Porosome: the Universal Molecular Machinery for Cell Secretion

Mol. Cells OS, 517-529, December 31, 2008 Molecules

Minireview and Cells

Bhanu P. Jena*

Porosomes are supramolecular, lipoprotein structures at vesicles following secretion. The journey leading to the discov- the cell plasma membrane, where membrane-bound secre- ery of the ‘éçêçëçãÉ’, a nanometer-size structure at the cell tory vesicles transiently dock and fuse to release inrave- plasma membrane -the universal secretory machinery, and its sicular contents to the outside during cell secretion. The structure, function, isolation, chemistry, and reconstitution into mouth of the porosome opening to the outside, range in lipid membrane, the molecular mechanism of secretory vesicle size from 150 nm in diameter in acinar cells of the exocrine swelling and fusion at the base of porosomes, is discussed. pancreas, to 12 nm in neurons, which dilates during cell The isolation of the porosome, and the determination of its secretion, returning to its resting size following completion biochemical composition, its structure and dynamics at nm of the process. In the past decade, the composition of the resolution and in real time, and its functional reconstitution into porosome, its structure and dynamics at nm resolution architecture lipid membrane (Cho et al., 2002a; 2002b; 2004; and in real time, and its functional reconstitution into artifi- 2007; Jena, 2005; 2007; Jena et al., 2003; Jeremic et al., 2003; cial lipid membrane, have all been elucidated. In this mini Schneider et al., 1997), has greatly advanced our understand- review, the discovery of the porosome, its structure, func- ing of the secretory process in cells. tion, isolation, chemistry, and reconstitution into lipid The establishment of continuity between the secretory vesi- membrane, the molecular mechanism of secretory vesicle cle membrane and the membrane at the porosome base, re- swelling and fusion at the base of porosomes, and how quires the participation of specific membrane proteins called this new information provides a paradigm shift in our un- SNAREs. At the nerve terminal for example, target membrane derstanding of cell secretion, is discussed. proteins SNAP-25 and syntaxin, collectively called t-SNAREs present at the base of the neuronal porosome complex, and synaptic vesicle-associated protein v-SNARE, are involved in INTRODUCTION fusion of synaptic vesicles at the porosome base. To understand SNARE-induced membrane fusion, required an under- The story of cell secretion, a fundamental process as old as life standing of the interaction and assembly of membrane- itself, occurs in all organisms- from the simple yeast to cells in associated v-SNARE and t-SNAREs. The structure and ar- humans. Secretion is responsible for numerous physiological rangement of membrane-associated t-/v-SNARE complex, was activities in living organisms, such as neurotransmission and first determined using AFM. Results from the study demon- the release of hormones and digestive enzymes. Secretory strate that t-SNAREs and v-SNARE, when present in opposing defects in cells are responsible for a host of debilitating dis- bilayers, interact in a circular array to form ring complexes or eases, and hence this field has been the subject of intense channels, each measuring a few nanometers (Cho et al., study for over half a century. In the past 15 years, primarily 2002c). The size of the ring complex is directly proportional to using the atomic force microscope -a force spectroscope, a the curvature of the opposing bilayers. In the presence of cal- detailed understanding of the molecular machinery and mecha- cium, the ring-complex enables the establishment of continuity nism of cell secretion has come to light. As opposed to the between compartments across the opposing bilayers. In con- commonly held belief that during cell secretion secretory trast however, in the absence of membrane, soluble v- and t- vesicles completely merge at the cell plasma membrane (to be SNAREs fail to assemble in such specific and organized pat- endocytosed later), it has become clear that secretory vesicles tern, nor form such conducting channels (Cho et al., 2002c; transiently dock, fuse, partially expel their contents, and disso- Cook et al., 2008). Once v-SNARE and t-SNAREs residing in ciate, allowing multiple such rounds. It has been further deter- opposing bilayers meet, the resulting SNARE complex over- mined that swelling of secretory vesicles is required for the come the repulsive forces between opposing bilayers, bringing expulsion if intravesicular contents during cell secretion. These them closer to within a distance of 2.8-3 Å, allowing calcium findings have led to a paradigm shift in our understanding of bridging of the opposing phospholipids headgroups, leading to cell secretion, and explains the generation of partially empty local dehydration and membrane fusion (Jeremic et al., 2004a;

Department of Physiology, Wayne State University School of Medicine, Detroit, MI 48201, USA *Correspondence: [email protected]

Received November 6, 2008; accepted November 10, 2008; published online November 17, 2008

Keywords: cell secretion, membrane fusion, porosome

518 The Molecular Machinery and Mechanism of Cell Secretion

2004b; Potoff et al., 2008). A B In this review, the discovery of the porosome as the universal secretory machinery in cells, its structure, dynamics, its isolation, composition, its functional reconstitution in artificial lipid membrane, and the molecular mechanism of SNARE-induced fusion of secretory vesicle at the porosome base, is discussed. The article is primarily focused on porosome in a slow secretory cell, i.e., the acinar cell of the exocrine pancreas, and in a fast secretory cell, the neuron. Similarly, results from studies on neuronal SNAREs, have been used to explain the molecular mechanism of SNARE-induced membrane fusion, i.e., the fusion of synaptic vesicle at the porosome base. C D Discovery of the ‘porosome’ -the universal secretory machinery in cells

Porosomes were first discovered in acinar cells of the exocrine pancreas (Schneider et al., 1997). Exocrine pancreatic acinar cells are polarized secretory cells possessing an apical and a basolateral end. This well characterized cell of the exocrine pancreas, synthesize digestive enzymes, which is stored within 0.2-1.2 μm in diameter apically located membranous sacs or secretory vesicles, called zymogen granules (ZG). Following a E F secretory stimulus, ZG’s dock and fuse with the apical plasma membrane to release their contents to the outside. Contrary to neurons, where secretion of neurotransmitters occurs in the millisecond time regime, the pancreatic acinar cells secrete digestive enzymes over minutes following a secretory stimulus. Being a slow secretory cell, pancreatic acinar cells were ideal for investigation of the molecular steps involved in cell secretion. In the mid 1990’s, AFM studies were undertaken on live pancreatic acinar cells to evaluate at high resolution, the structure and dynamics of the apical plasma membrane in both resting and following stimulation of cell secretion. To our surprise, iso- Fig. 1. Porosomes or previously referred to as ‘depression’ at the lated live pancreatic acinar cells in physiological buffer, when plasma membrane in pancreatic acinar cell and at the nerve termi- imaged using the AFM (Schneider et al., 1997), reveal new nal. (A) AFM micrograph depicting ‘pits’ (yellow arrow) and ‘poro- cellular structures. At the apical plasma membrane, a group of somes’ within (blue arrow), at the apical plasma membrane in a live circular ‘pits’ measuring 0.4-1.2 μm in diameter, contain smaller pancreatic acinar cell. (B) To the right is a schematic drawing de- ‘depressions’ were observed. Each depression measure be- picting porosomes at the cell plasma membrane (PM), where tween 100-180 nm in diameter, and typically 3-4 depressions membrane-bound secretory vesicles called zymogen granules (ZG), are found within a pit. The basolateral membrane in acinar cells, dock and fuse to release intravesicular contents. (C) A high resolu- are devoid of such structures. High-resolution AFM images of tion AFM micrograph shows a single pit with four 100-180 nm poro- depressions in live acinar cells further reveal a cone-shaped somes within. (D) An electron micrograph depicting a porosome morphology, and the depth of each cone measure 15-35 nm. (red arrowhead) close to a microvilli (MV) at the apical plasma Subsequent studies over the years, demonstrate the presence membrane (PM) of a pancreatic acinar cell. Note association of the of depressions in all secretory cells examined. Analogous to porosome membrane (yellow arrowhead), and the zymogen gran- pancreatic acinar cells, examination of resting GH secreting ule membrane (ZGM) (red arrow head) of a docked ZG (inset). cells of the pituitary (Cho et al., 2002b) and chromaffin cells of Cross section of a circular complex at the mouth of the porosome is the adrenal medulla (Cho et al., 2002d) also reveal the pres- seen (blue arrow head). (E) The bottom left panel shows an elec- ence of pits and depressions at the cell plasma membrane. The tron micrograph of a porosome (red arrowhead) at the nerve termi- presence of depressions or porosomes in neurons, astrocytes, nal, in association with a synaptic vesicle (SV) at the presynaptic β-cells of the endocrine pancreas, and in mast cells have also membrane (Pre-SM). Notice a central plug at the neuronal poro- been elucidated (Cho et al., 2004; 2007; Jena, 2004), demon- some opening. (F) The bottom right panel is an AFM micrograph of strating their universal presence (Figs. 1-3). a neuronal porosome in physiological buffer, also showing the cen- Exposure of pancreatic acinar cells to a secretagogue (mas- tral plug (red arrowhead) at its opening. It is believed that the central toparan) results in a time-dependent increase (25-45%) in both plug in neuronal porosomes may regulate its rapid close-open con- the diameter and relative depth of depressions. Studies dem- formation during neurotransmitter release. The neuronal porosome onstrate that depressions return to resting size on completion of is an order of magnitude smaller (10-15 nm) in comparison to poro- cell secretion (Cho et al., 2002a; Schneider et al., 1997). No some in the exocrine pancreas (Cho et al., 2005a). demonstrable change in pit size is detected following stimulation of secretion (Schneider et al., 1997). Enlargement of depression diameter and an increase in its relative depth after results in a 15-20% decrease in depression size and a conse- exposure to secretagogue, correlated with secretion (Fig. 4). quent 50-60% loss in secretion (Schneider et al., 1997). Re- Additionally, exposure of pancreatic acinar cells to cytochalasin sults from these studies suggest depressions to be the fusion B, a fungal toxin that inhibits actin polymerization and secretion, pores in pancreatic acinar cells. Furthermore, these studies

