Cell Tissue Res (1991) 263:529-534
Cell and Tissue Research 9 Springer-Verlag 1991
Ultrastructure of melatonin-responsive cells in the ovine pars tuberalis P.J. Morgan, T.P. King, W. Lawson, D. Slater, and G. Davidson Rowett Research Institute, Greenburn Road, Bucksburn, Aberdeen AB2 9SB, UK Accepted October 28, 1990
Summary. Functional receptors for melatonin have been localized and characterized on the pars tuberalis (PT) of a number of mammalian species, but the cell-type responsive to melatonin is unknown. The ultrastructure of the ovine pars tuberalis has been examined and these findings correlated with the functional response of the gland to melatonin. This study revealed that two secretory cell types predominate in the ovine PT, which differ in the abundance of dense-core granules. The most abundant of the cells are either agranular or very sparsely granulated and represent 90% of the total population, with the remaining 10% being composed of cells with abundant dense-core vesicles. Few follicular cells were observed. This ratio of secretory cell-types persisted in primary culture, with the two types non-separable by Percoll gradient centrifugation. Using forskolin, as a non-specific stimulant of adenylate cyclase, melatonin was shown to inhibit the formation of cyclic A M P by 80-90% in cells both before and after Percoll centrifugation. The results demonstrate that the agranular secretory cells of the ovine pars tuberalis are the melatonin responsive cell-type of this gland. Key words: Melatonin - Pituitary gland, pars tuberalis - Secretory cells - Cyclic A M P - Ultrastructure - Cell culture - Sheep
tonin (Morgan and Williams 1989), One of these sites, the pars tuberalis (PT) o f the pituitary, has been studied in some detail and provides a useful model system on which to study the mode of action of melatonin. Ultrastructural studies have revealed that the mammalian PT is composed of at least two cell-types, nonglandular (follicular) and glandular (secretory) (Dellman et al. 1974). The latter can be further sub-divided into secretory cells specific to the PT and glandular cells immunocytochemically similar to those of the pars distalis (PD), but the cellular composition of the PT appears to be species specific (Gross 1984; Stoeckel and Porte 1984). Intriguingly it has been shown in the Djungarian hamster, Phodopus sungorus, that both ultrastructural and immunohistochemical changes in the secretory cells can be correlated with changes in photoperiod (Wittkowski et al. 1984, 1988). In the sheep, Ovis ovis, it has been shown that the melatonin receptors located on the PT are membrane bound and linked through an inhibitory G-protein to the inhibition of adenylate cyclase, but the cell-type through which melatonin acts is unknown (Morgan et al. 1989a-c). The purpose of this study was to identify the cell-type responsive to melatonin by examining the ultrastructure of the ovine PT and correlating a functional response of melatonin.
Materials and methods The temporal synchronization of the breeding period with the appropriate season in periodically reproductive animals is driven by photoperiod (Lincoln and Short 1980). Melatonin provides the chemical translation of this stimulus through a proportional representation of the scotophase (Morgan and Williams 1989). Central high affinity receptors for melatonin have been identified in both the brain and pituitary, but as yet it is not possible to ascribe specific sites to particular actions of melaOffprint requests to: P.J. Morgan
