Endocrinology, doi:10.1210/en.2006-1644
Endocrinology Vol. 148, No. 8 3863-3874
Copyright © 2007 by The Endocrine Society
Cell Membrane Structures during Exocytosis
Pascale Savigny,
John Evans and
Kathryn M. McGrath
MacDiarmid Institute for Advanced Materials and Nanotechnology, School of Chemical and Physical Sciences (P.S., K.M.M.), Victoria University of Wellington, 6012 Wellington, New Zealand; and Department of Obstetrics and Gynaecology and Centre for Neuroendocrinology (J.E.), Christchurch School of Medicine and Health Sciences, University of Otago, 8140 Christchurch, New Zealand
Address all correspondence and requests for reprints to: Kathryn McGrath, MacDiarmid Institute for Advanced Materials and Nanotechnology, School of Chemical and Physical Sciences, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand. E-mail: kathryn.mcgrath{at}vuw.ac.nz.
 |
Abstract
|
|---|
Exocytosis is a key biological process that controls the neurotransmission and release of hormones from cells. In endocrine cells, hormones are packed into secretory vesicles and released into the extracellular environment via openings in the plasma membrane, a few hundred nanometers wide, which form as a result of fusion of the membranes of the granule and cell. The complex processes and dynamics that result in the formation of the fusion pore, as well as its structure, remain scantly understood. A number of different exocytosis mechanisms have been postulated. Furthermore, the possibility exists that several mechanisms occur simultaneously. We present here an investigation of the cell membrane dynamics during exocytosis in anterior pituitary cells, especially gonadotropes, which secrete LH, a hormone central to ovulation. Gonadotrope enrichment was achieved using immunolabeled magnetic nanobeads. Three complementary imaging techniques were used to realize a fine structure study of the dynamics of the exocytosis-like sites occurring during secretion. Living pituitary and gonadotrope-enriched cells were imaged with atomic force microscopy, as well as cells that had been fixed to obtain better resolution. Atomic force microscopy, along with scanning and transmission electron microscopy, studies of these cells revealed that there are at least two different site configurations: simple single fusion pores and a complex association of pores consisting of a simple primary site combined with secondary attachments.
|
Introduction
|
|---|
SECRETION OF COMPOUNDS from within a cell to its exterior is a process that is fundamental to the exchange of biochemical information between tissues and cells, and as such is at the center of biological functioning. One of the most common methods by which the delivery of intracellular compounds, such as proteins, to the extracellular space occurs is exocytosis.
However, the details of exocytosis are unclear, and a number of models have been proposed since the first was developed 50 yr ago (1). To date, three main mechanisms remain in contention ("kiss and collapse," "kiss and run," and "kiss and stay"), each term describing different secretory vesicle behavior after release of its content.
One of the impediments to understanding the process of exocytosis has been a lack of methods of visualizing sequential steps at a single site. Electron micrographs, using both scanning electron microscopy (SEM) and transmission electron microscopy (TEM), have provided snapshots that have been interpreted in a number of ways and aligned with one or more of the models.
In this project the atomic force microscope (AFM) has been applied to the investigation of pituitary cells to examine external structures of the cell membrane. The AFM, invented in 1986, scans a molecular tip, which is attached to a cantilevered spring, over the surface of a specimen. The deflections of the tip, caused by the interaction forces between the atoms of the tip and the specimen, are transduced to generate an image. The AFM can provide three-dimensional (3-D) images in nanoscale resolution. The tip is able to obtain signals in fluid and, thus, can image biological cells in culture, in real-time. The potential of the AFM in biology was recognized some time ago (2), and several studies on whole cells have produced important results on a variety of cell functions (3, 4). This study used the AFM, in association with the TEM and SEM, to consider in particular gonadotropes, cells of the reproductive axis that synthesize and release LH and FSH.
It is well accepted that regulated secretion in response to biochemical stimuli is at the core of homeostatic and reactive mechanisms in the body. As the understanding of endocrinology and neurochemistry increased, the details of cellular mechanisms of secretion accumulated in parallel. The delineation of the processes associated with exocytosis is important to completely define both physiology and pathophysiology.
In this study characteristics of pores were noted in a culture of dispersed mixed pituitary cells and cells in a population enriched with gonadotropes. Both simple and complex structures were observed.
|
Materials and Methods
|
|---|
Cell culture
Pituitary glands were collected from adult female Sprague Dawley rats. The anterior pituitary was placed in DMEM with GlutaMaxI (Invitrogen Corp., Paisley, UK) with high glucose, 3.6 g/liter HEPES, 3.0 g/liter BSA, and penicillin/streptomycin (dispersion medium). The gland was digested using trypsin, followed by addition of trypsin inhibitor. Dispersion was completed by trituration of the sample. Cells were pelleted by centrifugation, isolated, and resuspended in approximately 2 ml dispersion medium containing 300 pg/ml of estradiol. Cells were then plated and incubated at 37 C in 95/5 air/CO2. After approximately 3 h, fetal bovine serum was introduced to a final concentration of 10%, and the cells were incubated for at least overnight at 37 C.
LH-release stimulation
Cultured cells were washed in medium 199 with Earle’s salts with L-glutamine, containing 2.20 g/liter sodium bicarbonate, 5.95 g/liter HEPES, 1.0 g/liter BSA and penicillin/streptomycin, and 300 pg/ml of estradiol (incubation medium) for 1 h before stimulation. Preincubation of the cells in estradiol provides an estrogenized proestrus-like environment. Thus, although cells were from randomly cycling rats, they were imaged in a similar hormonal milieu. The supernatant was discarded, and the cells were incubated in medium containing either GnRH, or potassium chloride or no secretagogue.