Bhanu P. Jena 519

A B C Fig. 2. Neuronal fusion pore distribution, size and structure. Figure 2A shows the structure and distribution of fusion pores at the cytosolic compartment of a synaptosome. Inside-out synaptosome preparations when imaged in buffer using AFM, demonstrates inverted 12-16 nm cup-shaped

fusion pores, some with docked vesi- D F cles. Note one inverted cup-shaped fusion pore (green arrow heads), with a docked synaptic vesicle (red arrow heads), shown at higher magnification in Fig. 2B. (B) Atomic force micrograph shows a 37 nm synaptic vesicle docked to a 14 nm fusion pore at the cytoplas- mic compartment in the isolated synaptosomal membrane. (C) AFM measurement of the fusion pores (13.05 ± 0.91) and attached synaptic vesicles (40.15 ± 3.14) in the cytosolic compartment of synaptosome membrane is demonstrated in Fig. 2C (n = 15). (D) Schematic illustration of a neuronal fusion pore, showing the 8 vertical ridges and a central plug. (E) Photon correlation spectroscopy, further demonstrates fusion pores to measure 12-16 nm (Cho et al., 2007).

A B demonstrate the involvement of actin in regulation of both the structure and function of depressions. Similarly, depression in

resting GH cells measure 154 ± 4.5 nm (mean ± SE) in diame-

ter, and following exposure to a secretagogue results in a 40%

increase in depression diameter (215 ± 4.6 nm; p < 0.01), with

no appreciable change in pit size. The enlargement of depres-

sion diameter during cell secretion and subsequent decrease

accompanied by loss in secretion following exposure to actin

depolymerizing agents (Cho et al., 2002b), also suggested

them to be the secretory port holes.

A direct determination that depressions are indeed the port

holes via which secretory products are released from cells, was C D unequivocally demonstrated using immuno-AFM studies. Local- ization at depressions of gold-conjugated antibody to secretory proteins, finally provided a direct evidence that secretion occur through depressions (Cho et al., 2002a; 2002b). Zymogen granules contain the starch digesting enzyme amylase. AFM micrographs of the specific localization of gold-tagged amylase-specific antibodies (Fig. 5) at depressions, following stimulation of cell secretion (Cho et al., 2002a; Jena et al., 2003), conclusively demonstrated depressions as the cellular secretory porthole. Similarly, in somatotrophs of the pituitary gland, gold-tagged growth hormone-specific antibody found to selectively localize at Fig. 3. Nanoscale, three-dimensional contour map of protein the depression openings following stimulation of secretion (Cho assembly within the neuronal porosome complex. (A) Atomic et al., 2002b), established these sites too, to be the porthole in force micrograph of an immunoisolated neuronal porosome, these cells. Over the years, the term “fusion pore” has been reconstituted in lipid membrane. Note the central plug of the loosely referred to plasma membrane dimples that originate fol- porosome complex and the presence of approximately 8 globu- lowing a secretory stimulus, or to the continuity or channel estab- lar units arranged at the lip of the complex. (B) Negatively lished between opposing lipid membrane during membrane fu- stained electron micrographs of isolated neuronal porosome sion. Hence for clarity, the term “éçêçëçãÉ” was assigned to protein complexes. Note the 10-12 nm complexes exhibiting a depressions. circular profile and having a central plug. Approximately 8-10 The porosome structure, at the cytosolic compartment of the interconnected protein densities are observed at the rim of the plasma membrane in the exocrine pancreas (Jena et al., 2003), structure, which are connected to a central element via spoke- and in neurons (Cho et al., 2004), has also been determined at like structures. (C) Electron density maps of negatively stained near nm resolution in live tissue. To determine the morphology electron micrographs of isolated neuronal porosome protein of porosomes at the cytosolic compartment of pancreatic acinar complexes. (D) 3D topography of porosomes obtained from cells, isolated plasma membrane preparations in near physio- their corresponding electron density maps. The colors from logical buffered solution, have been imaged at high resolution yellow, through green to blue, correspond to the protein image using AFM. These studies reveal scattered circular disks density from lowest to the highest. The highest peak in each measuring 0.5-1 μm in diameter, with inverted cup-shaped image represents 27 Å (Cho et al., 2008). structures within (Jena et al., 2003). The inverted cups at the

520 The Molecular Machinery and Mechanism of Cell Secretion

Fig. 4. Depression or porosome dynamics A B C in pancreatic acinar cells, following stimulation of secretion. (A) Several porosomes within a pit are shown. The scan line across three porosomes in the top panel is represented graphically in the middle panel and defines the diameter and relative depth of each of the three porosomes. The porosome at the center, is represented by red arrowheads. The bottom panel represents % total cellular amylase release in the presence and absence of the secretagogue Mas7. (B) Notice an increase in porosome diameter and relative depth, correlating with an increase in total cellular amylase release at 5 min following stimulation of secretion. (C) At 30 min following a secretory stimulus, there is a decrease in diameter and relative depth of porosomes and no further increase in amylase release beyond the 5- min time point. No significant changes in amylase secretion or porosome diameter were observed in control cells in either the presence or absence of the non-stimulatory

mastoparan analogue (Mas17), throughout the experiment. High-resolution images of porosomes were obtained before and after stimulation with Mas7, for up to 30 min (Schneider et al., 1997).

Fig. 5. Porosomes dilate to allow expul- A B C D sion of vesicular contents. (A, B) AFM micrographs and section analysis of a pit and two out of the four depressions or porosomes, showing enlargement of porosomes following stimulation of secretion. (C) Exposure of live cells to gold conjugated-amylase antibody (Ab) results in specific localization of gold to these secretory sites. Note the localization of amylase-specific immunogold at the edge of porosomes. (D) AFM micrograph of pits and porosomes with immunogold localization is also demonstrated in cells immunolabeled and then fixed. Blue arrowheads point to immunogold clusters and the yellow arrowhead points to a depression or porosome opening (Cho et al., 2002a).

cytosolic compartment of isolated pancreatic plasma mem- The t-SNARE protein SNAP-23 had previously been reported brane preparations, range in height from 10-15 nm. On several in pancreatic acinar cells (Gaisano et al., 1997). A polyclonal occasions, ZG’s ranging in size from 0.4-1 μm in diameter, monospecific SNAP-23 antibody recognizing a single 23 kDa were observed in association with one or more of the inverted protein in immunoblots of pancreatic plasma membrane frac- cups, suggesting the circular disks to be pits, and the inverted tion, when used in immuno-AFM studies, demonstrated selec- cups to be porosomes. To further confirm that the cup-shaped tive localization to the base of the cup shaped structures. These structures are porosomes, where secretory vesicles dock and results demonstrate the inverted cup-shaped structures in in- fuse, immuno-AFM studies were performed. Target membrane side-out isolated pancreatic plasma membrane preparations, proteins, SNAP-23 (Oyler et al., 1989) and syntaxin (Bennett et are indeed porosomes, where secretory vesicles dock and fuse al., 1992) (t-SNARE) and secretory vesicle-associated mem- to release their contents during cell secretion (Jena et al., 2003). brane protein v-SNARE or VAMP (Trimble et al., 1988), are The morphology of the pancreatic porosome complex has part of the conserved protein complex involved in fusion of been further evaluated using transmission electron microscopy opposing bilayers in the presence of calcium (Cho et al., 2002c; (TEM) (Jeremic et al., 2003). TEM studies confirm the poro- 2005a; Cho and Jena, 2007; Jeremic et al., 2006; Malhotra et some to possess a cup-shaped structure, with similar dimen- al., 1988; Wilson et al., 1992). Since ZGs dock and fuse at the sions as determined from AFM measurement. Additionally, plasma membrane to release vesicular contents, it was hy- TEM micrographs demonstrate pancreatic porosomes to ex- pothesized that if porosomes are the secretory sites, then hibit a basket-like morphology, with three lateral and a number plasma membrane-associated t-SNAREs should localize there. of vertically arranged ridges. A ring at the base of the complex

