[19] Measurement of secretion in confocal microscopy

328

IMAGING OF PROCESSES

[ 191

Data from photomultiplier tubes can be channelized according to intensity and a mean and/or median fluorescence can be obtained. Data captured by the video camera can be contoured according to pixel numbers. The intensity of fluorescence within the pixel contours can then be calculated. Simply stated, the greater the fluorescence intensity per cell, the more phagocytosis per cell. In the protocol given earlier, only the green fluorescence, which represents phagocytosis, is analyzed. Dual-labeled cells, green/ orange in the protocol described earlier, are not analyzed as this represents binding. The green fluorescence of cells incubated consecutively with bacteria at 4° and then stained with the appropriate antibody (same as for the experiment) represents the background or nonspecific fluorescence control, as the cells should not phagocytose at 4°. Only green fluorescence signals exceeding this value should be included or alternatively subtracted from the experimental values. The isotypic antibody control is used to ensure that the dual fluorescence is indeed specific for organisms (particles) on the outside of cells and is not caused by the nonspecific binding of immunoglobulin molecules. Summary Confocal microscopy is an excellent tool to quantify phagocytosis. Depending on the particle used, phagocytosis can be determined by the simple manual counting of internalized particles. If a fluorescence probe is utilized, an analysis of fluorescence intensity can be used for quantification. The basic procedure can be altered in a number of areas to conform with the scientific needs of the investigator. This includes the use of different particles, cell types, fluorescence dyes, and even the degree of sophistication of the instrumentation. The major pitfall encountered when trying to quantify phagocytosis is the inability to separate external from internal (phagocytosed) particles. If not determined properly, data will be erroneous, usually indicating a much higher degree of phagocytosis than actually occurred.

[19] M e a s u r e m e n t

of Secretion in Confocal Microscopy

By AKIHISA SEGAWA Introduction Secretion is a dynamic biological activity, seen widely in glandular and nonglandular tissues, accomplished by the spatially and temporally organized movement of molecules and organelles within and between the cells.

METHODS IN ENZYMOLOGY, VOL. 307

Copyright © 1999 by Academic Press All rights of reproduction in any form reserved. 0076-6879/99 $30.00

[1 9]

MEASUREMENTOF SECRETION

329

Secretion of cellular p r o d u c t s (hormones, enzymes, and neurotransmitters, etc.) relies on the cascade of the biosynthetic p a t h w a y with the final release step exocytosis, 1 whereas secretion of b o d y fluids (sweat, tears, digestive juice, etc.) comprises the transepithelial passage of s e r u m c o m p o n e n t s across epithelial cells (transcellular p a t h w a y ) or the intercellular junction (paracellular p a t h w a y ) 2 (Fig. 1). Studies o n secretion have b e e n d o n e successfully using biochemical and electrophysiological approaches, and several crucial events involved in secretion have b e e n unraveled. O f exocytosis, these studies resolved the molecular and kinetic events at the level of single secretory granules, e.g., granule docking, priming, triggering, fusion/release, and removal, 3 each of which constitutes the rate-limiting process of secretion. 4 F o r epithelial transport, they also d e m o n s t r a t e d the existence of a " l e a k y " tight junction, not a " t i g h t " seal as recognized previously, that alters its permeability in response to physiological stimuli to allow passage of water, ions, nonelectrolytes, 5-7 and even m a c r o m o l e c u l a r substances 8 t h r o u g h the paracellular pathway. Morphologically, light and electron mic r o s c o p y on fixed cells have elucidated such processes in detail, 9-12 but only in static images. A n i m p o r t a n t challenge is to clarify their d y n a m i c aspects and to integrate m o r p h o l o g y directly with the physiological events that occur during secretion. Confocal m i c r o s c o p y of living cells is expected, and indeed has offered opportunities, to answer such d e m a n d . 13-22 This article