At the end of the incubation period, the supernatant was collected for RIA, the cells were collected for magnetic separation, or the cells were fixed for immunohistochemistry or SEM. The cells were also directly imaged by AFM during stimulation.
Gonadotrope enrichment
Gonadotropes comprise approximately 10% of the native pituitary cell population. Using a magnetic separation method, based on a modification of a previously used procedure (5), we have achieved at least a 5 times gonadotrope enrichment in the sample. To date, the number of cells retrieved in the gonadotrope-enriched fraction remains low.
Cells were dispersed as described previously and plated onto a sterile glass Petri dish to facilitate subsequent detachment. Cells were incubated overnight at 37 C in 10% fetal bovine serum/DMEM medium containing estradiol (300 pg/ml), detached from the Petri dish, and transferred with supernatant into a sterile tube. To maximize cell retrieval, a trypsin-based solution was used to facilitate detachment. The cell suspension was centrifuged at 1500 rpm.
Cells were washed in incubation medium at 37 C, rewashed and resuspended in sterile, 50 mmol/liter KCl in incubation medium, at 37 C, and then at room temperature. The use of K+-induced stimulation of hormone has been used previously in a gonadotrope-enrichment process using magnetic beads (5).
After stimulation, cells were pelleted and resuspended in 3 ml of a 1/300 anti-LH solution in degassed buffer solution [PBS (pH 7.2), 0.5% BSA, and 2 mM EDTA]. Cells were incubated with the antibody to LH (raised in rabbits) [National Hormone and Pituitary Program (NHPP) and A. F. Parlow] at 4 C, washed, and resuspended in buffer solution. Twenty microliters of microbeads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) coupled to goat antirabbit IgG were added. The whole mixture was mixed and incubated at 4 C, and washed with buffer solution. Cells and beads were finally dispersed in 500 µl buffer solution. The beads have a ferrite core and polysaccharide coating. The mix was loaded onto a column that was placed within a U-shaped magnet. Elution was achieved using the buffer solution. Those cells not containing magnetic beads pass through the column, whereas bead-containing cells (gonadotropes) remain trapped. Recovery of this gonadotrope-enriched fraction was achieved by elution subsequent to removing the magnet. Immediately after separation, each collected fraction was centrifuged, and the cells were dispersed in incubation medium. Cells were plated and incubated at 37 C.
RIA
Supernatant was collected after cells had been submitted to the appropriate experimental protocol and stored at –20 C until assay. Antibody to rat LH (NHPP and A. F. Parlow) was then added to sample aliquots, together with radioiodinated LH (NHPP and A. F. Parlow). Secondary antibody was added, and polyethylene glycol was then added to separate antibody bound and free fractions.
Immunohistochemistry
Immunohistochemistry was performed on cells that had been previously plated on chamber slides and cultured for at least 1 d. Before the experimental incubation, cells were washed with incubation medium for 1 h at 37 C. Cells were stimulated with GnRH or KCl as described previously. Subsequent to stimulation, cells were fixed in 2.5% glutaraldehyde for 1 h at room temperature. After fixation, cells were washed in phosphate-buffered sodium chloride containing 18 mg/ml NaCl (PBN). Cells were then immersed in 0.3% H2O2, followed by washing in PBN and incubation for 1 h in the presence of a blocking solution of Tris-HCl containing 3% BSA. Anti-LH (1/3000 diluted in 0.3% BSA in PBS; NHPP and A. F. Parlow) was added and incubated for 1 h at 37 C or overnight at 4 C. The cells were then washed with PBN. The secondary antibody (antirabbit IgG 1/3000 diluted in 0.3% BSA in PBS) was introduced into the chambers and incubated at room temperature for 1 h, followed by washing with PBN. Staining proceeded using alkaline phosphatase Vectorstain kit (Vector Laboratories, Burlingame, CA). Cells were finally washed with PBN and placed in distilled water. The chamber was then detached from the slide and the sample dehydrated by successive immersion into graded ethanol concentration solutions (50%, 70%, 90%, and 100%), followed by xylene before mounting under a coverslip. The percentage of stained cells was counted.
AFM
All AFM experiments were performed using a Bioscope from Digital Instruments, Veeco Metrology Group (Santa Barbara, CA), mounted on an inverted microscope (Olympus IX70; Olympus, Hamburg, Germany). Cells were plated onto Cell-Tak (BD, Franklin Lakes, NJ) precoated, plastic 35-mm diameter Petri dishes and incubated for approximately 2 d, enabling high-resolution images to be obtained. Depending on the type of experiment realized, cells were fixed, using 2.5% glutaraldehyde solution, or not before AFM imaging All images were obtained using tapping mode. Silicon nitride cantilevers with a nominal spring constant of 0.06 N/m, cantilever length 200 µm, and resonance frequency in fluid of 30–100 kHz were used. Typically, a scan rate of between 0.3 and 0.5 Hz was used.
SEM
All SEM micrographs were obtained using a JEOL 6500F high-resolution instrument (JEOL Ltd., Tokyo, Japan); accelerating voltages of 10–15 kV were used. Anterior pituitary cells or gonadotropes were plated onto 8-mm glass coverslips, previously coated with Cell-Tak, and incubated for 2 d. Cells were either unstimulated or had been exposed to GnRH 10–7 mol/liter, just before fixation. The fixation procedure was identical with that used for AFM experiments. Cell dehydration was by graded ethanol concentrations as described previously. Cells were then washed twice in 100% acetone and finally subjected to critical point drying. Samples were mounted using carbon tape onto SEM stubs and coated with gold, 4 nm, followed by a 7-nm carbon coating.