Bhanu P. Jena 521

is further identified (Jeremic et al., 2003), and is hypothesized A to represent t-SNARE present in a circular array. Studies using full length recombinant SNARE proteins and artificial lipid membranes demonstrated that t- and v-SNAREs located in opposing bilayers interact in a circular array to form conducting channels (Cho et al., 2002c). Since similar circular structures are observed at the base of the pancreatic porosome complex, and SNAP-23 immunoreactivity is localized to the same site, suggests the circular arrangement of proteins at the porosome base to be t-SNAREs. In the past decade, a number of studies demonstrate the involvement of cytoskeletal proteins in cell secretion, some implicating a direct interaction of cytoskeleton protein with SNAREs (Bennett, 1990; Cho et al., 2005b; Faigle et al., 2000; Goodson et al., 1997; Nakano et al., 2001; Ohyma et al., 2001). Furthermore, actin and microtubule-based cytoskeleton has been implicated in intracellular vesicle traffic. B Fodrin, which was previously implicated in exocytosis, has also been shown to directly interact with SNAREs (Nakano et al., 2001). Studies demonstrate α-fodrin to regulate exocytosis via its interaction with t-SNARE syntaxin family of proteins. The c- terminal region of syntaxin is known to interact with α-fodrin, a major component of the submembranous cytoskeleton. Simi- larly, vimentin filaments interact with SNAP23/25 and hence are able to control the availability of free SNAP23/25 for assembly of the t-/v-SNARE complex (Faigle et al., 2000). All these findings suggested that vimentin, α-fodrin, actin, and SNAREs may be part of the porosome complex. Additional proteins such as v-SNARE (VAMP or synaptobrevin), synaptophysin and myosin, may associate when the porosome establishes continuity with the secretory vesicle membrane. The globular tail domain C D of myosin V contains a binding site for VAMP, which is bound in a calcium independent manner (Ohyma et al., 2001). Further interaction of myosin V with syntaxin had been shown to re- quire both calcium and calmodulin. It had also been suggested that VAMP, may act as a myosin V receptor on secretory vesicles, and regulate formation of the SNARE complex (Nakano et E F al., 2001). Interaction of VAMP with synaptophysin and myosin V had also been reported (Ohyma et al., 2001). In agreement with these earlier findings, our studies (Fig. 6) (Cho et al., 2004; Jena et al., 2003) demonstrate the association of SNAP-23, syntaxin 2, cytoskeletal proteins actin, α-fodrin, and vimentin, and calcium channels β3 and α1c, together with the SNARE G H regulatory protein NSF, in the porosome complex (Cho et al., 2004; Jena et al., 2003; Jeremic et al., 2003). Additionally, chloride ion channels ClC2 and ClC3 were also identified as part of the porosome complex (Cho et al., 2004; Jena et al., 2003; Jeremic et al., 2003). Isoforms of the various other proteins identified in the porosome complex, have also been demon- Fig. 6. One (1D) and two dimensionally (2D) resolved proteins strated using 2D-BAC gels electrophoresis (Jeremic et al., immunoisolated from solubilized pancreatic plasma membrane 2003). Three isoforms each of the calcium ion channel and preparations, using a SNAP-23 specific antibody. The resolved vimentin were found in porosomes (Jeremic et al., 2003). Using proteins were silver stained and transferred to nitrocellulose mem- yeast two-hybrid analysis, studies confirm the presence and branes for immunoblot analysis. Note the identification in the immu- interaction of some of these proteins with t-SNAREs within the noisolates of eight protein bands in a silver-stained SDS-PAGE 1D- porosome complex (Cho et al., 2005b). resolved gel. (the far left panel in A). Immunoblot analysis of the 1D- The size and shape of the immunoisolated porosome com- resolved immunoisolated proteins, demonstrate the presence of plex has also been determined using both negative staining EM actin, vimentin, syntaxin-2, calcium channel, NSF, and chloride and AFM (Jeremic et al., 2003). The morphology of immunoiso- channels CLC-2 and CLC-3. (B) 2D resolution of the immunoisolated porosomes obtained using EM and AFM, were similar, lated proteins, provide eleven spots in a silver stained gel. Im- and found to be super imposable (Jeremic et al., 2003). The munoblot analysis of the 2D-resolved immunoisolated proteins immunoisolated porosome complex has also been both struc- further demonstrates the presence of isoforms of some of the pro- turally and functionally reconstituted into liposomes and lipid teins identified in a. (C) There appears to be at least three isoforms bilayer membrane (Cho et al., 2004; Jeremic et al., 2003). of the calcium channel: (D) one isoform of actin; (E) three isoforms Transmission electron micrographs of pancreatic porosomes of vimentin, (F) one NSF; (G) one each of the CLC-2; and (H) the reconstituted into liposomes, exhibit a 150-200 nm cup-shaped CLC-3. Arrows pointing to lower molecular weights may represent basket-like morphology, similar to what is observed in its native proteolytic cleavage products.

522 The Molecular Machinery and Mechanism of Cell Secretion

state when co-isolated with ZGs. To test the functionality of the A isolated porosome complex, purified porosomes obtained from exocrine pancreas or neurons were subjected to reconstitution in lipid membrane of the electrophysiological setup (EPC9), and challenged with isolated ZGs or synaptic vesicles. Electrical activity of the reconstituted membrane as well as the transport of vesicular contents from the Åáë to the íê~åë compartments of the bilayer chambers was monitored. Results from these experiments demonstrate that the lipid membrane-reconstituted B porosomes, are indeed functional (Cho et al., 2004; Jeremic et al., 2003), since in the presence of calcium, isolated secretory vesicles dock and fuse to transfer intravesicular contents from the cis to the trans compartment of the bilayer chamber (Fig. 7). ZGs fused with the porosome-reconstituted bilayer as demonstrated by an increase in capacitance and conductance, and a time-dependent transport of the ZG enzyme amylase from Åáë to the íê~åë compartment of the bilayer chamber (Jena, 2007; Jeremic et al., 2003). Amylase is detected using immunoblot analysis of the buffer in the cis and trans chambers, using immunoblot analysis (Jena, 2007; Jeremic et al., 2003). As observed in immunoblot assays of isolated porosomes, chloride channel activity is also present in the reconstituted porosome complex. Furthermore, the chloride channel inhibitor DIDS, was found to inhibit current activity through the porosome-reconstituted bilayer, demonstrating a requirement of the porosome-associated chloride channel activity for porosome function. Similarly, the structure and biochemical composition of the neuronal porosome, and the docking and fusion of synaptic vesicles at the neuronal porosome complex has also been elucidated (Figs. 1-3) (Cho et al., 2004; 2007). In summary, C these studies demonstrate porosomes to be permanent supramolecular lipoprotein structures at the cell plasma membrane, where membrane-bound secretory vesicles transiently dock and fuse to release intravesicular contents to the outside. Porosomes are therefore the universal secretory machinery in cells (Jena, 2004; 2005; 2007).

Participation of SNAREs in secretory vesicle fusion at the porosome base

V-SNARE and t-SNAREs need to reside in opposing membrane to appropriately interact and establish continuity across these membranes (Cho et al., 2002c). Purified recombinant t- and v-SNARE proteins, when applied to a lipid membrane, form globular complexes do not alter membrane current in the EPC9 bilayer. In contrast, when t-SNAREs and v-SNARE in opposing bilayers are exposed to each other, they interact and arrange in Fig. 7. Lipid bilayer-reconstituted porosome complex is functional. (A) circular pattern, forming channel-like structures (Figs. 8 and 9). Schematic drawing of the bilayer setup for electrophysiological meas- These channels are conducting, since some vesicles having urements. (B) Zymogen granules (ZGs) added to the cis side of the discharged their contents appear flat, measuring only 10-15 nm bilayer fuse with the reconstituted porosomes, as demonstrated by an in height as compared to 40-60 nm when filled. Since the t-/v- increase in capacitance and current activities, and a concomitant time SNARE complex lies between the opposing bilayers, the dis- dependent release of amylase (a major ZG content) to the trans side charged vesicles clearly reveal t-/v-SNAREs arranged in a of the membrane. The movement of amylase from the cis to the trans rosette, or a channel-like structure. On the contrary, unfused v- side of the chamber was determined by immunoblot analysis of the SNARE vesicles associated with the t-SNARE reconstituted contents in the cis and the trans chamber over time. (C) As demon- lipid membrane, exhibit only the vesicle profile. To further de- strated by immunoblot analysis of the immunoisolated complex, elec- termine if the channel-like structures were capable of establish- trical measurements in the presence and absence of chloride ion ing continuity between the opposing bilayers, lipid vesicles channel blocker DIDS, demonstrate the presence of chloride chan- reconstituted with t-SNAREs and the antifungal agent nystatin nels in association with the complex (Jeremic et al., 2003). and the cholesterol homologue ergosterol, where added to the cis compartment of the bilayer chamber. The bilayer membrane of the EPC9 setup was reconstituted with v-SNARE, for the Woodbury and Miller, 1990). When vesicles containing nystatin experiment. Nystatin, in the presence of ergosterol, forms a and ergosterol incorporate into an ergosterol-free membrane, a cation-conducting channel in lipid membranes (Cohen and current spike is observed since the nystatin channel collapses Niles, 1993; Kelly and Woodbury, 1996; Woodbury, 1999; as ergosterol diffuses into the lipid membrane (Cohen and Niles,

