteins such as neurofilaments, tyrosine hydroxylase and neuron specific enolase also exist in NE cells, giving rise to the possibility that the previously identified ovarian intrinsic neurons or neuron-like cells might instead possess a neuroendocrine phenotype. The purpose of the present work was therefore to determine whether the neuron-like cells in the ovary of the domestic fowl belong to this lineage.
Materials and methods
Left ovaries of White Leghorn post-hatching pullets between 10 and 14 days of age were used. Fertilized eggs from a commercial supplier (Aves Libres de Patógenos Específicos, ALPES, Tehuacán, Puebla, Mexico) were incubated at 37.5 °C in a humidified incubator with automatic turning for 21 days, until hatching. The chickens were then maintained at a constant temperature of 24 °C, under a 12 : 12 h light–dark cycle, with ad libitum access to food (Purina, Mexico) and tap water, until they reached the ages of 10–14 days. Animals were sacrificed using an intraperitoneal injection of sodium pentobarbital (a minimum of 85.8 mg kg−1) followed by decapitation at the specified ages. Animal handling and sacrifice followed the guidelines of the ‘CARE 308.01 Avian Euthanasia’ at the Cornell Center for Animal Resources & Education, Cornell University (Guanzini, 2004; based on the AVMA Guidelines on Euthanasia,) and the protocols were reviewed and approved by the institutional bioethics committee at Instituto de Investigaciones Biomédicas, UNAM.
Tissue sampling and immunocytochemistry
Ovaries were dissected, including the adrenal gland, and fixed by immersion in a 4% paraformaldehyde (PFA) solution diluted in phosphate buffer (PB; 0.1 m, pH 7.4) overnight at 4 °C. For cryoprotection, tissue samples were then washed in PB twice before being incubated overnight at 4 °C in a 30% sucrose solution in PB. The ovaries were included in Neg-50 Frozen Section Medium (Richard-Allan Scientific, USA) and frozen on dry ice. Transverse coronal slices, 30 or 50 μm, were obtained using a cryostat microtome (Microm HM550, Germany) at −20 °C, collected in a cryoprotecting solution (25% ethylene glycol, 25% glycerol in PB), and stored at −20 °C until used.
To evaluate the presence of neuronal and/or neuroendocrine cells by immunocytochemistry, endogenous peroxidase activity was inactivated by exposing tissue sections to 0.3% hydrogen peroxide (H2O2) in PB supplemented with 0.3% Triton X-100 (PBT; Sigma, USA) at room temperature (RT) during 30 min. The tissue slices were then washed in PB four times for 10 min at RT, followed by the incubation in BioSB Immuno Retriever (BioSB, Inc., USA) for 30 min at 70 °C. Following three washes in PB and one in PBT for 10 min each at RT, slices were incubated with different primary antibodies (Table 1) diluted in PBT for 40 h at 4 °C, under continuous gentle shaking. After this period and following a gentle wash, sections were incubated with the corresponding donkey-raised, biotin-conjugated secondary antibodies (1 : 500 in PBT; Chemicon, USA) for 120 min at RT, with continuous shaking. After four washes in PB for 10 min each at RT, samples were incubated with the avidin-peroxidase complex (Vectastain ABC Elite, Vector Laboratories, USA) for 90 min at room temperature. Finally, after four 10-min washes in PB at RT, peroxidase activity was developed with a DAB kit (Vector Laboratories, USA) at RT. DAB incubation times were determined for each primary antibody immunoreaction, according to their concentration and staining intensity. Slices were mounted on gelatinized glass slides and dried for 2 h at RT in a vacuum chamber. Mounted slices were re-hydrated with saline solution (0.9%) and counterstained with methyl green (0.05% w/v; Sigma, USA) in sodium acetate buffer (0.1 m, pH 4.2) for 2 min. The slides were thoroughly washed in deionized water to remove excess stain. Slides were air-dried for 15 min at room temperature, and covered using Cytoseal-60 mounting medium (Richard-Allan Scientific, Inc., USA) for observation and image recording. Samples were observed under bright field microscopy on a Nikon Eclipse 80i and images were recorded using the nis-elements® F3.0 software (Nikon, Japan).