TEM
A JEOL 2010 TEM (JEOL Ltd.) operating at an accelerating voltage of 200 kV was used. Both whole anterior pituitary gland tissue and dispersed pituitary cells incubated with magnetic beads were investigated. Samples were prepared as follows. Cells were fixed as described previously using a combination glutaraldehyde/paraformaldehyde solution, followed by several rinses in a sodium cacodylate-trihydrate (CAB)/sucrose buffer solution, to remove glutaraldehyde from the cells. Staining was achieved over a 2- to 3-h period using a 1% osmium tetroxide CAB/sucrose buffer, followed by washing with CAB/sucrose buffer and distilled water several times. The sample was dehydrated using an ethanol gradient, followed by washing twice in 100% propylene oxide, and then left overnight, open to ventilation in a combination 50% resin [Procure 812 (resin) (ProSciTech, Thuringowa, Queensland, Australia), NMA (methyl-5-norbornene-2, 3-dicarboxylic anhydride), DDSA (dodecenyl succinic anhydride), and DMP-30 (2, 4, 6-tridimethylaminomethyl phenol) (catalyst)]/50% propylene oxide mixture. The sample was transferred to 100% resin and left for 4 h. This process was repeated once. Samples were transferred to resin-filled electron microscopy capsules and cured overnight at 80 C. One to 2-µm thick sections were cut using a glass knife and placed on a glass slide. Samples were stained with toluene blue, to localize and identify the specimen cut. Ultrathin sections (60 to 90 nm thick) were obtained by cutting with a diamond knife. These sections were placed on a 200-mesh hexagonal copper grid and subsequently stained with uranyl acetate, rinsed with distilled water, and dried at 40 C. A second staining, using lead citrate, was performed. The sections were then rinsed with cooled preboiled deionized water and dried at 40 C.
|
Results and Discussion
|
|---|
Single and simple fusion pore configuration
The "single and simple fusion pore" configuration term is used to describe the simple depressions found on the cell membrane of stimulated or nonstimulated enriched or not-enriched cells. Pores consistent with this simple pore structure of exocytosis have been widely reported (1). In particular, this type of opening has been associated with gonadotropes, with fusion pore diameters ranging from 100–500 nm, with an increase in their prevalence after stimulation with GnRH 10–7 mol/liter as evidenced from low resolution SEM data (total scan size 
6 µm) (6).
Mixed pituitary cells.
Figure 1
shows SEM images of a simple fusion pore. Such images are commonly obtained for pituitary cells in general, regardless of whether the cells were stimulated or not before fixation, or which stimulation conditions were used (e.g. potassium, 10–7 mol/liter GnRH). It was found that across the sample, fusion pores of quite different size were seen ranging from approximately 50 nm up to more than 500 nm, both between cells and on a single cell. Because of this variability, it was not possible to make any specific conclusions concerning the variation of the fusion pore size with the nature of the stimulation because the size of the fusion pore might also depend on the type of cell and the size of its secretory vesicle. It was often observed that there were more single and simple fusion pores present on cells in basic culture conditions compared with those that had been exposed to secretagogue. Figure 1b reveals a detailed structure of the fusion pore, where a net border is visible around the hole. The protuberant contour might be due to the fusion of the vesicle membrane with the plasma membrane. This border was generally more visible and defined on cells that had flattened and spread during culture such as that shown in Fig. 1a.
It was apparent from the images garnered that the depth of the fusion pores varied, with some appearing to be considerably deeper than others. This might be explained by the difference in the sizes of the docked and fused vesicles, or by the kiss and collapse exocytosis mechanism. Indeed, in this latter case, the membrane of a secretory vesicle that has just released its contents is slowly incorporated into the cell membrane. Therefore, this event would appear as an increasingly shallow fusion pore, eventually seeming to disappear altogether. Favor for the occurrence of this explanation is gained from analysis of TEM micrographs of the anterior pituitary gland whole tissue.
Figure 2 shows a series of TEM micrographs within which a variety of fusion pores are evident. Two pores are present in Fig. 2a, one is quite narrow and deep (400 nm, note that it no longer includes its content), whereas the other is considerably more shallow. The diameter and depth of the pores vary, but in general, their depth is limited to the size of the vesicle itself. In Fig. 2b, a fusion pore with a very narrow opening (60 nm in diameter) is visible, potentially corresponding to the very beginning of the fusion of the two membranes. It is possible to distinguish in the magnified view of this fusion pore the protuberant portion of the plasma membrane around the depression, as already found on the SEM image shown in Fig. 1b
. This type of structure for the fusion pore has been described as a "basket-like morphology" (7). Accordingly, the t-SNARE would be located at the base of the basket, waiting to form a complex with the v-SNARE (soluble N-ethylmaleimide-sensitive fusion protein attachment receptor) of the vesicle membrane, inducing the fusion of the two membranes and, therefore, the extrusion of the plasma membrane outside the cell (as seen in the inset to Fig. 2b).
View larger version (75K):
[in this window]
[in a new window]
|
FIG. 2. TEM pictures of the anterior pituitary gland whole tissue. a and b, Several fusion pores are visible (arrows). Magnification, x6000. Scale bar, 200 nm. Inset magnification, x18000. Note in panel b that the arrow is pointing into the pore as for panel a.
|
|
In addition to the fixed cell systems investigated using SEM and TEM, AFM studies on live pituitary cells were also undertaken. Obtaining images of fusion pores present on live pituitary cells gives access to the dimensions of the depressions, especially to the real depth, which cannot be measured with SEM. Figure 3 shows images of fusion pores in living pituitary cells.