Bhanu P. Jena 523

A E Fig. 8. Opposing bilayers containing t- and v-SNAREs respectively, interact in a circular array to form conducting channels in presence of calcium. (A) Schematic diagram of F the bilayer-electrophysiology setup (EPC9). (B) Lipid vesicle containing nystatin channels (êÉÇ) and membrane bilayer with SNAREs, demonstrate significant changes in capacitance and conductance. When t- B SNARE vesicles were added to a v-SNARE membrane support, the SNAREs in opposing bilayers arranged in a ring pattern, forming pores (as seen in the AFM micro-

G graph on the extreme right) and there were seen stepwise in- C D creases in capacitance and conductance (-60 mV holding poten- tial). Docking and fusion of the vesicle at the bilayer membrane, opens vesicle-associated nystatin channels and SNARE-induced pore formation, allowing conductance of ions from Åáë=to the íê~åë side of the bilayer membrane. Then further addition of KCl to induce gradient-driven fusion resulted in little or no further increase in conductance and capacitance, demonstrating that docked vesicles have already fused. (C) t-/v-SNARE ring complex at low and high resolution (D) is shown. Bar = 100 nm. (E-G) The size of the t-/v-SNARE complex is directly proportional to the size of the SNARE-reconstituted vesicles. (E) Schematic diagram depicting the interaction of t-SNARE-reconstituted and v-SNARE-reconstituted liposomes. (F) AFM images of vesicle before and after their removal using the AFM cantilever tip, exposing the t-/v-SNARE-ring complex at the center. (G) Note the high correlation coefficient between vesicle diameter and size of the SNARE complex (Cho et al., 2002c; 2005a).

1993; Kelly and Woodbury, 1996; Woodbury, 1999). As a posi- complex or channel, however, the process may occur due to a tive control, a KCl gradient tests the ability of vesicles to fuse at progressive recruitment of t-/v-SNARE pairs as the opposing the lipid membrane (410 mM cis: 150 mM trans). The KCl gra- vesicles are pulled toward each other, until a complete ring is dient provides a driving force for vesicle incorporation that is established (Fig. 9), preventing any further recruitment of t-/v- independent of the influence of SNARE proteins. When t- SNARE pairs. SNARE vesicles are exposed to v-SNARE reconstituted bilayers, vesicles fused. Fusions of t-SNARE containing vesicles Disassembly of the SNARE complex with the membrane are observed as current spikes. To verify if the channel-like SNARE complex structure establishes continu- Studies demonstrate that the soluble kJÉíÜóäã~äÉáãáÇÉJëÉåëáíáîÉ= ity across the membrane, capacitance and conductance meas- Ñ~Åíçê (NSF) an ATPase, disassembles the t-/v-SNARE com- urements were taken. Phospholipid vesicles that come in con- plex in presence of ATP (Jeremic et al., 2006). This study was tact with the bilayer membrane do not readily fuse with the also the first conformation by direct physical observation that membrane. However, on addition of v-SNARE-reconstituted NSF-ATP alone can lead to SNARE complex disassembly. In phospholipid vesicles to the cis compartment of the bilayer this study, using purified recombinant NSF, and t- and v- chamber, a small increase in capacitance and a simultaneous SNARE-reconstituted liposomes, the disassembly of the t-/v- increase in conductance are observed. However, in the pres- SNARE complex was examined. Lipid vesicles ranging in size ence of calcium, when t-SNARE vesicles containing nystatin from 0.2-2 μm were reconstituted with either t-SNAREs or v- and ergosterol are added to the cis compartment of the bilayer SNARE. Kinetics of association and dissociation of t-SNARE- chamber, an initial increase in capacitance and conductance and v-SNARE-reconstituted liposomes in solution, in the pres- occurs, followed by a stepwise increase in both membrane ence or absence of NSF, ATP, and AMP-PNP (the non- capacitance and conductance, demonstrating the establish- hydrolyzable ATP analogue), were monitored by right angle ment of continuity between the bilayer chambers (Fig. 8). light scattering. Addition of NSF and ATP to the t/v-SNARE- These studies demonstrate that when opposing bilayers meet, vesicle mixture led to a rapid and significant increase in inten- SNAREs arrange in a ring pattern results in the formation of a sity of light scattering, suggesting rapid disassembly of the conducting channel in presence of calcium (Cho et al., 2002c). SNARE complex and dissociation of vesicles. Dissociation of t- Furthermore, membrane curvature, dictate the size of the /v-SNARE vesicles occurs on a logarithmic scale that can be SNARE ring complex (Fig. 8) (Cho et al., 2005a). expressed by first order equation, with rate constant k=1.1 s-1. What is the molecular mechanism of SNARE ring complex To determine whether NSF-induced dissociation of t- and v- formation when t-SNARE-vesicles and V-SNARE-vesicles SNARE vesicles is energy driven, experiments were performed meet? Unfortunately, there is no direct observation of the vari- in the presence and absence of ATP and AMP-PNP. No sig- ous steps involved, leading to the formation of the SNARE ring nificant change with NSF alone, or in presence of NSF-AMP-

524 The Molecular Machinery and Mechanism of Cell Secretion

Fig. 9. Schematic diagram depicting the pos- sible molecular mechanism of SNARE ring complex formation, when t-SNARE-vesicles and V-SNARE-vesicles meet. The process may occur due to a progressive recruitment of t-/v-SNARE pairs as the opposing vesicles are pulled toward each other, until a complete ring is established, preventing any further recruitment of t-/v-SNARE pairs to the complex. The top panel is a side view of two vesicles (one t- SNARE-reconstituted, and the other v-SNARE reconstituted) interacting to form a single t-/v- SNARE complex, leading progressively (from left to right) to the formation of the ring complex. The lower panel is a top view of the two interacting vesicles.