Table 1. List of neuroendocrine protein markers and antibodies utilized for the immuno-detection and characterization of NE cells in the developing Gallus gallus ovary
A pre-pro-protein present in dense-core secretory granules of sympathoadrenal cells, that can be cleaved to form pancreastatin, vasostatin, parastatin, chromostatin and ß-granin, among other peptides, involved in negative feedback loops
1 : 1000
Santa Cruz, Inc., USA
L-DOPA Decarboxylase (DDC)
Involved in the decarboxylation of L-3,4-dihydroxyphenylalanine (DOPA) to dopamine and L-5-hydroxytryptophan to serotonin. Participates in catecholamine and indolamine neurotransmitter biosynthesis pathways
1 : 1000
Santa Cruz, Inc., USA
Dopamine ß-Hydroxylase (DBH)
Catalyzes the conversion of dopamine to noradrenaline within the catecholamine synthesis pathway
1 : 4000
1 : 2000
Abcam, Plc., USA
Heavy-chain Neurofilament (NF-H)
Present in some neuroendocrine cells, reportedly identified in intrinsic neuron-like cells in mammalian gonads.
1 : 4000
A neurosecretory-granule membrane-associated protein implicated in the pre-synaptic or secretory vesicle genesis and recycling
1 : 4000
Abcam, Plc., USA
Tryptophan Hydroxylase (TPH)
First and rate-limiting enzyme of the indolamine neurotransmitter synthesis pathway, with the conversion of tryptophan to L-5-hydroxytryptophan
1 : 4000
Tyrosine Hydroxylase (TH)
First and rate-limiting enzyme for catecholamine neurotransmitter synthesis, with the conversion of L-tyrosine to L-3,4-dihydrophenylalanine (DOPA)
1 : 4000
1 : 8000
Co-localization of neuronal/neuroendocrine markers was recorded using double immunofluorescence labelling following the protocols for each marker under the conditions described for single immunocytochemical procedures, sequentially. A 30-min blocking step with ‘cold avidin’ (RT; Vector Laboratories, USA) was added prior to the secondary antibody incubation. Immunoreactivity was visualized using fluorescein- or Texas Red-conjugated avidin (1 : 200, 60 min at RT; Vector Laboratories, USA). Once thoroughly washed in PB, slices were counter-stained with 4′,6-diamidino-2-phenylindole (DAPI, 1 μg mL−1) for 30 s at RT, mounted on gelatinized glass slides and covered using DAKO fluorescence mounting medium (Dako, USA) for observation and image recording. Immunofluorescence samples were observed under a Disc Scanning Unit-coupled Olympus BX51 DSU (Olympus, USA) and images were recorded using the MBF bioscience stereo investigator® software (MicroBrightField, Inc., USA). In both immunocytochemistry and immunofluorescence, negative controls were processed simultaneously but the incubation with the primary antibodies was omitted (Supporting Information Fig. S1).
Transmission electron microscopy
Dissected ovaries were washed in 0.9% saline solution and sectioned transversely (1.0 mm width). Tissue samples were then fixed in 4% paraformaldehyde and 2.5% glutaraldehyde in sodium cacodylate buffer (0.1 m, pH 7.4) for 3 h at 4 °C. After three 10-min washes in PB, samples were post-fixed in 1.0% osmium tetroxide (Polysciences, Inc., USA) in PB for 2 h at RT, followed by two washes in PB for 10 min. Dehydration was carried out in 70, 95 and 100% ethanol and 100% propylene oxide (twice each step for 10 min) followed by a pre-embedding incubation in a 1 : 1 mixture of embedding medium (EMBed-812, Electron Microscopy Sciences, USA, prepared according to the manufacturer's guidelines) and propylene oxide, overnight at RT. Embedding was done by incubating in EMBed-812 medium for 24 h at 60 °C. Semi-thin and ultra-thin sections were obtained in a Reichert-Jung Ultracut E ultra-microtome (USA). Semi-thin sections were mounted on a glass slide and stained with toluidine blue. Once the regions of interest were identified, ultra-thin sections were obtained and observed, and images were photographed using an M-10 electron microscope (Carl Zeiss NTS GmbH, Germany).
Heavy-chain neurofilament (NF-H) and synaptophysin (SYP) immunoreactivity was observed in a population of small, single, isolated cells distributed throughout the cortical and medullar interstitial space of the immature fowl ovary. At the ultrastructural level, these cells displayed small to medium-sized dense-core secretory granules. These cells also showed positive immunoreactivity for chromogranin-A (ChgA), as well as the enzymes involved in monoamine biosynthesis, namely tyrosine hydroxylase (TH), L-DOPA decarboxylase (DDC), dopamine ß-hydroxylase (DBH) and tryptophan hydroxylase (TPH). All these features are consistent with their NE lineage (Fig. 1).
To determine whether the identified NE cells constituted a single or a heterogeneous population, we conducted immunofluorescence analyses, which revealed that almost all cells displayed positive co-localization of ChgA, NF-H and SYP (Fig. 2). These cells also possessed the enzymes required for catecholamine and serotonin (5-HT) biosynthesis (Figs 3 and 4, respectively). Interestingly, both pathways co-localized in the same cellular population (Fig. 5). Finally, the monoamine biosynthesis enzymes were co-expressed with SYP and ChgA (Figs 6 and 7, respectively).