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 3. Three-dimensional AFM images and section analysis of pores found on pituitary cells. a and b, Living cells in control medium and (c) living cells that have been exposed to 10–7 mol/liter GnRH. Horizontal distance measured in sectional analysis is 727.18 nm. Scale, x: 0.2 µm/div; z: 150 nm/div (a); and x: 0.2 µm/div; z: 120.354 nm/div (c).
|
|
From the images, it is revealed that the 3-D form of a simple fusion pore is on average spherical in shape. Sectional analysis of the fusion pore shown in Fig. 3a, found on a pituitary cell in control medium yields a diameter of approximately 700 nm at the opening and a total depth of 70 nm. It is apparent that, in contrast to the undercut features observed in the SEM figures (Fig. 1a), the AFM images reveal smooth depression edges, which could be due to tip convolution issues. Similar analysis of the pore formed on a cell exposed to GnRH (Fig. 3c) gives a depression 400-nm wide and 50-nm deep, showing once again the variety in size of the pores within the pituitary cell sample.
For direct comparison with the SEM images shown in Fig. 1, AFM images were also obtained of fixed cells. Figure 4 is a 3-D image, plus section analysis, of a fusion pore similar to that shown in Fig. 1b. The pore is spherical as for the living cells, and its dimensions (500-nm wide and 100-nm deep) are within the range found for pores on live cells.
View larger version (20K):
[in this window]
[in a new window]
|
FIG. 4. a, Three-dimensional AFM image (scale, x: 0.2 µm/div; and z: 256.2 nm/div) and (b) section analysis (diameter and depth of pore 527.65 and 94.36 nm, respectively) of a pore found on a pituitary cell exposed to 10–7 mol/liter GnRH and then fixed.
|
|
The images obtained from all three techniques are complementary and confirm that the single simple fusion pore configuration is a hole formed by the fusion of the secretory vesicle membrane with the cell membrane itself.
Gonadotrope-enriched samples.
Understanding LH release is central to treatments of certain types of infertility and, possibly, for the development of new types of contraceptives. To gain access to this central tenet of LH release, ideally, pure gonadotrope cells would be investigated. To work toward achieving this goal, we investigated an immunolabeled magnetic nanobead separation method to achieve gonadotrope enhancement in the cell population. Immunostaining was used to monitor the cells, and this coupled with monitoring secretion allowed cell viability and responsiveness to be checked.
From the immunohistochemistry results, it was determined that the control pituitary sample consisted of approximately 12% stained cells, which is consistent with the known proportion of gonadotropes (8, 9). The gonadotrope-depleted fraction obtained from the separation process was shown to have a depleted percentage of stained cells (1–2%). In contrast, in the gonadotrope-enriched fraction, the percentage of stained cells had increased to approximately 50%, indicating a 5 times enrichment in gonadotropes. However, it is possible that the fraction consisted of more than 50% gonadotrope cells because some of the gonadotropes may not have been detected because the level of LH may have been below the immunohistochemical threshold after the separation process. A reduced level of LH may be present due to the preliminary stimulation step or because there was nonspecific induction of release during enrichment.
The enriched fraction was responsive to GnRH stimulation of LH release in a manner similar to that of control pituitary cells. Therefore, conditions optimized for the mixed pituitary cells were used on the enriched samples.
In contrast to the mixed pituitary cell samples, the diameter of the fusion pores imaged with SEM on fixed gonadotrope-enriched samples was relatively constant, around 100 nm and never exceeding 150 nm (Fig. 5). These observations are consistent with the single main type of cell present in the samples after magnetic separation being gonadotropes. It was noted when examining stimulated vs. nonstimulated cell samples that the average size of the fusion pores was relatively similar after the two conditions. On the other hand, it has been suggested that the size of the fusion pores of somatotropes increases from 25–40% after stimulation during secretion (10, 11). For this reason, such findings must be viewed with extreme care, and the best way to characterize a change of size of the depressions after stimulation is to focus on a fusion pore and monitor it in real-time with the AFM. Despite the difficulty involved in this operation, this experiment has been successfully realized, and the results are presented later.
View larger version (111K):
[in this window]
[in a new window]
|
FIG. 5. a and b, SEM images of a fixed cell in a gonadotrope-enriched sample previously exposed to 10–7 mol/liter GnRH.
|
|
In addition to the fusion pores discussed previously, very tiny holes (with a 20- to 40-nm diameter) have also been found (Fig. 5b). These fusion pores might be the result of a mechanism that is part of basal or "constitutive" secretion of LH (12). In these circumstances, boluses of less LH are required, and processes that release only part of the contents of a single granule, such as kiss and run, may be prominent. Full fusion does not occur, and a small pore is formed.
The union of two adjacent fusion pores was a very frequent event (Fig. 6). Indeed, as was often observed and clearly evident in Fig. 5b, vesicles have an area of privilege for the release of their content through the cell membrane. If the size of the pore tends to increase during release (10, 11), two nearby depressions may tend to form one larger fusion pore.
View larger version (109K):
[in this window]
[in a new window]
|
FIG. 6. SEM images of a pituitary cell (a) and a cell from an enriched sample (b) showing the connection of two fusion pores.
|
|
All of the single fusion pores shown previously can illustrate equally well both kiss and collapse (also called full fusion), and kiss and run (also known as transient fusion) processes. These two mechanisms are well recognized and known, with the kiss and run mechanism apparently being more common (13). Indeed, this mechanism does not require the intervention of succeeding endocytotic events to remove the excess membrane that occurs in the kiss and collapse process from the integration of the vesicle membrane into the cell membrane. However, it has been reported that the incorporation of the vesicle membrane into the cell membrane allows the renewal of proteins expressed on the surface of the cell (14), and so it is suggested that both mechanisms coexist. However, the following results give evidence for a fusion pore configuration (as indicated in Fig. 7) that supports the most recently suggested exocytosis mechanism: the kiss and stay mechanism, which involves successive exocytotic events (15, 16, 17, 18).