PNP, was observed demonstrating that t-/v-SNARE disassem- α-helical content. The t-SNAREs (Figs. 10Aii and 10Bii), shows bly is an enzymatic and energy-driven process. clearly defined peaks at both these wavelengths, consistent The ability of NSF-ATP in the disassembly of the t-/v-SNARE with a higher degree of helical secondary structures formed complex was further confirmed immunochemically. It has been both in buffered suspension and in membrane, at ca. 66 and demonstrated that v-SNARE and t-SNAREs form an SDS- 20%, respectively (Table 1). Again, the membrane-associated resistant complex (Jeong et al., 1999). NSF binds to SNAREs SNARE exhibits less helical content than when in suspension. and forms a stable complex when locked in the ATP-bound Similarly, there appears to be a dramatic difference in the CD state (ATP-NSF). Thus, in the presence of ATP + EDTA, signal observed in t-/v-SNARE complexes in suspension, and VAMP antibody has been demonstrated to be able to immuno- those complexes that are formed when membrane-associated precipitate this stable NSF-SNARE complex (Jeong et al., SNAREs interact (Figs. 10Aiii and 10Biii). Interestingly, there is 1999). Therefore, when t- and v-SNARE vesicles were mixed in no increase of secondary structure upon complex formation. the presence or absence of ATP, NSF, NSF + ATP, or NSF + Rather, the CD spectra of the complexes are identical to a AMP-PNP, and resolved using SDS-PAGE followed by im- combination of individual spectra. Moreover, membrane asso- munoblots using syntaxin-1 specific antibody, t-/v-SNARE dis- ciated t-/v-SNAREs are less folded than the purified SNARE assembly was found to be complete only in the presence of complex. This data supports previous AFM results that lipid is NSF-ATP. Direct observation of the t-/v-SNARE complex dis- required for proper arrangement of the SNARE proteins in assembly was further assessed using AFM. When purified membrane fusion. Addition of NSF to the t-/v-SNARE complex recombinant t-SNAREs and v-SNARE in opposing bilayers results in an increase in the unordered fraction (Figs. 10Aiv and interact and self-assemble to form supramolecular ring com- 10Biv; Table 1), which may be attributed to an overall disor- plexes, they disassembled when exposed to recombinant NSF dered secondary structure of the NSF, and not necessarily and ATP, as observed at nm resolution using AFM (Jeremic et unfolding of the t-/v-SNARE complex. In contrast, activation of al., 2006). Close examination of the NSF-ATP-induced disas- NSF by the addition of ATP almost completely abolishes all α- sembled SNARE complex by AFM, demonstrates NSF to func- helical content within the multi-protein complex (Figs. 10Av and tion as a right-handed molecular motor (Cho et al., 2007). 10Bv). This direct observation of the helical unfolding of the SNARE complex using CD spectroscopy under physiologically CD spectroscopy confirm the requirement of relevant conditions (i.e. in membrane-associated SNAREs), membrane for appropriate t-/v-SNARE complex confirms earlier AFM reports on NSF-ATP-induced t-/v-SNARE assembly, and that NSF-ATP alone can mediated complex disassembly (Jeremic et al., 2006). In further agree- SNARE disassembly ment with previously reported studies using the AFM, the con- sequence of ATP addition to the t-/v-SNARE-NSF complex is The overall secondary structural content of full-length neuronal disassembly, regardless of whether the t-/v-SNARE+NSF v-SNARE and t-SNAREs, and the t-/v-SNARE complex, both in complex is membrane-associated or in buffered suspension. In suspension and membrane-associated, were determined by earlier AFM studies, 0.16-0.2 μg/mL of SNARE proteins were CD spectroscopy using an Olis DSM 17 spectrometer (Cook et used, as opposed to the 800-1000 μg/mL protein concentration al., 2008). Circular dichroism spectroscopy reveals that v-SNARE required for the current CD studies. To determine if t-SNARE in buffered suspension (Fig. 10Ai), when incorporated into lipo- and v-SNARE interact differently at higher protein concentra- somes (Fig. 10Bi), exhibit reduced folding (Table 1). This loss of tions, both membrane-associated and in-suspension v- and t- secondary structure following incorporation of full-length v- SNARE complexes used in CD studies, were imaged using the SNARE in membrane may be a result of self-association of the AFM. In confirmation to previously reported AFM studies, re- hydrophobic regions of the protein in absence of membrane. sults from the CD study demonstrated the formation of t-/v- When incorporated into liposomes, v-SNARE may freely unfold SNARE ring complexes, only when t-SNARE-liposomes are without the artifactual induction of secondary structure, as reflec- exposed to v-SNARE-liposomes. Hence, higher SNARE pro- tive of the lack in CD signals at 208 and 222 nm, distinct for tein concentrations are without influence on the membrane-

Bhanu P. Jena 525

A SNAREs bring opposing bilayers closer, enabling calcium bridging and membrane fusion

Diffraction patterns of non-reconstituted vesicles and t- and v- SNARE-reconstituted vesicles in the absence and presence of 2+ 5 mM Ca have been performed in the 2-4 Å diffraction range (Jeremic et al., 2004a), having broad pattern spanning 2θ ranges approximately 23-48° or d values of 3.9-1.9 Å with sharp drop off intensity on either sides of the range. Relatively, broad feature of the diffractogram indicate multitude of contacts between atoms of one vesicle as well between different vesicles during collision. However, two broad peaks are visible on the diffractogram, the stronger one at 3.1 Å and a weaker one at 1.9 Å, indicating that the greatest number of contacts be- 2+ tween them have these two distances. Addition of Ca or incorporation of SNAREs at the vesicles membrane or both, influence both peaks within the 2.1-3.3 Å intensity range. How- 2+ ever, the influence of Ca , SNAREs or both is more visible on B peak positioned at 3.1 Å in form of an increased Imax of arbi- trary units and 2θ. This increase of Imax at the 3.1 Å can be explained in terms of increased vesicle pairing and/or a decrease in distance between apposed vesicles. Incorporation of t- and v-SNARE proteins at the vesicle membrane allows for tight vesicle-vesicle interaction, demonstrated again as an Imax 2+ shifts to 30.5° or 2.9 Å from 3.1 Å. Ca and SNAREs work in manner that induces a much higher increase of peak intensity with appearance of shoulders at 2.8 Å. 2+ Hydrated calcium [Ca(H2O)n] has more than one shell 2+ around it, and the first hydration shell around the Ca has six water molecules in an octahedral arrangement (Bako et al., 2002). Calcium drives SNARE-induced fusion of opposing bilayers (Jeremic et al., 2004a; 2004b). SNARE interactions allow opposing bilayers to come close within a distance of approximately 2.8 Å. Using light scattering and x-ray diffraction experiments involving SNARE-reconstituted liposomes, it became clear that fusion proceeds only when Ca2+ ions are available Fig. 10. Circular dichroism data reflecting structural changes to between the t- and v-SNARE-apposed bilayers (Jeremic et al., SNAREs, both in suspension and in association with membrane. 2004b). Since t-SNAREs and v-SNARE in opposing bilayers Structural changes, following the assembly and disassembly of interact in a circular array to form conducting channels in the the t-/v-SNARE complex is further shown. (A) CD spectra of presence of calcium (Cho et al., 2002c), would necessitate Ca2+ purified full-length SNARE proteins in suspension and (B) in ions to be present between the SNARE-apposed bilayers, to membrane-associated; their assemblyand (NSF-ATP)-induced allow bridging of the opposing membranes. Once calcium disassembly is demonstrated. (i) v-SNARE; (ii) t-SNAREs; (iii) t- forms such a bridge, it can no longer hold its water shells, lead- /v-SNARE complex; (iv) t-/v-SNARE + NSF and (v) t-/v-SNARE ing to water expulsion, membrane destabilization, and fusion. + NSF + 2.5 mM ATP, is shown. CD spectra were recorded at To further test the above results, atomistic molecular dynamic 25°C in 5 mM sodium phosphate buffer (pH 7.5), at a protein simulations in the isobaric-isothermal ensemble using hydrated concentration of 10 μM. In each experiment, 30 scans were dimethylphosphate anions (DMP-) and calcium cations, were averaged per sample for enhanced signal to noise, and data performed (Potoff et al., 2008). DMP- was chosen for the study were acquired on duplicate independent samples to ensure since it represented the smallest molecular fragment of typical reproducibility (Cook et al., 2008). membrane phospholipids that retained properties of the phopholipid head-group, while providing a significant reduction in the computational complexity and enhance accuracy of the directed self-assembly of the SNARE complex (Cook et al., study. Furthermore, the strategy of using the DMP- rather than 2008). In summary, the CD results demonstrate that v-SNARE full phospholipids in the simulation, facilitates the search for in suspension when incorporated into liposomes, exhibits re- spontaneously formed Ca2+-phospholipid structures, which may duced folding. Similarly, t-SNAREs which exhibit clearly defined bridge the head groups of opposing phospholipids bilayers. peaks at CD signals of 208 and 222 nm wavelengths, consis- Results from the study clearly demonstrated that DMP and tent with a higher degree of helical secondary structure in both calcium, form DMP-Ca2+ complexes and the consequent re- the soluble and liposome-associated forms, exhibit reduced moval of water, supporting the hypothesis. As a result of Ca2+- folding when membrane-associated. ATP-induced activation of DMP self-assembly, the distance between anionic oxygens NSF bound to the t-/v-SNARE complex, results in disassembly between the two DMP molecules is reduced to 2.92 Å (Potoff et of the SNARE complex, eliminating all α-helices within the al., 2008), which is in agreement with the 2.8 Å SNARE- structure. In addition, these studies are a further confirmation of induced apposition established between opposing lipid bilayers, earlier reports (Jeremic et al., 2006) that NSF-ATP is sufficient reported from x-ray diffraction measurements (Jeremic et al., for the disassembly of the t-/v-SNARE complex. 2004a). The above findings provide for the first time, a molecu-

526 The Molecular Machinery and Mechanism of Cell Secretion

Table 1. Secondary structural fit parameters of SNARE complex formation and dissociation Suspension (100 × fa) Membrane-associated (100 × f) Proteinb α β O U Fitc α β O U Fit v-SNARE 4 36 18 43 0.19 0 30 32 38 0.21 t-SNAREs 66 34 0 0 0.02 20 15 21 44 0.84 v-/t-SNAREs 48 52 0 0 0.02 20 19 56 5 0.38 v-/t-SNAREs+NSF 20 25 0 55 0.07 18 6 8 68 0.2 v-/t-SNAREs+NSF+ATP 3 39 18 40 0.22 1 27 34 38 0.23 aAbbreviations used: Ñ, fraction of residues is a given conformational class; α, α-helix; β, β-sheet; l, other (sum of turns, distorted helix, distorted sheet); r, unordered. bProtein constructs: îJpk^ob (VAMP2); íJpk^obë (SNAP-25 + syntaxin 1A); kpc, N-ethylmaleimide Sensitive Factor. ATP, adenosine triphosphate. cFit: goodness of fit parameter expressed as Normalized Spectral Fit Standard Deviation (nm) (Cook et al., 2008).