1 G. Palade, Science 189, 347 (1975). 2 j. A. Young, D. I. Cook, E. W. van Lennep, and M. Roberts, in "Physiology of the Gastrointestinal Tract," 2nd ed., p. 773. Raven Press, New York, 1987. 3 T. F. J. Martin, Trends Cell Biol. 7, 271 (1997). 4 y. Ninomiya, T. Kishimoto, T. Yamazawa, H. Ikeda, Y. Miyashita, and H. Kasai, EMBO J. 16, 929 (1997). 5 E. Fr6mter and J. Diamond, Nature New Biol. 235, 9 (1972). 6 j. L. Madara, Cell 53, 497 (1988). 7 S. Citi, J. Cell Biol. 121, 485 (1993). 8 j. R. Garrett, in "Glandular Mechanisms of Salivary Secretion," Vol. 10 of Frontiers of Oral Biology, p. 153. Karger, Basel, 1998. 9 W. W. Douglas, Br. J. Pharmacol. 34, 453 (1968). 10A. Amsterdam, I. Ohad, and M. Schramm, J. Cell Biol. 41, 753 (1969). 11N. Takai, Y. Yoshida, and Y. Kakudo, J. Dent. Res. 62, 1022 (1983). iz M. R. Mazariegos and A. R. Hand, J. Dent. Res. 63, 1102 (1984). 13A. Segawa, S. Terakawa, S. Yamashina, and C. R. Hopkins, Eur. J. Cell Biol. 54, 322 (1991). 14y. Kawasaki, T. Saitoh, T. Okabe, K. Kumakura, and M. Ohara-Imaizumi, Biochim. Biophys. Acta 1067, 71 (1991). 15A. Nakamura, T. Nakahari, T. Senda, and Y. Imai, Jpn. J. Physiol. 43, 833 (1993). 16A. Segawa, J. Electr. Microsc. 43, 290 (1994). t7 M. Terasaki, J. Cell Sci. 108, 2293 (1995). 18T. Whalley, M. Terasaki, M.- S. Cho, and S. Vogel, J. Cell Biol. 131, 1183 (1995). 19A. Segawa and A. Riva, Eur. J. Morph. 34, 215 (1996).

330

[ 19]

IMAGING OF PROCESSES

fusion/release

A

docking priming

Secretory ~" 0 granules

l

triggering removal oo-" o o ~ o ~ Golgi apparatus

@ Exoeytosis

TranscellularPathway

B

ParacellularPathway

Tight Junction

[Basolateralspace] Epithelial Transport Fio. 1. Two major events in secretion: Exocytosis (A) and epithelial transport (B).

[1 91

MEASUREMENT OF SECRETION

331

describes the m e a s u r e m e n t of exocytosis and the paracellular pathway in rat salivary glands, using confocal microscopy combined with the fluorescent tracer technique. Salivary acini, the secretory end piece of this gland, are composed of highly polarized acinar cells capable of secreting enzymes and fluid; the secretion is regulated distinctively under different autonomic receptor control. 2'12'1639'22'23Methods of cell dissociation, selection of fluorescent tracers, and analytical procedures are described. A general method of confocal microscopy of living cells has been described previously by other authors, 24 and is not described here in detail. P r e p a r a t i o n of Cells Probably the ideal way to visualize glandular secretion is to observe the gland in situ. However, the technique of confocal microscopy for this purpose has not been established and, m o r e importantly, it is difficult for such a preparation to obtain a transmitted light or N o m a r s k y image, which is necessary to correlate with the confocal fluorescence image to locate the sites of secretory events exactly in the ceils and tissues. Use of tissue slices is thus a better choice, but this requires skilled techniques to obtain clear images. Slices must be cut intactly as thin as possible, preferably less than 150/zm. Practically, cell dissociation is the c o m m e n d a b l e way to image ceils with high resolution. Medium Eagle's minimal essential m e d i u m ( M E M ) is supplemented with 25 m M H E P E S (N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid), p H 7.3. Bubble the m e d i u m with O2 gas at least 10 rain before use to avoid anoxia of the cell. In rat salivary acinar cells, anoxia induces a " v a c u o l e " formation in the cytoplasmJ 6'15 Cell D i s s o c i a t i o n Dissect salivary glands after sacrifice of the rats, and place the excised glands on a plastic dish. R e m o v e capsules, connective tissues, and lymph 20C. B. Smith and W. J. Betz, Nature 380, 531 (1996). 21A. Segawa, Bioimages 5, 153 (1977). 22A. Segawa and S. Yamashina, in "Glandular Mechanisms of Salivary Secretion," Vol. 10 of Frontiers of Oral Biology, p. 89. Karger, Basel, 1998. 23B. J. Baum, Ann. N.Y. Acad. Sci. 694, 17 (1993). 24M. Terasaki and M. E. Dailey, in "Handbook of Biological Confocal Microscopy," 2nd ed., p. 327. Plenum Press, New York, 1995. 25R. L. Tapp and O. A. Trowell, J. Physiol 188, 191 (1967).