View larger version (31K):
[in this window]
[in a new window]
|
FIG. 7. Three-dimensional image of a 5 x 5-µm surface area of a fixed cell that has been exposed to 10–7 mol/liter–1 GnRH. a, Two simple fusion pores are visible (arrows), and a complex structure is also present (circle). Scale, x: 1.0 µm/div; z: 256.2 nm/div. b, The section analysis indicates: 1) a simple fusion pore between the green arrowheads, pore diameter 625.0 nm; and 2) the presence of a complex pore structure with a pit, between the two red arrowheads, which itself contains two smaller depressions within the white oval. Diameter of pit is 488.38 nm.
|
|
The "pit and fusion pores" configuration
Structure and dimensions in both pituitary and gonadotrope-enriched cells.
The 3-D AFM image shown in Fig. 7a indicates the existence of two different types of sites within a relatively small area of the cell membrane: two fusion pores that appear as quite deep simple holes, and one depression containing two smaller ones. This last complex structure of a site has been found using each of the three different techniques used here (AFM, SEM, and TEM) and can be described as "a pit and its fusion pores." Cho (10), Jeremic (19), and Schneider (20) et al. have also observed this configuration. Using AFM imaging coupled with immunolabeling, they confirmed that the depressions found on the pits are fusion pores involved in the secretion of hormone for instance [or neurotransmitters (21)]. For pancreatic acinar cells, the average size of the pits was 0.5–2 µm and 100–200 nm for the pores. However, detailed SEM images of such configurations have never before been published, and the following images (Figs. 8 and 9) present different aspects of this configuration. It should be noted that the pit and fusion pores structure is quite rarely found compared with the single fusion pore configuration.
A pit and fusion pores exocytosis-like site configuration on a cultured pituitary cell is shown in Fig. 8. The structure is similar to a simple fusion pore, but in this case, the main hole contains several other smaller depressions. Images of pit and fusion pores taken on flattened areas of cells (during culturing cells attach to the base of the culture vessel, and extend their cytoplasm and flatten their height) possess increased detail and offer a good view of the structure of the site. Such configurations were also observed for round cells (Fig. 9), however, the structure was less obvious. The lack of clarity of the configuration is partly due to the difficulty of imaging internal to the pit using SEM, especially if the pit is not well orientated with respect to the electron beam. Despite this, the pit and fusion pores configuration was observed in both extended and round cells.
The images shown in Fig. 10 offer a remarkable illustration of a pit and its fusion pores. The pit measures approximately 130 nm in diameter and contains four fusion pores, two of them being very obvious and having a diameter of slightly more than 50 nm. The two other pores are smaller in size, with a diameter of 25 nm or less. In general, when using SEM imaging, two to four fusion pores were found per pit, with the pit size varying from less than 100 nm to approximately 500 nm, and the pore diameter varying between 20 and 300 nm in the biggest pits. Most of the pits observed in the enriched cell samples were approximately 100 to 150 nm wide on average.
View larger version (80K):
[in this window]
[in a new window]
|
FIG. 10. a–c, A cell exposed to GnRH, showing a pit and fusion pores exocytosis-like site. The pit has in this case four internal fusion pores (white arrows).
|
|
Similar images of the fixed cells were also obtained using AFM (Fig. 11). These images are magnifications of the pit and fusion pores shown in Fig. 7. One of the depressions is bigger and deeper than the other one, as also seen by the section analysis presented in Fig. 7. This irregularity in the size of the fusion pores contained in the same pit was also seen with SEM imaging.
View larger version (31K):
[in this window]
[in a new window]
|
FIG. 11. AFM images of pit and fusion pores. a, Detailed 3-D AFM image of the pit and fusion pores shown in Fig. 7 (x: 0.2 µm/div; z: 247.32 nm/div). b, Cell membrane of a fixed cell from an enriched population that had been incubated in basal conditions showing several pits, one of which contains three fusion pores (x: 0.5 µm/div; z: 132.708 nm/div).
|
|
Figure 12 shows AFM images depicting several pits and fusion pores on cells separated by the immunomagnetic method, exposed to GnRH and fixed. We have not been able to obtain images of the pit and fusion pore configuration for living cells. This is due to the extreme difficulty of imaging living cells generally, but especially in the case when such fine structure details are being examined. Fusion pores in pits are on average only 30 to 180 nm maximum in diameter, as shown in the section analyses presented in Fig. 12, and such high resolution was not possible for living cells due to the dynamic nature of the cell.
View larger version (49K):
[in this window]
[in a new window]
|
FIG. 12. Two areas on two different fixed cells from an enriched population that had been incubated in basal conditions that exhibited several pits and fusion pores. a, Four pits are highlighted showing many secondary attachment sites (scale, x: 1.0 µm/div; z: 98.532 nm/div). b, Section analysis of highlighted region from panel a, total pit diameter 859.15 nm, individual pore size 182.69 nm. c, Two and 3-D (x: 0.2 µm/div; z: 92.769 nm/div) images. d, Section analysis of two secondary pores with diameters of 31.668 and 63.336 nm, red and green arrowheads, respectively.
|
|
The two general views shown in Fig. 12 illustrate that these pits, which are approximately 0.3–1 µm in diameter, seem relatively frequent in certain areas of the cell membrane. A few pits are evident that do not exhibit fusion pores, and this could be due to the limitation of resolution during AFM imaging, or because no secretion was occurring in these pits. The range of size for both pits and fusion pores is similar in images obtained from both AFM and SEM, though generally dimensions obtained from SEM images are smaller than the corresponding AFM images. This could be due to the additional dehydration process required for SEM imaging.