A B C D Fig. 11. The swelling dynamics of ZGs in live pancreatic acinar cells. (A) Electron

micrograph of pancreatic acinar cells

showing the basolaterally located nucleus

(N) and the apically located ZGs. The

apical end of the cell faces the acinar

lumen (L). Bar = 2.5 μm. (B-D) The apical

ends of live pancreatic acinar cells were

imaged by AFM, showing ZGs (red and

green arrowheads) lying just below the

apical plasma membrane. Exposure of the E cell to a secretory stimulus using 1 μM carbamylcholine, resulted in ZG swelling within 2.5 min, followed by a decrease in ZG size after 5 min. The decrease in size of ZGs after 5 min is due to the release of secretory products such as α-amylase, as demonstrated by the immunoblot assay (E) (Kelly et al., 2004).

lar understanding of SNARE-induced membrane fusion in cells. membrane fusion and in the expulsion of intravesicular contents, an electrophysiological ZG-reconstituted lipid bilayer Swelling of secretory vesicles is required for the fusion assay (Jeremic et al., 2003) has been employed. The expulsion of vesicular contents during cell secretion ZGs used in the bilayer fusion assays were characterized for their purity and their ability to respond to a swelling stimulus. Isolated live pancreatic acinar cells in near physiological buffer ZGs were isolated (Jena et al., 1997) and their purity assessed when imaged using AFM at high force (200-300 pN), demon- using electron microscopy. As previously reported (Abu- strate the size and shape of the secretory vesicles (zymogen Hamdah et al., 2004; Cho et al., 2002f; Jena et al., 1997) expo- granules or ZGs) lying immediately below the apical plasma sure of isolated ZGs to GTP results in ZG swelling. Similar to membrane of the cell (Fig. 11). Within 2.5 min of exposure to a what is observed in live acinar cells (Fig. 11), each isolated ZG physiological secretory stimulus (1 μM carbamylcholine), the responded differently to the same swelling stimulus. This differ- majority of ZGs within cells swell (Fig. 11), followed by a de- ential response of isolated ZGs to GTP has been further as- crease in ZG size (Fig. 11) by which time most of the release of sessed by measuring changes in the volume of isolated ZGs of secretory products from within ZGs had occurred (Fig. 11). different sizes (Kelly et al., 2004). ZGs in the exocrine pancreas These studies reveal for the first time in live cells, intracellular range in size from 0.2 to 1.3 µm in diameter (Jena et al., 1997), swelling of secretory vesicles following stimulation of secretion not all ZGs are found to swell following a GTP challenge (Kelly and their deflation following partial discharge of vesicular con- et al., 2004). Most ZGs volume increases were between 5-20%, tents (Kelly et al., 2004). No loss of secretory vesicles is ob- however, larger increases (up to 45%) were observed only in served throughout the experiment. Measurements of intracellu- vesicles ranging from 250 to 750 nm in diameter. These studies lar ZG size further reveal that different vesicles swell differently, demonstrate that following stimulation of secretion, ZGs within following a secretory stimulus. For example, the ZG marked by pancreatic acinar cells swell, followed by a release of intrave- the red arrowhead swelled to show a 23-25% increase in di- sicular contents through porosomes (Jeremic et al., 2003) at ameter, in contrast to the green arrowhead-marked ZG, which the cell plasma membrane, and a return to resting size on increased by only 10-11% (Fig. 11). This differential swelling completion of secretion. On the contrary, isolated ZGs stay among ZGs within the same cell, may explain why following swollen following GTP exposure, since there is no release of stimulation of secretion, some intracellular ZGs demonstrate the intravesicular contents. In acinar cells, little or no secretion the presence of less vesicular content than others, and hence is detected 2.5 min following stimulation of secretion, although have discharged more of their contents (Cho et al., 2002e). To the ZGs within them were completely swollen (Fig. 11). How- determine precisely the role of swelling in vesicle-plasma ever at 5 min following stimulation, ZGs deflate and the intrave-

Bhanu P. Jena 527

A membrane results in a small increase in capacitance (Fig. 12B), possibly due to the increase in membrane surface area contributed by incorporation of porosomes, ranging in size from 100-150 nm in diameter (Jeremic et al., 2003). Isolated ZGs when added to the cis compartment of the bilayer chamber, fuse at the porosome-reconstituted lipid membrane (Fig. 12A) and is detected as a step increase in membrane capacitance (Fig. 12B). However, even after 15 min of ZG addition to the cis compartment of the bilayer chamber, little or no release of the intra-vesicular enzyme α-amylase is detected in the trans compartment of the chamber (Figs. 12C and 12D). On the contrary, exposure of ZGs to 20 mM GTP, induced swelling (Abu- Hamdah et al., 2004; Cho et al., 2002f; Jena et al., 1997) and B results both in the potentiation of fusion as well as a robust expulsion of α-amylase into the trans compartment of the bilayer chamber (Figs. 12C and 12D). These studies demonstrate that during secretion, secretory vesicle swelling is required for the efficient expulsion of intravesicular contents. C Within minutes or even seconds following stimulation of secretion, empty and partially empty secretory vesicles accumulate within cells (Cho et al., 2002e; Lawson et al., 1975; Plattner et al., 1997). Following addition of ZGs to the cis compartment of the bilayer chamber, membrane capacitance continues to increase, however, little or no detectable secretion occurred even

after 15 min (Fig. 11), suggesting that either variable degrees of vesicle swelling or repetitive cycles of fusion and swelling of the same vesicle or both, may operate during cell secretion. Under these circumstances, empty and partially empty vesicles could be generated within cells following secretion. To test this hy- D pothesis, two key parameters were examined. One, whether the extent of swelling is same for all ZGs exposed to a certain concentration of GTP, and two, whether ZG is capable of swelling to different degrees? And if so, whether there is a correla- Fig. 12. Fusion of isolated ZGs at porosome-reconstituted bilayer tion between extent of swelling and the quantity of intravesicu- and GTP-induced expulsion of α-amylase. (A) Schematic diagram lar contents expelled. The answer to the first question is clear, of the EPC9 bilayer apparatus showing the cis and trans chambers. that different ZGs respond to the same stimulus differently (Fig. Isolated ZGs when added to the cis chamber, fuse at the bilayers- 11). Our study revealed that different ZGs within cells or in iso- reconstituted porosome. Addition of GTP to the cis chamber in- lation undergo different degrees of swelling, even though they duces ZG swelling and expulsion of its contents such as α-amylase are exposed to the same stimuli (carbamylcholine for live pan- to the trans bilayers chamber. (B) Capacitance traces of the lipid creatic acinar cells) or GTP for isolated ZGs (Figs. 11B-11D). bilayer from three separate experiments following reconstitution of The requirement of ZG swelling for expulsion of vesicular con- porosomes (green arrowhead), addition of ZGs to the cis chamber tents is further confirmed, when the GTP dose-dependently (blue arrowhead), and the red arrowhead represents the 5 min time increased ZG swelling (Kelly et al., 2004) translated into in- point after ZG addition. Note the small increase in membrane ca- creased secretion of α-amylase. Although higher GTP concen- pacitance following porosome reconstitution, and a greater increase trations elicited an increased ZG swelling, the extent of swelling when ZGs fuse at the bilayers. (C) In a separate experiment, 15 between ZGs once again varied. min after addition of ZGs to the cis chamber, 20 μM GTP was intro- To determine if a similar or an alternate mechanism is reduced. Note the increase in capacitance, demonstrating potentia- sponsible for the release of secretory products in a fast secretion of ZG fusion. Flickers in current trace represent current activity. tory cell, synaptosomes and synaptic vesicles from rat brain (D) Immunoblot analysis of α-amylase in the trans chamber fluid at has been used in studies (Kelly et al., 2004). Since synaptic different times following exposure to ZGs and GTP. Note the unde- vesicle membrane is known to possess both Gi and Go pro- tectable levels of α-amylase even up to 15 min following ZG fusion teins, we hypothesized GTP and Gi-agonist (mastoparan) me- at the bilayer. However, following exposure to GTP, significant diated vesicle swelling. To test this hypothesis, isolated synap- amounts of α-amylase from within ZGs were expelled into the trans tosomes were lysed to obtain synaptic vesicles and synapto- bilayers chamber (Kelly et al., 2004). somal membrane (Kelly et al., 2004). Isolated synaptosomal membrane when placed on mica and imaged by the AFM in near physiological buffer, reveal on the cytosolic side the pres- sicular α-amylase released from the acinar cell was detected, ence of 40-50 nm in diameter synaptic vesicles still docked to suggesting the involvement of ZG swelling in secretion. the presynaptic membrane. Similar to the ZG’s, exposure of In electrophysiological bilayer fusion assays, immunoisolated synaptic vesicles to 20 μM GTP, results in an increase in syn- porosomes from the exocrine pancreas, functionally reconsti- aptic vesicle swelling (Kelly et al., 2004). However, exposure to tuted (Jeremic et al., 2003) into the lipid membrane of the bi- Ca2+, results in the transient fusion of synaptic vesicles at the layer apparatus, where membrane conductance and capaci- presynaptic membrane, expulsion of intravesicular contents, tance can be continually monitored (Fig. 12A), has been used and the consequent decrease in size of the synaptic vesicle. (Kelly et al., 2004). Reconstitution of the porosome into the lipid Additionally, as observed in ZG’s of the exocrine pancreas, not