Neuroendocrine cells constitute a heterogeneous population distributed in most organs of the vertebrate body; however, it has been unclear whether these cells are found in the ovaries. Although the presence of neuron-like cells in the immature ostrich ovary was reported previously (Kimaro & Madekurozwa, 2006; Madekurozwa, 2008), the identity of these cells could not be conclusive, as neuron specific enolase and protein gene product 9.5 are not exclusively neuronal markers but may also be present in cells of neuroendocrine lineage (Perez et al. 1990; Martín et al. 2000; Portela-Gomes et al. 2004). Here, we describe ovarian interstitial cells that were positively labelled for NF-H, SYP, ChgA, TH, TPH, DDC and DBH, as well as displaying cytoplasmic dense-core secretory granules. Taken together, the cellular features observed in immature chick ovaries are common traits of the NE phenotype (Pearse, 1986; Flatmark, 2000; Toni, 2004), and suggest these ovarian cells belong to the NE lineage. Hence, our observations differ from previous reports claiming the existence of neuron-like cells in bird ovaries (Kimaro & Madekurozwa, 2006; Madekurozwa, 2008). Our findings also suggest the need to re-examine the idea of the existence of intrinsic neurons in mammalian ovaries (Dees et al. 1995; D'Albora & Barcia, 1996; Mayerhofer et al. 1996; Anesetti et al. 2001; D'Albora et al. 2002), given that the phenotypical characterization supporting such claims might also be incomplete.
In our study, all the enzymes involved in catecholamine and indolamine biosynthesis were co-localized within NE cells. This could mean that NE cells of the domestic fowl ovary have the ability to produce, and possibly release, both monoamine groups. In support of this possibility, there is an increasing body of evidence that NE cells synthesize and release catecholamines and/or serotonin (Pearse, 1986; Flatmark, 2000; Li et al. 2011). However, it is also possible that the NE cells of the fowl immature ovary present this dual potential because they have not yet committed to a differentiated state, as has been shown for mammalian peripheral adrenergic and cholinergic neurons during early stages of development (von Bartheld & Rubel, 1989; Landis, 1990). Future ontological as well as comparative studies could address these possibilities.
The presence of NE cells in the immature fowl ovary raises interesting questions regarding their function and embryonic origin. It is tempting to hypothesize that their secretory products might regulate early ovarian histogenesis directly, as has been shown for the development of the respiratory tract (Cutz, 1982; Pan et al. 2004, 2006; Evsyukova, 2006), or indirectly, by modifying ovarian steroid secretion (Méndez-Herrera et al. 1993). In relation to their possible origin, all NE cells were originally considered to be derivatives of the neural crest, although it is now known that NE cells of different organ systems are not specified in the same germ layers of the embryo. Regarding the gonads, the proposed neural crest origin of ‘ovarian neurons’ (Dees et al. 2006) contrasts with evidence that testicular Leydig cells, displaying a neuroendocrine phenotype (Davidoff et al. 1993, 2005), arise from the local vascular smooth muscle cells and pericytes (Davidoff et al. 2004). Future studies should explore whether NE cells in the fowl ovary originate from local or distant embryonic tissues.
The authors thank María Verónica Gasca Ramírez and Patricia Bizarro Nevares for their technical support. The authors also thank Tamil Kendall, Doctoral Candidate, University of British Columbia, for proofreading the present manuscript. This work was funded by the National Council of Science and Technology (Consejo Nacional de Ciencia y Tecnología, CONACyT), grant 82879, and the Program to Support Research and Technological Innovation of the National Autonomous University of Mexico (Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica –UNAM), grants IN203912-3 and IN215208.
Pablo G. Hofmann: Designed the study plan and carried out the tissue sampling, processing, and the immunodetection and co-localization studies of the selected neuroendocrine markers. Drafted and edited the manuscript.
Armida Báez-Saldaña: Participated in the study design, interpretation of the data and discussion of the results.
Teresa Fortoul-Van-Der-Goes: Carried out the observation and image recording with electron microscopy for the identification of the representative cells of neuroendocrine lineage at an ultrastructural level.
Margarita González-del-Pliego: Conducted preliminary observations of a cellular population immunoreactive for some neuropeptides in the immature chick ovary. These observations motivated the present study.
Gabriel Gutiérrez-Ospína: Evaluated and revised the study design and supervised the execution of the research. Instrumental in the revision and editing of the submitted manuscript.