332

IMAGING OF PROCESSES

[ 19]

nodes with fine forceps and a pair of scissors under the binocular macroscope. Cut tissues into small pieces (approximately 1-2 mm 2) with scalpels, put them into a tissue culture flask (bottle size 25 ml with double seal cap, Iwaki Glass, Japan), and incubate with 10 ml medium containing 20-30 mg of collagenase (Wako Pure Chemical Industries, Osaka, Japan). Hyaluronidase and bovine serum albumin (BSA) may be added to the medium to increase the cell viability. After capping tightly with the screw cap, incubate the culture flask in a water bath kept at 37° for 45 min with constant shaking at 130 oscillations per minute. Every 15 min during digestion, pipette tissues gently with a Pasteur pipette to facilitate mechanically the cell dissociation. Following digestion, centrifuge the cell suspension at 1000 rpm for 1 min, discard medium, and resuspend the cells in MEM. Wash twice. This procedure yields cell aggregates having a well-preserved acinar structure. They exhibit a high secretory response as intact tissue slices. 26 Keep the cell suspension on ice until use. Prepared cells should be used within 1 hr. Later, secretory response decreases and many "vacuoles" appear. Fluorescent Tracers For the measurement of paracellular pathway (tight junctional permeability), fluid-phase fluorescent tracers of varying molecular weights are used. 16 These include Lucifer yellow (molecular weight 457), dextrans labeled with FITC, RITC, or Texas Red (molecular weight 3, 10, 40, 70 and 500 K are available from Molecular Probes Inc., Eugene, OR). They are dissolved in MEM at concentrations of 0.5-2 mg/ml. When perfused, these tracers flood the extracellular space of the tissue and, under laser excitation, reveal bright fluorescence against the nonfluorescent acinar cell cytoplasm. As will be described, fluorescence is detectable in the basolateral (interstitial) space, but not the luminal space, of salivary acini unless a tight junction permeates the tracer into the lumen. The measurement of exocytosis includes four different categories of fluorescent approaches. One is the use of fluid-phase tracers to stain extracellular space as described earlierJ 3-15'17-19'21'22 If exocytosis occurs, the exocytosed granule space is flooded with the tracer and reveals bright fluorescence. The second approach is to stain all the cytoplasm. On exocytosis, granules lose their fluorescence. BCECF-AM [2',7'-bis(2-carboxyethyl)-5,6-carboxyfluorescein, Dojin, Japan] has been used at 3/~M for this purpose. 15The third approach is to stain the contents of secretory granules. As in the second approach, exocytosed granules lose fluorescence. Use of 26 A. Segawa, N. Sahara, K. Suzuki, and S. Yamashina, Z Cell Sci. 78, 67 (1985).

[19]

MEASUREMENT OF SECRETION

333

acridine orange (1 tzM) has been reported. 14The fourth approach is to stain plasma membrane. Following exocytosis, the secretory granule membrane becomes continuous with the plasma membrane, thus revealing ring-shaped fluorescence. FM1-43 (2/zM) 17'18,2° and TMA-DPH (1 /~M) 14 have been used. For the second and third approaches, preincubation of the cells with fluorochromes is necessary to load the dye into the cell. The loaded cells are washed, placed in the chamber, and perfused with MEM. The first and fourth approaches do not require preincubation. Instead, cells are placed in the chamber and perfused with MEM containing the fluorescent dyes. The combined use of different fluorescent tracers is possible when the emission wavelengths of tracers are discriminated distinctively by the detector.