528 The Molecular Machinery and Mechanism of Cell Secretion

all synaptic vesicles swell, and if they do, the swell to different induced and Go-mediated synaptic vesicle swelling in neurons extents even though they had been exposed to the same (Jeremic et al., 2005). stimulus. This differential response of synaptic vesicles within In conclusion, áå=îáîç and áå=îáíêç measurements of secretory the same nerve ending may dictate and regulate the potency vesicle dynamics, demonstrate that vesicle swelling is required and efficacy of neurotransmitter release at the nerve terminal for the expulsion of intravesicular content from cells during se- (Kelly et al., 2004). To further confirm synaptic vesicle swelling cretion. It is demonstrated that the amount of intravesicular and determine the swelling rate, light scattering experiments contents expelled, is directly proportional to the extent of secre- have been performed. Light scattering studies demonstrate a tory vesicle swelling. This unique capability provides cells with mastoparan-dose dependent increase in synaptic vesicle swell- the ability to precisely regulate the release of secretory products ing. Mastoparan (20 μM) induces a time-dependent (in se- during cell secretion. The direct observation in live cells using conds) increase if synaptic vesicle swelling, as opposed to the the AFM, the requirement of secretory vesicle swelling in cell control peptide (Mast-17). Results from this study show that secretion, also explains the appearance of empty and partially following stimulation of secretion, ZGs, the membrane-bound empty vesicles following cell secretion. secretory vesicles in exocrine pancreas swell. Different ZGs swell differently, and the extent of their swelling dictates the CONCLUSION amount of intravesicular contents expelled. ZG swelling is therefore a requirement for the expulsion of vesicular contents In this article, the current understanding of the molecular ma- in the exocrine pancreas (Kelly et al., 2004), and similar to ZG’s, chinery and mechanism of cell secretion is presented. Poro- synaptic vesicles swell enabling the expulsion of neurotransmit- somes are specialized plasma membrane structures universally ters at the nerve terminal. This mechanism of vesicular expul- present in secretory cells, from exocrine and endocrine cells, to sion during cell secretion may explain why partially empty vesi- neuroendocrine cells and neurons. Since porosomes in exo- cles are observed in secretory cells (Cho et al., 2002e; Lawson crine and neuroendocrine cells measure 100-180 nm, and only et al., 1975; Plattner et al., 1997) following secretion. The pres- 20-35% increase in porosome diameter is demonstrated follow- ence of empty secretory vesicles could result from multiple ing the docking and fusion of 0.2-1.2 μm in diameter secretory rounds of fusion-swelling-expulsion cycles a vesicle may un- vesicles, it is concluded that secretory vesicles “transiently” dergo, during the secretory process. These results reflect the dock and fuse at the base of the porosome complex to release precise and regulated nature of cell secretion, both in the exo- their contents to the outside. The discovery of the porosome, crine pancreas and in neurons. and an understanding of its structure and dynamics at nm resolution and in real time in live cells, its composition, and its func- Molecular mechanism of secretory vesicle swelling tional reconstitution in lipid membrane, has greatly advanced our understanding of cell secretion. It is evident that the secre- Our understanding of the molecular mechanism of secretory tory process in cells is well coordinated, highly regulated, and a vesicle swelling has greatly advanced in the last decade. Iso- finely tuned biomolecular orchestra. Clearly, these findings lated secretory vesicles and reconstituted swelling-competent could not have advanced without the AFM, and therefore this proteoliposomes have been utilized (Abu-Hamdah et al., 2004; powerful tool, has greatly contributed to a new understanding of Cho et al., 2002c; Jena et al., 1997; Jeremic et al., 2005; Kelly the cell. The AFM has enabled the determination of live cellular et al., 2004) to determine the mechanism and regulation of structure-function at sub nanometer to angstrom resolution, in vesicle swelling. Isolated ZGs from the exocrine pancreas swell real time, contributing to the birth of the new field of Nano Cell- rapidly in response to GTP (Cho et al., 2002f; Jena et al., 1997; Biology. Kelly et al., 2004), suggesting rapid water gating into ZGs. Re- sults from studies demonstrate the presence of the water chan- ACKNOWLEDGMENTS nel aquaporin-1 (AQP1) at the ZG membrane (Cho et al., The author thanks the many students and collaborators who 2002c) and its participation in GTP-mediated water entry and have participated in the various studies discussed in this article. vesicle swelling. Further, the molecular regulation of AQP1 at Support from the National Institutes of Health (USA), the Na- the ZG membrane has been studied (Cho et al., 2002c; 2002e), tional Science Foundation (USA), and Wayne State University, providing a general mechanism of secretory vesicle swelling. is greatly appreciated. Detergent-solubilized ZGs, immunoisolated using monoclonal AQP-1 antibody, co-isolates AQP-1, PLA2, Gαi3, potassium REFERENCES channel IRK-8, and the chloride channel ClC-2 (Abu-Hamdah et al., 2004). Exposure of ZGs to either the potassium channel Abu-Hamdah, R., Cho, W.J., Cho, S.-J., Jeremic, A., Kelly, M., Ilie, A.E., and Jena, B.P. (2004). Regulation of the water channel blocker glyburide, or the PLA2 inhibitor ONO-RS-082, blocks aquaporin-1: isolation and reconstitution of the regulatory com- GTP-induced ZG swelling. Red blood cells known to possess plex. Cell Biol. Int. OU, 7-17. AQP-1 at the plasma membrane also swell on exposure to the Bako, I., Hutter, J., and Palinkas, G. (2002). Car-Parrinello molecu- GαI agonist mastoparan, and responds similarly to ONO-RS- lar dynamics simulation of the hydrated calcium ion. J. Chem. 082 and glyburide, as do ZGs (Abu-Hamdah et al., 2004). Arti- Phys. NNT, 9838-9843. Bennett, V. (1990). Spectrin-based membrane skeleton: a multipo- ficial liposomes reconstituted with the AQP-1 immunoisolated tential adaptor between plasma membrane and cytoplasm. complex from solubilized ZGs also swell in response to GTP. Physiol. RevK=TM, 1029-1065. Glyburide or ONO-RS-082 is found to abrogate the GTP effect Bennett, M.K., Calakos, N., and Schller, R.H. (1992). Syntaxin: A in reconstituted liposomes. AQP-1 immunoisolate-reconstituted synaptic protein implicated in docking of synaptic vesicles at pre- planar lipid membrane demonstrate conductance, which is synaptic active zones. Science=ORT, 255-259. sensitive to glyburide and an AQP-1 specific antibody. These Cho, S.-J., Quinn, A.S., Stromer, M.H., Dash, S., Cho, J., Taatjes, D.J., and Jena, B.P. (2002a). Structure and dynamics of the fu- results demonstrate a G -PLA2 mediated pathway and potas- αi3 sion pore in live cells. Cell Biol. Int. OS, 35-42. sium channel involvement in AQP-1 regulation at the ZG mem- Cho, S.-J., Jeftinija, K., Glavaski, A., Jeftinija, S., Jena, B.P., and brane (Abu-Hamdah et al., 2004), contributing to ZG swelling. Anderson, L.L. (2002b). Structure and dynamics of the fusion Similarly, AQP-6 involvement has been demonstrated in GTP- pores in live GH-secreting cells revealed using atomic force mi-