For all experiments ovine PT were collected from sheep, Ovis ovis, killed at a local abattoir. PT tissue samples were prepared for examination by electron microscopy by fixation in 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.3, followed by 4 washes in phosphate buffer. PT were postfixed in 0.2% osmium tetroxide in phosphate buffer for 15 h, washed in distilled water and dehydrated through a graded series of alcohols. The tissue was cleared in propylene oxide and embedded in Araldite resin. Ultra-thin sections were cut and stained in 5% uranyl acetate followed by lead citrate and examined in a Jeol 1200 EXB. Dispersed cells were prepared as described above, except that they were embedded in 1.5% agarose in water prior to post fixation. For light microscopy, 1 gin-thick sections were cut and stained with toluidine blue. Cells
530 from primary culture were examined by phase-contrast microscopy using a Leitz Ortholux microscope. Primary cultures of ovine PT cells were prepared as described previously (Morgan et al. 1989a). Routinely 40 PT would be collected for the preparation of one batch of cells, providing ca. 120170 x 106 cells (70-80% viable) after enzymatic digestion. These cells were seeded into T-75 flasks (Sterilin, UK) and incubated overnight in a CO2 incubator (5% CO2) at 36.5 ~ C. On the second day the cells were harvested using a rubber policeman (50% recovery), and these used in all experiments. Cell viability was assessed by their ability to exclude trypan blue. Density gradient centrifugation of dispersed cells was performed on a discontinuous Percoll (Sigma Chemicals, Poole, Dorset, UK) gradient. Stock Percoll was diluted 1:10 with Hanks' balanced salt solution (HBSS, Gibco, Paisley, Scotland) and this was then diluted to different percentage strengths of Percoll with HBSS. Starting with 90% Percoll, 1 ml aliquots of decreasing strength (10% increments) were layered into a 12 ml conical centrifuge tube (Sterilin, UK) up to 20% Percoll. Cells were layered above the 20% Percoll in 1 ml of HBSS. The tube was then centrifuged at 400 g for 15 rain at 4 ~ C. The Percoll, cells and solutions were kept on ice throughout the preparation procedure. For the measurement of cyclic A M P production 250000 cells were incubated in 250 gl of Dulbecco's modified Eagle's medium (DMEM, Gibco) containing the required concentration of drug(s) at 37 ~ C for 10 min. Melatonin and forskolin (both obtained from Sigma Chemical Co.) were used as final concentrations of 10 nM and i gM respectively. An incubation was started by the addition of cells in a volume of 100 ~tl to each tube, and terminated by placing tubes in a boiling water bath for 5 rain followed by quenching on ice. Complete cell lysis was ensured by sonication; then
the tubes were centrifuged at 12000 g for 2 rain, the supernatants removed to clean tubes and stored at - 2 0 ~ C until assay. Cyclic AMP levels in samples were assayed in duplicate by acetylation RIA using a validated antibody against cyclic AMP (anti-cAMP serum no. 388 from D.C. Klein/A.K. Ho N.I.H.) and the radioligand 12SI-succinyl cyclic AMP-tyrosyl methyl ester, prepared by the method of Steiner (1979) and purified by HPLC. For assay 100 gl of sample or standard were acetylated by the addition of 5 pl of acetic anhydride/triethylamine (1 : 2, v: v), followed by the addition of 100 gl of tracer (10000 cpm) and 100 gl of antibody. Assay buffer was 0.05 M acetate buffer (pH 5.8). After overnight incubation at 4 ~ C bound and free tracer were separated by the addition of 100 gl of BSA (10%) and 2 ml of ice-cold ethanol followed by centrifugation at 1800 g for 5 min at 4 ~ C. The supernatant was decanted and the pellet counted by gamma counting. Both intra- and inter-assay coefficients of variation were below 10%. Statistical comparisons were made using Student's t-test.