Perfusion Coat the cover slides (24 × 60 mm, for the observation with inverted microscope) or the slide glass (for the upright microscope) with CELL TAK (Collaborative Biomedical Products, Two Oak Park, Bedford). Put the cell suspension on the slide and allow it to settle for several seconds for cells to adhere onto the glass surface (Fig. 2). Wipe the medium with filter paper so that nonadherent cells are removed. If cells do not adhere sufficiently, centrifuge specimens with Cytospin (Shandon, Cheshire, UK) at 200 rpm for a few minutes. For observation with the inverted microscope, specimens can be seen without a cover slide. Simply drop the perfusion medium on the specimen. Wipe it using filter paper. This application method is especially recommended when the response occurs quite rapidly, in the order of milliseconds to seconds, after the addition of secretagogues.27 For the longer observation, apply medium at least every 2-3 min to avoid anoxia of the cells. Observation with the upright microscope and observation of tissue slices need a cover slide. To make the perfusion space, strips of vinyl tape are attached on lateral sides of the glass. Place the specimen in the center, cover with cover slides (24 × 24 mm), and seal with petrolium jelly along the vinyl tape. Pour medium from one side into the chamber and wipe it from the other side with the filter paper. Adjust the thickness of the chamber so as to allow passage of perfusion medium while retaining the specimens (tissue slices) in a fixed position.

27 M. Yamamoto-Hino, A. Miyawaki, A. Segawa, E. Adachi, S. Yamashina, T. Fujimoto, T. Sugiyama, T. Furuichi, M. Hasegawa, and K. Mikoshiba, J. Cell Biol. 141, 135 (1998).

334

IMAGING OF PROCESSES

A

[ 19]

Lj Coverslide

Perfusion

I

/

~

y

/

7"

SealedwithVaseline (Petroleum ,/7 jely) /

'

~

~

Slideglass

B pipet ~medium

/ (dissS~p~CaitmanScini ~ )

FIG.2. Methodsof perfusionfor an uprightmicroscope(A) and an invertedmicroscope(B). Confocal Microscopy Place the sample on the microscope stage warmed to 37°. Using the ordinary microscope mode, confirm the location, shape, and visibility of cells. Choose the best cells. Change to the confocal microscope mode. Adjust brightness and the plane of focus (height of optical sectioning) by the simultaneous fluorescence and transmitted light or Nomarsky imaging. Select the box (image) size. A large box allows imaging with high resolution but requires longer scanning time. A small box enables rapid acquisition of images, up to three to four frames per second (128 x 128 pixel) for Bio-

[191

MEASUREMENTOF SECRETION

335

Rad MRC 1024. For more rapid observation, a real confocal microscope (Nikon RCM 8000, Noran Oz and Yokogawa CSU10) obtains images faster than the video rate (33 msec/frame). To increase the signal-to-noise ratio, average several images. Averaging three improves the image quality greatly. Before starting the image aquisition, decide the time interval. The rapid acquisition of images consumes a lot of computer memory, which limits the observation period to a short time. The acquisition of large images also consumes memory and reduces the temporal resolution. Thus for ordinary observation, we collect images every 1.5 to 5 sec with a moderate image size (768 × 512 pixel). Secretory Stimulation Rat parotid acinar cells possess two distinct secretion mechanisms regulated by different receptor-signaling systems: (1) enzyme release by exocytosis through the activation of fl-adrenergic receptor generating cyclic AMP and (2) fluid secretion activated by muscarinic, a-adrenergic, and substance P receptors via an increase in cytosolic calcium. 2'23For stimulation of exocytosis, DL-isoproterenol (1-20 ~M) is dissolved in the peffusion medium to activate fl receptor. Fluid secretion is stimulated by carbachol (10 /~M), which activates the muscarinic receptor, a major regulator of fluid secretion. Measurement of Tight Junctional Permeability: Paracellular Pathway The basic structure of salivary acini is described first to evaluate properly the results of experiments. As mentioned previously, a tight junction adjoins neighboring cells to separate the lumen from the interstitial (basolateral) space (Fig. 1). In parotid acini, the configuration of lumen is not simple but exhibits complex narrow (approximately 1 ~m in diameter) canalicular extensions called the intercellular canaliculi (Fig. 3). In the sectioned image, they reveal a very small ring or tubular appearance within the acini. Intercellular canaliculi also provide the exclusive site of exocytosis in salivary acini. Therefore, it is very important to define the location of intercellular canaliculi in the confocal images (Fig. 4). Apply medium containing the fluid-phase fluorescent tracers. In untreated cells, bright fluorescence can be detected in the basolateral space but not in the luminal space (Fig. 4). This indicates that the tight junction of parotid acini does not allow permeation of fluorescent tracers into the lumen. Stimulate cells by adding secretagogues in the perfusion medium and observe if the fluorescence appears in the lumen. If it appears, then