Bhanu P. Jena 529

croscopy. Endocrinology NQP, 1144-1148. machinery in cells. Cell Biol. Int. OO, 657-670. Cho, S.-J., Kelly, M., Rognlien, K.T., Cho, J., Hörber, J.K., and Jeremic, A., Kelly, M., Cho, S.-J., Stromer, M.H., and Jena, B.P. Jena, B.P. (2002c). SNAREs in opposing bilayers interact in a (2003). Reconstituted fusion pore. Biophys. J.=UR, 2035-2043. circular array to form conducting pores. Biophys. JK= UP, 2522- Jeremic, A., Kelly, M., Cho, J., Cho, S.-J., Hörber, J.K., and Jena, 2527. B.P. (2004a). Calcium drives fusion of SNARE-apposed bilayers. Cho, S.J., Wakade, A., Pappas, G.D., and Jena, B.P. (2002d). New Cell Biol .IntK=OU, 19-31. structure involved in transient membrane fusion and exocytosis. Jeremic, A., Cho, W.J., and Jena, B.P. (2004b). Membrane fusion: Ann. N Y Acad. Sci.=VTN, 254-256. what may transpire at the atomic level. J. Biol. Phys. Chem.=Q, Cho, S.-J., Cho, J., and Jena, B.P. (2002e). The number of secre- 139-142. tory vesicles remains unchanged flowing exocytosis. Cell Biol. Jeremic, A., Cho, W.J., and Jena, B.P. (2005). Involvement of water Int. OS, 29-33. channels in synaptic vesicle swelling. Exp. Biol. Med. OPM, 674- Cho, S.-J., Sattar, A.K., Jeong, E.H., Satchi, M., Cho, J., Dash, S., 680. Mayes, M.S., Stromer, M.H., and Jena, B.P. (2002f). Aquaporin Jeremic, A., Quinn, A.S., Cho, W.J., Taatjes, D.J., and Jena, B.P. 1 regulates GTP-induced rapid gating of water in secretory vesi- (2006). Energy-dependent disassembly of self-assembled cles. Proc. Natl. Acad. Sci. USA VV, 4720-4724. SNARE complex: observation at nanometer resolution using Cho, W.J., Jeremic, A., Rognlien, K.T., Zhvania, M.G., Lazrishvili, I., atomic force microscopy. J. Am. Chem. Soc.=NOU, 26-27. Tamar, B., and Jena, B.P. (2004). Structure, isolation, composi- Kelly, M.L., and Woodbury, D.J. (1996). Ion channels from synaptic tion and reconstitution of the neuronal fusion pore. Cell Biol. Int.= vesicle membrane fragments reconstituted into lipid bilayers. OU, 699-708. Biophys. JK=TM, 2593-2599. Cho, W.J., Jeremic, A., and Jena, B.P. (2005a). Size of supramolecu- Kelly, M., Cho, W.J., Jeremic, A., Abu-Hamdah, R., and Jena, B.P. lar SNARE complex membrane-directed self-assembly. J. Am. (2004). Vesicle swelling regulates content expulsion during Chem. Soc.=NOT, 10156-10157. secretion. Cell Biol. Int. OU, 709-716. Cho, W.J., Jeremic, A., and Jena, B.P. (2005b). Direct interaction Lawson, D., Fewtrell, C., Gomperts, B., and Raff, M. (1975). Anti- between SNAP-23 and L-type calcium channel. J. Cell. Mol. immunoglobulin-induced histamine secretion by rat peritoneal MedK=V, 380-386. mast cells studied by immunoferritin electron microscopy. J. Cho, W.J., Jeremic, A. Jin, H., Ren, G., and Jena, B.P. (2007). Exp. Med. NQO, 391-401. Neuronal fusion pore assembly requires membrane cholesterol. Malhotra, V., Orci, L., Glick, B.S., Block, M.R., and Rothman, J.E. Cell Biol. IntK=PN, 1301-1308. (1988). Role of an N-ethylmaleimide-sensitive transport compo- Cho, W.J., Ren, G., and Jena, B.P. (2008). EM 3D contourmaps nent in promoting fusion of transport vesicles with cisternae of provide protein assembly at the nanoscale within the neuronal the Golgi stack. Cell=RQ, 221-227. porosome complex. J. Microscopy OPO, 106-111. Nakano, M., Nogami, S., Sato, S., Terano, A., and Shirataki, H. Cho, W.J., and Jena, B.P. (2007). N-ethymaleimide sensitive factor (2001). Interaction of syntaxin with α-fodrin, a major component is a right-handed molecular motor. J. Biomed. Nanotech.=P, 209- of the submembranous cytoskeleton. Biochem. Biophys. Res. 211. Commun.=OUU, 468-475. Cohen, F.S., and Niles, W.D. (1993). Reconstituting channels into Ohyama, A., Komiya, Y., and Igarashi, M. (2001). Globular tail of planar membranes: a conceptual framework and methods for myosin-V is bound to vamp/synaptobrevin. Biochem. Biophys. fusing vesicles to planar bilayer phospholipid membranes.= Res. Commun.=OUM, 988-991. Methods Enzymol.=OOM, 50-68. Oyler, G.A., Higgins, G.A., Hart, R.A., Battenberg, E., Billingsley, Cook, J.D., Cho, W.J., Stemmler, T.L., and Jena, B.P. (2008). Cir- M., Bloom, F.E., and Wilson, M.C. (1989). The identification of a cular dichroism (CD) spectroscopy of the assembly and disas- novel synaptosomal-associated protein, SNAP-25, differentially sembly of SNAREs: the proteins involved in membrane fusion in expressed by neuronal subpopulations. J. Cell BiolK NMV, 3039- cells. Chem .Phys. LettK=QSO, 6-9. 3052. Faigle, W., Colucci-Guyon, E., Louvard, D., Amigorena, S., and Plattner, H., Atalejo, A.R., and Neher, E. (1997). Ultrastructural Galli, T. (2000). Vimentin filaments in fibroblasts are a reservoir organization of bovine chromaffin cell cortex-analysis by cryo- for SNAP-23, a component of the membrane fusion machinery. fixation and morphometry of aspects pertinent to exocytosis. J. Mol. Biol. Cell NN, 3485-3494. Cell Biol. NPV, 1709-1717. Gaisano, H.Y., Sheu, L., Wong, P.P., Klip, A., and Trimble, W.S. Potoff, J.J., Issa, Z., Manke, C.W., Jr., and Jena, B.P. (2008). Ca2+- (1997). SNAP-23 is located in the basolateral plasma mem- Dimethylphosphate complex formation: providing insight into brane of rat pancreatic acinar cells. FEBS LettK=QNQ, 298-302. Ca2+ mediated local dehydration and membrane fusion in cells. Goodson, H.V., Valetti, C., and Kreis, T.E. (1997). Motors and Cell Biol. IntK=PO, 361-366. membrane traffic. Curr. Opin. Cell Biol.=V, 18-28. Schneider, S.W., Sritharan, K.C., Geibel, J.P., Oberleithner, H., and Jena, B.P. (2004). Discovery of the Porosome: revealing the mo- Jena, B.P. (1997). Surface dynamics in living acinar cells im- lecular mechanism of secretion and membrane fusion in cells. J. aged by atomic force microscopy: identification of plasma mem- Cell. Mol. Med. U, 1-21. brane structures involved in exocytosis. Proc. Natl. Acad. Sci. Jena, B.P. (2005). Molecular machinery and mechanism of cell USA VQ, 316-321. secretion. Exp. Biol. Med.=OPM, 307-319. Trimble, W.S., Cowan, D.W., and Scheller, R.H. (1988). VAMP-1: A Jena, B.P. (2007). Secretion machinery at the cell plasma mem- synaptic vesicle-associated integral membrane protein. Proc. brane. Curr. Opin. Struct. Biol.=NT, 437-443. Natl. Acad. Sci. USA=UR, 4538-4542. Jena, B.P., Schneider, S.W., Geibel, J.P., Webster, P., Oberleith- Wilson, D.W., Whiteheart, S.W., Wiedmann, M., Brunner, M., and ner, H., and Sritharan, K.C. (1997). Gi regulation of secretory Rothman, J.E. (1992). A multisubunit particle implicated in vesicle swelling examined by atomic force microscopy. Proc. membrane fusion. J. Cell Biol.=NNT, 531-538. Natl. Acad. Sci. USA VQ, 13317-13322. Woodbury, D.J. (1999). Nystatin/ergosterol method for reconstitut- Jena, B.P., Cho, S.-J., Jeremic, A., Stromer, M.H., and Abu- ing ion channels into planar lipid bilayers. Methods Enzymol.= Hamdah, R. (2003). Structure and composition of the fusion OVQ, 319-339. pore. Biophys. JK=UQI=1-7. Woodbury, D.J., and Miller, C. (1990). Nystatin-induced liposome Jeong, E.-H., Webster, P., Khuong, C.Q., Abdus Sattar, A.K., fusion. A versatile approach to ion channel reconstitution into Satchi, M., and Jena, B.P. (1999). The native membrane fusion planar bilayers. Biophys. J.=RU, 833-839.