Table 1. Count of secretory cell-types in ovine pars tuberalis a
Intact PT Band 4 cells
Granular
Agranular (A)
% (A)
114 85
753 811
87 90
a Data from transmission electron microscope sections cut from 10 blocks of intact PT and 10 blocks of Percoll band 4 cell suspensions
Fig. 1. a Light micrograph of the ovine PT showing the columnar arrangement of cells, b Electron micrograph of agranular secretory cells of the ovine PT. R E R Rough endoplasmic reticulum; G Golgi apparatus, a x 310. Bar: 25 pm; b x 5000. Bar: 2 g m
531 120
Table 2. Separation of ovine PT cells on Percoll Band Distance from meniscus (mm)
% Percoll
1 2 3
20 30 40
4
9 18 24
31
50
5
39
60
6
55
80
Observation
Mostly debris with a few dead cells Debris with mainly dead cells Mix of small and medium size cells of reasonable viability (ca. 60-70%) Mainly small cells in good condition (viability ca. 90%) Band not always seen, few cells and debris Few cells, mainly rbc's
Results The ovine P T is c o m p o s e d o f two secretory cell-types, which differ in the a b u n d a n c e o f dense-core granules. The first o f these glandular or secretory cell-types is approximately 12 g m in diameter and is characterized by the presence o f a b u n d a n t dense-core granules ( D C G ) o f diameter 200 nm. This cell-type constitutes only 1015% o f the total p o p u l a t i o n o f secretory cells in the ovine P T (Table 1). The remaining secretory cells are characterized by the absence or extremely low abundance o f dense-core granules (Table 1). These cells, which are also ca. 12 ~tm in diameter are s h o w n at the light-microscope level a r r a n g e d in clusters f o r m i n g a colu m n a r configuration (Fig. 1 a). T h e y are characterized by a large nucleus relative to the volume o f the cytoplasm, well-developed r o u g h endoplasmic reticulum ( R E R ) , Golgi a p p a r a t u s and a b u n d a n t m i t o c h o n d r i a (Fig. I b). These cells are also well vascularized. Follicular cells were n o t f o u n d to be abundant.
I
100-
g ~E "b.
80
E
60
/ / / / / /
o.
40 C) d )C)
c)
20
/
0
C F NS
F/M
C F
F/M
Band 2
C V F/M Band 3
C
F/M'
Bond 4
Fig. 2. Responsiveness of ovine PT cells to melatonin before and after Percoll separation, measured as the ability of melatonin to inhibit forskolin-stimulated cyclic AMP production per million viable cells. C non-stimulated cells; F cells stimulated with forskolin (1 gM); F/M cells stimulated with forskolin (1 pM) in the presence of melatonin (10 nM). NS are cells not separated on Percoll, whereas bands 2, 3, and 4 correspond to the cells after Percoll separation described in Table 2. The data are the means (_+ SEM) of triplicate measurements from a single experiment, but the experiment was repeated three times with identical results, In all cases the melatonin significantly inhibits forskolin-stimulated cyclic AMP production (P
Cell and Tissue Research 9 Springer-Verlag 1991
Ultrastructure of melatonin-responsive cells in the ovine pars tuberalis P.J. Morgan, T.P. King, W. Lawson, D. Slater, and G. Davidson Rowett Research Institute, Greenburn Road, Bucksburn, Aberdeen AB2 9SB, UK Accepted October 28, 1990
Summary. Functional receptors for melatonin have been localized and characterized on the pars tuberalis (PT) of a number of mammalian species, but the cell-type responsive to melatonin is unknown. The ultrastructure of the ovine pars tuberalis has been examined and these findings correlated with the functional response of the gland to melatonin. This study revealed that two secretory cell types predominate in the ovine PT, which differ in the abundance of dense-core granules. The most abundant of the cells are either agranular or very sparsely granulated and represent 90% of the total population, with the remaining 10% being composed of cells with abundant dense-core vesicles. Few follicular cells were observed. This ratio of secretory cell-types persisted in primary culture, with the two types non-separable by Percoll gradient centrifugation. Using forskolin, as a non-specific stimulant of adenylate cyclase, melatonin was shown to inhibit the formation of cyclic A M P by 80-90% in cells both before and after Percoll centrifugation. The results demonstrate that the agranular secretory cells of the ovine pars tuberalis are the melatonin responsive cell-type of this gland. Key words: Melatonin - Pituitary gland, pars tuberalis - Secretory cells - Cyclic A M P - Ultrastructure - Cell culture - Sheep
tonin (Morgan and Williams 1989), One of these sites, the pars tuberalis (PT) o f the pituitary, has been studied in some detail and provides a useful model system on which to study the mode of action of melatonin. Ultrastructural studies have revealed that the mammalian PT is composed of at least two cell-types, nonglandular (follicular) and glandular (secretory) (Dellman et al. 1974). The latter can be further sub-divided into secretory cells specific to the PT and glandular cells immunocytochemically similar to those of the pars distalis (PD), but the cellular composition of the PT appears to be species specific (Gross 1984; Stoeckel and Porte 1984). Intriguingly it has been shown in the Djungarian hamster, Phodopus sungorus, that both ultrastructural and immunohistochemical changes in the secretory cells can be correlated with changes in photoperiod (Wittkowski et al. 1984, 1988). In the sheep, Ovis ovis, it has been shown that the melatonin receptors located on the PT are membrane bound and linked through an inhibitory G-protein to the inhibition of adenylate cyclase, but the cell-type through which melatonin acts is unknown (Morgan et al. 1989a-c). The purpose of this study was to identify the cell-type responsive to melatonin by examining the ultrastructure of the ovine PT and correlating a functional response of melatonin.