Similar dimensions were obtained from TEM images of whole anterior pituitary tissue. Figure 13 illustrates in addition to three simple fusion pores (with a diameter of about 180 nm), a large pit of approximately 400 nm in diameter, in which one depression (of 70 nm in diameter and 50 nm in depth) is clearly visible, whereas two others are less well defined. The content that has been released is still noticeable as three different masses that probably initiated from the three depressions. The central mass is distinct from the other two and has probably been released through the central fusion pore that appears clear, whereas the two others are quite diffuse with the secretion possibly still occurring during fixation. In these latter cases, the secretory vesicles were still docked to the cell membrane, explaining why the fusion pore is not very distinct.
View larger version (118K):
[in this window]
[in a new window]
|
FIG. 13. TEM picture showing simple fusion pores and a pit and its fusion pores found on a pituitary cell. An exocytotic event involving a pit and its fusion pores occurs in the free space between two cells (b), whereas the simple fusion pores, in panel a (black arrows), face a blood capillary (BC). a, Whole cell magnification, x3000 (scale bar, 500 nm). b, Part of cell showing pit magnification, x8000 (scale bar, 200 nm).
|
|
Potential origins of, and explanations for, the "pit and fusion pores" configuration.
The pit and fusion pores configuration has been found on cells from the enriched population and also on mixed pituitary cells, and has been verified by AFM, SEM, and TEM analyses. In brief, the pits observed have a maximum diameter of 1 µm and contained two to four fusion pores, measuring between 20 and 300 nm. This arrangement was relatively rare compared with the simple fusion pores that tended to be very frequent. Three different possibilities exist for the development of the pit and fusion pores site.
The first process by which fusion pores may form a pit results from where a balance between exocytotic and endocytotic events occur to maintain a constant cell membrane quantity (18, 22, 23). Furthermore, it has been reported that "GnRH-induced exocytosis, endocytosis and recycling in gonadotropes maintain dynamic equivalence" (24). The endocytotic sites were found to be generally located at the bottom of an exocytotic site and to be involved in the removal of exocytosed membrane. Because the size of the endocytosed vesicles is generally smaller than the secretory granules, the term of micropinocytosis, which is a microscaled form of endocytosis with engulfing and digestion of dissolved substances, has been used (22). The pit and fusion pores configuration conforms with exocytosis involving "full fusion" of the secretory granule into the cell membrane, being directly followed by endocytosis. The average size of the fusion pores (30–180 nm) is similar to the 50-nm diameter proposed for the endocytosed vesicles (18). However, in 1992, it was demonstrated by Monck and Fernandez (25) that transient fusion can occur, providing a possible explanation as to why the number of secretory vesicles remains unchanged after secretion (26). Since this finding, the kiss and run mechanism of exocytosis has become widely accepted, thus suppressing the idea of the necessity of endocytosis after an exocytotic event.
The second mechanism involves the notion of a persistent and continuous fusion pore. The kiss and run mechanism also called transient fusion can be depicted by the pit and fusion pores shown in Fig. 13b, in which the secretory vesicle that has just released its content through the central fusion pore has already left this site and returned deeper into the cell, whereas the depression remains like this until another secretory vesicle docks, fuses, and secretes its content. Therefore, the fusion pore is always present, and only its opening diameter changes during the secretion (7, 11). This implies that all fusion pores located in pits are in fact persistent depressions that increase their diameter while releasing the content of the vesicle docked at their sites.
If this were the case, the depressions would always remain present on the surface of the cell membrane. However, the images shown in Fig. 14, representing still shots of topographical real-time modifications of the cell membrane of a living cell stimulated with 10–7 mol/liter GnRH from the enriched pituitary fraction, show the opposite trend. Here, we observe the total disappearance of the fusion pores rather than their shrinking after secretion. Twenty minutes after the first scan of a pit and its fusion pores (Fig. 14a), the fusion pores on the left-hand side of the pit are no longer present (Fig. 14b), i.e. the diameter is not observed to just decrease, but the pores are seen to totally disappear. This time of opening for a fusion pore (
20 min) correlates well with the fact that pituitary cells release hormones that do not need to be secreted quickly and used immediately. This real-time observation strongly suggests that at least some of the vesicles tend to be incorporated into the cell membrane.
View larger version (55K):
[in this window]
[in a new window]
|
FIG. 14. Three-dimensional AFM images of a pit and fusion pores found on a living cell from an enriched population that was incubated with 10–7 mol/liter GnRH, and imaged (a) 1 h after exposure (scale, x: 0.5 µm/div; z: 100.0 nm/div) and (b) 1 h 20 min after the exposure (scale, x: 0.5 µm/div; z: 150.812 nm/div). The cell was imaged using tapping mode in liquid with a scan rate of 0.2 Hz. Note that some thermal drift has occurred during the 20 min, as is normal in AFM imaging, meaning that the primary pit and fusion pore of interest is displaced to the left in panel b. The green overlay in the section analysis of panel b corresponds to that shown in panel a, accounting for the thermal drift, with the arrow showing the membrane displacement over time.
|
|
A third potential process of exocytosis incorporates two different aspects, including: 1) the creation of the so-called pit, formed via a primary granule fusing with the cell membrane; and 2) the notion of sequential exocytosis involving internal vesicle-vesicle fusion.
Total fusion of the primary vesicle.
Real-time AFM imaging (Fig. 14) revealed that some fusion pores disappear as a function of time, implying the incorporation of the associated vesicles into the cell membrane. Thus, the pit shown in Fig. 13 could result from the fusion of a large primary granule that has released the noticeable large, central mass of molecules. This site is then able to become a privileged cell membrane area for further exocytotic events, explaining the presence of several fusion pores per pit. This correlates well with the relatively large size of the pit seen (180 nm) and the appearance that the vesicles are beginning to be incorporated into the cell membrane. Indeed, the fusion pores are much larger than the depressions found at the bottom of the pit. This mechanism of exocytosis is the kiss and stay mechanism that involves the notion of sequential exocytosis (15, 16, 17, 18). In this version of the secretion process, a long fusion pore opening is observed corresponding to a first granule that fuses with the cell membrane. This original site is then maintained and used as a conduit for further secretion. The observation that the first fused vesicle has a propensity to flatten has been reported previously (17).