f

Lumen uct

Tight Junction

Acinus ~

9 O 0 I Lj o O

"~

Intercellular

~Canaliculi

Fie. 3. Parotid acini shown by electron micrograph [modified from A. Segawa, J. Electr. Microsc. 43, 290 (1994)]. (A) Unstimulated. (B) Stimulated with isoproterenol in vitro for 30 min. Arrowheads denote the intercellular canaliculi. Secretory granules (SG) are numerous in A but depleted in B, where the exocytotic profiles are seen at the intercellular canaliculi. L, lumen. Bar: 5 ~m. (C) Schematic representation.

[ 19]

MEASUREMENTOF SECRETION

337

Fic. 4. Dynamic changes in tight junctional permeability shown by confocal microscopy. Reproduced from A. Segawa, J. Electr. Microsc. 43, 290 (1994). Tissue slices of rat parotid glands were perfused with Lucifer yellow and stimulated with isoproterenol for 1 (A) and 5 (B) min. Simultaneous transmitted light (left) and confocal fluorescence (right) imaging. Arrowheads indicate the intercellular canaliculi. The intensity of fluorescence in the intercellular canaliculi is weak in A but high in B. Bar: 10/zm. examine the size-selectivity characteristics. A p p l y m e d i u m without fluorescent tracers to wash out the luminal fluorescence and reintroduce m e d i u m containing tracers having larger molecular sizes. In rat parotid acini, a tight junction exhibits different size-selectivity characteristics in response to different secretory stimuli: W e detected up to molecular weight 40 K for isoproterenol and 10 K for carbachol w h e n fluorescent dextrans were

338

IMAGING OF PROCESSES

[ 191

used as a monitoring tracer. 16 This observation coincides with the results of previous experiments carried out by biochemical and morphological approaches. 8'12,28 The existence of size-selectivity characteristics also rules out the possibility that the luminal fluorescence originates by the retrograde diffusion of tracers from the duct. Advantages of this method include (1) applicability to the glandular epithelia whose lumen is too narrow to allow for the insertion of electrode, which is necessary for the electrophysiological approach; (2) distinction of para- from transcellular pathways involved in transepithelial transport, which is difficult in the biochemical approach; and (3) consecutive observation of morphological changes occurring in the same cell, which is impossible for electron microscopy. N o t e : Enzymatic digestion sometimes damages salivary tight junction; luminal fluorescence often appears in the dissociated acini, even though they do not receive secretory stimulation.I6 Thus the use of tissue slices is preferable to the study of tight junction dynamics. Anoxia also seems to increase the permeability of tight junction as anoxic "vacuoles" were found to exhibit fluorescence. 16 To avoid this, carefully keep perfusing tissues with a well-oxygenated medium. Measurement of Exocytosis Secretion by exocytosis involves the cycles of fusion and the removal of secretory granule membranes at the cell surface (exocytosis-removal cycle). This section describes the visualization of the exocytosis-removal cycle of single secretory granules using fluid-phase fluorescent tracers. 21 Place dissociated acini on a slide glass. Perfuse cells with MEM containing fluorescent tracers without secretagogues. Using simultaneous fluorescence and transmitted light or Nomarsky imaging, observe for a few minutes to confirm that cells exhibit little changes. Start acquisition of images (Fig. 5). Apply fluorescent medium containing isoproterenol. Ten to 15 sec later, omega-shaped fluorescent spots appear abruptly along the intercellular canaliculi. This represents the exocytotic response. It is difficult to predict the sites exhibiting exocytosis until the cells receive secretory stimuli. Therefore, it is recommended to image a wide area with a relatively large box size (768 x 512 pixels) to capture as many morphological events as possible. Continue image collection for at least 2 min to follow the fate of fluorescent spots. Granule movement and the dynamics of membrane reorganization during exocytotic secretion are analyzed later. Image analyses are performed by merging the confocal fluorescence 28L. C. U. Junqueira, A. M. S. Toledo, and R. G. Ferri, Arch. Oral Biol. 10, 863 (1965).