Materials and methods The temporal synchronization of the breeding period with the appropriate season in periodically reproductive animals is driven by photoperiod (Lincoln and Short 1980). Melatonin provides the chemical translation of this stimulus through a proportional representation of the scotophase (Morgan and Williams 1989). Central high affinity receptors for melatonin have been identified in both the brain and pituitary, but as yet it is not possible to ascribe specific sites to particular actions of melaOffprint requests to: P.J. Morgan
For all experiments ovine PT were collected from sheep, Ovis ovis, killed at a local abattoir. PT tissue samples were prepared for examination by electron microscopy by fixation in 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.3, followed by 4 washes in phosphate buffer. PT were postfixed in 0.2% osmium tetroxide in phosphate buffer for 15 h, washed in distilled water and dehydrated through a graded series of alcohols. The tissue was cleared in propylene oxide and embedded in Araldite resin. Ultra-thin sections were cut and stained in 5% uranyl acetate followed by lead citrate and examined in a Jeol 1200 EXB. Dispersed cells were prepared as described above, except that they were embedded in 1.5% agarose in water prior to post fixation. For light microscopy, 1 gin-thick sections were cut and stained with toluidine blue. Cells
530 from primary culture were examined by phase-contrast microscopy using a Leitz Ortholux microscope. Primary cultures of ovine PT cells were prepared as described previously (Morgan et al. 1989a). Routinely 40 PT would be collected for the preparation of one batch of cells, providing ca. 120170 x 106 cells (70-80% viable) after enzymatic digestion. These cells were seeded into T-75 flasks (Sterilin, UK) and incubated overnight in a CO2 incubator (5% CO2) at 36.5 ~ C. On the second day the cells were harvested using a rubber policeman (50% recovery), and these used in all experiments. Cell viability was assessed by their ability to exclude trypan blue. Density gradient centrifugation of dispersed cells was performed on a discontinuous Percoll (Sigma Chemicals, Poole, Dorset, UK) gradient. Stock Percoll was diluted 1:10 with Hanks' balanced salt solution (HBSS, Gibco, Paisley, Scotland) and this was then diluted to different percentage strengths of Percoll with HBSS. Starting with 90% Percoll, 1 ml aliquots of decreasing strength (10% increments) were layered into a 12 ml conical centrifuge tube (Sterilin, UK) up to 20% Percoll. Cells were layered above the 20% Percoll in 1 ml of HBSS. The tube was then centrifuged at 400 g for 15 rain at 4 ~ C. The Percoll, cells and solutions were kept on ice throughout the preparation procedure. For the measurement of cyclic A M P production 250000 cells were incubated in 250 gl of Dulbecco's modified Eagle's medium (DMEM, Gibco) containing the required concentration of drug(s) at 37 ~ C for 10 min. Melatonin and forskolin (both obtained from Sigma Chemical Co.) were used as final concentrations of 10 nM and i gM respectively. An incubation was started by the addition of cells in a volume of 100 ~tl to each tube, and terminated by placing tubes in a boiling water bath for 5 rain followed by quenching on ice. Complete cell lysis was ensured by sonication; then
the tubes were centrifuged at 12000 g for 2 rain, the supernatants removed to clean tubes and stored at - 2 0 ~ C until assay. Cyclic AMP levels in samples were assayed in duplicate by acetylation RIA using a validated antibody against cyclic AMP (anti-cAMP serum no. 388 from D.C. Klein/A.K. Ho N.I.H.) and the radioligand 12SI-succinyl cyclic AMP-tyrosyl methyl ester, prepared by the method of Steiner (1979) and purified by HPLC. For assay 100 gl of sample or standard were acetylated by the addition of 5 pl of acetic anhydride/triethylamine (1 : 2, v: v), followed by the addition of 100 gl of tracer (10000 cpm) and 100 gl of antibody. Assay buffer was 0.05 M acetate buffer (pH 5.8). After overnight incubation at 4 ~ C bound and free tracer were separated by the addition of 100 gl of BSA (10%) and 2 ml of ice-cold ethanol followed by centrifugation at 1800 g for 5 min at 4 ~ C. The supernatant was decanted and the pellet counted by gamma counting. Both intra- and inter-assay coefficients of variation were below 10%. Statistical comparisons were made using Student's t-test.