SEM images shown in Fig. 15 provide further considerable detail of the external structure of pits that might well be representative of the kiss and stay mechanism. All show the incorporation of a first vesicle into the cell membrane, as well as the presence of other fusion pores at the same site. Such images have not been previously reported. The internal structure of the pit and fusion pores can be extremely complex, with the appearance of cavity networks. In Fig. 15b, a very small fusion pore is evident at the bottom of the pit. Figure 15c is consistent with successive exocytosis events occurring at the same site with full fusion of the first granules because three different levels are clearly visible and indicate the presence of fusion pores at the bottom of previously incorporated vesicles. Integration of the first fused vesicle into the cell membrane may also allow the incorporation or renewal of the signaling molecules required to be expressed by the cell.
View larger version (81K):
[in this window]
[in a new window]
|
FIG. 15. a–c, Detailed SEM images of complex exocytosis sites found on pituitary cells incubated in basal conditions.
|
|
Vesicle-vesicle fusion.
The second proposal concerns the kiss and run mechanism, in which the fusion pore always remains present at the bottom of a pit. If the so-called pit in the previous pit and fusion pores images is actually a vesicle that has previously fused with the cell membrane, then vesicle-vesicle fusion becomes a requisite. Previous studies have shown that vesicle-vesicle fusion can occur (17, 18), a notion supported by fluorescence confocal microscopy and total internal reflection microscopy data from pancreatic acinar cells or islet ß-cells. These techniques allow tracking of the vesicles themselves inside a living cell.
Evidence for such a process occurring here was provided by TEM data. Two examples of potential vesicle connection, which could increase secretion efficiency, are shown in Fig. 16. Figure 16a depicts in particular a fusion pore and the primary granule that is released, as well as several vesicles at different levels that could be involved in sequential exocytosis. This idea is reinforced by the discovery in 1998 of intergranular bridges in anterior pituitary cells that are involved in the vesicle-vesicle fusion (27).
View larger version (109K):
[in this window]
[in a new window]
|
FIG. 16. TEM micrographs showing potential vesicle connections on pituitary tissue (a) and a dispersed pituitary cell (b). Scale bar, 200 nm.
|
|
The micropinocytosis concept, which involves the full fusion of the secretory vesicle membrane, has the benefit of allowing both the renewal of the proteins in the cell membrane and the ingestion by the cell of any useful substance present in the extracellular environment. The depressions found in this investigation have the correct range of sizes to be involved in the formation of endocytotic vesicles, and the structure and sizes of pits and fusion pores found on both fixed and live cells in this project correspond to those reported previously (11).
Although the two previous mechanisms could potentially explain the "pit and fusion pores" configuration, our results are most consistent with the sequential exocytosis mechanism. This mechanism includes the notion of incorporation of the first fused vesicle into the cell membrane at one place that becomes the favored site for further exocytotic events. Vesicle-vesicle fusion subsequently occurs and results in a more efficient secretion, with a saving of energy due to the avoidance of transporting vesicles to the cell membrane, and finally a limited cell membrane deformation.
Conclusions
A protocol for the enrichment of gonadotropes in an anterior pituitary cell sample was developed. The technique used immunolabeled magnetic nanobeads and achieved at least a 5 times enrichment of gonadotropes in the cell population.
The combination of AFM, SEM, and TEM techniques allowed a complete and detailed study of different types of membrane structures. The first and most common type that was observed was the so-called "single and simple fusion pore," whereas the complex structure "pit and fusion pores" was more unusual. These two configurations were detected on both mixed pituitary cells and cells in the enriched population, either fixed or alive, and are believed to be involved in different secretion mechanisms. On the one hand, the kiss and collapse and kiss and run machineries might be responsible for the simple depressions. On the other hand, a sequential exocytosis combined with the incorporation of the first granules into the cell membrane would explain the pit and fusion pores configuration. In this last case, vesicle-vesicle fusion occurs and improves the efficiency of the secretion.
|
Acknowledgments
|
|---|
We thank David Flynn for transmission electron microscopy sample preparation and Maan Alkaisi for discussions. We also thank the National Hormone and Peptide Program for its generous assistance, and A. F. Parlow for supplying immunological reagents.
|
Footnotes
|
|---|
Financial support for this work is gratefully acknowledged from the MacDiarmid Institute and the Canterbury Medical Research Foundation. Ethical approval was obtained from The Victoria University of Wellington and the Christchurch School of Medicine animal ethics committees.
This work was supported by grants from the Centre of Research Excellence, The MacDiarmid Institute for Advanced Materials and Nanotechnology.
Disclosure Summary: The authors have nothing to disclose.
First Published Online May 10, 2007
Abbreviations: AFM, Atomic force microscopy; CAB, cacodylate-trihydrate; 3-D, three-dimensional; div, division; SEM, scanning electron microscopy; TEM, transmission electron microscopy.
Received December 7, 2006.