[19]

MEASUREMENTOF SECRETION

339

FIG. 5. Exocytosis in living cells shown by confocal microscopy. Dissociated rat parotid acini were perfused with FITC-dextran (molecular weight 3000). Simultaneous fluorescence and transmitted light images were taken every 5 sec and displayed consecutively. Isoproterenol was added at image 1, so that acini moved (asterisk). Arrows indicate the appearance of fluorescent spots. Some images are pseudocolored in blue (B), green (G), and red (R) for image analyses (see Figs. 6A and 6B). Bar: 10/zm.

image with transmitted light or Nomarsky image (Fig. 6C, see color insert). Confirm that the size and shape of the fluorescent spots are comparable to those of single secretory granules. Analyze granule movement and membrane dynamics by displaying images in a time sequence (Fig. 5). Find critical images showing dynamic changes. Merge them to make the timeresolved changes clearer. MRC 1024 makes it possible to merge images obtained at three different time points. Pesudocolor each image in blue, green, and red and merge into one (Figs. 6A and 6B, see color insert). The unchanged area summates all three colors and is displayed in gray scale. In Fig. 6B, only the granule-shaped fluorescent spots are displayed in the color image. Other areas are displayed mostly in gray scale. This is important as it indicates that the observed changes in fluorescent spots are not caused by the movement of the specimen as a whole. The transmitted light image also shows that the granules are immobile before exhibiting the exocytotic response. This indicates the presence of docked granules in salivary exo-

340

IMAGING OF PROCESSES

[20]

cytosis, as in neurons and neuroendocrine cells. 3'22 Select an area where the exocytotic response is recognized (Fig. 6B). Enlarge and display the area sequentially (Fig. 6D). This demonstrates the appearing and disappearing changes of fluorescent spots, representing the exocytosis-removal cycle of a single secretory granule. Fluorescent spots appear 10-15 sec after the addition of isoproterenol. This latency period is likely to represent the time required for priming and triggering. Following their appearance, the fluorescent spots maintain their round shape for several seconds to tens of seconds, diminish gradually, and finally disappear, with a total detectable period of 40-70 sec in many instances. Animated demonstration of these changes helps recognize the dynamic image. 29 There are many softwares, such as"Confocal assistant," which present a motion picture on the personal computer and also enables the animated demonstration of merged transmitted and fluorescence image.

29 A. Segawa and M. Ono, in "Image Analysis & 3D Reconstruction CD-ROM Series, Vol. 1," Purdue University Cytometry Laboratories, West Lafayette, USA (1998).

[20] R e c e p t o r - L i g a n d

By GUIDO

Internalization

ORLANDINI, NICOLETTA RONDA, RITA GATTI,

GIAN CARLO GAZZOLA, and ALBERICO BORGHETrI Introduction The interaction between biologically active compounds and target cells has been studied extensively by various quantitative and qualitative approaches as it is the crucial first step in the chain of events leading to the final effect. Additionally it is a suitable process to be studied for pharmacological purposes. 1 The visualization of receptor-ligand binding and internalization has been studied by immunocytochemistry techniques for both light and electron microscopy on fixed, permeabilized tissues or cells. The information achievable by these methods is limited and static, as the localization observed in fixed samples might not correspond to the actual binding site

1 H. Lodish, D. Baltimore, A. Berk, S. L. Zipursky, P. Matsudaira, and J. Darnell, in "Molecular Cell Biology" (J. Darnell, ed.), 3rd ed. Scientific American Books, New York, 1995.

METHODSIN ENZYMOLOGY,VOL.307

Copyright© 1999by AcademicPress All rightsof reproductionin any formreserved. 0076-6879199 $30.00