Table 1. Count of secretory cell-types in ovine pars tuberalis a
Intact PT Band 4 cells
Granular
Agranular (A)
% (A)
114 85
753 811
87 90
a Data from transmission electron microscope sections cut from 10 blocks of intact PT and 10 blocks of Percoll band 4 cell suspensions
Fig. 1. a Light micrograph of the ovine PT showing the columnar arrangement of cells, b Electron micrograph of agranular secretory cells of the ovine PT. R E R Rough endoplasmic reticulum; G Golgi apparatus, a x 310. Bar: 25 pm; b x 5000. Bar: 2 g m
531 120
Table 2. Separation of ovine PT cells on Percoll Band Distance from meniscus (mm)
% Percoll
1 2 3
20 30 40
4
9 18 24
31
50
5
39
60
6
55
80
Observation
Mostly debris with a few dead cells Debris with mainly dead cells Mix of small and medium size cells of reasonable viability (ca. 60-70%) Mainly small cells in good condition (viability ca. 90%) Band not always seen, few cells and debris Few cells, mainly rbc's
Results The ovine P T is c o m p o s e d o f two secretory cell-types, which differ in the a b u n d a n c e o f dense-core granules. The first o f these glandular or secretory cell-types is approximately 12 g m in diameter and is characterized by the presence o f a b u n d a n t dense-core granules ( D C G ) o f diameter 200 nm. This cell-type constitutes only 1015% o f the total p o p u l a t i o n o f secretory cells in the ovine P T (Table 1). The remaining secretory cells are characterized by the absence or extremely low abundance o f dense-core granules (Table 1). These cells, which are also ca. 12 ~tm in diameter are s h o w n at the light-microscope level a r r a n g e d in clusters f o r m i n g a colu m n a r configuration (Fig. 1 a). T h e y are characterized by a large nucleus relative to the volume o f the cytoplasm, well-developed r o u g h endoplasmic reticulum ( R E R ) , Golgi a p p a r a t u s and a b u n d a n t m i t o c h o n d r i a (Fig. I b). These cells are also well vascularized. Follicular cells were n o t f o u n d to be abundant.
I
100-
g ~E "b.
80
E
60
/ / / / / /
o.
40 C) d )C)
c)
20
/
0
C F NS
F/M
C F
F/M
Band 2
C V F/M Band 3
C
F/M'
Bond 4
Fig. 2. Responsiveness of ovine PT cells to melatonin before and after Percoll separation, measured as the ability of melatonin to inhibit forskolin-stimulated cyclic AMP production per million viable cells. C non-stimulated cells; F cells stimulated with forskolin (1 gM); F/M cells stimulated with forskolin (1 pM) in the presence of melatonin (10 nM). NS are cells not separated on Percoll, whereas bands 2, 3, and 4 correspond to the cells after Percoll separation described in Table 2. The data are the means (_+ SEM) of triplicate measurements from a single experiment, but the experiment was repeated three times with identical results, In all cases the melatonin significantly inhibits forskolin-stimulated cyclic AMP production (P