Accepted for publication May 3, 2007.
|
References
|
|---|
- Fernandez-Peruchena C, Navas S, Montes MA, Alvarez de Toledo G 2005 Fusion pore regulation of transmitter release. Brain Res Brain Res Rev 49:406–415[CrossRef][Medline]
- Lal R, Scott JA 1994 Biological applications of atomic force microscopy. Am J Physiol 266(1 Pt 1):C1–C21
- Alonso JL, Goldmann WH 2003 Feeling the forces: atomic force microscopy in cell biology. Life Sci 72:2553–2560[CrossRef][Medline]
- Frederix PLTM, Hoogenboom BW, Fotiadis D, Muller DJ, Engel A 2004 Atomic force microscopy of biological samples. MRS Bull 29:449–455
- Valenti S, Sarkissian A, Giusti M, Giordano G, Dahl KD 1995 A technique for sorting rat gonadotropes using anti-LH or anti-FSH antibodies covalently attached to magnetic beads. J Neuroendocrinol 7:673–679[CrossRef][Medline]
- Mizuki J, Masumoto N, Tahara M, Fukami K, Mammoto A, Tasaka K, Miyake A 1995 Scanning electron microscope assessment of exocytotic changes in purified gonadotrophs. J Endocrinol 144:193–200[Abstract/Free Full Text]
- Jena BP, Cho S-J, Jeremic A, Stromer MH, Abu-Hamdah R 2003 Structure and composition of the fusion pore. Biophys J 84:1337–1343[Medline]
- Childs GV, Miller BT, Miller WL 1997 Differential effects of inhibin on gonadotropin stores and gonadotropin-releasing hormone binding to pituitary cells from cycling female rats. Endocrinology 138:1577–1584[Abstract/Free Full Text]
- Evans JJ, Youssef AH, Abbas MM, Schwartz J 1999 GnRH and oxytocin have nonidentical effects on the cellular LH response by gonadotrophs at pro-oestrus. J Endocrinol 163:345–351[Abstract]
- Cho S-J, Jeftinija K, Glavaski A, Jeftinija S, Jena BP, Anderson LL 2002 Structure and dynamics of the fusion pores in live GH-secreting cells revealed using atomic force microscopy. Endocrinology 143:1144–1148[Abstract]
- Jena BP 2003 Fusion pore or porosome: structure and dynamics. J Endocrinol 176:169–174[Abstract]
- Farnworth PG 1995 Gonadotropin secretion revisited. How many ways can FSH leave a gonadotroph? J Endocrinol 145:387–395[Abstract/Free Full Text]
- Lindau M, Alvarez de Toledo G 2003 The fusion pore. Biochim Biophys Acta 1641:167–173[Medline]
- Alberts B, Bray D, Lewis J, Raff M, Roberts K, Watson JD 1994 Molecular biology of the cell. 3rd ed. New York: Garland Publishing, Inc.; 477–525
- Allersma MW, Wang L, Axelrod D, Holz RW 2004 Visualization of regulated exocytosis with a granule-membrane probe using total internal reflection microscopy. Mol Biol Cell 15:4658–4668[Abstract/Free Full Text]
- Leung YM, Sheu L, Kwan E, Wang G, Tsushima R, Gaisano H 2002 Visualization of sequential exocytosis in rat pancreatic islet b cells. Biochem Biophys Res Commun 292:980–986[CrossRef][Medline]
- Nemoto T, Kimura R, Ito K, Tachikawa A, Miyashita Y, Iino M, Kasai H 2001 Sequential-replenishment mechanism of exocytosis in pancreatic acini. Nat Cell Biol 3:253–258[CrossRef][Medline]
- Thorn P, Fogarty Kevin E, Parker I 2004 Zymogen granule exocytosis is characterized by long fusion pore openings and preservation of vesicle lipid identity. Proc Natl Acad Sci USA 101:6774–6779[Abstract/Free Full Text]
- Jeremic A, Kelly M, Cho S-J, Stromer MH, Jena BP 2003 Reconstituted fusion pore. Biophys J 85:2035–2043[Medline]
- Schneider SW, Sritharan KC, Geibel JP, Oberleithner H, Jena BP 1997 Surface dynamics in living acinar cells imaged by atomic force microscopy: identification of plasma membrane structures involved in exocytosis. Proc Natl Acad Sci USA 94:316–321[Abstract/Free Full Text]
- Cho WJ, Jeremic A, Rognlien KT, Zhvania MG, Lazrishvili I, Tamar B, Jena BP 2004 Structure, isolation, composition and reconstitution of the neuronal fusion pore. Cell Biol Int 28:699–708[CrossRef][Medline]
- Kurosumi K, Inoue K 1980 Surface pits of typical gonadotrophs and castration cells of the rat anterior pituitary suggestive of exocytosis and micropinocytosis. Arch Histol Jpn 43:373–382[Medline]
- Segawa A, Loffredo F, Puxeddu R, Yamashina S, Testa Riva F, Riva A 1998 Exocytosis in human salivary glands visualized by high-resolution scanning electron microscopy. Cell Tissue Res 291:325–336[CrossRef][Medline]
- Masumoto N, Tasaka K, Mizuki J, Tahara M, Miyake A, Tanizawa O 1993 Dynamics of exocytosis, endocytosis and recycling in single pituitary gonadotropes. Biochem Biophys Res Commun 197:207–213[CrossRef][Medline]
- Monck JR, Fernandez JM 1992 The exocytotic fusion pore. J Cell Biol 119:1395–1404[Free Full Text]
- Lee J-S, Mayes MS, Stromer MH, Scanes CG, Jeftinija S, Anderson LL 2004 Number of secretory vesicles in growth hormone cells of the pituitary remains unchanged after secretion. Exp Biol Med (Maywood) 229:632–639[Abstract/Free Full Text]
- Senda T, Yamashita K, Okabe T, Sugimoto N, Matsuda M 1998 Intergranular bridges in the anterior pituitary cell and their possible involvement in Ca2+-induced granule-granule fusion. Cell Tissue Res 292:513–519[CrossRef][Medline]
Top
Abstract
Introduction
